Duty: Memoirs of a Secretary at War

I knew I wouldn’t be able to coast through my last six months as secretary, but as I flew back to Washington, D.C., from Christmas vacation, I had no idea how hard it would be right up to the last days. The Arab revolutions beginning in January and our subsequent military operations against Libya were daunting enough. But there were still big internal fights coming over the next steps in both Iraq and Afghanistan; another budget battle looming; major issues with China, Russia, and the Middle East; getting the president’s agreement on a new chairman of the Joint Chiefs; and, conducting a daring—and dangerous—raid into Pakistan. I had no choice but to sprint to the finish line.

CHINA, RUSSIA, AND THE MIDDLE EAST

When Chinese defense minister Liang invited me in October 2010 to return to China, as I said earlier, he explicitly asked me to make the trip before President Hu Jintao’s state visit to the United States in late January. Liang’s restrained behavior at that fall meeting of Asian defense ministers in Hanoi and in our bilateral discussions there indicated that the People’s Liberation Army had been told to help set a positive atmosphere for Hu’s trip. When I arrived in Beijing on January 9 (thirty years after my first visit), it was obvious the Chinese were pulling out all the stops to make my visit a success. From closed-off roads and highways to banquet sites, I was given head-of-state treatment. They had been standoffish for three years because of our arms sales to Taiwan, but now they welcomed me warmly.

In every meeting, I emphasized the importance of strengthening the military-to-military relationship, including a strategic dialogue covering nuclear weapons, missile defense, space, and cyber affairs. An on-again, off-again relationship served no one’s interests. Sustained and reliable ties insulated from political ups and downs were, I said, essential to reduce miscommunication, misunderstanding, and miscalculation. I also warned that North Korea’s nuclear and missile programs had reached a point where the president had concluded they represented “a direct threat to the United States,” and we would react accordingly if they did not stop. I said that after thirty years of patiently enduring North Korea’s lethal provocations, public opinion in South Korea had changed with the sinking of their warship and shelling of their islands. They intended to react forcefully to such provocations in the future, and that raised the risk of escalating hostilities on the Korean peninsula. The Chinese should weigh in with North Korea to stand down. I also made clear our view that China’s continued aggressive response to operations of U.S. aircraft and ships operating in international airspace and waters in the South China Sea could lead to an incident that neither country wanted. We were within our rights, and they should back off. Of course, I couched all I said in diplomatic terms full of sweetness and light (I could do that when the occasion demanded), but they understood what I was saying.

All my interlocutors supported strengthening the military-to-military relationship in principle but were hesitant about a sustained, formal, high-level diplomatic-military strategic dialogue, arguing that there were already multiple mechanisms for such discussions. Given the sensitive agenda I had proposed, I think the PLA leaders were reluctant to sign on to a dialogue that would include Chinese civilian officials from the party and the Foreign Ministry. (It reminded me of the Soviet general in the strategic arms talks in the early 1970s who complained to the U.S. delegation head that he should stop talking about the detailed capabilities of Soviet missiles and nuclear weapons because the civilians on the Soviet side weren’t cleared for that information.) They didn’t want to dampen the atmosphere of my visit, so they didn’t reject the strategic dialogue idea; they just said they’d study it.

While each senior Chinese official was careful to frame his comments on other topics positively, they had some tough messages of their own. Liang said the military relationship had been “on again, off again” for thirty years. There had been “six ons and six offs.” The offs were due to U.S. arms sales to Taiwan and to “harmful discriminating actions against China,” such as our surveillance operations. On these two issues, he said, “there is no space for compromise or discretion when it comes to our core interests.” We should make mutual respect, trust, reciprocity, and benefit the guiding principles of our military relationship, he said, and “mutual respect means accommodating our core interests.” I got the point about core interests.

I went through all the familiar points about Taiwan. On surveillance, I told him we did it near many countries worldwide, including Russia, and that the Russians did it to us, and neither country considered these activities as hostile acts. I said the United States did not consider China an enemy or Cold War–style rival, but I warned that since August 2010, PLA aircraft had on several occasions come very close to our planes—I showed him a photograph of a PLA fighter closing to within thirty feet of one of our aircraft—which raised the risk of a serious incident. We then faced the press together, and Liang was exceptionally positive. He said we had reached consensus on a number of issues; the talks had been positive, constructive, and productive; and a healthy military relationship was in both our interests. He announced that the chief of the PLA general staff—Admiral Mullen’s counterpart—would visit the United States that spring. I essentially said ditto.

The chosen “bad cop” for my visit was the foreign minister, who treated me to a long, condescending, and occasionally threatening diatribe that covered all the bases: Taiwan, surveillance, North Korea, U.S. naval deployments around Korea, and China’s need to build up its military defenses. I responded in kind.

Vice President Xi Jinping (then President Hu’s likely successor) responded to my concerns about North Korea with some candor by acknowledging that the situation had become a concern for both China and the United States and, further, that the recent escalation of tension and continued enrichment of uranium “had put the six-party talks in a grim and grave situation.” Xi said China had made every effort to mediate and to keep the United States informed of those efforts. He added that a denuclearized and stable Korean peninsula was in everyone’s interest.

He raised U.S. arms sales to Taiwan almost in a perfunctory way. He, like others, downplayed China’s strength and economic success, saying that while China’s economy was the second largest in the world, GDP per person was one-tenth that of the United States, and that the gap between rural and urban China was even bigger. Liang had commented that China’s military was two to three decades behind “advanced” militaries—meaning the United States and our strongest NATO allies—and was “not a military threat to the world.”

Despite President Hu’s desire to have my visit be picture-perfect to pave the way for his state visit to Washington just a little over a week later, in a remarkable display of chutzpah, the PLA nearly wrecked both trips. Just hours before my meeting with Hu, the PLA rolled out for the first time publicly its new J-20 stealth fighter. Photos of the plane hit the Chinese press about two hours before my session with Hu. As one of my China policy experts insightfully expressed it, “This is about as big a ‘fuck you’ as you can get.” There was some talk among my team about canceling the rest of the visit or part of it, or ignoring the insult. U.S. ambassador to China Jon Huntsman, seconded by my senior China policy expert, Michael Schiffer, came up with the best approach: as I had been embarrassed, I should turn the tables and embarrass the PLA.

I met with President Hu midafternoon on January 11 in the Great Hall of the People, in a reception room roughly the size of Grand Central Station. The two of us sat at the head of a horseshoe-shaped arrangement of overstuffed easy chairs in which we and our colleagues wallowed, a setting that required us both to use microphones. After Hu’s opening pleasantries and recital of standard Chinese talking points, I noted to Hu that everyone had been focused on ensuring as positive an atmosphere as possible for his visit to Washington, but I had noticed in the Chinese media a few hours earlier reports of the rollout of the PLA’s new stealth fighter. I told Hu the U.S. press was trying to figure out the significance of this test in the middle of my visit and just before his trip. I said I was worried that the U.S. press would present the test as a negative development in the relationship and asked the president of China to advise me on how to explain the test to them. Hu laughed nervously as he turned to his military aides and asked, Is this true? A furious discussion broke out on the Chinese side involving Liang, his deputy General Ma, and others. The Chinese civilians in the room had known nothing about the test. A Chinese admiral seated farthest from Hu passed word back up the line that it had been a “scientific research project.” After several minutes of chatter on the Chinese side, Hu adamantly assured me that the rollout had been a “previously scheduled scientific test” having nothing at all to do with my visit—or his. I suspected the PLA would have given me a different explanation. That the PLA would pull such a politically portentous stunt without telling Hu in advance was worrying, to say the least.

Central Military Commission vice chairman General Xu (earlier my guest at Lincoln’s cottage in Washington) hosted a dinner for me in the same guesthouse where Hu had hosted President Obama, with several of China’s most famous singers as the entertainment. Baijiu, Chinese “white lightning,” flowed as toasts were made. Both Xu’s and Liang’s wives were present, as was Becky, and decorum was largely maintained. Our entire crew visited the Great Wall the next day, the highway shut down by troops the entire distance for my motorcade. One of the traveling press bought a small backpack at a gift shop near the wall with Obama’s picture on it dressed in a Mao jacket and wearing a PLA hat. I persuaded the journalist to sell it to me, and I presented it to the president upon my return. I told him it would validate what a lot of Republicans already thought about him. He laughed.

Hu’s visit to the United States began a week later and went off without a hitch. But high-level cordiality and professions of cooperation cannot mask the reality that the U.S.-Chinese relationship faces serious challenges. China continues to invest a growing portion of its budget in new military capabilities and technologies—including highly accurate antiship cruise and ballistic missiles, diesel and nuclear submarines, antisatellite capabilities, and stealth fighters—designed to keep U.S. air and naval assets well east of the South China Sea and Taiwan. They are building a navy that, while far inferior to that of the United States globally, could be a serious problem for us in Northeast and Southeast Asia. Beijing learned from the Soviet experience, I believe, and has no intention of matching us ship for ship, tank for tank, missile for missile, and thereby draining China financially in a no-holds-barred arms race with the United States. They are investing selectively in capabilities that target our vulnerabilities, not our strengths. The Chinese are becoming increasingly aggressive in asserting territorial claims over much of the South China Sea and islands close to Japan. And they continue to challenge U.S. air and naval surveillance missions, even though we operate in international airspace and waters. Their cyber-attack capabilities are advanced and getting better, and they are targeting both our military and our civilian networks every day. All in all, this is a relationship that will require careful and skilled long-term management by leaders on both sides if we are to sustain our partnership in some areas (for example, economic) and keep competition in other areas from becoming adversarial. A robust American air and naval presence in the Pacific, especially in East Asia, will continue to be necessary to reassure our friends and allies but also to ensure peaceful resolution of disputes.

When I arrived in Russia for the last time as secretary a day after the bombing in Libya started, I began in St. Petersburg, capital city of the Russian Empire from its founding on the Baltic Sea by Emperor Peter the Great in 1703 until the Bolshevik revolution in 1917. First stop was the Russian Naval Museum, to give a lecture to about 200 middle-grade Russian naval officers. The atmosphere was barely more welcoming than for my speech at the Russian General Staff Academy in October 2007; no applause when I was introduced and tepid applause when I finished. This time, though, the questions were not confrontational but curious. What did we see as the greatest threat? Was I streamlining the U.S. Defense Department? What role would the Navy play in U.S. security? What about joint operations and joint combat training with Russia? How about Russian naval officers attending U.S. military institutions? What was the most significant event for me as secretary? I left the session somewhat heartened by the prospect for future U.S.-Russian military exchanges and cooperation.

I then motorcaded to the Peter and Paul Fortress, the original citadel of the city, where I had been invited to fire the “noon cannon,” set off daily since the days of Peter the Great. Following the ceremony, I visited the Peter and Paul Cathedral on the grounds, burial place of most Russian tsars. As someone who had studied Russian history all my adult life, seeing these sights was a pleasure that had been denied me for decades because of the Cold War and my CIA career.

The next day, March 22, I flew to Moscow to meet with Defense Minister Serdyukov and President Medvedev. Putin was traveling. Libya was on everyone’s mind, especially in light of an unusual public difference of opinion between Putin and Medvedev. The day before, Putin had told some factory workers in central Russia that the UN resolution on Libya “reminds me of a medieval call for a crusade.” Medvedev had taken issue with that statement: “Under no circumstances is it acceptable to use expressions that essentially lead to a clash of civilizations—such as ‘crusade’ and so on.” He also defended his decision not to veto the Security Council resolution.

The Russians later firmly believed they had been deceived on Libya. They had been persuaded to abstain at the UN on the grounds that the resolution provided for a humanitarian mission to prevent the slaughter of civilians. Yet as the list of bombing targets steadily grew, it became clear that very few targets were off-limits and that NATO was intent on getting rid of Qaddafi. Convinced they had been tricked, the Russians would subsequently block any such future resolutions, including against President Bashar al-Assad in Syria.

Both Serdyukov and Medvedev expressed concern about growing civilian casualties in Libya as a result of our air strikes. I urged them not to believe Qaddafi’s claims about large-scale civilian deaths. We were taking every possible precaution to avoid such casualties and believed that very few Libyan civilians had been hurt or killed by our aircraft and missiles. I wanted the Russians to know that we believed Qaddafi was forcing civilians into buildings that were obvious targets and also that he was placing the bodies of people he had executed at the bombing sites. Medvedev said he was not happy to see NATO jets and missiles operating in Libya, but these actions were “the result of Qaddafi’s irresponsible behavior” and his “blunders.” He expressed concern that the conflict would go on indefinitely but was “not convinced things will calm down while Qaddafi is in power.” Medvedev then repeated what he had told Vice President Biden in Moscow just two weeks earlier: “Land operations in Libya may have to be considered.” He said Biden had told him that was impossible. Medvedev then worried aloud that “if Libya breaks up and al Qaeda takes root there, no one will benefit, including us, because the extremists will end up in the north Caucasus” part of Russia.

Missile defense was the other main subject of discussion during my visit. Medvedev had made new proposals for NATO-Russian cooperation in this area at the NATO summit in Lisbon the preceding November, and he had followed up with a letter to Obama. Serdyukov began our discussion by noting that Medvedev’s letter had said it was “high time” for a breakthrough in this area. Among other things, Medvedev had proposed a “sectoral approach”—that is, Russian missile defense systems would protect Russia and “neighboring states,” thus “minimizing the negative impact of the U.S. system on Russia’s nuclear forces.” There should be a legally binding agreement assuring that U.S.-NATO missile defenses would not weaken or undermine Russia’s nuclear deterrent. I told Serdyukov we were interested in the proposals Medvedev made at Lisbon. Building on Serdyukov’s suggestion for operational data exchanges, I proposed that we establish two missile defense data centers, one in Russia and one in western Europe, where both Russian and NATO officers would be assigned. The centers could do collaborative planning, establish rules of engagement for missile defense, develop preplanned responses to various missile threat scenarios, and carry out joint exercises focused on countering common missile threats.

I met with Medvedev that evening at his modernistic dacha outside Moscow. He insisted that Russia needed legal guarantees that missile defenses were not aimed at Russia. “Either we reach agreement or we increase our combat potential,” he said. I repeated what I had told Serdyukov about the impossibility of getting a legal agreement ratified by the Senate and that the Baltic states would never accept Russian responsibility for their security. I knew that the Russians’ concern over Obama’s new missile defense approach was focused on the danger posed by future improvements to our SM-3 missile systems. I told Medvedev I understood their concerns. He and I both knew the early phases were of no concern to Russia, but as the United States continued to develop more advanced capabilities, “over time we can persuade you that nothing we have in mind will jeopardize Russia’s nuclear or ballistic missile capabilities.”

Medvedev said he was grateful that Obama was president, that “I can work with him, make deals, and respect each other when we disagree.” He acknowledged that the Iranian threat was real. As we parted, he wished me “success in this part of your life and the next one. May they both be interesting.”

My program in Russia concluded with a dinner cruise that evening on the Moscow River, hosted by Serdyukov. It was an elegant affair, reciprocating a similar cruise I had hosted for him on the Potomac the previous year. On my last night in Russia as secretary of defense, as we glided by the Kremlin, I thought about the remarkable path I had followed during the forty-three years since I began work as a junior Soviet analyst at CIA two days before the USSR invaded Czechoslovakia.

Had Putin allowed Medvedev to run for reelection as president in 2012, the prospects for the Russian people and for the U.S.-Russian relationship would be far brighter. I felt that Medvedev understood Russia’s deep internal problems—economic, demographic, and political, as well as the absence of the rule of law, among others—and had realistic ideas about how to deal with them, including the need to more closely align Russia with the West and to attract foreign investment. However, Putin’s lust for power led him to shoulder Medvedev aside and reclaim the presidency. I believe Putin is a man of Russia’s past, haunted by lost empire, lost glory, and lost power. Putin potentially can serve as president until 2024. As long as he remains in that office, I believe Russia’s internal problems will not be addressed. Russia’s neighbors will continue to be subject to bullying from Moscow, and while the tensions and threats of the Cold War period will not return, opportunities for Russian cooperation with the United States and Europe will be limited. It’s a pity. Russia is a great country too long burdened and held back by autocrats.

I flew from Moscow to Egypt, a visit described earlier, and then on March 24 to Israel. The day before, there had been a terrorist attack on a bus in Jerusalem, leaving one dead and thirty-nine injured. Rocket attacks on Israeli towns from Gaza were continuing, and a little over a week before my visit the Israelis had seized a ship carrying fifty tons of rockets and missiles to Gaza, including missiles from Iran. Political unrest across the Middle East had not interrupted the security threats to Israel.

I had not been to Israel since July 2009, although Defense Minister Ehud Barak visited me in Washington every two or three months. As I said earlier, we had developed a close relationship and were very candid with each other. After a formal welcoming ceremony at the ministry in Tel Aviv (seeing the Stars and Stripes and the Star of David flying together always moved me), we went to Barak’s office to meet privately. I was there primarily to reassure the Israelis of American steadfastness in the midst of the political earthquake under way in the Middle East.

I opened the conversation by expressing condolences over the terrorist attack, to which Barak simply replied, “We will respond shortly to what happened.”

For once, we had more than Iran to discuss. He was interested in my meetings in Egypt, and the bombing of Libya, which had begun just a few days before. He was, naturally, very concerned about developments in the region. He told me Egypt was losing its grip on the Sinai peninsula and hoped it was only temporary because of the potential for large-scale smuggling of weapons into Gaza. I told him that both Tantawi and the prime minister in Cairo had reaffirmed to me their commitment to the Egyptian peace treaty with Israel and said that they would continue to work with the Israeli government. Speaking as a friend, I said now was the time for Israel not to hunker down but to act boldly in the region—to move on the peace process with the Palestinians, to reconcile with Turkey, and to help Jordan. I added that the good news about the turmoil in the region was that it was not about Israel or the United States—“No one is burning U.S. or Israeli flags, yet”—but about internal problems in the Arab countries, and we needed to make sure that that remained the focus. Barak said the best approach on Libya would be to keep hitting the military until they turned on Qaddafi. He hoped the regional turmoil would spread to Iran, where he said the mullahs were celebrating Mubarak’s fall and the increase in oil prices because of the broad unrest. We need to accelerate the sanctions, he continued, in order “to help this earthquake to reach Tehran.”

Barak asked about Obama’s view of events in the region, and I told him that while some in the United States thought the president wasn’t tough enough in international affairs, I totally disagreed. Obama had sent 60,000 troops to Afghanistan and had now attacked Libya. He was aggressively pursuing al Qaeda. While he was willing to talk with adversaries such as Iran, I said, “when push comes to shove, he is willing to push back and protect the interests of the United States and our allies.”

At a subsequent joint press conference, Barak said that the security relationship between Israel and the United States had never been stronger, and that cooperation between his ministry and the U.S. Department of Defense was unprecedented. On the unrest across the Middle East, he said that nothing like what was going on had been seen since the collapse of the Ottoman Empire, and it was a moving and inspiring phenomenon. He added, though, that “a pessimist in the Middle East is an optimist with experience.”

The next morning we drove up the Israeli coast to Caesarea for a breakfast meeting with Prime Minister Netanyahu. Caesarea was built by King Herod the Great a few years before the birth of Christ, and I would have liked to explore some of the ruins, but business left no time for pleasure. There were about twenty people at the breakfast, so both Netanyahu and I stuck pretty close to our script, although the prime minister understandably took a very tough line on the need to respond forcefully to the recent terrorist attacks. We talked about the continuing problem of Iran and, of course, about Libya and the political unrest across the region. The Israelis clearly were nervous about events, seeing considerable potential for trouble and little opportunity for outcomes that were in Israel’s interest. As I had with Barak, I urged Netanyahu not to go into a defensive crouch but to seize the moment with bold moves in the peace process. Bibi wasn’t buying.

I ended my visit with an eighty-minute motorcade to Ramallah in the Palestinian West Bank to meet with Prime Minister Salam Fayyad. It was the first time a U.S. secretary of defense had made such a trip. As I left Israeli-controlled territory, the motorcade rolled into a large fenced area and then, inside that, a sizable enclosure with high concrete walls. Everybody but me had to transfer to Palestinian armored vehicles for the drive to Ramallah. I suppose I was allowed to continue on in my own vehicle as a courtesy. Even so, my security team was pretty edgy at this point. When I met with Fayyad, he complained that although Palestinian security had never been better—we had trained them—there had been an increase in Israeli military incursions. Further, he said, violence by Israeli settlers—including “outright terrorism”—had been on the rise, but the Israeli authorities had “done nothing to rein it in.” I shared with Fayyad what I had told Netanyahu about using the regional turmoil to take bold steps for peace, adding that progress would require bold steps by the Palestinians as well. I thought my comments had about the same effect on Fayyad as they had on Netanyahu.

Less than two weeks later, I made my last visits to Saudi Arabia, Iraq, and the UAE. After a particularly productive conversation in 2010, King Abdullah of Saudi Arabia had asked me to stop by and see him whenever I was in the region, “even if only for an hour.” I had done just that in early March, and now I was back again less than a month later. We met for nearly two hours at his palace in Riyadh, a huge white marble building. His office was about ten times the size of my Pentagon office and ornately decorated with dark wood and eight crystal chandeliers. In a meeting attended by many, we agreed that the bilateral military relationship was strong and affirmed that the $60 billion arms sale he and I had concluded was on track. The king said modernization of their eastern navy (in the Persian Gulf) was the next project.

Pleasantries done, the king excused virtually everyone else, and the two of us, and the Saudi ambassador to the United States, Adel Al-Jubeir, who interpreted, privately turned to Egypt and Iran. The king, in his late eighties, was not in good health physically—he still enjoyed smoking cigarettes—but was very sharp mentally. I walked into the meeting knowing that he was very upset with the United States for what he saw as our abandonment of Mubarak and our failure to fully support other longtime friends and allies, such as Bahrain, facing similar unrest. In fact, there had been loose talk by some senior Saudis about fundamentally altering the relationship with the United States and developing closer relations with other big powers such as China and Russia.

Reading from notes, Abdullah had a stark message for me and for the president:

He concluded by saying that his message was intended to be supportive.

While we favored democratic reform, I said, the United States had not been the cause of the uprisings in Tunisia, Egypt, Libya, or Bahrain. These were the protests of people who had been forced to live too long under autocratic governments. I said our only advice to the Egyptian government, and to the protesters, had been to avoid violence and to embrace peaceful reform. I had told the king of Bahrain that stability there required reform led by the royal family. And while Iran had not caused the protests, I said, it was exploiting them for its own purposes.

After a long discussion of the unrest, the king again said the Gulf region’s leaders were bothered by the way the United States had turned its back on Mubarak and that, in light of talk of putting him on trial, the United States should protect him. I was noncommittal.

As we parted, Abdullah said he had “heard rumors I hope are not true—that you are leaving.” I said that I was leaving in a few months, to which the king replied, “Make it a few years.” I joked that President Obama insisted that I still looked healthy, but I had told him that was just on the outside. And then we parted company for the last time.

MILITARY SUCCESSION

One of the most significant responsibilities of the secretary of defense is recommending to the president officers to fill the highest positions in the military. It is a complicated business, involving not just picking the right person for each job but ensuring that the appointments are equitably distributed among the four services and dealing with the “daisy chain” of vacancies that cascade from each appointment. These senior personnel decisions usually are made months in advance because of the need to identify replacements and the uncertainty of every confirmation process.

As I said earlier, in July 2009, the president told me that he wanted to talk to Hoss Cartwright about succeeding Mike Mullen in 2011 as chairman of the Joint Chiefs of Staff. Obama, like Bush, had quickly come to admire Cartwright. The White House staff and the NSS also liked working with him. I let the chairman’s succession issue simmer for nearly a year. However, by early summer 2010 time was growing short for me to act, inasmuch as I thought I would be leaving by the end of the year. I thought there was zero chance Obama would nominate Petraeus as chairman. The White House didn’t trust him and was suspicious that he had political ambitions. An alternative candidate for chairman I had in mind was Army General Marty Dempsey, then leading the Army’s Training and Doctrine Command. Previously he had commanded a division in Baghdad during the bloody first year of the Iraq occupation, led the training of Iraqi security forces, and had served superbly as the deputy commander and then acting commander at Centcom. I wanted very much to ensure that the next chairman or vice chairman had commanded in Iraq or Afghanistan. For my own job, my short list included Hillary, Colin Powell, Panetta, and New York mayor Michael Bloomberg.

Obama and I seriously discussed the succession issue in an hour-long private meeting on October 1, 2010. He began by asking yet again whether there was any chance of my staying on longer. I simply said, “Please don’t.” I asked him, once more, if Petraeus was not a possiblity as chairman. Obama replied that pulling Petraeus out of Afghanistan would be a problem, especially with drawdowns set to begin in July 2011. I told Obama that Cartwright was willing to stay on either as national security adviser or as chairman. A big fan of Cartwright’s, I nevertheless felt obligated to share again with the president my concerns about Cartwright’s relationships with the other chiefs and his propensity to hold information close. Cartwright had told me he would prefer the national security adviser’s job as a new and different challenge. Obama said he needed “to talk to him,” and he would do so on several occasions. At the end of the meeting, I again urged the president to think about Panetta as my successor.

The president was sold on Cartwright as the next chairman, and as so often, I was being difficult. I kept thinking about having assigned Dave McKiernan to a job that did not play to his strengths and worried about doing the same to Cartwright.

On April 4, 2011, the president told me I could still change my mind about leaving. During our meeting he said that Hillary had told him the previous day, “You’re not leaning hard enough on Bob.” I told him, “I’m spent. I’m just out of gas.” I then recommended that he nominate Marty Dempsey as chairman. The previous fall, not knowing how the chairman’s succession would play out, I had recommended that the president nominate Dempsey to be the new chief of staff of the Army, and he had done so. I proposed he nominate Panetta as my successor and Petraeus to take his place at CIA. (Petraeus had surprised me shortly before by expressing his interest in the CIA job.) I told Obama I thought he could wait until mid-May to announce the military choices, but that I wanted to go public with my firm departure date by the end of April.

Dempsey was sworn in as Army chief of staff on April 11. I called him to my office the next day to tell him I was recommending him to the president to be the next chairman of the Joint Chiefs. He was flabbergasted. I shared with him the challenges I believed he would face, particularly with the budget, and said that he would need to lead the chiefs as a team, maintain their cohesion, and help the new secretary manage the relationship between the senior military and the president.

On April 28, in the East Room of the White House (where Abigail Adams once hung the presidential laundry), the president announced that I would leave on June 30 and be replaced by Panetta. Panetta would be replaced at CIA by Petraeus, and Petraeus as commander in Afghanistan by Marine General John Allen. Eikenberry would be replaced as ambassador to Afghanistan by Ryan Crocker. We were all on the dais with the president, along with the vice president, Hillary, and Mullen. The president invited each of us involved in the changes to say a few words, and we were all quite disciplined. For my part, I thanked President Obama for “asking me to stay on—and on and on.”

On Memorial Day, I stood with the president in the Rose Garden at the White House as he announced his intention to nominate Dempsey and Admiral Sandy Winnefeld as chairman and vice chairman of the Joint Chiefs, and Ray Odierno as chief of staff of the Army. Two weeks later I announced that I would recommend Admiral Jon Greenert as the next chief of naval operations. He would be my last personnel recommendation to the president.

When it was all done, I felt I had left the president with the strongest possible team of military leaders to face the daunting challenges ahead. It was a legacy that made me proud.

THE BIN LADEN RAID

During my first three and a half years as secretary of defense, the hunt for Osama bin Laden had been dormant as far as senior policy makers were concerned. While there was lip service to the priority of finding him, there were seemingly no new leads, and our focus in Afghanistan was on fighting the Taliban, not on finding Bin Laden. When Obama early in his presidency directed a more concentrated effort to get the world’s most notorious terrorist, I thought it was an empty gesture without new intelligence information on his whereabouts. In the summer and early fall of 2010, I did not know that a small cell of analysts at CIA had acquired a lead on a courier thought to be in contact with Bin Laden. In the end, he would be found not through the $25 million reward or a new agent with firm evidence of his location, and certainly not through any help from the Pakistanis. Bin Laden was found through old-fashioned detective work and long, painstaking analysis by CIA experts. There would be a lot of heroes in the Bin Laden raid and even more people in Washington who would take credit for it, but without those extraordinary analysts at CIA, there would have been no raid.

The story of the raid by now has been told countless times. Here is my perspective. Sometime in December 2010, Panetta came to see me and privately informed me of his analysts’ belief they had found Bin Laden’s location. Leon would update me from time to time, and then in February 2011 he invited the commander of the Joint Special Operations Command, Vice Admiral Bill McRaven, to CIA headquarters to begin a collaborative effort to strike the suspect compound in Abbottabad, Pakistan. McRaven’s special operators had been carrying out similar raids virtually every night for years inside Afghanistan to capture or kill Taliban commanders, and had the requisite skills and experience to carry out the strike successfully.

The president and his seniormost national security team met multiple times in March and April to debate whether to strike the compound. Joe Biden and I were the two primary skeptics, although everyone was asking tough questions. Biden’s primary concern was the political consequences of failure. My highest priority was the war in Afghanistan, and so my greatest worry was that no matter what happened during the raid, as a result the Pakistanis might well shut down our vital supply line from Karachi to Afghanistan (carrying 50 percent of our fuel and 55 percent of our cargo), withdraw permission for us to overfly Pakistan, and take other steps that would have a dramatically negative impact on the war effort. A successful raid would be a humiliation of the worst kind for the Pakistani military. The Abbottabad compound was thirty-five miles from the Pakistani capital of Islamabad, six miles from a nuclear missile facility, and within a couple of miles of the Pakistan Military Academy (their West Point), the boot camps and training centers for two storied Pakistani regiments, a Pakistani intelligence office, and a police station.

I was also concerned that the case for Bin Laden being at the compound was entirely circumstantial. We did not have a single piece of hard evidence he was there. As we probed the analysts about how confident they were Bin Laden was in the Abbottabad house, the estimates ranged from 40 to 80 percent. As a former CIA analyst, I knew those numbers were based on nothing but gut instinct. As the president said at one point, “Look, it’s a fifty-fifty proposition no matter how you look at it.” From my vantage point, we were risking the war in Afghanistan on a crapshoot.

Our discussion of the raid was influenced by the arrest in late January of a CIA security officer named Raymond Davis in Lahore, Pakistan. His car was full of weapons, spy gear, and pictures of Pakistani military installations when he was stopped by two motorcyclists who pointed guns at him. Davis shot and killed both. He was arrested at the scene. By mid-March, a deal had been struck, payments were made to the families of the two men Davis had shot, and Davis was released. But white-hot public anger in Pakistan at the United States had not abated. Another such infringement on Pakistani sovereignty would almost certainly get very ugly. And we were thinking about a beaut.

There were three possibilities for a strike at Abbottabad—a special operations raid, bombs, and a limited, small-scale strike from a drone. The advantage of the last two options was that they posed the lowest risk of a Pakistani reaction. One big disadvantage was that we would not know if we had actually killed Bin Laden. The military planners initially proposed a massive air strike using thirty-two 2,000-pound bombs. Even though we persuaded them to scale that down, there was still a high likelihood of civilian casualties in the surrounding residential neighborhood. The drone attack was attractive because any damage could be confined to the compound, but it still would require a high degree of accuracy and, importantly, the drone had not been fully tested. The special operations raid, the riskiest option, also offered the greatest chance of knowing for sure we had gotten Bin Laden and offered an opportunity for gathering up all the intelligence about al Qaeda operations he might have with him. I had total confidence in the ability of the SEAL team to carry out the mission. My reservations lay elsewhere.

I laid out my concerns in detail at a meeting with the president on April 19. Succeed or fail, the raid would jeopardize an already fragile relationship with Pakistan and thus the fate of the war in Afghanistan. I said that while I had complete confidence in the raid plan, I was concerned that the case for Bin Laden’s presence in the compound was purely circumstantial. “It is a compelling case,” I said, “for what we want to do. I worry that it is compelling because we want to do it.” I worried that Pakistani Inter-Services Intelligence (ISI) was aware of where Bin Laden was and that there might be rings of security around the compound that we knew nothing about or, at minimum, that ISI might have more eyes on the compound than we could know.

The worst-case scenario was that the Pakistanis could get a number of troops to the compound quickly, prevent extraction of our team, and take them prisoner. When I asked Vice Admiral McRaven what he planned to do if the Pakistani military showed up during the operation, he said the team would just hunker down and wait for a “diplomatic extraction.” They would wait inside the compound and not shoot any Pakistanis. I then asked what they would do if the Pakistanis breached the walls: “Do you shoot or surrender?” I said that after the Davis episode, and given the high level of anti-Americanism in Pakistan, negotiating the release of the team could take months or much longer, and meanwhile we’d have the spectacle of U.S. special operators in Pakistani custody and perhaps even show trials. Our team couldn’t surrender, I said. If the Pakistani military showed up, our team needed to be prepared to do whatever was necessary to escape. After considerable discussion, there was broad agreement to this, and as a result, additional MH-47 helicopters and forces were assigned to the mission. McRaven later expressed his appreciation to me for raising the issue.

I expressed caution about the operation based on personal experience and the historical record. I recalled the Son Tay raid in 1970 to rescue some 500 American prisoners of war in North Vietnam; despite a well-executed mission, the intelligence was flawed, and no U.S. prisoners were at the camp. I had been executive assistant to CIA director Stansfield Turner in the spring of 1980 when the attempt was made to rescue the hostages at the American embassy in Tehran. Operation Eagle Claw, a failure in the desert that left eight American servicemen dead, was aborted because of helicopter problems and then became a disaster when a helicopter crashed into a C-130 refueling aircraft on the ground. I had gone to the White House with Turner the night of the mission, and it was a searing memory. I remembered a cross-border mission into Pakistan by U.S. forces in the fall of 2008 that was supposed to be a quick and clean in-and-out, but the team ended up in an hours-long firefight and barely made it back across the border into Afghanistan. The Pakistani reaction had been so hostile that we had not undertaken another such operation since. In each case, a great plan, even when well executed, had led to national embarrassment and, in the case of Eagle Claw, a crushing humiliation that took years for our military to overcome.

I believe Obama thought from early in his presidency that my long experience in the national security arena was an asset for him. Now I told him in front of the rest of the team that perhaps in this case my experience was doing him a disservice because it made me too cautious. He forcefully disagreed, saying my concerns were exactly what he needed to take into account as he weighed the decision.

No one thought we should ask the Pakistanis for help or permission. In every instance when we had provided a heads-up to the Pakistani military or intelligence services, the target was forewarned and fled, or the Pakistanis went after the target unilaterally, prematurely, and unsuccessfully. We all knew we needed to act pretty quickly, whatever we did; everyone was scared to death of a leak. There was considerable discussion about whether to wait and see if CIA could get more proof that Bin Laden was at the compound, but the experts told us that was highly unlikely.

Who should have overall authority for executing the raid was never in question. If it was carried out under Defense Department authority, the U.S. government could not deny our involvement; CIA, on the other hand, could. To preserve at least a fig leaf—granted, a very small leaf—of deniability, we all agreed that when the time came, the president would authorize Panetta to order the operation. Defense periodically would loan—“chop”—forces to CIA for operations, so this was a familiar practice.

The final meeting was on April 28. The plan, if approved, was to launch the raid two days later. Most of us, including the president, were scheduled to attend the White House Correspondents’ Dinner that night, one of Washington’s springtime rituals in which press, politicians, and officials all dress up and pretend to like one another for at least a few hours. Someone raised the question of how it would look if all of us rose from our tables and left at the same time because of something that had happened relating to the raid. The point was also made that yukking it up when our servicemen were risking their lives in a daring operation was not desirable. Hillary was forceful in saying there should be no change in the plan and those of us going to the dinner should do so. The president strongly agreed. (As it turned out, weather forced delay of the raid by a day, and we would all later get credit for our poker faces at the dinner.)

Finally, the president went around the table and asked each person for his or her recommendation. Biden was against the operation. Cartwright and I supported the drone option. Panetta was in favor of the raid. Everyone else acknowledged it was a close call but also supported the raid. The president said he would make a decision within twenty-four hours.

The next morning Undersecretaries Michèle Flournoy and Mike Vickers came to my office to try to persuade me to support the raid option. There were no two people whose judgment I trusted more, so I listened closely. After they left, I discussed the raid with Robert Rangel. I then shut the door to my office to think about everything the three of them had said. After a few minutes, I called Donilon and asked him to inform the president that I now supported the raid. The president had made the decision to go ahead an hour or so earlier.

Midday on Sunday we gathered in the Situation Room. We were all tense, bantering nervously. The stakes involved were enormous, and yet at this point, we all knew we were just spectators. For such a sensitive operation, it seemed to me there were a lot of people in the room. Panetta remained at CIA to monitor the action. Across the hall, in a small conference room, Air Force Brigadier General Marshall Webb was monitoring a video feed of the Abbottabad compound, and an Army sergeant was keeping a detailed log of audio reports he was hearing over headphones. Someone had told the president about the video feed, and he crossed the hall to the small room, grabbed a chair, and sat in a corner, just to Webb’s right. As soon as the rest of us realized where he had gone, we joined him. Biden, Clinton, Denis McDonough, and I sat at the table, with Mullen, Donilon, Daley, John Brennan, Jim Clapper, and others standing around the edges.

When early in the raid a helicopter went down, I cringed as I remembered the attempted Iranian rescue mission thirty years before. At first, we feared disaster, but the pilot skillfully managed the crash-landing, and all the SEALs aboard were okay; the mission continued. We could track every move until the team entered the house, and then in the most critical moments of the raid, we could see and hear nothing. After an unimaginably long fifteen or so minutes, we heard the message “Geronimo—EKIA,” enemy killed in action. McRaven had told us earlier that the only way Bin Laden would be taken alive was if he greeted the SEALs naked and with his hands up. Other than a shared sigh of relief, there was little reaction in the room. The SEAL team still had to get out of the compound and get back across the border to Afghanistan, which involved a helicopter-refueling stop in a dry streambed.

After nearly forty minutes, the SEALs were headed out of the compound, some escorting women and children beyond the walls for safety as others took time to plant explosives and blow up the downed helicopter. It was a huge blast, and we could be confident not many Pakistanis anywhere close were now still asleep. And then the team was on its way, one helicopter carrying the remains of Bin Laden, another carrying the forensic evidence that proved who he was—and what turned out to be a mound of intelligence. Even after the helicopters had returned safely, there was no celebration, no high-fives. There was just a deep feeling of satisfaction—and closure—that all the Americans who had been killed by al Qaeda on September 11, 2001, and in the years before, had finally been avenged. I was very proud to work for a president who had made one of the most courageous decisions I had ever witnessed in the White House.

As on nearly all such dramatic occasions, there was a light moment. When the SEALs got Bin Laden to the base in Jalalabad, McRaven wanted to measure his height as part of making sure we had the right man. When no one had a tape measure, he had a six-foot-tall SEAL lie down beside the body. The president would later quip that McRaven had no problem blowing up a $60 million helicopter but couldn’t afford a tape measure. He would later present the admiral with one attached to a plaque.

Before we broke up and the president headed upstairs to tell the American people what had just happened, I reminded everyone that the techniques, tactics, and procedures the SEALs had used in the Bin Laden operation were used every night in Afghanistan and elsewhere in hunting down terrorists and other enemies. It was therefore essential that we agree not to release any operational details of the raid. That we killed him, I said, is all we needed to say. Everybody in that room agreed to keep mum on details. That commitment lasted about five hours. The initial leaks came from the White House and CIA. They just couldn’t wait to brag and to claim credit. The facts were often wrong, including details in the first press briefing. Nonetheless the information just kept pouring out. I was outraged and, at one point, told Donilon, “Why doesn’t everybody just shut the fuck up?” To no avail.

Soon after the raid was over, the White House released the now-famous photo of all of us watching the video in that small conference room. Within hours, I received from a friend a Photoshopped version with each of the principals shown dressed in superhero costumes: Obama was Superman; Biden, Spiderman; Hillary, Wonder Woman; and I, for some reason, was the Green Lantern. The spoof had an important substantive effect on me. We soon faced a great hue and cry demanding that we release photos of the dead Bin Laden, photos we had all seen. I quickly realized that while the Photoshop of us was amusing, others could Photoshop the pictures of Bin Laden in disrespectful ways certain to outrage Muslims everywhere and place Americans throughout the Middle East and our troops in Afghanistan at greater risk. Everyone agreed, and the president decided the photos would not be released. All the photos that had been circulating among the principals were gathered up and placed in CIA’s custody. As of this writing, none has ever leaked.

The Pakistani reaction was bad, although not as bad as I had feared. There were public anger and demonstrations, but probably the biggest impact was the humiliation of the Pakistani military. The one respected institution in the country was considered by many Pakistanis to have been either complicit in the raid or incompetent. The fact that our team had penetrated 150 miles into Pakistan, carried out the raid in the middle of a military garrison town, and then escaped without the Pakistani military being the wiser was an awful black eye. Pakistani investigations of the raid focused far more on who in Pakistan had helped us than on how the world’s most notorious terrorist had lived with impunity on their soil for five years. The supply lines to Afghanistan remained open.

Four days after the raid, I visited the SEAL team that had carried it out, and they gave me a detailed briefing. (It was my second meeting with many of them.) I congratulated them and said I had wanted to thank them in person for their extraordinary achievement. I told them that earlier in the day I had encountered the mother of one of the seventeen sailors who had been killed in the al Qaeda attack on the USS Cole. She had told me that if I met with the SEAL team, she wanted me to thank them for avenging her son. I did so. The SEALs shared with me their concerns about the leaks, particularly the fact that reporters were nosing around their communities trying to find them. They were worried about their families. I said we would do whatever was necessary to protect them—although I thought to myself that a reporter who approached one of these guys’ families likely would find himself in the middle of his worst nightmare.

I shared with them my respect for the president’s courage in making the decision to go forward with the mission. I reminded them that President Carter had perhaps gambled his presidency on such a mission in 1980, and it had failed. Obama had taken a significant risk, and thanks to those in that room, he had succeeded. I said I knew they had just returned from a deployment to Afghanistan a few months before the raid and would be heading out again in the summer. I thanked them and asked them to thank their families for me for “supporting you and your service.” I concluded by saying that the SEALs in that room truly gave meaning to George Orwell’s observation that “people sleep peaceably in their beds at night only because rough men stand ready to do violence on their behalf.”

A final observation on the raid: its success was the result of decisions and investments made over the preceding thirty years. Lessons learned from the disaster in Iran in 1980 led to the creation of the Joint Special Operations Command and development of the training and equipment that undergirded the success at Abbottabad. In 1986, as deputy director in charge of analysis at CIA, I agreed to provide more than a dozen analysts to the new Counterterrorism Center in the clandestine service, an unprecedented and controversial assignment of analysts to help inform and plan counterterrorist operations. I had no idea then that we had laid the foundation for such a historic success by those analysts’ successors twenty-five years later.

CUTTING THE DEFENSE BUDGET: MATH, NOT STRATEGY

Halfway through FY2011, Congress as usual had not passed an appropriations bill—we were still operating on a continuing resolution (CR), which meant a budget of about $530 billion instead of the $548 billion the president had requested. As I wrote earlier, that year we would have six different continuing resolutions and finally a yearlong CR representing an $18 billion cut that we would have to absorb in the last few months of the fiscal year. The world’s largest and most complex organization was being funded hand to mouth, living paycheck to paycheck. It was also apparent that Congress would not jump our budget from $530 billion in FY2011 to $553 billion in FY2012, as we were requesting.

On March 15, 2011, I gathered the senior military and civilian leadership of the department to begin planning for the dire prospects ahead. I described a stark future:

I added that I thought we would not be doing our job right if we obscured the consequences of big reductions by quietly making thousands of small cuts—“salami-slicing”—across the whole department. Significant choices and decisions needed to be made. We had to force the politicians, I said, to face up to the strategic military consequences of their budget math. For once, we had to abandon the military’s traditional “can do” culture and make clear what we “can’t do.”

On April 12, I was summoned to Bill Daley’s office for a meeting with him and OMB director Jack Lew. My budget people had learned from OMB that we were going to be hit with another major cut. Lew told me the president was going to make a speech on the budget and deficit reduction the next day and wanted to announce he would cut Defense by $400 billion over ten years. I was furious. I pointed my finger at Daley and said, “This White House’s word means nothing!” I reminded them that my December 2009 agreement with Emanuel and Orszag had been thrown out, and now the new agreement I had reached with Lew and the president just four months earlier was being thrown out as well. I reminded Daley of his unfulfilled promise regarding funding for the Libyan operation. “You didn’t get us a fucking dime,” I told him.

Again, a decision with monumental consequences was being driven by a presidential speech of which I was given one day’s advance notice. I told Daley and Lew this was math, not strategy. It would have a big impact on the morale of the forces and send a big strategic message abroad: “The United States is going home, cut a deal with Iran and China while you still can.” The only way to cut that much, I continued, was to get rid of both people and equipment, at a time when we needed to refurbish worn-out equipment from two wars, replace aging Reagan-era ships and planes, and buy a new Air Force tanker. I proposed that the president should be vague in his speech. He should say something to the effect that Defense had cut nearly $400 billion in programs over the past two years and would be asked to do the same again over the next ten to twelve years. Have him ask us, I urged, to assess our strategy, mission, and force structure and make recommendations for his decisions based on that review.

That afternoon I met with the president. He described the desperate economic circumstances facing the country and said that, as he cut domestic spending, Medicare, and Medicaid (little or none of which he has done as of this writing), he couldn’t leave out defense. He said that the Republicans would be okay with that, but not the Democrats. The best politics, he said, would be for him to lie back and stay out of the budget fight. There was no gain in it politically for him. (How many times over the years had I heard presidents, beginning with Richard Nixon, say they were doing the politically hard thing for the good of the country, when in reality it was obvious they were doing the politically easy thing?) I made the same argument to Obama that I had made to Daley and Lew: What did he want us to stop doing? I advised him to keep in mind that the enemy always gets a vote. Suppose, I said, once you make these cuts, “Iran forces you into a real war”? I spoke from the heart: “The way we will compensate for force cuts today in the next war is with blood—more American kids will die because of our decisions.”

Obama told me that he was not asking Defense to match domestic cuts dollar for dollar, maybe one for ten. (Such a one-sided ratio was never in the cards.) Arguing further was pointless, so I shifted focus to how the defense budget would be cut. In the hours between my meetings with Lew and Daley and the president, my staff and I had drafted a paragraph for the president to use in the speech that I thought made his points but mine as well. There was some jockeying over the language, but given where he was headed on our budget, what he said the next day came out as well as I might have hoped:

As I walked out of the Oval Office, I thought to myself that I wouldn’t be around to complete this review, but I had to make the fight on the cuts. With defense spending at 15 percent of all federal expenditures (it had been over 50 percent when Eisenhower made his speech about the military-industrial complex), the lowest percentage since before World War II, I was convinced the defense budget was a very modest part of the nation’s fiscal problems. Political reality demanded that the military be cut, but at what cost to the troops and to our national security? Did those playing with the math ever consider that? Were they looking at what was going on in the rest of the world?

During my remaining weeks in office, I played both sides of the street. Internally, I organized the comprehensive review with a hoped-for completion by the end of summer. We structured the review to address four broad areas of possible reductions: more “efficiencies,” a combination of further reductions in overhead costs but also cutting programs that were not critical to the future; reducing or cutting capabilities for highly specialized but lower-priority purposes where generic capabilities could be made to work, and also cutting specialized missions such as counternarcotics and building security capacity in developing countries; changes in the size and composition of our forces, which would be the most difficult internally—could we accept reductions in forces that would make fighting two simultaneous conflicts tougher? Should we reduce our ground forces?; and last, a bundle of possible changes in military compensation and benefits. While I told the department’s senior leadership that I was not comfortable with a defense budget that would grow at only the rate of inflation for ten years, I went on to ask, “With every other agency of government on the chopping block, can we credibly argue that a $400 billion cut (or 7 percent) from the over $6 trillion presently planned for defense over the next decade is catastrophic and not doable?” I began including Panetta in meetings on these issues in early June, since he would lead the effort as of July 1. Happily, Leon and I saw eye to eye on the comprehensive review.

While fulfilling my responsibilities to the president inside the Pentagon, I used my last public speeches to warn Americans about the consequences of significant reductions in defense capabilities. In a commencement speech at Notre Dame on May 22, I said that we must not diminish our ability or our determination to deal with the threats and challenges on the horizon because ultimately they must be confronted. “If history—and religion—teach us anything,” I warned, “it is that there will always be evil in the world, people bent on aggression, oppression, satisfying their greed for wealth and power and territory, or determined to impose an ideology based on the subjugation of others and the denial of liberty to men and women.” I noted my strong support of “soft” power, of diplomacy and development, but reminded the audience that “the ultimate guarantee against the success of aggressors, dictators, and terrorists in the twenty-first century, as in the twentieth, is hard power—the size, strength, and global reach of the United States military.”

Two days later I spoke at the American Enterprise Institute (AEI), a conservative think tank in Washington, where various scholars had been critical of the earlier program cuts I had made. Ironically, I was at AEI to warn against further cuts to defense. I told the audience I had spent the last two years trying to prepare our defense institutions for the inevitable decrease of the defense budget. When looking at our modernization programs, I said, “the proverbial ‘low-hanging fruit’—those weapons and other programs considered most questionable—have not only been plucked, they have been stomped on and crushed.” What remained was needed capabilities. Those programs, I warned, should be protected “unless our country’s political leadership envisions a dramatically diminished global security role for the United States.” I urged that across-the-board cuts—“the simplest and most politically expedient approach both inside the Pentagon and outside of it”—be avoided and that future spending decisions be based on hard choices focused on priorities, strategy, and risks. The worst outcome would be to cut the budget deeply while leaving the existing force structure in place, an approach I said would lead to the kind of “hollowed-out” military of the late 1970s: ill trained, ill equipped, and ill prepared. The Pentagon had to be honest with the president, Congress, and the American people that a smaller military would be able to go fewer places and do fewer things. “To shirk this discussion of risks and consequences—and the hard decisions that must follow,” I asserted, “I would regard as managerial cowardice.”

The comprehensive review would be completed under Panetta and Dempsey’s leadership and provide a road map for further cuts. It would prove critically important because the Budget Control Act passed by Congress and signed by the president in August 2011 would reduce defense spending by $485 billion over the ensuing ten years and, through a “sequestration process,” expose the military to an additional potential cut of nearly $600 billion. Math, not strategy, had prevailed after all.

As I suggested earlier, the global security environment is becoming more complex, more turbulent, and in some instances, more dangerous—and the military challenges more diverse. The military capabilities of our longtime allies are shrinking fast, and those of potential adversaries are growing. Yet our security needs and responsibilities remain global. Earlier significant cuts in defense spending following major conflicts, including after the Cold War, were made because the world scene had changed significantly for the better, at least in the short term. But in 2011, neither the state of the world nor the state of our military justified significantly less spending on defense.

The problem with the defense budget, as I saw it, is not its size but how it gets spent. It’s not that we have too many planes, warships, submarines, tanks, and troops; rather, we load up every possible piece of equipment with every possible technology, and then they are so expensive, we can buy only a small number. Defense is not disciplined about eliminating programs that are in trouble, overdue, and over budget. The Pentagon spends far too much money on goods and services that make only a tangential (if any) contribution to military capabilities—overhead, or “tail.” Congress requires the military services to keep excess bases and facilities and to buy equipment that is no longer needed or is obsolete. And the personnel costs of the all-volunteer force are skyrocketing: health care costs alone have risen in a decade from about $12 billion to nearly $60 billion.

My effort between 2009 and 2011 to cut or cap weak, failing, or unnecessary programs and to find “efficiencies” was all about spending the defense budget more wisely, to force more dollars into actual military capabilities. If the budget is slashed and the problems I just described are not addressed, disaster and tragedy lie ahead. And when the next war comes, as it surely will, our men and women in uniform will pay the price for managerial cowardice, political parochialism, and shortsightedness.

MY LAST FIGHTS: ENDING TWO WARS

I had lost the argument on Libya. I had lost on the budget. I had had a tough but—I thought—successful run for four years. The last six months were turning out very differently.

As 2011 began, we were wrestling with continuing internal Iraqi disagreement on formation of a government, an increase in attacks on our embassy and other targets by powerful Iranian-provided improvised rocket-assisted munitions (IRAMs), and planning for the post-2011 U.S. presence in Iraq.

The IRAMs were not going to threaten the security gains that had been made, but they had the potential to cause a lot of American casualties, and they reflected increased targeting by Iranian-supported extremists of our troops and diplomats. The Iraqis were making little effort to stop the attacks. In January, I asked General Austin, our commander in Iraq since the preceding September, in a videoconference if he had the authority to go out and kill those firing the IRAMs. He said he was trying to get the Iraqis to do it but would use our troops if he had to. I responded, “If you get the opportunity to kill them, do it.” I asked for a menu of possible actions we could take against the Iranians and their minions in Iraq. I was cautious about going to war with Iran over its nuclear program, but I wouldn’t stand for Iranians killing our troops in Iraq.

The principal question surrounding Iraq that spring, however, was the size of the U.S. military and diplomatic presence after December 31, when—according to the agreement Bush 43 had concluded with Maliki—all our troops had to be out of the country. Any continuing U.S. military presence would require a new agreement with the Iraqis. I met with Mike Mullen, Austin, our ambassador to Iraq, Jim Jeffrey, and others on January 31. Jeffrey and Austin said that Maliki wanted a U.S. troop presence after December but was doubtful he could get the Council of Representatives to approve a status of forces agreement (legal protections for our troops when stationed in foreign countries). In fact, all the key Iraqi leaders wanted a continuing U.S. military presence, Austin said, but as in 2008, no one wanted to take the political risk of saying so publicly or leading the political fight. Jeffrey said he was looking at a post-December State Department presence of about 20,000, many of them for security.

On February 2, in the middle of the Egypt crisis, the principals were to meet in the Situation Room to discuss all this. I believed that 40,000 Americans—20,000 civilians, 20,000 troops—would be a very hard sell, both in Washington and in Baghdad. Mullen said that Austin was trying to get the numbers down, but we were still looking at a three-to-five-year transition in Iraq. We agreed that if we stayed, we needed to keep our capabilities for intelligence, air defense, logistics, and counterterrorism.

At the principals’ meeting later that day, I said “Whoa” when we quickly dived into the details. Basic questions had to be answered first, including whether we all agreed we wanted a U.S. military presence in Iraq after December 31? (I did.) To what extent did State’s plans after December 31 depend upon a U.S. military presence? What if Congress wouldn’t approve the money for State? As so often, I said, the NSS was already in the weeds micromanaging before basic questions had been addressed.

To a certain extent, as in the Afghan debate in the fall of 2009, I found myself in a different place from both the White House advisers and the military commanders. Recognizing the huge political roadblocks, I believed a substantial U.S. military presence was needed post-2011 to help keep Iraq stabilized, to continue training and supporting their security forces, and to signal our friends in the region—and Iran—that we weren’t abandoning the field. Accordingly, I asked Austin to prepare force options below 20,000. He came back in mid-March with options for 15,000 troops (which would forgo any U.S. presence in southern Iraq) and 10,000 (which would severely limit the support we could provide for the embassy). The lower option would result in virtually no U.S. troops on the political fault line around Kirkuk between the central government and the Kurds, an area of continuing potential confrontations. Sustaining helicopter support both for our forces and especially for the embassy was vital since it was still too dangerous for civilians to move around Iraq in vehicles.

I made my fourteenth and last visit to Iraq in early April. I couldn’t help but reflect on how far we had come in four and a half years—and on the cost of that progress. I flew into Baghdad from Saudi Arabia and helicoptered to the distinguished visitors’ quarters. As we flew over the city, I marveled at how much had changed since December 2006. The security forces and police were all Iraqi now. There were traffic jams. The parks were filled with families. The markets were bustling. Life had returned to the city.

With each successive visit, my basketball-court-size bedroom had one or another new amenity, like hangers in the closet. The primitive plumbing, however, was the same. After showering the next morning, I looked in the mirror, and to my horror, my white hair had turned yellow. There had been something strange in the shower water, and now, with a full day of meetings and an interview with Katie Couric of 60 Minutes ahead of me, I looked like someone had peed on my head. Iraq continued to surprise me in new and different ways until the very end.

Apart from wanting to thank the troops, the primary purpose of my trip was to tell the Iraqi leaders they had to make some decisions quickly about whether they wanted us to stay after the end of the year. I reviewed with Prime Minister Maliki the areas in which his forces were deficient: counterterrorism, intelligence, air defense, logistics, training, and capabilities for external defense. Noting that most Iraqi leaders had privately expressed their support for a post-2011 American presence, I asked if he would describe for me his strategy for building support in the Council of Representatives. At the end of our meeting, I warned him that if U.S. soldiers kept getting killed by extremist groups and he did not approve operations to capture or kill those responsible, I had directed General Austin to exercise our right to self-defense under the security agreement and to go after them unilaterally. I had the same messages for Sunni deputy prime minister Saleh al-Mutlaq and for President Talabani. “The clock is ticking,” I said. “Time is short. You need to figure out whether you want some U.S. troops to remain after December. You can’t wait until October or even this summer to figure it out.” I also told Talabani that Iraq’s leaders needed to reach private agreement to support one another on this issue in public.

In my meetings with junior enlisted troops, they asked me about numerous news reports on the latest budget crisis in Washington and rumors that the troops might not get paid. I told them, “Let me just say you will get paid. All smart governments throughout history always pay the guys with guns first.”

By mid-April, the president asked Austin to explore the feasibility and risks of having 8,000 to 10,000 troops remain in Iraq. There was some grumbling in Defense over the low number; I thought we could make that work. But the thumb twiddling continued in both Baghdad and Washington, and in June, as I prepared to leave, the number of troops that might stay on as well as the size of our embassy post-December were totally up in the air.

I don’t know how hard the Obama administration—or the president personally—pushed the Iraqis for an agreement that would have allowed a residual U.S. troop presence. In the end, the Iraqi leadership did not try to get an agreement through their parliament that would have made possible a continued U.S. military presence after December 31. Maliki was just too fearful of the political consequences. Most Iraqis wanted us gone. It was a regrettable turn of events for our future influence in Iraq and our strategic position in the region. And a win for Iran.

As you will recall, the president had put all of us on notice in the late fall of 2010 that, while he wanted a low-key and swift review of the Afghan strategy in December, he intended to return to the subject in the spring. He didn’t wait that long. He gathered Biden, Clinton, Mullen, Donilon, Lute, and me (and other White House and NSS staff) in the Oval Office on January 20 to begin the strategy review. The key subjects were the troop drawdowns in July and determining what our presence should be in Afghanistan after 2014. Did we want bases? Would we continue to conduct counterterrorism operations? What is “Afghan good enough”? How big should the Afghan national security forces be? How much would they cost, and who would pay for them? Petraeus and the Defense Department were proposing an Afghan force level between 352,000 and 378,000. The president expressed his displeasure that those numbers had leaked, again making it look like the military was trying to “jam” him. He wondered how our strategy for pursuing “reconciliation” with the Taliban might play out and fit with Karzai’s and Pakistani general Kayani’s view. Obama said we needed a political strategy to accommodate or work around Karzai and Kayani.

It was as if we had never stopped arguing since 2009. The vice president jumped in aggressively, saying the strategy in Afghanistan could never succeed, there was no government, corruption was rampant, and Pakistan was still providing sanctuaries. He proclaimed that neither Karzai nor Kayani wanted a big Afghan army. I countered.

The internal fight heated up again on March 1, when Biden convened a meeting at his residence to push for a dramatic troop drawdown. The residence is a big Victorian-era house on the grounds of the Naval Observatory, first occupied by Vice President Nelson Rockefeller in the mid-1970s. As always, Biden was warm in welcoming us, a cordial host. When we got down to business, he asked whether the strategy had succeeded enough so we could “think bigger about transition sooner.” Could we meet our strategic goals with less “input” over the next two years? He argued again that no one wanted an Afghan army of 300,000 or more and that our commitment in Afghanistan was limiting our ability to deal with both Iran and North Korea. He contended that both public opinion and Congress were becoming more negative about the war. (In my view, virtually no effort had been made by the White House to change that attitude during the fifteen months since the president’s decisions on the Afghan surge.)

The temperature of the Afghan debate rose further a few days later, provoked in no small part by a cable from the U.S. ambassador to NATO, Ivo Daalder, reporting that Petraeus had told a NATO meeting that the transition to Afghan security leadership would “commence” everywhere by the end of 2014, a statement that seemed to contradict the president’s intention that the security transition be completed by then. When the president saw that cable, it looked to him like another case of military insubordination. As a result, the president opened an NSC meeting on March 3 with a blast: “I am troubled by people popping off in the press that 2011 doesn’t mean anything…. My intention is to begin the security transition in July 2011 and complete it by the end of 2014. We will think through the glidepath [of troop drawdowns], but I will push back very hard if anyone proposes moving the drawdowns to the right [delaying them]. I prefer to move to the left [accelerating them]. I don’t want any recommendations trying to finesse the orders I laid out.” He concluded, “If I believe I am being gamed…” and left the sentence hanging there with the clear implication the consequences would be dire.

I was pretty upset myself. I thought implicitly accusing Petraeus (and perhaps Mullen and me) of gaming him in front of thirty people in the Situation Room was inappropriate, not to mention highly disrespectful of Petraeus. As I sat there, I thought: The president doesn’t trust his commander, can’t stand Karzai, doesn’t believe in his own strategy, and doesn’t consider the war to be his. For him, it’s all about getting out. Biden continued to egg him on, and his staff missed no oppertunity to pass him inflammatory news clips and other information raising questions about Petraeus and the senior military leaders.

I called Donilon two days later to express my concern that the vice president was poisoning the well with the president with regard to Petraeus and Afghanistan. I said I thought Biden was subjecting Obama to Chinese water torture, every day saying, “the military can’t be trusted,” “the strategy can’t work,” “it’s all failing,” “the military is trying to game you, to screw you.” I said we couldn’t operate that way. I asked how the Daalder cable could be sent in to the president without someone checking its accuracy. I said, if he or the president had been concerned about the cable, why didn’t they call me instead of posturing in front of thirty people “who will inevitably leak how the president imposed his will on the military” and about mistrust of the military in the White House?

My fuse was really getting short. It seemed like I was blowing up—in my own, quiet way—nearly every day, and no longer just in the privacy of my office with my staff. As we’ve seen, I had blown up at Donilon and the vice president at a meeting on Libya on March 2 and at House Defense Appropriations chair Bill Young on the third, had come close to openly arguing with the president in the NSC meeting that same day, and had gone off on Donilon again on the fifth. Partly, I think, I was just exhausted from the daily fights.

As the debate in Washington over the pace of troop drawdowns cranked up, I wanted to get a firsthand report on how the campaign was going. I also needed to talk with Karzai about the overall relationship and our post-2014 relationship. In addition, I wanted to reassure Afghans that the drawdowns beginning in July would be gradual, that there would still be many American troops fighting in the fall.

Tensions between the United States and Karzai were running particularly high when I arrived on March 7, following the deaths the preceding week of nine young Afghan boys in an American air strike. I had a long private meeting with him late that first afternoon. I apologized profusely for the deaths of the boys and, as I had so often before, described for him the extraordinary measures we were taking to avoid civilian casualties. With regard to the security transition, I told him I shared his concerns about foreign governments and organizations operating independently of the Afghan government, creating parallel structures. I also recognized the intrusiveness of ISAF troops and operations on the daily lives of Afghans. The solution, I said, was for the Afghans gradually to assume leadership for security. While NATO would provide recommendations on which places were ready to transition, I said, Karzai should have the final approval authority.

I said the Afghan security forces were critical for transition. The United States had budgeted $12.8 billion to train, equip, and sustain those forces for the coming year, but how, I asked him, could that be sustained long term? Maybe over time Afghanistan could maintain a small regular army plus a large national-guard-type organization. I told Karzai I believed a long-term U.S. presence in Afghanistan would be important for his country but also in the interest of regional stability. We did not want permanent bases, I made clear, but perhaps we could share some facilities with the Afghan security forces. He had spoken of a binding agreement between us, but I told him it had taken Congress five years just to ratify defense-technology-sharing agreements with the British and Australians. What we needed was a mutual commitment to an enduring U.S. presence.

Partnerships must be of mutual value to last, I said, raising the level of my intensity. He and I had been working together for more than four years, I said, and I had been his advocate and defender throughout. “I have listened to you” on civilian casualties, on more respect for Afghans, on respect for Afghan sovereignty, on private security contractors, and most recently, on the provincial reconstruction teams. “But my efforts are not helped when you blame us for all of Afghanistan’s problems. We are your ally and partner. We protect your government, and we saved your life. Your criticisms are making a long-term relationship more difficult to sustain in the United States and elsewhere.” Looking ahead, I said, we needed to work together on transition and the Kabul Bank. In February, Dexter Filkins had published a devastating exposé of the looting of the bank in The New York Times. I told Karzai he could not ignore it or blame earlier audits or the United States or the international community. I told him that if he did not address the bank problem or continued to blame us, it would undermine efforts to agree on any strategic partnership. I said the bank issue provided “an opportunity for you to stand up for your people.” Not for the first time, I warned him that he had people around him who exploited his worries and concerns, who tried to get him angry and upset at us, and who propounded all kinds of ridiculous conspiracy theories.

Karzai’s responses in the meeting and then at dinner led me to wonder if he listened to anyone but the conspiracy-minded. He said he had heard that the United States wanted to weaken Afghanistan, to create many small states in its place. U.S. efforts to build the Afghan Local Police (ALP) and to work with local leaders could “be very destabilizing,” he added. In the war on terror, he claimed, it was never clear whether the United States wanted Pakistan strong or weak. The Chinese view, he said, was that the United States wanted to strengthen Afghanistan against Pakistan and to use India against China. What is the “real” American agenda? he asked. He carried on at some length about the “radicalization” of the Pashtuns, wondering who was behind it. The Indians, he said, thought it might be the United States or the United Kingdom. To all this and more, knowing the futility—and risks—of challenging him in front of a roomful of people, I responded only that he “needed to get his relationship with the United States straight in the very near future.”

Before and after the Karzai meeting, I met at length with Petraeus, Rodriguez, the operational commander, and others to pose questions I felt would be at the heart of the White House discussions in the coming weeks. I asked about their expectations for the spring and summer campaigns, whether the Pakistanis were actually making a difference, and how we might discourage our allies from pulling out of Afghanistan prematurely.

The day after the Karzai encounter was an emotional one. I flew to Camp Leatherneck in southern Afghanistan and visited the medevac unit there. Pilots, medics, and doctors described what they had been able to do with the additional assets we had provided them, the lives they had been able to save. Every day those crews put their lives on the line to save our troops; to say they are heroic doesn’t do them justice. Talking with them fueled my gratitude for what had been accomplished but reignited my fury at those in the Pentagon who had fought the medevac initiative with such vigor.

I flew to Sangin in northern Helmand province, scene of some of the toughest fighting of the Afghan War. At Forward Operating Base Sabit Qadam, I met with Marines of the 3rd Battalion, 5th Marine Regiment. Twenty-nine Marines had been killed and 175 wounded in five months clearing Sangin, the heaviest losses of any battalion in the entire war. Accompanying me was my new senior military assistant, Marine Lieutenant General John Kelly. Kelly’s son Robert had been one of those twenty-nine killed. Kelly met privately with the Marines of his son’s platoon, who gave him a picture of Robert taken a few hours before he was killed and signed by all the Marines in the platoon.

The commanders in Sangin expected a resurgence of violence during the summer. I told the press, “The Taliban will try to take back much of what they have lost, and that in many respects will be the acid test.” Rodriguez told reporters accompanying me, “We think they’ll be returning this spring to a significantly different environment than when they left last year.” I told the Marines at Sabit Qadam that they had written “in sweat and blood” a new chapter in the Marine Corps’ roll of honor. I added, “Every day I monitor how you are doing. And every day you return to your base without a loss, I say a little prayer. I say a prayer on the other days as well.”

During the worst of the battalion’s fight in Sangin, when they were taking such significant casualties, some in the Pentagon had suggested the unit be pulled out of the line. Commanders in the field strongly recommended against it, and I had deferred to their judgment. I thought to myself at Sabat Qadam that pulling them out would possibly have been one of my worst mistakes as secretary of defense. These Marines had been hit hard, very hard. But despite their terrible losses, they were very proud they had succeeded where so many others had failed. And justifiably so.

My last troop visit on the trip was to Combat Outpost Kowall, just north of Kandahar. The area had long been a Taliban stronghold. I walked a few hundred yards to a nearby village to meet with the elders and take a look at the Afghan Local Police unit there. The ALP, mentioned previously, was an initiative pushed by Petraeus that recruited men from local villages and trained them as a local security force to keep the Taliban away. There were farm animals around and, more significantly, lots of children and women out and about, unique in my visits to rural Afghanistan. There were a number of troops lining the road about twenty-five yards apart, and again, for the first time in my experience, about half were Afghans. The village council greeted me, and I met with about twenty of the ALP. I was encouraged when told by the village council that the ALP had worked so well, other nearby villages were starting to participate.

After leaving Kowall, I told the press with me that I was very encouraged and felt that “the pieces were coming together.” “The closer you get to the fight, the better it looks.” I thought but didn’t say that I only wished some of the skeptics working in the White House and the NSS would get a little closer to the fight and be a little less reliant on Washington-based intelligence assessments and press reports.

After the killing of Bin Laden, there was a frenzy of commentary about whether that success would allow us to get out of Afghanistan faster, and if not, why not. There were several Principals Committee meetings on issues relating to Afghanistan and Pakistan in April, but the decisive discussion about how many troops to withdraw and at what pace was not to occur until June.

I made my twelfth and final visit to Afghanistan in early June. Support for the war was steadily dropping at home. On May 26, House Democratic leader Nancy Pelosi had given a speech in which she said Americans had done our job to help the Afghans, and “it is time to come home.” That posture wasn’t new for her, but twenty-six Republicans joining most House Democrats in voting for an amendment calling for an exit strategy and accelerated withdrawal was. The measure failed by 215–204. One of my goals during this last trip was to make the case, through the press accompanying me, for a gradual drawdown of troops, so as to not jeopardize the troops’ hard-won gains. I warned Karzai how fragile support for the war was in Washington and mentioned his “constant criticism.”

The main purpose of the visit, though, was to say thanks and good-bye to the troops. I visited five different forward bases over two days, including spending the anniversary of D-Day, June 6, with units of the 4th Brigade of the 101st Airborne Division. I shook hands and had photos taken with some 2,500 soldiers and Marines. Choking back tears, at the end of each visit I said the same thing:

I participated in the first White House session on drawdowns by videoconference. It was very discouraging. Briefers talked about the weakness of the Afghan central government, the poor performance of the Karzai government, the dependence of the Afghan forces on ISAF, and the lack of progress on reconciliation. To my chagrin, both Panetta and Clapper said that another year or two of effort still would not lead to a satisfactory outcome. Petraeus was recommending that the final elements of the surge be withdrawn in December 2012. Biden said the president should withdraw 15,000 troops by the end of 2011 and the remaining 15,000 of the surge by April 2012 or July at the latest, before the next “fighting season.”

I responded on the video screen by asking whether the strategy was to get out of Afghanistan at all costs or to achieve some level of success for the president and the country. I said the critics were too focused on Karzai and the central government, that the situation was far better than it had been a year earlier. The war was not open-ended, I said. The surge would end in 2012, and we faced a deadline of 2014.

Donilon was so concerned that Biden had convinced the president to withdraw the entire surge by April or July 2012 that he helped me get a private session with Obama a few days after my return from Afghanistan. I started with my bottom line: I recommended he announce a drawdown of 5,000 between July and December 2011 and the return of all surge troops by the end of summer 2012—late September. This would, I said, be consistent with his decision to surge for between eighteen and twenty-four months. I said the full surge had been in place only since late summer 2010—just nine months. I said the strategy was working and that you couldn’t generalize across the entire country. A quarter of the Afghan population was under Afghan security, and our troops had a much more positive view of the Afghan army than did intelligence assessments. Kabul was now safer than Baghdad, and the Afghans had primary security responsibility there. Karzai was a challenge, I said, but we had a new start with Ryan Crocker as ambassador, and in any event, we had successfully managed around Karzai when necessary. The coalition was strong, our costs were declining sharply—from $40 billion in FY2012 to $25–$30 billion in FY2013. This most certainly was not an endless war. We needed to retain our confidence, act in a measured way. Afghanistan would be messy, just as Iraq still was, but the commanders in the field, many journalists, and NATO leaders believed we were achieving success. “The more time you spend in Afghanistan,” I told the president, “the closer to the front you get, the more optimistic people are.”

I then tackled Biden’s proposal to withdraw all surge forces by April 2012. I reminded Obama that the vice president had never accepted the 2009 decision, had never thought about the consequences if his approach failed. I said if Obama were to announce the withdrawal of all 30,000 surge troops by April, he would signal to the Afghans, the Taliban, the Pakistanis, the allies, and the world that the United States had concluded it could not be successful and was pulling the plug nine months into the surge. I continued that if he opted for April or even July, the entire force would be focused on withdrawing, on choosing which areas to leave exposed, on defense, and on meeting the deadline. The surge effort would largely be over before the first soldier came out. I reminded him that the logistics were complicated and that we would be looking at pulling 40,000 to 50,000 troops out of Iraq by December, another 15,000 from Afghanistan by then as well, and another 15,000 by April. I said I thought the troops would feel betrayed by such a decision: “After all their losses, they are convinced they are winning and thus would consider their sacrifices in vain.” Confidence and morale were high, I said, but a precipitous withdrawal of the surge troops would lead them to believe that their successes were neither understood nor appreciated. “I know you want to end this war,” I told him, “but how you end it is of critical importance. To pull the entire surge out before the end of summer would be a tragic mistake.”

Biden was relentless during those few days in pushing his view and in attacking the integrity of the senior military leadership. A White House insider told me he was telling the president, They’ll screw you every time. Biden was said to be pushing Donilon really hard, accusing him of being “too fucking even-handed.” I considered that a high compliment for a national security adviser. Tom continued to be deeply suspicious of the military, but he wanted to do what was in the best interests of the president and the country. His willingness and courage to challenge both Obama and Biden when he thought they were mistaken was a great service, including to them.

The president met with the team on June 17. I repeated most of what I had told him privately. Clinton argued forcefully that withdrawing the surge by April or July 2012 would signal we were abandoning Afghanistan. It would, she said, shock the Afghans, encourage the Taliban, and encourage ordinary Afghans to hedge their bets. We would have to cede control of some areas, and it would take all the pressure off the Taliban to seek a political solution. She recommended withdrawing 8,000 troops by December 2011, with the rest of the surge coming out by December 2012, the “pace and structure… linked to political negotiations.” We had to leverage the drawdown, she said, to pressure both the Taliban and the Afghan government. At the end, the president said he wasn’t sure we needed the surge for another fighting season, especially since we would still have nearly 70,000 troops there. He would ponder on it.

The decisive meeting was on June 21, nine days before I stepped down. The president walked into the Situation Room, sat down, and declared that he intended to withdraw 10,000 troops by the end of December and the remaining 20,000 of the surge by July 2012. “You’re welcome to try to change my mind,” he said. Petraeus and Mullen described the risks associated with that timetable. I said that a July completion of the surge withdrawal meant the surge troops would have no 2012 summer fighting season at all since the planning for their withdrawal would pull them out of the line in April and May. The vice president argued strongly for July—“though I prefer earlier”—because the fighting season would still be under way in September and therefore finishing the surge then would be illogical. (I didn’t ask what that would make a July withdrawal.)

Hillary said with conviction that the entire State Department team preferred December but she could live with September because I had suggested it. (I knew Obama wouldn’t agree to Petraeus’s proposed December timeline; at least late September would get us through much of the fighting season.) Panetta said that CIA’s analysts unanimously agreed the surge troops should stay until then. Leon then put on his experienced “Washington hand” hat and told Obama that “speaking politically,” all of Defense, State, and CIA were recommending September or later. “Do you really want to go forward with July against all that?” The president then went around the room. Clinton, Mullen, Petraeus, Panetta, Donilon, McDonough, and I all supported the end of September. Biden, Blinken, Lute, Rhodes, and Brennan supported July or earlier. Not one person outside the White House favored a withdrawal by July or earlier.

The president decided to withdraw 10,000 troops by the end of December 2011 and the remainder of the surge by the end of summer 2012. He turned to me, Hillary, Mullen, and Petraeus and asked, “If I decide this, will you support it publicly?” All but Petraeus said yes. He said that he had a confirmation hearing for CIA director in two days and that he was certain he would be asked his professional military judgment about the decision. He intended to say that the scheduled withdrawal was “more aggressive” than he liked. The president said that was okay and in fact would be helpful. But he then asked Dave, Will you say it can succeed? How can you have confidence the plan will succeed if I decided December but not September? Dave got argumentative with Obama at that point, and I came within a whisker of telling him to shut up. He had gotten most of what he wanted, and I believed we had avoided a much worse outcome.

The next day the president announced his decisions. He spoke encouragingly of the progress of the war:

Eight days later I resigned as secretary of defense. My first fight as secretary had been over Iraq. My last was over Afghanistan. My entire tenure was framed by war. I had served longer than all but four of my predecessors and had been at war every single day. It was time to go home. My wars were finally over.