One of the most celebrated diplomats of his generation, Richard Holbrooke helped normalize U.S. relations with China; served as U.S. ambassador to a newly reunified Germany and then to the United Nations; and, most famously, negotiated the 1995 Dayton peace agreement that ended the war in Bosnia. But he began and ended his career struggling with how to resolve two American wars: first in Vietnam, then in Afghanistan.

Richard Holbrooke was six feet one but seemed bigger. He had long skinny limbs and a barrel chest and broad square shoulder bones, on top of which sat his strangely small head and, encased within it, the sleepless brain. His feet were so far from his trunk that, as his body wore down and the blood stopped circulating properly, they swelled up and became marbled red and white like steak. He had special shoes made and carried extra socks in his leather attaché case, sweating through half a dozen pairs a day, stripping them off on long flights and draping them over his seat pocket in first class, or else cramming used socks next to the classified documents in his briefcase. He wrote his book about ending the war in Bosnia—the place in history that he always craved, though it was never enough—with his feet planted in a Brookstone shiatsu foot massager. One morning he showed up late for a meeting in the secretary of state’s suite at the Waldorf Astoria in his stocking feet, shirt untucked and fly half zipped, padding around the room and picking grapes off a fruit basket, while Madeleine Albright’s furious stare tracked his every move. During a videoconference call from the U.S. mission to the United Nations, in New York, his feet were propped up on a chair, while down in the White House Situation Room their giant distortion completely filled the wall screen and so disrupted the meeting that President Bill Clinton’s national security adviser finally ordered a military aide to turn off the video feed. Holbrooke put his feet up anywhere, in the White House, on other people’s desks and coffee tables—for relief, and for advantage.

Near the end, it seemed as if all his troubles were collecting in his feet—atrial fibrillation, marital tension, thwarted ambition, conspiring colleagues, hundreds of thousands of air miles, corrupt foreign leaders, a war that would not yield to the relentless force of his will.

But at the other extreme from his feet, the ice-blue eyes were on perpetual alert. Their light told you that his intelligence was always awake and working. They captured nearly everything and gave almost nothing away. Like one-way mirrors, they looked outward, not inward. No one was quicker to size up a room, an adversary, a newspaper article, a set of variables in a complex situation—even his own imminent death. The ceaseless appraising told of a manic spirit churning somewhere within the low voice and languid limbs. Once, in the 1980s, he was walking down Madison Avenue when an acquaintance passed him and called out, “Hi, Dick.” Holbrooke watched the man go by, then turned to his companion: “I wonder what he meant by that.” Yes, his curly hair never obeyed the comb, and his suit always looked rumpled, and he couldn’t stay off the phone or TV, and he kept losing things, and he ate as much food as fast as he could, once slicing open the tip of his nose on a clamshell and bleeding through a pair of cloth napkins—yes, he was in almost every way a disorderly presence. But his eyes never lost focus.

So much thought, so little inwardness. He could not be alone—he might have had to think about himself. Maybe that was something he couldn’t afford to do. Leslie Gelb, Holbrooke’s friend of 45 years and recipient of multiple daily phone calls, would butt into a monologue and ask, “What’s Obama like?” Holbrooke would give a brilliant analysis of the president. “How do you think you affect Obama?” Holbrooke had nothing to say. Where did it come from, that blind spot behind his eyes that masked his inner life? It was a great advantage over the rest of us, because the propulsion from idea to action was never broken by self-scrutiny. It was also a great vulnerability, and finally, it was fatal.

SOUTH VIETNAM, 1963

In 1963, Holbrooke was a 22-year-old U.S. Foreign Service officer on his first diplomatic posting, to South Vietnam. The State Department detailed Holbrooke to the U.S. Agency for International Development in Saigon and a small, unconventional entity called Rural Affairs. It was an odd place for a young diplomat to land—unheard of, really. Holbrooke and a colleague were going to be the first Foreign Service officers sent into the field as aid workers. The agency would put them out among peasants in Vietcong strongholds where the war was being fought and have them hand out bulgur wheat, cement, fertilizer, and barbed wire. As bachelors, they were considered relatively expendable. It was an early experiment in counterinsurgency.

Within just a couple of months of arriving in Vietnam, Holbrooke had maneuvered his way into running the Rural Affairs operation in the province of Ba Xuyen, down in the Mekong Delta. Ba Xuyen was the end of the earth. It was almost all the way to Ca Mau, and Ca Mau was the terminal point of the Asian continent, “the southernmost province of North Vietnam,” the New York Times correspondent David Halberstam once called it, because Ca Mau and the lower delta were the heartland of the Vietcong, the communist guerrillas who had been lurking for years among the hamlets and canals and rice paddies and mangrove forests. Ba Xuyen was a province of more than half a million, eight or nine hours’ drive from Saigon down Route 4, across the interminable wet flatness of the delta, nothing but flooded paddy fields mile after mile all the way to the horizon—in mid-September, when Holbrooke arrived in the town of Soc Trang, the rice shoots were still golden, not yet the emerald green of the harvest—though more often he would fly, since there was a daily milk run on an Air America Caribou between Tan Son Nhut airport and airstrips around the delta, and driving was risky by day and out of the question after dark.

His room was on the second floor of a clay-colored colonial guesthouse, with a balcony overlooking the town square, across from the provincial headquarters and its tennis court. Next door to the guesthouse was a dance club called the Bungalow, except that the government of South Vietnam had banned dancing in order to protect the honor of Vietnamese women, so the Bungalow was now just a bar where local soldiers could go drink and pick up girls. Holbrooke’s neighbors, also newly arrived, were a young Christian couple from Rhode Island, George and Renee McDowell. George was an aggie with International Voluntary Services—he was introducing local farmers to a strain of enormous watermelons from Georgia. Holbrooke made it known that he wasn’t interested. He and McDowell once went to the Soc Trang airstrip to meet some officials visiting from Saigon, and Holbrooke introduced himself: “I’m Richard Holbrooke, the AID man here in Ba Xuyen.” He gestured to McDowell, who was three years older. “This is George McDowell, the IVS boy.”

Holbrooke’s thing was strategic hamlets. There were 324 of them in Ba Xuyen—at least, that was what he arrived believing. When he asked to visit a few of the farther-flung hamlets he was told that it was too dangerous. He went anyway, in his white short-sleeve button-up shirt, with his sunglasses case clipped to the breast pocket, and found that the strategic hamlets consisted of punji sticks stuck in a moat and a barely armed local militia. The Vietcong were overrunning and destroying them at will. There were 3,000 hard-core cadres in the province, according to the intelligence reports. Saigon had permanently conceded half the provincial territory to the guerrillas, who had their own district chiefs, tax collectors, and schools. At night only the towns belonged to the government. Nonetheless, in Saigon and Washington there were 324 strategic hamlets in Ba Xuyen, putting 61 percent of the population under the government’s theoretical control.

In Soc Trang the war was very close. The airstrip was often hit by mortar fire. Holbrooke lost 15 pounds in the heat. His room had no air conditioning or fan, no working toilet or shower, and he could never get away from the mosquitoes, so he spent a good deal of time at a compound a block toward the canal that was occupied by Americans from the Military Assistance Advisory Group. They were among the 15,000 U.S. troops supporting the South Vietnamese army, often in combat. The advisers had a small projector and showed movies such as Seven Brides for Seven Brothers and Satan Never Sleeps, for which Holbrooke had a bottomless appetite. On weekends he tried to get back to Saigon.

Holbrooke was a good writer, never better than in his youth. He wrote hundreds of letters. Let him tell it.

I wish I could tell it all to you—the poorly lit room and bar that I am now sitting in, where the MAAG men sit and wait their tours out; the playmates from Playboy on the walls here, somehow very much out of place; the stacks of old magazines and paperbacks, the other hints of home that the US Army flies into the Vietcong’s homeland to make us feel a little less lost; the water everywhere, rising, raining, so that literally this province, even the ground around our building, is under water; the waiting; the ugliness, the cruelty, the tragedy. And in Saigon a regime so totally bankrupt and disgusting it is hard to describe.

There is something different about the Delta. Flying over it begins to give you some idea of the problems. It is completely flat, and ⅔ of it is under water right now. Yet it is the great VC [Vietcong] stronghold, which may be the last to fall. How is it possible? Where can they possibly be? Many are in the marshes and inaccessible swamps of the far south, but the fact is that for most, this day means being sheltered in someone’s house and in one of the hamlets right below us.

My job as civilian advisor to the province chief and overseer of the aid program here puts me continually in the position of advocate of plans and projects which would seek to make a reality out of the clichés that everyone pays lip service to. I don’t mind this (actually enjoy it) but it is sometimes tiring to try to get the Vietnamese to do something which is, after all, for their own good (or so we think . . .). On the other hand, when I step back just a little to look at everything, it seems to me that the Vietnamese have taken our overbearing presence rather well over the last few years. We arrive here with no knowledge of the country or of the situation and immediately start giving advice, some of which we can really turn almost into orders because of the materials and money and transportation that we fully control. I think that no American would stand for such a deep and continuing interference in our affairs, even if it appeared that survival was at stake. Yet the Vietnamese accept it, and with rather good grace.

At 0500 this morning the news came in that the VC had attacked and possibly overrun the furthest out outpost in the southeastern district of Ba Xuyen. It is a Cambodian post, located just three kilometers from a mangrove forest which forms the point where the lower branch of the Mekong meets the South China Sea. The mangrove forest is a VC haven, as almost all mangrove forests are. The post protects a huge and critical hamlet, also Cambodian, which was originally scheduled to be visited by [Secretary of Defense Robert] McNamara today before the schedule was cut. Anyway, by helicopter we flew out over the area for about an hour, circling at around 1500 feet, and from that height it could be clearly seen that the post had been destroyed. What the situation was on the ground could not yet be known—we did not go any lower, since we were getting shot at from time to time as we moved over the area. We refueled at Soc Trang, and joined an Eagle Flight moving out over the area now. An Eagle is a group of about 6 to 10 choppers, which fly very low over bad areas, hoping to draw fire, after which they pounce. We were above the main force choppers, which carry Vietnamese army. Finally, after the infantry had reached the hamlet and post, we went in.

On the ground was one of the worst sights I ever hope to see. The VC had apparently dug in with recoilless 75mm fire only 50 yards away, and leveled the post before moving a man against it. (Such a weapon is definitely from China—they never were used here by either US, French or VN.) Unlike most posts which fall here, it was apparently not an inside job. This may in part be due to the fact that these were Cambodians, and they are the best fighters around.

The fort was a shambles, of the 31 men in it 10 were dead, as were 7 children and 4 women, who live with their men in these terrible traps. The bodies were being assembled as we came in, and the noise of the women wailing, plus the horrible air and stench that overlay everything, was . . . One sees pictures of people picking their way through the war-torn rubble of Europe and Japan, and we have seen this sort of thing often in the histories of our times, but going in on the ground like this is still something new. One doesn’t know quite what his reactions will be. Mine were not as bad as I was afraid they might be; perhaps little by little I have been working up to this anyway. (There have been so many similar to this, and Vietnam is such a cruel country to begin with, but this was the worst I have yet been in immediately afterwards.)

But afterwards it has been harder to put away the pictures of Can Nganh post. In a way, so unreal, since the birds still flew around, and the children in the nearest houses, less than 50 yards away, played games and seemed normal. But there were the women crying over the torn bodies of their husbands, and legs sticking out here and there grotesquely.

I have my doubts, getting deeper and deeper, about our basic approach here. Recent discussions and hints I have got from various sources would indicate that out of the McNamara visits came added weight for the exponents of Victory through Air Power—the Air Force, and the armed helicopters. I feel that this is a terrible step, both morally and tactically. Of course, it would never do to actually attack policy on moral grounds in the American community here, which is a basically tough and getting tougher community (“War is hell,” justifies any horror). However, the decision to fight the VC from the air can be quite easily attacked on the simple grounds of stupidity (or as Talleyrand once said, “Sir, it is worse than a crime, it is a blunder”). The VC, I am convinced, often fire on our planes merely to draw artillery and air destruction down upon hamlets. This may sound amazing, but it is a generally accepted fact, and the reason for it that once we have committed such an act, the VC can make great propaganda hay out of it.

So, anyway, if by air power we mean to win this war, thousands of Vietnamese will die and the enemy will resist far longer; we will be making a grave mistake and I am not happy about it. Of course the irony of the whole thing is overwhelming, if one is ever stupid enough to stop and think about it. Today, in Vietnam, we are using by far worse weapons and worse—less humane—tactics than the enemy. I have no doubt at all that we kill more civilians than the VC, and with what might generally be admitted are less selective, less “right” tactics. I suppose that we are on the right side in the long run here. There is no doubt in my mind that if we lose here we will be fighting this war in other countries in Latin America and Asia within a few years. But right now, we are fighting wrong, and it hurts. In the short run terms, we really should be on the other side. Take away the ties to Hanoi and Peking and the VC are fighting for the things we should always be fighting for in the world. Instead we continue to defend a class of haves which has not yet shown its real ability to understand that the have-nots must be brought into the nation. Let that be shown, and perhaps there will be an improvement in the situation, not of our making, but to our benefit.

The whole damn thing makes me slightly ill. (Or is it my throat?) This is the most exciting assignment in the world, and I will always be grateful for having it. But I do not think I will be sorry to leave. One friend of mine just got his next assignment: Luxembourg. It seems almost a joke, but it is true. There are such places. I think I am beginning to see war, which goddamn it this really is, in the least glorified of lights. That is when the fight sometimes doesn’t even seem worth it, so bloody is the cost. But there is no choice, really, is there?

Counterinsurgency isn’t for everyone—it’s a sophisticated taste. In Vietnam it attracted the idealists. This attraction wasn’t what got Americans into the war. We fell into Vietnam and kept on sinking out of a mistaken belief that the policy of containment required us to stake our security and credibility on not losing another square mile of Asia to communism even though the enemy were nationalists. But counterinsurgency was part of the lure. It was what kept Holbrooke and Americans like him there.

We prefer our wars quick and decisive, concluding with a surrender ceremony, and we like firepower more than we want to admit, while counterinsurgency requires supreme restraint. Its apostles in Vietnam used to say, “The best weapon for killing is a knife. If you can’t use a knife, then a gun. The worst weapon is airpower.” Counterinsurgency is, according to the experts, 80 percent political. We spend our time on American charts and plans and tasks, as if the solution to another country’s internal conflict is to get our own bureaucracy right. And maybe we don’t take the politics of other people seriously. It comes down to the power of our belief in ourselves. If we are good—and are we not good?—then we won’t need to force other people to do what we want. They will know us by our deeds, and they will want for themselves what we want for them.

There was a Peanuts comic strip that circulated among Holbrooke and his friends in Vietnam. Charlie Brown’s baseball team has just gotten slaughtered, 184–0. “I don’t understand it,” Charlie Brown says. “How can we lose when we’re so sincere?!”

WASHINGTON, 1967

Years later, Holbrooke would describe an almost inevitable sequence of doubt and disillusionment that took place in the minds of certain Americans in Vietnam. First, they would begin to question official assessments of the war. Then, they would start to question U.S. tactics, and then, the strategy.

By 1967, Holbrooke had entered the fourth and final stage of doubt. He began to question the American commitment in Vietnam. He had returned home and taken a position as a senior aide to Undersecretary of State Nicholas Katzenbach. Nine thousand miles away from Vietnam, he could see that the true threat was on the home front, that the war was tearing his country apart. He was coming to the conclusion that the United States could never win, at least not on terms that Americans would accept. But for the few doves in government, that didn’t mean, “Let’s get the hell out of Vietnam.” It meant, “What the hell do we do now?” That was about as far as skepticism could take someone while he was still inside. The process of disenchantment was excruciatingly slow. Later on, people would backdate their moment of truth, their long-deferred encounter with the glaringly obvious. This was often inadvertent—they honestly couldn’t believe that they were so wrong for so many years. And when they finally did begin to lose faith, they kept it to themselves and a few sympathetic friends.

Katzenbach, number two in the State Department, was having his own doubts. He began to meet with a dozen senior people from around the government every Thursday afternoon at five o’clock in his office on the seventh floor. For 90 minutes they would sit in a circle of chairs and have drinks and talk about Vietnam. Katzenbach called it “the Non-Group,” because there was no agenda, no paper trail, and no one was allowed to quote anyone to outsiders. The Non-Group became a safe place to explore alternative policies—that was how deep the lying and fear ran throughout the Johnson administration. Secretary of State Dean Rusk knew but never attended so that he wouldn’t be tainted by talk of peace. Holbrooke walked uninvited into Katzenbach’s office and badgered him so many times that Katzenbach, who found Holbrooke’s boyish enthusiasm refreshing, finally agreed to let him join the Non-Group. Holbrooke’s neckties were too loud and his manner too flip for some of his colleagues, but he kept quiet unless one of his superiors asked him a question. Thus he was allowed priceless time with senior members of the foreign policy establishment, such as Averell Harriman, Walt Rostow, and McNamara’s deputy, Cyrus Vance. Holbrooke was the only one of them with any experience in Vietnam.

On the evening of November 1, 11 elder statesmen of the Cold War assembled at the State Department for drinks, dinner, and a briefing on Vietnam. McNamara was there; he had just submitted a long memo to President Lyndon Johnson presenting a bleak view of the war, and he couldn’t conceal his gloom. But Rusk remained a good soldier, and the briefing was upbeat—body counts and captured documents showed that the United States was winning. The next morning, the Wise Men filed into the Cabinet Room and, one by one, told Johnson what he wanted to hear—stay the course. The president was greatly reassured.

Katzenbach wasn’t. He thought the briefing of the Wise Men had been misleading and their validation of Johnson all wrong. Holbrooke thought so, too, and he offered to write up a dissenting memo for his boss to give to the president. Government service tends to turn written prose to fog and mud because it’s far better to say nothing intelligible than to make a mistake. Not in the case of Holbrooke. In 17 pages, he laid out the strategic problem by turning to history:

Hanoi uses time the way the Russians used terrain before Napoleon’s advance on Moscow, always retreating, losing every battle, but eventually creating conditions in which the enemy can no longer function. For Napoleon it was his long supply lines and the cold Russian winter; Hanoi hopes that for us it will be the mounting dissension, impatience, and frustration caused by a protracted war without fronts or other visible signs of success; a growing need to choose between guns and butter; and an increasing American repugnance at finding, for the first time, their own country cast as “the heavy” with massive fire power brought to bear against a “small Asian nation.”

North Vietnam couldn’t defeat half a million American troops, but it could drain the American public of the will to go on fighting. So Johnson had two choices. He could turn all of North and South Vietnam along with parts of Cambodia and Laos into a free-fire zone and try to knock out the enemy before dissent at home grew too strong. Or he could win back the center at home, and thus more time—not with patriotic slogans and false hopes, but by reducing the United States’ commitment. The first option was unlikely to work, because Hanoi’s will to fight was inexhaustible. The second option might work, but it would require several steps.

Johnson should change the United States’ objective—from victory over communism to a South Vietnamese government that could survive and deal with an ongoing communist threat. The United States should demand more of the South Vietnamese, militarily and politically. It should look to its own moral values and stop using airpower and artillery that killed large numbers of civilians or turned them into refugees in order to eliminate a few Vietcong: “Too many people are appalled by the brutality of the war. They feel that to fight a war of insurgency with vastly superior fire power is immoral and counter-productive. . . . Some feeling (more abroad than in the United States) is based on a feeling that the United States is calloused where non-whites are concerned.” And Johnson should announce a bombing halt over most of North Vietnam, which could lead to negotiations. “Time is the crucial element at this stage of our involvement in Viet-Nam,” Holbrooke concluded. “If we can’t speed up the tortoise of demonstrable success in the field we must concentrate on slowing down the hare of dissent at home.”

The memo didn’t call for unilateral withdrawal, or even negotiated withdrawal. It made an argument for a way to buy more time. The war in Vietnam would go on. But on the spectrum of official opinion, the view was far dovish. In vivid and uncompromising language, the 26-year-old author said that the United States could not win the war. For this reason Katzenbach hesitated to put his name to the memo. But since he agreed with it and thought its analysis brilliant, he finally signed it on November 16. He didn’t show the memo to Rusk until a copy had been sent to the White House. When Rusk read it, he told Katzenbach, “I always try to find out what the president thinks before I give my advice.” No word came back from the White House. Johnson didn’t want to hear it.

WASHINGTON, 2009

Right after taking office in 2009, President Barack Obama had to make a decision on the U.S. military’s request to send 17,000 additional combat troops and 4,000 trainers to Afghanistan. According to the Pentagon, the increase was necessary to stave off growing chaos in the south and provide security for the Afghan election in August. Secretary of State Hillary Clinton had appointed Holbrooke to a position created especially for him: special representative for Afghanistan and Pakistan. He would report through her to the president. Obama was already a historic figure, a democratic prince, the John F. Kennedy of a new generation. Holbrooke had worked for every Democratic president since Kennedy. He badly wanted to win the trust of this one.

He thought that the president should approve the troops, not just because of the eroding situation in Afghanistan but to make good on his campaign rhetoric about the need to win in Afghanistan. Holbrooke also thought that the military was trying to squeeze the new president with deceptive numbers and a rushed decision.

He kept thinking about 1965. That was the year when Johnson, after being elected, increased the number of troops in Vietnam from 23,000 to 184,000. The parallels with 2009 and Obama were uncanny.

On February 13, Holbrooke was in Kabul on his first trip to the region since his appointment. In the Situation Room, the president and his advisers were meeting to make a final decision on the troops. Clinton was giving a speech at the Asia Society and had asked Holbrooke to fill in for her. He sat in a darkened room in the U.S. embassy, connected by secure videoconference to the White House. It was past midnight in Kabul and Holbrooke was tired. When Obama called on him, he began to read from notes he’d written down in a lined copybook.

“Let me speak on Secretary Clinton’s behalf, and at her direct instructions, in support of Option 2.” This was the option to send 17,000 combat troops in one deployment rather than splitting them up into two tranches. “We do so with reluctance, and mindful of the difficulties entailed in any troop deployment. This is a difficult decision, especially at a time when Afghanistan faces a political and constitutional crisis over its own elections that further complicates your decision. As your first decision to send troops overseas and into combat—as opposed to Iraq—this decision lies at the savage intersection of policy, politics, and history.”

“Who talks like this?” Obama murmured. He sounded genuinely puzzled. Everyone around the Situation Room table heard him, but Holbrooke, 7,000 miles away, didn’t hear and kept going.

“It is in many ways strange to send more American troops into such a potentially chaotic political situation. If we send more troops, of course we deepen our commitment, with no guarantee of success. And the shadow of Vietnam hovers over us.”

Obama interrupted him. “Richard, what are you doing? Are you reading something?”

Holbrooke, onscreen, explained that the secretary had wanted to be sure the president heard her views accurately. He continued, “But if we do not send more troops, the chances of both political chaos and Taliban success increase.”

“Why are you reading?” Obama insisted.

Holbrooke stopped to explain again. He managed to get through the rest of his notes, which could have been summed up in a couple of lines. But he had lost the president. He didn’t understand what he’d done wrong, only that Obama sounded annoyed and ignored him for the rest of the meeting.

Holbrooke regretted reading his notes aloud. He’d done so in order not to ramble on, but it had sounded like a speech or a first draft of his memoirs. A few younger people seated back against the walls found it exciting to hear this old lion talk about savage intersections, but no one around the table wanted to be addressed like that, and when Obama expressed irritation they could only conclude that Holbrooke was already out of favor with the new president. Which meant that nobody had to worry about him. After the meeting, Obama told his national security adviser, James Jones, that he would tolerate Holbrooke in the Situation Room only if he kept his remarks short, and that he wanted to be in Holbrooke’s presence as little as possible.

The heart of the matter was Vietnam. Holbrooke brought it up all the time. He couldn’t resist. He passed around copies of a book he’d recently reviewed, Lessons in Disaster, about the fatally flawed decisions that led to escalation. He invoked the critical months of 1965 so portentously that Obama once asked him, “Is that the way people used to talk in the Johnson administration?” It wasn’t just that Holbrooke was becoming a Vietnam bore, a sodden old vet staggering out of the triple-canopy jungle to grab strangers by the shirtfront and make them listen to his harrowing tale. Obama actually didn’t want to hear about Vietnam. He told his young aides that it wasn’t relevant, and they agreed: Vietnam was ancient history. Obama was three years old in July 1965.

And what was Obama supposed to do with the analogy? It didn’t tell him how many more troops could make a difference in Helmand Province. It told him that his presidency might be destroyed by this war. It was the note of doom in the Situation Room. It turned Holbrooke into a lecturer, condescending to the less experienced man, and that was as intolerable to Obama as flattery. He liked young, smart, ultraloyal staffers. He didn’t like big competitive personalities.

The divide between the two men began with temperament, widened with generation, and ended in outlook. Obama—half Kenyan, raised in Indonesia, Pakistani friends in college—saw himself as the first president who understood the United States from the outside in. He grasped the limits to American power and knew that not every problem had an American solution. The Bush administration, and Clinton’s before it, had fallen prey to the hubris of a lone superpower. Then came the Iraq war and the economic collapse of 2008, and a reckoning required the country to sober up.

Obama wouldn’t say so, but his task was to manage American decline, which meant using power wisely. He embodied—his long slender fingers pressed skeptically against his cheek as he listened from the head of the table in the Situation Room—the very opposite of the baggy grandiosity that thought the United States could do anything and the craven fear of being called weak for not trying. Obama probably wasn’t thinking of the Berlin airlift or the Dayton peace accords, which Holbrooke had negotiated and which had ended the Bosnian war; Obama was thinking of the impulses that had sunk the United States in Vietnam and Iraq. The president and his aides believed these were Holbrooke’s impulses too, when in fact he was only saying, “Be careful. It could happen to you.” Obama didn’t want to hear it—couldn’t hear it, because the speaker kept distracting him with theatrics and bombast worthy of Johnson himself. So Obama told Jones, and Jones told Clinton, and Clinton told Holbrooke: stop it with Vietnam.

“They don’t think they have anything to learn from Vietnam,” she said.

“They’re going to make the same mistakes!” Holbrooke replied.

Holbrooke confessed to his friend Gelb that even Clinton wasn’t interested.

He tried to stop, but it was impossible. How could he not be haunted? There was nothing new under the sun. Somehow, after a half-century excursion across the heights of American greatness, the country had returned to the exact same place. All the questions in Afghanistan had been the questions in Vietnam. Could the United States transform Afghan society? If not, could Americans still win the war? Did our very effort make it less likely? What leverage did we have? Should we get rid of the Afghan leader? Could we talk our way out?

“It is beyond ironic that 40+ years later we are back in Vietnam,” Holbrooke wrote in his diary. “Of course, everything is different—and everything is the same. And somehow, I am back in the middle of it, the only senior official who really lived it. I had not thought much about it for years, now it comes back every day. Every program has its prior incarnation—mostly unsuccessful. . . . I think we must recognize that military success is not possible, + we must seek a negotiation. But with who? The Taliban are not Hanoi, + their alliance with Al Qaeda is a deal-breaker.”

Here was the paradox: he knew from Vietnam that what the United States was doing in Afghanistan wouldn’t work—but he thought he could do it anyway. And there was something else. If he applied the real lesson of Vietnam—don’t—he would be out of a job. And then who would he be?

Over time, he learned to save Vietnam for his staff. One day, as he sat through another White House meeting on Afghanistan, listening to another optimistic military briefing, a quote surfaced from the deep past, and he scribbled it down on a scrap of paper and took it back to the office to show his young aides, who of course had no idea where it came from: “How can we lose when we’re so sincere?”

In the fall of 2009, Obama faced another decision on troops. His new commander in Afghanistan, General Stanley McChrystal, was asking for 40,000 troops in addition to the earlier 21,000. The latest increase would put the total number of U.S. forces in Afghanistan at more than 100,000. McChrystal had been in Afghanistan since June, traveling around the country, learning the state of the war, and he had come to a conclusion: without a surge, Afghanistan would go into what he called “a death spiral.” McChrystal’s troop request had leaked, and Obama and his advisers felt boxed in again by the military.

Over ten weeks in the fall of 2009, Obama presided at no fewer than nine sessions of his National Security Council, two or three hours at a time. In his diary, Holbrooke once called the Situation Room “a room that, to me, symbolizes the problem; a windowless below-ground room in which the distance from real knowledge to people is at its very greatest—very high-ranking people who know very little make grand (or not so grand) decisions, or maybe (as in the Clinton years so often) no decisions at all.” There had been an Afghanistan strategy review in the last months of the Bush administration, and there had been another in Obama’s first weeks in office, and here they were again, this time a marathon review: a sure sign of a troubled war, like the many fact-finding missions Kennedy had sent to South Vietnam.

The discussion ran up against the fundamental contradictions of the war. Obama knew them as well as anyone. Around and around they went in the Situation Room as the weeks dragged on and Obama, crisp and lawyerly, listened and asked hard questions.

Let’s get started.

Why are we in Afghanistan?

Because al Qaeda attacked us from Afghanistan. Our objective is to prevent another attack, and ultimately to destroy al Qaeda.

But al Qaeda is in Pakistan.

If the Taliban take power again in Afghanistan, al Qaeda could regain its safe haven there.

But al Qaeda already has a safe haven in western Pakistan—not to mention in Somalia and Yemen and the African Sahel. Why do we need 100,000 troops and a counterinsurgency campaign in Afghanistan to go after 100 al Qaeda members in the tribal areas of Pakistan?

Pakistan, our supposed ally, is actually supporting our enemies. The Pakistanis won’t stand for American troops on their soil. All we can do is covert ops, intelligence collection, drone strikes in the tribal areas against militants, some of whom are attacking Pakistani targets—even that is very unpopular.

What do we really know about the Taliban? Are we sure they will allow al Qaeda back into Afghanistan?

No, but they refuse to renounce al Qaeda.

Why not do a counterterrorism campaign: drones and a few thousand Special Forces and spies going after the hard-core bad guys?

That’s what we’ve been trying since 2001, and it hasn’t worked. Only counterinsurgency will give the Afghan government the breathing space to win the support of the people and gain strength until it can defend itself.

But classic counterinsurgency requires hundreds of thousands of troops.

So we’ll limit ourselves to protecting population centers and key lines of communication until the Afghan army gets bigger and better.

What if the enemy keeps getting bigger and better?

We might need to send more troops in a year or two.

What if our presence makes it bigger and better?

We’ll begin to transfer responsibility to the Afghan government in two to three years.

What if the Afghan president, Hamid Karzai, wants us to stick around for the fat contracts and the combat brigades while his government continues to prey on the people? Counterinsurgency can only succeed with a reliable partner, and the election did Karzai’s legitimacy great harm. What if the Afghan government lacks the ability or will to win the support of the people?

There’s no good answer.

And what if the Pakistani military will never change its strategy?

There’s no good answer.

Holbrooke sat at the far end of the table, next to General David Petraeus with his four stars, and took notes. Among his notes were private interjections. When McChrystal showed a slide that changed his definition of the American goal from “defeat the Taliban” to “the Taliban-led insurgency no longer poses an existential threat to the government of Afghanistan,” without changing the number of troops, Holbrooke wrote: “Wow! Words can be used to mean whatever we want them to mean.” Susan Rice, the U.S. ambassador to the United Nations, proposed joint U.S.-Chinese aid programs in Pakistan: “NONSENSE.” Robert Gates, the secretary of defense, argued that civilian aid to Pakistan might cause a backlash against the United States: “THIS IS NONSENSE!” Vice President Joe Biden said that every one of Pakistan’s interests was also America’s interest: “HUH?”

Holbrooke kept the caustic skepticism to himself. He no longer gave speeches or read from notes. He complimented the president less often. He spoke very little, and when he did, it was on subjects that were part of his job but peripheral to the main discussion—agriculture and police corruption. He advocated a “civilian surge”—the State Department’s plan to recruit more than a thousand American experts and deploy them to Afghanistan’s cities and districts. The civilian surge gave Holbrooke a place at the table and credibility with the generals, who were always complaining that the civilian effort lagged behind. So at the White House he was careful not to say what he really thought—but back at the office, when his adviser on aid, Sepideh Keyvanshad, who did not believe that more was better in Afghanistan, asked him, “Why are we sending all these people? It won’t make any difference,” Holbrooke shot back, “You don’t think I know that?”

In the 1990s, during meetings on the war in Bosnia, Holbrooke had said whatever he believed—hadn’t hesitated to contradict his boss, Secretary of State Warren Christopher, or even President Clinton, when he thought they were wrong. Now, in the 47th year of his career, he grew careful. He felt that he didn’t have the standing with Obama to go up against the military, least of all the famous general sitting just to his left. He had no supporters in the room except Hillary Clinton, and because he was wounded, and his need for her was existential, he couldn’t allow a glimmer of light or a breath of air between them. And she was with the generals. As a result, almost no one knew what Holbrooke thought of the surge. He kept it from his colleagues and his staff.

On Columbus Day weekend, he stayed up one night till four in the morning drafting a nine-page memo for Clinton. He rewrote it several times in the following days, still not satisfied. It goes straight back to the memo he wrote for Johnson in the fall of 1967, the one about Napoleon’s Russia campaign. It has the same clarity, the same ice-blue gaze at a difficult reality.

Like you, I believe in the possibilities of American leadership, and I am not a pessimist by nature. I hope my judgments are wrong. In 1965, over the course of a week, Lyndon Johnson had the same kind of discussions we are having now, but came up with the wrong answers. In 2002–3 George W. Bush never even really consulted his own Secretary of State before committing himself to the Iraq war. Now it is our turn, and Barack Obama deserves credit for having lengthy discussions and listening to everyone before making his decisions. But the parameters of the debate have been defined almost entirely by the military, and I do not believe the full political, regional, and global implications of McChrystal’s requests have been adequately discussed.

Holbrooke believed that counterinsurgency would never succeed in Afghanistan. Historically it had worked in colonial wars, where it required a lot of coercion, and in wars where the enemy had no cross-border sanctuary. In Iraq, Petraeus’ counterinsurgency strategy had depended on specific political developments in the Shiite and Sunni communities. The analogy for Afghanistan was none of these. It was Vietnam, the war that had been barred from discussion.

Rather than securing the Afghan population, 100,000 U.S. troops would only confirm the Taliban narrative of an infidel army of occupation supporting a puppet government. Everyone said that this was a political war, but Holbrooke pointed out that the review had ignored politics—the election disaster, the cancer of corruption, Karzai’s illegitimacy. The discussions had focused almost entirely on troop numbers—but what kind of government would tens of thousands of new troops be sent to support? “The current government does not have sufficient legitimacy and appeal to motivate hundreds of thousands of Afghans to die for it,” he wrote. “While a substantial portion of the Afghan population is strongly motivated to fight the Taliban, their principal motivation is usually ethnic and tribal, not any commitment to the values supposedly represented by the government in Kabul.”

He wasn’t arguing against sending more troops—not in a memo to Clinton, anyway. (He told Gelb privately that if it were up to him, they’d send just 4,500 advisers, but he couldn’t tell Clinton that, not even discreetly.) A U.S. withdrawal from Afghanistan would “set off a cycle of uncontrollable events that could seriously damage our most vital interests,” he wrote. It was a kind of soft domino theory—not that neighboring governments would topple one after another, but the whole region stretching from the Middle East to India, with nuclear weapons and numerous insurgencies and jihadist groups, would be destabilized. Instead of a way out, Holbrooke was seeking a policy that allowed the United States to stay.

The country didn’t want to hear this, and neither did Obama, but Americans needed to be long-distance runners in Afghanistan. That was why Holbrooke kept saying it would be the longest American war. A big surge promised too much, to both Americans and Afghans, and would soon play out in predictable ways, with calls for yet more troops or a rapid departure. A more modest number—Holbrooke settled on 20,000 to 25,000, just one combat brigade and the rest trainers and advisers to the Afghan army—would hold off the Taliban and the American public while giving a new political strategy time to work. “And time, the commodity we need most to succeed, is in the shortest supply.” More time—that had been the theme of his Napoleon-in-Russia memo, too.

What would a political strategy look like? That part wasn’t clear—solutions for Afghanistan were never as persuasive as critiques. Holbrooke included a brief, vague paragraph on “reintegration and reconciliation”—“the biggest missing piece of our policy.” Reintegration meant bringing in low-level Taliban defectors. Reconciliation meant talking to the Taliban leadership. But Clinton didn’t want to hear of peace talks, and neither did the military, and neither did the White House. Talking to the enemy—the only way to end the war—was never part of the strategy review.

NEW YORK AND WASHINGTON, 2010

Yesterday I went to the final performance of the revival of South Pacific at Lincoln Center. A fantastic production, which I found immensely moving. Men were crying, myself included. I tried to understand why that show had such an enormous emotional impact on us. For me it was the combination of the beauty of the show and its music, and the capturing in that show of so many moments in American history, the show itself opening in New York at the height of New York’s greatness, 1949, the theme—Americans at war in a distant land or islands in the South Pacific—the sense of loss of American optimism and our feeling that we could do anything. The contrast with today—it was very powerful, and I kept thinking of where we were today, our nation, our lack of confidence in our own ability to lead compared to where we were in 1949 when it came out, evoking an era only five years or seven years earlier, when we had gone to the most distant corners of the globe and saved civilization.