To End a War

by Richard Holbrooke

  With Croat and Bosnian forces advancing against the Serbs in western Bosnia, there should have been a marked improvement in the situation in Mostar, far to the south. But there wasn’t. Along the line dividing the two halves of Mostar, there were only rubble and tension. The hostility of the people was evident from their stares and gestures. The Croat portion of the city was held by organized gangsters. Each part of the city was patrolled by heavily armed men in police and paramilitary uniforms. When we had toured the “confrontation line” in 1994, Boyd and I had walked between armed Croats and Muslims men only ten or fifteen feet apart who drank and talked to one another, joked, and even played cards—but with the safeties off their weapons. It was one of the saddest and most tense walks of my life.

  Now, a year later, little had changed. As our heavily armed convoy weaved through the streets, around wrecked vehicles and barricades, to reach the Muslim sector, where our meeting would be held, I was again appalled by the senselessness of the war.

  The meeting with Izetbegovic was even worse than we expected. Christopher had called Izetbegovic earlier in the day, urging him to support the agreement, but the Bosnian President told the Secretary he wanted to withhold judgment until he saw us in Mostar. He showed no appreciation that the long siege of his capital city was over. He would prefer to let the people of Sarajevo live under Serb guns for a while longer if it also meant that the NATO bombing would continue.

  Haris Silajdzic showed even greater fury. For the first time I saw in the normally urbane Prime Minister a tendency to explode that would re-emerge, sometimes disastrously, at tense moments in the negotiations. Calling the cease-fire “totally unacceptable,” Silajdzic demanded that the bombing continue. As Silajdzic continued to complain vigorously, Izetbegovic signaled me to leave the room with him. Once alone, he told me he understood why the United States had taken its position, and would reluctantly support us. But, he said, he could not publicly endorse a bombing suspension yet. (There was a large press corps assembled outside our meeting place.) First, he would have to return to Sarajevo and, as he put it, “work with my people.” He would have to show them that he had forced us to produce something “better” than the present agreement. In other words, he wanted us to return to Belgrade and “strengthen” the agreement.

  I said that I sympathized with his dilemma. I told him, in confidence, that the bombing would have ended within days anyway and that his choices, like mine, were therefore limited. We returned to the larger meeting, where Silajdzic demanded several changes in the agreement. I agreed to negotiate all of these with Milosevic as soon as I returned to Belgrade, after a Contact Group meeting scheduled for the next day in Geneva. We parted amid confusion, in a mud-filled alley surrounded by journalists using long “boom” microphones to try to pick up our farewell comments. True to their promise, the Bosnians were reserved in their public comments, but they did not go so far as to attack the agreement.

  Geneva. The next morning, I asked Bob Owen and Chris Hill to drive to Sarajevo with Silajdzic to reassure him and begin discussion of a postwar constitution. For both Americans it was their first trip over Mount Igman, and when they reached the scene of the accident, they got out for a moment to pay tribute to their fallen comrades and, as Hill told me later, take a look at that “godforsaken and worthless place.”

  The rest of us flew to Geneva for another round of Contact Group ritual. With the pace of negotiations so intense, I wanted to postpone the meeting, but we were locked in because we had agreed to let the Russians host it in lieu of a meeting in Moscow. The meeting was large and messy, but it received heavy publicity, which was all the Russians cared about. Igor Ivanov, the Russian Deputy Foreign Minister, chaired the meeting. Usually affable, Ivanov occasionally exploded in anger, but regained his composure quickly. The German Contact Group representative, Wolfgang Ischinger, helped smooth things over.

  All four nations pressed us to convene an international peace conference. I replied that the differences between the sides were still too great, but we would get there eventually. Instead, I suggested we invite the three Foreign Ministers to another meeting, similar to the one in Geneva, to be held during the United Nations General Assembly session in New York near the end of September. We needed an agreement on the creation of central governmental institutions—the “connective tissue” between the Serbs and the Federation—that had eluded us at Geneva. If we could achieve that, we could begin to plan “the big one”—a full-fledged peace conference with the three presidents.

  Belgrade. We flew back to Belgrade on the afternoon of September 16 to convince Milosevic to accept changes in the cease-fire agreement. The bombing had been suspended since 10:00 A.M. on September 14, but we made it clear to Milosevic that it could resume. President Clinton had made a strong and simple public statement, at our request, to underline this threat: “Let me emphasize that if the Bosnian Serbs do not comply with their commitments, the air strikes will resume.”

  Milosevic asked General Momcilo Perisic, the Yugoslav Army chief of staff, to join the meeting. Perisic was a sullen chain-smoker who looked like a living Cold War relic. Milosevic said that Mladic was in the hospital for removal of kidney stones. I silently hoped that they would be the kind of medical problem that Chinese and Soviet leaders sometimes discovered in their political opponents, stones from which there is no recovery. Perhaps reading my thoughts, Milosevic offered to let us visit Mladic in the hospital to prove he really was sick. I declined.*

  We presented to Milosevic and Perisic the requirements for continuing the halt in the bombing: first, we wanted the French Rapid Reaction Force to protect the roads into Sarajevo; second, “humanitarian” goods would henceforth mean all civilian goods, including cement, glass, shoes, and radios, which the Serbs had previously prevented from reaching the Bosnian capital; third, we needed assurance that a drafting error made during the long night of September 12–13 concerning the size of artillery that must be removed would be corrected. This mistake, the result of fatigue by a member of our military support group, had already been reported around the world as a “major concession” by the American negotiators. Finally, we told the Serbs that henceforth the United States and NATO, not the U.N., would decide if they were in compliance.

  Throughout this long discussion, I shuttled between the smoke-filled dining room in the villa and an American military field telephone on the patio, through which we had opened a continuous line to Sarajevo and General Rupert Smith, the British commander of all U.N. forces inside Bosnia.* Milosevic and Perisic argued over some of our demands, but eventually they agreed to all of them. Still, the same issue that had undermined so many previous cease-fires remained: making sure the orders agreed to at one level were carried out at another. The Serbs had become expert at pretending that they could not control their field commanders.

  To prevent this, we demanded the name of a Serb field commander in the Sarajevo area with whom General Rupert Smith could negotiate starting the next morning in Sarajevo. Perisic offered the name of General Dragomir Milosevic (no relation to the Serbian President), who, he promised, would appear for discussions in Sarajevo the next morning. From Sarajevo, over a poor telephone connection, Smith told me dryly that he doubted he would ever see the Serb general.

  Smith and I spent over an hour speaking on the ancient field telephone that night, and Wes Clark took over whenever I had to return to the villa. The obsolete military telephone system—there were no direct telephone lines between Belgrade and Sarajevo—was difficult to use, and several times the system got so overheated that we had to wait in silence while it cooled down. From Sarajevo, General Smith impatiently questioned our negotiations. I understood why he might mistrust Serb promises—so did we—but, like Janvier, Smith seemed slow to realize that this new situation offered a unique opportunity to break the Serb siege of Sarajevo.

  We had been with Milosevic for almost seven hours. As Owen and Hill waited for me to conclude my conversation with Gene
ral Smith, they tried to talk to Milosevic about constitutional issues. But no matter how hard the polite but persistent Owen tried, Milosevic avoided the subject. Finally, Milutinovic pulled me into a corner of the dining room. “Listen,” he told me, “tell your colleagues that my President will not discuss these issues in front of General Perisic. Hold off, and he will talk about them later.” Such deep distrust among close compatriots was as common as plum brandy in the Balkans.

  When Milosevic accepted all of our demands that evening, NATO’s bombing was truly over. Although it could have started again if the Serbs challenged the agreement, the threat of resumption kept the Serbs in line. We got our first proof of this the very next day: to General Smith’s surprise, General Milosevic appeared on schedule, and the withdrawal of Serb heavy weapons from the Sarajevo area began shortly thereafter.

  The following day, to dramatize the end of the siege of Sarajevo, we did something that had not been attempted during the war: we visited all three Balkan capitals and met all three Presidents on a single day. This was more than a stunt; we were entering a new phase of the negotiations, where the ability to visit all three capitals in a single day was essential.

  Our first stop on Sunday morning was Zagreb, where the topic was the gathering momentum of the Federation offensive—although, of course, the Federation was never mentioned; as far as the Croatians were concerned, this was their operation. Two more important towns had fallen: Bosanski Petrovac and Jajce. The Bosnian Serb communications network in the west remained out of commission, although the Serbs were struggling to repair it.

  With many of the roads toward the Serb strongholds lying open before his forces, Tudjman had to decide: should he continue the offensive, and, if so, how far should he go? His government was receiving mixed signals from the United States, and he was confused.

  Tudjman’s confusion about the American view was understandable. Two days before I returned to Zagreb, Galbraith had presented to Defense Minister Susak a formal message—a démarche, in State Department jargon—asking the Croatians to halt the military campaign. Galbraith, who did not agree with the démarche, had asked for a revision, but his appeal was overruled by Washington, and he unhappily delivered it. At almost the same time, in separate meetings with Croatian Foreign Minister Mate Granic in Washington on September 12, Lake and Christopher recommended that the advance stop as soon as the confrontation lines were “stabilized.”

  I did not agree, as John Kornblum told Washington. Simply stated, after four years of Serb aggression, the Federation forces were finally gaining territory lost at the beginning of the war. As a matter of both simple justice and high strategy, we should not oppose the offensive unless it either ran into trouble or went too far.

  Like so many issues, the policy dispute began with a flawed intelligence assessment. Almost every morning’s “daily intelligence report” brought to top Washington officials new warnings of the dangers posed by the offensive. The “experts” predicted that the more successful the Croatian-Bosnian offensive, the greater the chance that the regular Yugoslav Army would re-enter the war. These opinions were based not on secret intelligence of Yugoslav plans, but on a long-standing belief in the intelligence community about the military superiority of the Serbs and their cohesiveness.

  By mid-September, having spent more time with the Serbian leadership than any other Americans, we had come to a different opinion. We concluded that Milosevic had virtually written off the Bosnian Serbs—as he had the Krajina Serbs—and would not intervene militarily to save them. In our opinion, there was only one move that might bring Yugoslav troops back into Bosnia: closing the narrow five-kilometer-long corridor at Brcko, an action that would physically cut off the majority of the Bosnian Serb population from Serbia.

  Washington’s desire to stop the offensive became public just as we met privately with Tudjman. In a front-page article in The New York Times filed from Belgrade—an article which, significantly, quoted Washington officials—Chris Hedges wrote,

  United Nations and American officials said they feared that the assault could draw Serbia directly into the war…. “All the lights have been red, irrevocably red,” the [Washington] official said. “It risks blowing the whole thing out of the water.” The message from Washington, this official said, was “quit while you’re ahead.”

  The next day, White House Press Secretary Mike McCurry put the same message on the record. “There’s fighting going on in western Bosnia,” he said. “We wish they would suspend that fighting and turn their attention to the discussions that Ambassador Holbrooke has been conducting.” I would have preferred a different message from the White House, but McCurry, a highly skilled press secretary, would not have made these remarks without guidance from the NSC.

  While Washington wanted the offensive to stop, we never had a clear instruction, only the general sense of our senior colleagues, who left to us the exact calibration of the signal. Remembering again how Harriman and Vance had been “overinstructed” during their negotiations with the North Vietnamese in 1968, I was grateful that Washington was giving us such flexibility and support. Later, Tom Donilon told me that most of the credit for protecting our flexibility was owed to Warren Christopher, who, despite his own views, argued that Washington should back its negotiators.

  Galbraith and I saw Tudjman on September 14. Tudjman wanted clarification of the American position. He bluntly asked for my personal views. I indicated my general support for the offensive, but delayed a more detailed exchange for a second meeting so that I could discuss it with my colleagues and Washington.

  Galbraith and I met Tudjman alone again on September 17. At the same time, by prearrangement, Clark, Hill, Kerrick, and Pardew met with Susak. Peter and I sat side by side on an ornate sofa, embroidered with gold trim, while Tudjman sat at my right in a Louis Quinze armchair.

  I told Tudjman the offensive had great value to the negotiations. It would be much easier to retain at the table what had been won on the battlefield than to get the Serbs to give up territory they had controlled for several years. I urged Tudjman to take Sanski Most, Prijedor, and Bosanski Novi—all important towns that had become worldwide symbols of ethnic cleansing. If they were captured before we opened negotiations on territory, they would remain under Federation control—otherwise it would be difficult to regain them in a negotiation.

  Banja Luka, I said, was a different matter. As we spoke the road to this largest Bosnian Serb city appeared to lie open to the Croatian offensive, although it was not at all certain that the city could be taken. We knew that Susak wanted to go for it as quickly as possible. On the other hand, I told Tudjman, the city was unquestionably within the Serb portion of Bosnia. Even if it were captured, the Federation would have to return it to the Serbs in any peace negotiation. Finally, capturing Banja Luka would generate over two hundred thousand additional refugees. I did not think the United States should encourage an action that would create so many more refugees. I concluded my comments with a blunt statement: “Mr. President, I urge you to go as far as you can, but not to take Banja Luka.”

  Since we were encouraging military action in three specific areas while objecting to it in Banja Luka, I was conscious, of course, that we could be accused of applying a double standard. But these three towns were smaller and less charged with emotional and historical baggage, and they could be retained in a negotiation. And the number of refugees that would be created weighed heavily on my mind.

  Even while encouraging the offensive, Galbraith and I expressed great concern over the many refugees already displaced. We told Tudjman that there was no excuse for the brutal treatment of Serbs that followed most Croatian military successes. The abuse of Serb civilians, most of whom had lived in the area for generations, was wrong. Using a provocative phrase normally applied only to the Serbs, I told Tudjman that current Croatian behavior might be viewed as a milder form of ethnic cleansing. Tudjman reacted strongly, but did not quite deny it; if our in
formation was correct, he said, he would put an immediate stop to it. On the critical question of whether or not to take Banja Luka, Tudjman was noncommittal, although he made a strange and troubling proposal—that we “trade” Banja Luka for Tuzla, the most Muslim city in Bosnia. Galbraith leaned over to me and whispered, “This is one of his obsessions. No one else agrees with it.” I told him it was inconceivable, and it was never revived—but it had provided a momentary glimpse into his heart.

  The Western Offensive, August-September 1995

  Tudjman’s proposal reflected his deep hatred of the Muslims and his dream to unite all Croats in one country, under one flag—under his leadership. He knew he could not rearrange international boundaries while the war continued, but he was testing the idea of a substantial land swap that would restructure the entire region. Under this scheme, Zagreb would gain de facto control of much of western Bosnia, which was closer physically and economically to Zagreb than to Sarajevo, while the Serbs would control much of eastern Bosnia, leaving the Muslims with a landlocked ministate around Sarajevo. We called this the “Stalin-Hitler” scenario, recalling the division of Poland in August 1939. We had repeatedly asked Tudjman to repudiate rumors of such a deal—one version of which had received wide publicity after he had discussed it informally at a dinner in London in May 1995.

  Tensions were growing again between the Croats and Muslims. That same day we received alarming news: after taking the town of Bosanski Petrovac, the two sides had turned on each other, and three Croats had been killed by Muslim soldiers. Something had to be done immediately.