Peace Is Still Possible in Ethiopia
How to Avoid a Balkan-Style Catastrophe in the Horn of Africa
The extravagant lurches of the U.S. intervention in Afghanistan—from a $1 trillion surge to total withdrawal, culminating in the reestablishment of a Taliban government 20 years after the 9/11 attacks—must rank among the most surreal and disturbing episodes in modern foreign policy. At the heart of the tragedy was an obsession with universal plans and extensive resources, which stymied the modest but meaningful progress that could have been achieved with far fewer troops and at a lower cost. Yet this failure to chart a middle path between ruinous overinvestment and complete neglect says less about what was possible in Afghanistan than it does about the fantasies of those who intervened there.
The age of intervention began in Bosnia in 1995 and accelerated with the missions in Kosovo, Afghanistan, and Iraq. Over this period, the United States and its allies developed a vision of themselves as turnaround CEOs: they had the strategy and resources to fix things, collect their bonuses, and get out as soon as possible. The symbol of the age was the American general up at 4 AM to run eight miles before mending the failed state.
Had the same U.S. and European officials been seeking to improve the lives of people in a poor ex-coal town in eastern Kentucky or to work with Native American tribes in South Dakota, they might have been more skeptical of universal blueprints for societal transformation, paid more attention to the history and trauma of local communities, and been more modest about their own status as outsiders. They might have understood that messiness was inevitable, failure possible, and patience essential. They might even have grasped why humility was better than a heavy footprint and why listening was better than lecturing.
Yet in the Balkans, Afghanistan, and Iraq—places far more traumatized, impoverished, and damaged than anywhere at home—U.S. and European officials insisted that there could be a formula for success, a “clearly defined mission,” and an “exit strategy.” Any setback, they reasoned, could be blamed only on a lack of international planning or resources.
These ideas were damaging in Bosnia and Kosovo. But in the interventions in Afghanistan and Iraq—unstable hybrids of humanitarianism and counterterrorism that soon became even more unstable hybrids of state building and counterinsurgency—they proved fatal. From the very beginning, the international plans were surreally detached from the local reality. The first draft of the development strategy for Afghanistan, written by international consultants in 2002, described the Afghans as committed to “an accountable, broad-based, multi-ethnic, representative government” based on “respect for human rights.” That same year, then U.S. National Security Adviser Condoleezza Rice claimed that terrorism from Afghanistan posed “an existential threat to our security.”
Such hyperbolic untruths, which multiplied with each new strategy or plan, were designed to win resources and defend the intervention at home. By exaggerating both the potential for success and the risks of failure in Afghanistan, they made it difficult to resist calls for more troops. And when troops were killed (and more of them were killed than at any time since the Vietnam War), domestic politics dictated ever more strident mission statements, increasingly inflated plans, and additional troop deployments.
Eventually, the rhetorical Ponzi scheme collapsed. But having failed to fulfill their fantasies and realize their power as saviors, the United States and its allies now seemed unable to recognize or value the progress that was actually occurring on the ground—in part, because it was slow, unfamiliar, and often not in line with their plans. Political leaders had so overstated their case that once they were revealed to be wrong, they could not return to the moderate position of a light footprint and instead lurched from extreme overreach to denial, isolationism, and withdrawal. In the end, they walked out, blaming the chaos that followed on the corruption, ingratitude, and the supposed cowardice of their former partners.
The obsession with universal plans backed by heavy resources that led to the failures in Afghanistan and Iraq stemmed in part from a misunderstanding of an earlier success. The first act in the 20-year age of intervention, the NATO operation in Bosnia, was largely effective. Not only did it end the war and preserve the peace for decades at almost no cost to the United States and its NATO allies, but it achieved things that not long before had seemed impossible: the protection of civilians, the demobilization of vicious militias, the safe return of refugees to ethnically cleansed areas, and the imprisonment of war criminals. Today, the Bosnian state remains fragile, ethnically divided, and corrupt—but also peaceful.
This success, which emerged from a large but very restrained international presence, was misinterpreted as an argument for bold international interventions grounded in universal state-building templates and backed by overwhelming resources. Paddy Ashdown, the British politician who was the senior international representative in Bosnia and Herzegovina, asserted that Bosnia demonstrated seven “pillars of peace-making” that “apply more or less universally” and provided a plan to create everything from security to water supplies, prisons, and an efficient market-based economy. In his view, an international administration with absolute executive power was needed to achieve these things. Local elections or consultations should be avoided. The intervening powers should, he said, “go in hard from the start,” establishing the rule of law as quickly and decisively as possible, “even if you have to do that quite brutally.”
Many embraced Ashdown’s vision and developed similar blueprints. James Dobbins, a former U.S. special envoy to Bosnia and a future special representative for Afghanistan and Pakistan, co-authored The Beginner’s Guide to Nation-Building, published by the RAND Corporation, which asserted that “heavy” peace-enforcement operations required 13 soldiers for every 1,000 inhabitants and “light” peacekeeping operations required two. The future president of Afghanistan, Ashraf Ghani, matched this with a co-authored textbook titled Fixing Failed States that defined ten functions of a state and laid out a universal state construction scheme that could be applied from the Horn of Africa to the Urals.
At the heart of the tragedy in Afghanistan was an obsession with universal plans and extensive resources.
In Kosovo and Iraq, ever-greater power was deployed to advance such plans. In Kosovo, the UN administration assumed the authority to jail anyone, change the constitution, appoint officials, and approve the government’s budget (although it used these powers relatively cautiously). In Iraq, Paul Bremer, the American administrator of the Coalition Provisional Authority, assumed full executive power and sent American and British officials—I was one of them—to govern the Iraqi provinces. They rewrote university curricula, remade the army, and fired hundreds of thousands of members of Saddam Hussein’s Baath Party and detained tens of thousands more.
Afghanistan—the third of the four great interventions of the age—was the exception. There, the senior UN official, Lakhdar Brahimi, and U.S. Secretary of Defense Donald Rumsfeld proposed a light footprint. Although they came from very different political traditions (Brahimi was an anticolonial independence leader in Algeria), they both mocked Kosovo as a neocolonial farce. Both feared that a heavy footprint in Afghanistan would make the government too dependent on foreign money and troops and provoke an insurgency. Rumsfeld initially authorized only 2,000 U.S. troops and forbade any nation building. No attempt was made to create anything comparable to the mission in Kosovo or, later, that in Iraq. And in order to ensure that his idealistic UN staff was not tempted into running Afghanistan, Brahimi blocked the opening of UN field offices in many of the provinces. Instead, the lead was given to the Afghan transitional government under President Hamid Karzai.
By 2004, three years into the intervention, most of Afghanistan was safer, freer, and more prosperous, with better services and opportunities than it had had in 30 years. But there was a dark side to this story: the corruption was far worse than during the Soviet occupation or Taliban rule, the police were brutal, and the judicial system worked only for those who could afford the bribes. The production of opium poppies—which had been nearly eliminated by the Taliban by 2000—soared, with profits flowing to the most senior government officials.
Helmand Province was perhaps the most extreme failure. It was controlled by local strongmen—confirmed in government positions by Karzai—whose families had run the province in the 1980s and early 1990s and who used their newfound power to reignite a decades-long civil war over land and drugs. (Helmand was then producing 90 percent of Afghanistan’s opium and much of the heroin that found its way to Europe.) Regularly robbed and tortured by these commanders, Afghans in some parts of the province became nostalgic for the Taliban.
Many commentators blamed these setbacks on the light footprint, arguing that the United States had been distracted by Iraq, had failed to plan properly, and had not deployed enough resources or troops. UN officials, counternarcotics agents, journalists, and human rights and anticorruption campaigners all called for the toppling of the warlords. Academics warned that the lack of good governance would alienate the local population and undermine the credibility of the Afghan government. Practically everyone assumed that there was a realistic plan to fix governance in Afghanistan—and that the missing ingredients were more resources and international troops. As one 2003 RAND report on nation building argued: “The United States and its allies have put 25 times more money and 50 times more troops, on a per capita basis, into post-conflict Kosovo than into post-conflict Afghanistan. This higher level of input accounts in significant measure for the higher level of output measured in terms of democratic institutions and economic growth.”
These ideas led NATO to launch what was in effect a second, heavier intervention: a regime-change operation aimed this time not at the Taliban but at the power structures that had been established by the coalition’s ally Karzai. By 2005, NATO “provincial reconstruction teams” had sprouted up across the country, the UN had begun to disarm and demobilize the warlords and their militias, and the number of NATO troops had begun to climb. General John Abizaid, the head of U.S. Central Command, predicted that 2005 would be “the decisive year.”
As the troop counts rose, the problem of good governance became a problem of insurgency.
By 2006, the most powerful warlords had been stripped of their posts in Helmand, and the United Kingdom had deployed thousands of troops to the province. Their aim was not to fight the Taliban, perceived at the time as a weak force. Rather, the troops focused on improving governance and justice and on stamping out corruption and drugs. This plan, dubbed “the comprehensive approach,” demanded an ever-heavier international footprint. Few seemed to doubt its feasibility. The commander of the NATO-led operation, British General David Richards, insisted that the mission was “doable if we get the formula right, and it is properly resourced.” He increased the number of troops under his command from 9,000 to 33,000 and claimed that 2006 would be “the crunch year.”
But as the troop counts rose, the problem of good governance became a problem of insurgency. In 2006, the number of Taliban bomb attacks increased fivefold, and the number of British casualties increased tenfold. This, too, was blamed on an imperfect plan and insufficient resources. In 2007, a new general announced another strategy, requiring still more resources. The same thing happened in 2008. NATO troop increases were followed by U.S. troop increases. In 2009, U.S. General Stanley McChrystal announced a new plan for 130,000 U.S. and NATO soldiers, claiming he was “knee-deep in the decisive year.”
By this point, tens of thousands of Afghans and thousands of international troops had been killed, and Afghanistan was considerably less safe than it had been in 2005. But the interveners still insisted that somewhere out there was a formula for state building and counterinsurgency that could succeed. Counterinsurgency experts began to suggest that perhaps 700,000 troops would do it.
As the U.S. presence in Afghanistan increased, so did the temperature of the political rhetoric in Washington. In 2003, when 30 U.S. service members were killed in Afghanistan, it was possible to justify the mission as one of a number of small U.S. operations stretching from Asia to the Horn of Africa. But by 2008, with five times as many U.S. soldiers dying per year and tens of billions of dollars being spent, more extreme justifications were demanded. Officials now argued that if Afghanistan fell to the Taliban, Pakistan would, too, and extremists would get their hands on nuclear weapons. Catching Osama bin Laden, President Barack Obama insisted, required “winning” in Afghanistan. Failure was not an option.
None of this was true, of course. Pakistan and much of the Middle East were more important threats in terms of terrorism and regional instability. Catching bin Laden required only catching bin Laden. But the savage and changeable winds of public opinion demanded ever more paranoid and grandiose statements. U.S. plans for state building and counterinsurgency became tissues of evasion and euphemism, justified with contorted logic, dressed in partial statistics, and decorated with false analogies. They were inflexible, simplistic, overly optimistic, and shrilly confident. And because these plans remained obsessed with fixing the Taliban-dominated areas of southern Afghanistan, they diverted investment from the stable, welcoming areas of central and northern Afghanistan, where significant development progress was still possible.
Many of these optimistic plans contained barely concealed prophecies of failure. McChrystal, for example, maintained that no amount of U.S. military power could stabilize Afghanistan “as long as pervasive corruption and preying upon the people continue to characterize governance.” Obama himself acknowledged that such misconduct was unlikely to change—but he nonetheless authorized a slightly pared-down version of McChrystal’s request for almost 40,000 additional troops.
While the United States continued to refine its plans, the Taliban implemented their own vision for how to establish security, governance, and the rule of law. They called it sharia, and they sold it not from a military fort but from within tribal structures, appealing to rural habits and using Islamic references, in Pashto. And the more military power the interveners deployed against them, the more they could present themselves as leading a jihad for Afghanistan and Islam against a foreign military occupation.
To the Americans and their allies, it seemed impossible that the U.S. military, with its fleets of gunships and cyberwarfare capabilities, its cutting-edge plans for counterinsurgency and state building, and its billions of dollars in aid and investment, could be held off by a medieval group that lived in mud huts, carried guns designed in the 1940s, and rode ponies. The interveners continued to believe that the international community could succeed in nation building anywhere in the world, provided that it had the right plan and enough resources.
This view reflected a tragic misreading of the experience in Bosnia, which was a much more cautious and constrained intervention than many recall. The number of international troops was higher there than in the early days of the war in Afghanistan, but both foreign soldiers and foreign civilians in Bosnia were severely limited in what they could do. (Ashdown’s vision of an omnipotent international state builder, overruling local voices and implementing the perfect plan, was what he wished for, not what he found.)
Scarred by memories of Vietnam and the more recent failed intervention in Somalia, senior U.S. and European officials did not wish to be drawn into the long history of ethnic strife in the Balkans and so approached the conflict with immense caution. When the United States belatedly mounted a military intervention, it was focused on air operations to bomb the Bosnian Serb artillery around Sarajevo. The ground fighting was conducted by the Sarajevo-based Bosnia authority and by Croatian soldiers, who received their training from U.S. contractors. When international troops were deployed after the Dayton peace accords, they spent most of their time on their bases. More U.S. soldiers were injured playing sports than in action.
The Office of the High Representative for Bosnia and Herzegovina had much less power than its equivalent would be given in Kosovo and could not order military or police officers to enforce its decrees. The Dayton agreement handed 49 percent of the country’s territory to the Bosnian Serb aggressors and enshrined their power in areas that they had ethnically cleansed. The cautious international presence also initially left the Croatian and Serbian paramilitaries, special police forces, and intelligence services in place and did not disarm them. Instead of doing the equivalent of “de-Baathifying,” as Bremer did in Iraq, or toppling the warlords, as U.S. and coalition forces did later in southern Afghanistan, the high representative for Bosnia and Herzegovina was required to work with the war criminals. The party of the Bosnian Serb leader Radovan Karadzic, who was responsible for the massacre in Srebrenica, was allowed to participate in elections (and won the first postwar one, in 1996).
Success in Bosnia was due not to the strength of the international presence but to its comparative weakness.
Bosnia was ultimately transformed not by foreign hands but by messy and often unexpected local solutions that were supported by international diplomacy. The first breakthrough came when Bosnian Serb President Biljana Plavsic split from her mentor, the war criminal Karadzic, and then requested international support. Plavsic was herself a war criminal who had described Bosnian Muslims as “genetically deformed material.” But the international forces worked with her to disarm the special police forces, Bosnian Serb units that acted as de facto militias. Later, the death of Croatian President Franjo Tudjman and the toppling of Serbian President Slobodan Milosevic fatally weakened their proxies in Bosnia. Neither of these events was part of a planned strategy by the international community, but both helped what had initially been a tiny and apparently toothless war crimes tribunal in The Hague expand its operations, leading eventually to the capture and prosecution not only of Karadzic but also of Plavsic herself. Cautious compromises ultimately led not to appeasement but to justice.
The reversal of the ethnic cleansing in Bosnia also owed very little to international plans. Despite the Dayton agreement’s commitment to refugee return, many international experts considered it reckless to allow refugees to go back to villages that had been burned to the ground and occupied by hostile militias. Nonetheless, small groups of Bosnians tried to move back to their homes. Some were ejected immediately by armed groups, but others held on and persuaded international troops to follow and protect them. These small Bosnian-led initiatives—improvised, incremental, and following no international plan—opened the door for the return of over a million refugees.
Within a decade of the intervention, more than 200,000 homes had been given back to their owners, over 400,000 soldiers from three armies had been disarmed, and Bosnia had built a unified army of 15,000 soldiers. All the major war criminals were caught and tried, and Bosnia’s homicide rate fell below that of Sweden. All of this was achieved at a cost of almost zero American and NATO lives. And as Gerald Knaus, the chair of the European Stability Initiative, a European think tank specializing in the Balkans, has argued, such successes were due not to the strength of the international presence but to its comparative weakness: a relatively restrained intervention forced local politicians to take the lead, necessitated often uncomfortable compromises, and made foreign civilians and troops act cautiously to reinforce unexpected and improvised local initiatives.
Could a light footprint in Afghanistan have eventually led to similar successes? Perhaps, but with greater difficulty. Afghanistan was much poorer when the United States invaded than Bosnia was at the time of the NATO intervention: adult life expectancy was about 48, one in seven children died before the age of five, and most men (and almost all women) were unable to read or write. Afghan communities were far more conservative, religious, and suspicious of foreigners than Bosnian communities had been (thanks in part to CIA efforts to develop their identity as heroic resisters of foreign occupation during the Soviet period). But the initially limited and restrained international presence in Afghanistan still enabled far more progress than most critics of the war have acknowledged.
The violence and poor governance—particularly in Helmand, elsewhere in southern Afghanistan, and in eastern Afghanistan—that were used to discredit the light-footprint approach were not representative of all of rural Afghanistan. In Bamiyan, for example, a province of three million people in the center of the country, military strongmen retained power, but there was peace. Between 2001 and 2004, locals established excellent schools, even in outlying settlements, providing most girls with their first experience of formal education and laying the foundation for some of them to attend college. The people of Bamiyan—long a marginalized community—began to take senior positions in universities, the media, ministries, and other government agencies. The government extended paved roads and electricity to villages that had never seen them before. Life was much better than it had been under the Taliban, which had led genocidal attacks against Bamiyan communities. (In the winter of 2001–2, I walked through village after village that had been burned to the ground by the Taliban.) All this progress occurred with only a few dozen foreign soldiers in the province and no international civilian administrators.
There also was progress in other central regions and in areas to the north, including in Herat, much of Mazar-e Sharif, the Panjshir Valley, the Shomali Plain, and Kabul. In all these places, a light international footprint meant fewer international casualties, which in turn reduced the pressure on American and European politicians and generals to make exaggerated claims. It also compelled the international community to engage in a more modest discussion with the Afghan people about what kind of society they themselves desired and to accept ideas and values that Americans and Europeans did not always share. In short, it forced a partnership.
By 2005, the Afghan economy was almost twice as big as it had been in 2001. The population of Kabul had quadrupled in size, and new buildings were shooting up. On television, young female and male presenters had the confidence to satirize their rulers. And the progress was not confined to the capital: across the country, 1.5 million girls went to school for the first time. Mobile phones spread like wildfire. Health and life expectancy improved. There was less violence than at any point in the previous 40 years, and no insurgency remotely comparable to what had exploded in Iraq. Perhaps most encouraging of all was that although millions of people had fled in the wake of the U.S.-led invasion of Iraq, millions of Afghan refugees were choosing to return home during this period.
A light and sustained footprint modeled on the Bosnian intervention should have been the approach for Afghanistan.
What would have happened if the United States and NATO had tried to retain a light footprint and a restrained approach beyond 2005? What if they had deployed fewer troops, invested in generous development aid, and resisted fighting the drug trade, toppling warlords, and pursuing a counterinsurgency campaign against the Taliban? The answer would have depended to a great extent on the initiatives of local actors and the competition among them, the developments in neighboring countries, and luck—just as the outcome in Bosnia did. In many parts of Afghanistan, there would have been poverty, a lack of democratic representation, and strongman rule. In regions controlled by drug lords and racked by Pashtun infighting and Pakistani meddling, there probably would have been continued horror, especially if U.S. special operations forces and their proxies had continued to hunt for terrorists. But across much of the country, from Bamiyan to Panjshir, there could have been continued improvements in health, education, and employment—particularly if an overambitious surge had not diverted development funds away from these regions and to the insurgency areas. And for millions of people in Herat and Kabul, this progress could have been combined with an increasingly open and democratic civil society.
Most important, however, many of the problems caused by the heavier international presence and the surge would have been avoided. Well meaning though they were, the attempts to depose local warlords in the name of good governance created power vacuums in some of the most ungovernable regions of the country, alienated and undermined the elected government, and drove the warlords and their militias to ally with the Taliban. The counternarcotics campaigns alienated many others who lost their livelihoods.
The United States did attempt to return to a lighter footprint in 2014, but by then, immense damage had been done. The surge had formed an Afghan army that was entirely reliant on expensive U.S. aircraft and technology, created a new group of gangster capitalists fed from foreign military contracts, and supercharged corruption. Military operations had killed thousands of people, including many civilians, deepening hatred. And the presence of more than 100,000 international troops in rural villages had allowed the Taliban—which had been a weak and fragile group—to present themselves as fighting for Afghanistan and Islam against a foreign occupation. In 2005, under the light footprint, a British intelligence analyst told me there were between 2,000 and 3,000 Taliban fighters in Afghanistan. Six years later, after tens of thousands of Afghans had been killed and half a trillion dollars had been spent, General Richard Barrons of the British army estimated that there were 36,000 Taliban fighters in the country.
But just as the initial light footprint was better than the surge, so the later light footprint was better than a total withdrawal. A few thousand international troops, supporting air operations, were still capable of preventing the Taliban from holding any district capital—much less marching on Kabul. And by preventing a Taliban takeover, the troops were able to buy valuable time for health and educational outcomes to improve, development assistance to continue, income and opportunity to grow, and rights to be more firmly established for millions of Afghans.
Although the cost of the surge had been immense, the cost of remaining beyond 2021 would have been minimal. The United States could have supported 2,500 soldiers in Afghanistan almost indefinitely—and with little risk. So long as U.S. airpower and support for the Afghan air force remained in place, the Taliban would have posed a minimal threat to U.S. troops in their heavily defended air bases. (Eighteen U.S. service members were killed in 2019, perhaps the fiercest year of the fighting, before the cease-fire agreement.) The Taliban were not on the verge of victory; they won because the United States withdrew, crippled the Afghan air force on its way out, and left Afghan troops without air support or resupply lines. In other words, the decision to withdraw was driven not by military necessity, the interests of the Afghans, or even larger U.S. foreign policy objectives but by U.S. domestic politics.
Yet many Americans welcomed the end of the U.S. war in Afghanistan because their leaders had not properly explained to them how light the U.S. presence had become or what it was protecting. Politics in the West seems to abhor the middle ground, swinging inexorably from overreach and overstatement to isolationism and withdrawal. A light and sustained footprint modeled on the Bosnian intervention should have been the approach for Afghanistan—and, indeed, for interventions elsewhere in the world. Yet instead of arguing that failure in Afghanistan was not an option, former U.S. President Donald Trump behaved as though failure had no consequences. He showed no concern for how a U.S. withdrawal from Afghanistan would affect the United States’ reputation and alliances, regional stability, terrorism, or the lives of ordinary Afghans. And he responded to exaggerated claims about Afghanistan’s importance not with moderate claims but with a refusal to maintain even the smallest presence there or to bear the slightest cost.
President Joe Biden has followed Trump’s Afghan policy in every detail, despite having famously advocated a light footprint—and argued against the surge—when he was Obama’s vice president. Somehow, over the years, he seems to have convinced himself that such an approach had failed. But the light footprint did not fail. What failed was the political culture of the West and the imagination of Western bureaucrats. The United States and its allies lacked the patience, realism, and moderation needed to find the middle path.
Two Decades of Mistakes, Misjudgments, and Collective Failure