From 1995 to 1996, Daalder served as director for European Affairs at the National Security Council and coordinated U.S. policy in Bosnia. Currently a senior fellow at the Brookings Institution in Washington, Daalder is the co-author of the forthcoming book Winning Ugly: NATO's War to Save Kosovo..

The administration's experience in Bosnia was the single most defining element in how it approached the pending crisis in Kosovo. First, there was a conviction that you had to act early, and you had to act before having massive casualties, or massive ethnic cleansing taking place. Second, you had to act in a way that was consistent and forceful against the main perpetrator of the conflict, which was believed to be Mr. Milosevic. Third, you had to have a clear sense of where you wanted to go regarding the final political outcome of the Kosovo conflict.

It is here that the problem arises, because, in the end, the administration was not prepared to sanction an independent Kosovo. It had, in fact, adopted a political position that was very similar to Mr. Milosevic's, which was that Kosovo had to remain part of Serbia; it had to remain part of the former Yugoslavia. So, on the one hand, they had a policy of opposing Milosevic with the threat and use of force, if necessary. But on the other hand, they had a policy that relied on his underlying political demands, and accepted those demands as the correct ones. That conflict was never really resolved until the war started.

So, right from the beginning, we're in a bit of a political conundrum here.

Absolutely. There is a political conundrum with believing, on one hand, that Milosevic is the problem and that you have to oppose him, and on the other hand, supporting a political solution that is much closer to that which Milosevic supports than what the Kosovars support.

Once we're in Bosnia, how does that affect our interests in Kosovo?

In Bosnia, once we have committed ourselves, not only to be there in large numbers and troop strength, but also to maintain a Bosnia that is united between different factions within Bosnia--then we see Kosovo politically through the lenses of Bosnia. Solutions that might have been possible-- partition, or independence for Kosovo--are suddenly no longer possible, because of the consequences those solutions would have for Bosnia. For example, if Kosovo became independent, what is the argument we can use against the Bosnian Serbs and the Bosnian Croats? They, too, would like to join their brethren to their east and west. The argument becomes very difficult. If we would have supported partition in Kosovo, why not in Bosnia?

Quite apart from Bosnia, independence for Kosovo could have had ramifications for the wider region. Macedonia has a large Albanian population that might have desired to become part of Kosovo, or also to leave Macedonia. How do you deal with a large Albanian population in Kosovo and in Albania? Do you merge them? Do you make them two independent states? At what point does the unraveling of states in eastern and central and southeastern Europe stop, once you start in Kosovo? So the argument was, let's draw the line here. That meant that the line we drew at Dayton would stand, which said that Bosnia had to remain unified. Also, the consequences of an independent Kosovo for the rest of the region were simply unacceptable. ...

For understandable reasons, they didn't deal with the issue of Kosovo at Dayton. How did that affect the Albanians, their thinking, their goals, and the means to achieve those goals?

The failure to deal with Kosovo in Dayton was understandable. The focus there was on ending a war that was very real in Bosnia, as opposed to preventing a potential conflict that was not yet real in Kosovo. The failure to deal with Kosovo in Dayton led the Albanians to conclude that the one way to get western attention and a Dayton-like conference, and to get the president of the United States to pay attention to you, is to use violence. That violence begets international attention, and therefore, one should start violence. The Kosovars had been pursuing a policy of non-violent opposition since 1989. Suddenly, that became less and less viable. As time goes by, more and more people realize or conclude that the way you get the West involved is to start killing people.

And in February of 1998 that happens--they start killing people. We have a special envoy, Robert Gelbard, in the area then. He makes some comments. What does he say, and what's the significance?

Bob Gelbard, the US envoy to the region, is in Belgrade as the troubles start brewing inside Kosovo. He says to Mr. Milosevic, in his very clear message, both publicly but particularly in private, that there is a very big twist that Mr. Milosevic faces.

On the one hand lies integration in the rest of Europe at one fork of the road; at the other, is utter darkness. The determining factor is how Milosevic will deal with the Kosovo Liberation Army, which is beginning to engage in what he calls terrorist acts--killing law officers and policemen inside Serbia. He tells Milosevic, "You can deal with this terrorist group in a way that is consistent with dealing with terrorism, but don't go after the population--find a way to resolve the Kosovo problem politically--give the Kosovars more political rights, greater autonomy, and more rights over their own destiny. Then the United States and its allies will continue to engage Serbia, and will allow Serbia to emerge as part of the community of nations. But if you don't, if you go after the population, if you don't deal with them politically," there is, as he said, "utter darkness" at the end of the tunnel.

Days later, the Serb forces go hard after a particular clan, which is connected with one of the KLA strongholds at the time. That gets the attention of Secretary Albright immediately. She sees this, and reacts immediately. Tell me first of all, what does it mean to have Secretary Albright as secretary of state at that point in time? What does she bring to the table?

Since the beginning of her presence in the Clinton administration, Secretary Albright been perhaps the most forceful advocate for strong forceful opposition to the kinds of policies that Milosevic conducted, in Croatia, then in Bosnia, and by February, 1998, inside Kosovo. Her constant refrain was that the only language Milosevic understands is the language of force--that we have to threaten force, and if necessary use it--engage in air strikes, clobber him over the head like a schoolyard bully, and put him in his place . . . and then the problem gets solved.

The formula is to make the case over and over and over again that you need to be forceful. She has a particular knack for putting this in highly rhetorical and forceful language--go after Milosevic, make clear that this will not stand--hope to force the administration to come along, convince the public and the American Congress to come along, and in the end, give the allies no choice but to join what the United States is about to embark upon.

Tell me about what's happening in the White House, as she's pushing her rhetoric. What's the appetite, at that moment, for another adventure in the Balkans?

In the early part of 1998, the White House is preoccupied with very different things. The Monica Lewinsky story has just broken. There's the notion that, at this point, if you engage in another foreign adventure, it would have been portrayed that you start using force in a "Wag-the-Dog" scenario. That was not generally supported in the White House

The White House also had a different perspective than Madeleine Albright about how you would resolve this conflict. Their perspective generally was that we had in Milosevic somebody that we could perhaps make a deal with. It was a perspective that Richard Holbrooke brought to fruition in Dayton in 1995. Milosevic is a man who, however odious his behavior, however wrong his policies, if you deal with him with the right kind of carrots and the right kind of sticks, you get a deal. There's a strong belief in the White House, and in other parts of the administration, that forceful rhetoric--threats without really having a policy behind it, such as those was coming from the State

Department--was the wrong kind of policy at the wrong time.

And at that time, as it's heating up in Kosovo, one of the first questions that comes up is the Christmas warning. The Clinton administration reiterated the Christmas warning on a number of occasions, but it didn't act on the warning.

The Christmas warning was reiterated by the Clinton administration shortly after President Bush first issued it in early 1993. As originally issued, the warning was if Mr. Milosevic engages in threats against the Kosovars, or engages in military action against the Kosovars, the US will unilaterally act against you, including bombing in Serbia. By 1996, 1997, a realization emerged within the administration that the Christmas warning, originally issued as a unilateral US threat, could no longer be implemented.

It was very clear from the moment that violence started that the Christmas warning was off the table as far as Sandy Berger was concerned, as far as the president of the United States was concerned, and as far as Madeleine Albright was concerned. She may have wanted it to be on the table, but there was a clear decision not to have it on the table. The reason was simple. First, if we started to use force against Mr. Milosevic in these circumstances, we would not know what would happen inside Bosnia. Bosnia had now emerged as a major issue in how we dealt with Mr. Milosevic, in a way that didn't exist in 1992 or 1993. Second, the Christmas warning was a unilateral US threat to use unilateral US force. NATO had now been committed to dealing with the Balkans, because the United States had forced NATO into the position that it had in Bosnia. So the threat of force had to be a NATO threat.

So, from the perspective of this official US policy, the Christmas warning was no longer on the table. When asked publicly, the answer always was that all options are on the table. When asked at the Congress specifically, is the Christmas warning still in effect, Madeleine Albright, Bob Gelbard and others said "our policy is to have every option on the table." The warning was a specific threat--if you engage in violence, we will bomb you. That is not the same as saying all options are on the table.

What would Milosevic take from the fact that the warning had been in place and then it was not acted on? What would he take from that?

Clearly he didn't believe that the warning was any longer in effect. The moment he engages in the violence exactly defined by the Christmas warning, and nothing happens, means to him that, "Obviously, whatever I do, they're not going to pay attention. To the extent that they do pay attention, they're not going to bomb me, so I can continue."

Instead, the administration calls on Richard Holbrooke. They bring Holbrooke back to do his magic. Tell me about that. What does that mean? What's the significance of Richard Holbrooke coming on to the scene?

What's interesting is that the administration goes through an evolution of policy. In the first month of the crisis, the policy led by Madeleine Albright says we need to act early, and we need to act forcefully. She goes to London within days of the massacre and gets the contact group together, gets a very forceful statement out of it, threatens sanctions, and is willing to keep at least some form of military threat still on the table. She pushes the allies to go along with this course. She then has problems with the allies. She also turns out to have problems at home, because in the end, neither the allies nor the White House is dependent on, or prepared, to go in the direction that she wants to go.

Therefore, Richard Holbrooke emerges. Whether it is by request or whether it is by his forceful personality, Holbrooke says, "This is not the way to do this. The way to solve this problem is to deal directly with Milosevic. I'm the person who can deal with Milosevic. I've done it at Dayton. I'll do it again." At that point, Madeleine Albright disappears from the scene. Her policy perspective, which is forceful strong action against Milosevic, gets pushed aside in favor of a policy perspective that says we need to deal with Milosevic, we need to find a negotiating process, and we need to get the Serbs and the Kosovars to talk and deal directly with each other.

In mid-May, we see the first public meeting between Milosevic and Rugova. Both sides need to get some benefits in order to get that meeting staged. Milosevic gets sanctions relief--sanctions we have just slapped on three or four days beforehand. Rugova gets a meeting to the White House. Most people who go to the White House and meet the president are heads of state. The meaning for Mr. Rugova of getting a White House meeting is recognition that this guy has real power and real control over his state, his country, his territory called Kosovo. That's the meaning of getting Rugova to the White House and getting Milosevic sanctions relief. But in terms of settling the conflict, what we have is a photo op. but no coming together of the two sides on the issue in question. By just recognizing Rugova, you have not settled the question of who talks for who when it comes to the Kosovars.

And in recognizing Rugova, we have taken on the Kosovo cause. Now we're in there.

The meeting between the president and Rugova is recognition that what the Kosovars are demanding, a greater degree of autonomy, is at very least something that the United States will support. But it is not the intention of the United States, even though that is how people may well have read it, to get in bed with the Kosovars. The United States does not support the Kosovars' fundamental demand, which is independence. Indeed, the United States supports what Mr. Milosevic wants, which is to maintain Kosovo as an integral part of the Federal Republic of Yugoslavia, if not of Serbia itself.

In June, NATO begins some military planning, and they put together a lot of different plans. One of the first things they do is send some planes up and around Kosovo, the "Balkan Air Show." What's going on with that? What's that message, and who pushes that?

By May, and into June, NATO is ready to get into the act. Two things happen very early on. One is a series of military planning exercises that really range from the banal--having a couple of trips along the border--to the extreme, which is an invasion of Yugoslavia, and taking control of the whole country as exercise. But before you can move in that direction, NATO wants to demonstrate that it has real potential to do harm inside Kosovo or inside Serbia, and it launches this "Balkan Air Show." They have a bunch of aircraft flying over the border of Albania and Kosovo. They're flying low enough to make a lot of noise, and in effect, to tell the Serbs we're here, we can do damage, and perhaps more importantly, to tell the Kosovars we're here, we can do damage. But in the end, the fact that they only do an air show . . . demonstrates NATO's weakness in essence, rather than its strength.

Coming after a Christmas warning that has not been implemented, let alone reiterated publicly, the NATO air show once again tells Milosevic that the United States and NATO really aren't in this thing to fight--they are here basically to demonstrate that they're powerful, but not in order to go after the schoolyard bully directly with the use of force. So Milosevic takes from that that he can continue.

And by the summer, even though the Serbs had backed off a little bit, the changing dynamic on the ground is significant. The KLA has really moved out now, and they have taken over a lot of territory. That's an important point, because they take more territory and become a bigger force,. It changes the political calculations about who we are fighting for, and who we are helping here.

What happens in the summer of 1998 is an interesting change in the dynamics. The KLA moves out and becomes stronger and stronger militarily. In fact, they gain territory from the Serbs. By July, they claim control of about 40% of all of Kosovo, and that changes the dynamic in two very different ways. First, the KLA becomes the political leadership of Kosovo. It displaces Rugova, and demonstrates to the Kosovars that a policy of non-violent resistance is not working--that violence, and threats, and force are what would get Kosovo what it wants--autonomy and independence.

Second, there's a change in dynamic for the international community, particularly the Europeans, but also many in the United States, who believe that the KLA are a bunch of thugs, and these thugs are now winning. All of a sudden, a policy designed to oppose Milosevic creates a situation where the more we oppose Milosevic, the more these thugs will win--at what point are the KLA thugs the problem? At what point are we going to oppose these guys? So the notion of intervening militarily is suddenly problematic. "We don't want to be the air force of the KLA," is the standard stock answer of the allies to bombing the Serbs. At that point, the allies decide to stand back, to say, wait a minute, we've got to find a political solution--not because Milosevic might otherwise do bad things--but otherwise, the KLA will win.

By late summer, the Serbs are trying to take back some of that territory. NATO had many different options. But there was really only one option that they were developing, that was even on the table, which was the phased air plan. Why was that the only one?

A very interesting part of NATO's decision-making structure was going through the options. In June, the NATO defense ministers tell the NATO military authorities to look at all the options in which NATO military force might be brought in to improve and change the situation inside Kosovo. As good military men, the NATO military authorities go out, and they create a series of plans about what it is that we can do. This includes a whole series of plans that involve using ground forces to occupy Kosovo, and even to subjugate the former Yugoslavia by going into Belgrade. As the plan is being developed, and the United States military representatives and the people at SHAPE, at the NATO headquarters, are fully engaged in it, something changes in mid-air.

Suddenly the question of considering ground troops becomes one that the United States says, "No. We don't even want to look at this option anymore, we don't even want to present this option to the political authorities. We only want to look at air power. We don't want to look at any use of ground troops in any circumstances." Right in the middle of the planning process, there is a big fight within NATO, and there's a question whether the plans that had been drawn up can be presented to the civilian authorities. In the end, they are. But once presented to the civilian authorities, there is an immediate decision by those authorities that we're only going to look at two air options. One is a phased air campaign. The other one is a limited air response, consisting of a few tens of Cruise missiles meant to send a signal, and a phased air campaign built mostly on what we did in Bosnia in 1995. We would phase up air strikes over a period of time, designed to get Milosevic to a political solution.

So there wasn't a stomach among the allies for anything more than dropping a few bombs, and in particular among . . .

There were some allies, notably the British, who concluded by the early part of the summer that no solution to this problem could ever happen without military force, and that that military force would have to include a ground component. But there was a signal sent from the United States by July of 1998 that no, there is no way we are ever going to consider any deployment of ground troops, NATO or US, in this situation. The only thing that we are willing to do, and even to look at seriously, is the question of air strikes.

By September 30, there is a massacre in the village of Donji Obrinje. It's on the front page of the New York Times when they cover it. There's a pretty gruesome picture, and a big story about it. That day, there is a principals' committee meeting at the White House. The New York Times is in the situation room, standing witness to the decision making there. How important is that? How important is the fact that those images are constantly on the TV, in the papers? Images of these massacres are right there, front and center. How important is that to this administration in driving their decision making?

Images of the kind of horrible atrocities that were committed in late September are heart wrenching for any human being, whether you're the national security adviser, the secretary of state, or just simple Joe down the street. Those kinds of images will influence your policy. If the front page of the New York Times is showing the kind of atrocities that were committed

. . . If that's on the table when you are considering what to do, you're more likely to decide to do something, rather than to do nothing. It is very easy once you have these kinds of images to say, "That does it. Here we've got to go, and the decision is that we've got to threaten serious force."

That afternoon, Albright, Berger and Cohen go up to Capitol Hill to brief the senators and the congressional leaders there. Why were they going up? What was the message that they were carrying? Tell me about them travelling up to the Hill.

Once the US administration decides to press the allies to threaten serious use of force, and, if necessary, implement a phased air campaign, it is necessary to sell it on the Hill. It is particularly necessary now, because the president is in trouble politically over the Lewinsky scandal. They've got to get the Hill on board. Albright, Berger, Shelton and Cohen travel to the Hill, and they have two messages. One is, "This is a mess. We've got to do something about it. We have a really good air campaign. We're going to threaten to use air power in order to meet the demands that the UN Security Council has set. If necessary, we are going to implement a phased air campaign, and here's how that would look." The second message is, "Don't worry, we're not sending combat troops under any circumstances--not to implement an agreement, and not to force Milosevic to come to the table."

Why are they so adamant? First of all, Secretary Cohen is pretty adamant about this. But who is pushing it, and why are they so adamant about not sending combat troops?

Within the White House, the Pentagon, and the State Department is the underlying assumption that, if this policy ends in sending American troops to Kosovo, Congress won't support it. Therefore, we have to reassure Congress and the American people that, while we are seized with this policy, while we are willing to use force, we are drawing a clear red line. There will be no American combat troops in Kosovo under any circumstances. The attempt here is to reassure the Congress that we know our limits.

What happens is that the Congress says, "You guys don't know what you're talking about. You don't have a Plan B. You don't know what will happen if you use force from the air, and if Milosevic doesn't meet your demands, then you don't have a Plan B. If you want to go to war, go to war, but don't come to us and say we'll do some air strikes, but have no answer to the question of what if it doesn't work." That's the question that Sandy Berger had been asking right up to that point.

And that very week, there are important actions up on Capitol Hill. The House is actually beginning the voting to start impeachment of the president.

Clearly, when you get in a situation where the House and the Senate are considering impeaching the president, that's not a good time to go to war. This is especially so where the interests of the United States are not as clear-cut as they would be for instance, in the Persian Gulf, or if the United States or one of its allies was directly attacked. This is a question about interfering in the internal affairs of a state that we recognize to be internal affairs. There have been large-scale humanitarian atrocities, although not many people have been killed yet. It's a difficult issue for the president to consider the use of force, even without impeachment hanging over him. With impeachment hanging over him, it's nearly impossible to consider in a serious manner.

Berger seems to buy in, and the question isn't answered. As you said, he keeps raising the question "What if?" The question isn't answered. But Albright seems to win the day, and push for the bombing. What happens to Berger here?

When they agree to push for the threat of bombing, they genuinely believe that threats will work, and that it will be sufficient to get Milosevic to agree to the demands set by the UN Security Council. They believe if they send Dick Holbrooke down to Belgrade, he will find a deal where Milosevic agrees.

Remember, what the UN demands is not very much. It asks for a cease-fire. Winter is not a very good time to fight in Kosovo. It asks for a withdrawal of troops down to the level of February. Nobody really knows how many troops were there in February. But it's about 15,000-20,000--not an insignificant number in a province the size of Delaware. It asks for granting assistance to the international agencies to help and assist the people and refugees there. Milosevic should have no problem with the international community paying for the upkeep of refugees and people in Kosovo. It's not that they're demanding that much, and there is a general belief that threatening force, and demonstrating that NATO is united, is more than sufficient to get him to sign up to this. So there's really no need to consider what would happen if, in fact, you had to bomb.

Holbrooke does go over, and he does get a deal with Milosevic. You are brought over to the White House, and you're briefed on the deal. Tell me about that.

As often happens when the deal is set, the White House or others call in experts, who they believe to have opinions and expertise to share with the White House. And, more importantly, they are people who happen to be on television or who get quoted in the newspapers, who may be able to help explain what is going on. These are opinion leaders, and they help shape the public opinion. So the White House invites people like me, at times, to come over, and to listen to the deals that are being addressed there, and to carry the message forward if we agree with it. We also provide input to the kinds of deals that are being struck, and the kinds of policies that are being developed.

So you hear about the deal. They're briefing you about the deal Holbrooke got. Tell me about your reaction.

My reaction to the briefing I got at the White House was the same as my reaction when I first heard about the deal. I thought it was a bad deal. It was a deal that might have momentarily solved a pending humanitarian crisis, which was real. But the solution did so in a way that was certain to have this conflict reappear in the spring. We were about to send in 2,000 unarmed observers, in the hope that they could prevent the kind of atrocities that had been happening before. I didn't believe then, nor believe now, that that was very effective.

What did you tell them?

I told them, as I had said publicly, that if you send in unarmed observers, you are sending in de facto hostages. Once you send in hostages, NATO wouldn't be able to fulfill the threat to bomb if Milosevic violated the agreement he said he would comply with. You have taken away the one threat that you had that could ensure the implementation of this agreement.

How does Milosevic read this? We're not going to send ground troops--and we will send monitors?

Milosevic reads this the same way he's been reading it all along. We have a Christmas warning, which we don't implement, because we've all of a sudden forgotten about it. We have NATO planning on the question of using force, and we have a US veto on the consideration of ground troops. We have a threat of air strikes in order to get him to agree to the UN demands, and then the agreement he signs doesn't have any details of how to agree with those UN demands. He signs up to those details and immediately violates every single one of them. Nothing happens. So the way Milosevic reads it is that the international community is concerned about what happens in Kosovo, is willing to ruffle some feathers, but they're not going to use force against him.

The Serbs did, to some extent, comply initially with this. They did pull their troops back, and I've talked to people about this--they did put them in garrison. The KLA didn't necessarily play by the rules there.

Here you had a very interesting negotiation. You had a negotiation between an international mediator, Mr. Holbrooke, and one party, to a conflict that has two parties. We made a deal with Milosevic. We made no deal with the KLA. We never had the KLA sign on to a cease-fire. We never had them sign on to behaving in a certain way. We didn't even deal with them. So the reaction of the KLA was, hey, these troops are pulling back, great, we're pulling in. This was very predictable. Anybody who had any sense about this conflict knew that every checkpoint abandoned by the Serbs would be taken over by the KLA. Every village in which Serb troops left would be taken over by the KLA. Of course, that happened. The KLA is in here for a purpose. This is not a game to them--they want to win. This is independence for them. They want to get all of the Serbs out, and they will use any possible means to do so. In fact, the NATO air threat becomes their ally. They move in, take over checkpoints, and take the villages. At some point, they're going to provoke the Serbs into some kind of reaction. They're going to provoke the Serbs into doing things that violate the agreement that Holbrooke, Clark, and Naumann have negotiated with Milosevic. NATO has promised to then use air power. So the KLA has every incentive to provoke the Serbs into the kind of reactions that we see happening by December of 1998, in the hope that that would finally bring about NATO air power. That's the situation. Everybody knows it's the situation, and what do we do? We just hope it doesn't happen.

In mid-January, it's clear that there are big problems. The agreement has fallen apart. There is a lot of fighting, and concern about Milosevic shredding the agreement. There is a meeting at the White House on January 15. Tell me about that meeting. Who's there, why did they come together, and what are they looking at?

By Christmas, it's clear that the agreement Holbrooke negotiated is unraveling. Serb troops are starting to go back in numbers far exceeding what they were allowed to be. Serb troops that were supposed to be in garrisons are out there. There is fighting all across Kosovo, particularly in the north and east. The KLA has taken advantage of the Serb withdrawal, and re-established major positions. This thing is about to explode. Everybody knows it's about to explode. A meeting is called in the White House to ask what are we going to do--what is our future policy?

A policy paper called October Plus basically re-affirms the fundamental nature of the Holbrooke-Milosevic agreement. October Plus strengthens it a little bit, re-invigorates the shuttle that we're doing between the Albanians and Milosevic to get a political agreement. It sends some bodyguards to some of the OSCE monitors down there, and a variety of other means to set this up. There's general agreement within the government that that's what we're going to do. Madeleine Albright comes to the meeting and says, "No. You know, this is not working. This agreement is unraveling. The Serbs are not complying. The shuttle diplomacy is just not going to work. There is no way the Kosovars and the Serbs are ever going to agree to anything under these kinds of circumstances, and we basically have three options. We can say that this is a conflict we can't solve. We can pull back. We can say that this is just too hard, and we're not going to do this. We can continue to muddle through, knowing full well that come April, this thing is going to explode. Or we can take decisive action. We can launch air strikes. We can threaten air strikes and see if we can get a deal. What's the decision?" The decision by the principals is, "No, we'll muddle through. We'll go with October Plus. We can't leave our commitment. Our credibility is at stake here, and really can't leave. But decisive action--we just can't stomach it."

After all, on January 15, the Senate just started a week before a trial of the president, who was impeached by the House on December 19. This is no time to go to war. Therefore, the decision by the principals on January 15 is to continue with October Plus, muddle through, and find a way to solve this problem down the road.

That day is a massacre at Racak. Four days later, Sandy Berger pulls everybody back in to the White House. On January 19, there is a meeting. What's it about, and what are the dynamics there? What do they decide?

Just as the principals meet on January 15 in the White House, Serb forces are engaged in yet another massacre of 45 people in Racak. That's on the front page of the newspaper. The next day, the OSCE monitors are there on the scene, and report what happened. The head of the OSCE monitors says this is a crime against humanity. We again have pictures of bodies, of heads torn off, of torsos. Within four days, there is an immediate agreement in the White House that the very option that Madeleine Albright put on the table four days earlier, which was not acceptable, now becomes acceptable. We have to have decisive action. Muddling through is no longer possible. We can't postpone the moment of deciding what to do. On January 19, the decision is to have decisive action, start threatening force, and to do that to get an agreement.

Who is pushing for that? Who is on the fence, and who is opposed to it?

By January 19, Albright is the leading advocate of what becomes the Rambouillet strategy. On the fence, to a certain extent, is Sandy Berger. But he realizes that the politics are moving in the direction that Albright wants to go. On the fence, and opposed to this move, but also knowing that the politics are moving in the opposite direction, is the Pentagon. At that stage, the Pentagon still says they're still not convinced that we should send troops to implement a possible agreement as Albright is proposing. But in the end, everybody signs off on the strategy, and goes ahead with it.

The NFC staff is putting together the memorandum to the president, laying out the recommendation of the principals to threaten serious use of force. At that time, the president is travelling in his limousine up to Capitol Hill to give a State of the Union address at the moment of the highest political crisis of his presidency. The House impeached him a month before. The Senate is in the midst of a trial to remove him from office. Here is a president standing in front of the country and in front of all those people who voted against him, who impeached him, and who were about to vote against him in the Senate. He has to stand there and make a speech about the state of the union, and also about deciding whether to use serious force with regard to Kosovo. It's not an easy situation for anybody to be in.

On January 21, the president and Prime Minister Blair have a conversation about this use of force. What are they concerned about? What do they talk about, and what's their big concern about having to drop bombs?

The big concern in Clinton and Blair's conversation on January 21 is whether the threat of air power will get Milosevic to deal. There is no certainty about that. But if air power has to be used, where does it end? Are we going to send in ground troops to implement an agreement that we may achieve because of the threat or use of air power? Are we going to send invading ground troops to bring this about? It's very clear that, at that point, Clinton rules that out, and Blair does so as well.

From the beginning, Tony Blair is very, very clear about what needs to be done. This is a conflict that will not be solved without the use of force. That's been the underlying assumption of British policy from day one. He's clear that, at some point, use of force will have to include the use of ground troops. You don't fight wars without ground forces, and you don't implement agreements without combat troops on the ground. From about July, 1998, and onwards, that is the leitmotif of British policy. British policy from that moment on is unwavering with regard to the threat of force and the use of force. What Britain wants to do is to lead the effort in a forceful way. It doesn't want to do what it did in Bosnia, in which British policy was to wait, to halt, and to put a brake on any development that might lead increasing use of force. In this case, Britain was committed. There was a humanitarian crisis in the midst of Europe that was unacceptable, that had to be stopped. Milosevic was a bully who had to be stopped. Britain was unwavering in its support within the Cabinet, and from that point on, throughout this entire period.

Blair has a special relationship with Clinton. How does that play out?

Blair has a special relationship with Clinton in the sense that they are two peas from the same pod. They are both centralists who used to be on the left--they have this third way. They are deeply politically inclined, they are interested in politics, they are interested in each other's politics, and they are clearly close friends. Blair's role becomes to buck up Clinton, not only personally in terms of taking the necessary steps, but to buck him up against his advisers. They're constantly telling him, "Mr. President, we can't really do this. This is too much. The Congress won't do it," and Blair says, "Mr. President, this is the right thing to do."

But the president has still not committed on ground troops, even as peacekeepers. They are saying, "We are not making a decision on that until February 13." The president gives a radio address; he indicates a commitment of troops. Tell me about the radio address, and any connection to the events of the day before, February 12.

Policy in Kosovo goes through a very strange period in mid-January to mid-February. In the Senate, day in and day out, senators are debating whether or not to remove President Clinton from office on the impeachment recommendation of the House. At the same time, the United States is pushing a policy on the allies, on the Russians, and on everybody else that says we need to get an interim agreement on autonomy in Kosovo, and the way to get that agreement is to be able to threaten to use force against the Serbs. Because there is a trial in the Senate, the United States is unwilling and unable to commit to providing ground troops, to help implement that agreement, until February 13, which is one day after the Senate votes and fails to remove President Clinton from office. Once the Lewinsky scandal ended, once the final political step in that torturous year-long process is over, the president feels able to commit to the deployment of ground troops. He wasn't able to commit beforehand.

So what's going on here? I've talked to people in the administration, who said they always knew that US troops would be a part of the ground force, that they knew it from day one. Yet we have here a president publicly saying no, we will not go, we will not go. The day after he is acquitted, boom . . .

They had privately always agreed that American troops would be part of a force to help implement an agreement to be reached at Rambouillet, but they had not said so publicly. The president was not going to say publicly that he was going to put American men and women in a very dangerous situation while he was still standing trial in the Senate over the Lewinsky scandal. It was only the day after he was acquitted from that trial, on February 13, that he publicly announced to do so.

But what message was sent? On the one hand, the United States may have known that this was all along going to be the case, but the allies didn't know. The Kosovars, who had to be reassured by the fact that the troops would be there, were not reassured by the delay. Milosevic reads this as hesitation by American policy towards its own strategy, towards its own ability to implement what it said it would implement. So Milosevic believed that you could push, and the United States would not follow through.

Let's talk about Rambouillet. The Albanians have some surprises when they come, and one of those surprises is who they pick as the leader of their delegation. Who did they pick, and why was that a surprise? What's the significance?

They pick a 29-year-old nobody called Hashim Thaci, who is a political leader in the KLA, rather than their own elected president, Ibrahim Rugova. The surprise here even for the United States, which had put together this whole delegation, is that Thaci becomes the leader of the game. It signals that the Albanians in Rambouillet will not be as easy to maneuver into the situation that we want as we had thought. Thaci was close to the hard-liners in the KLA. He'd talk over the mobile phones that they all carried with those hard-liners day in and day out, to make sure that Kosovo was not going to be sold down the tubes for a success of American diplomacy at Rambouillet.

Secretary Albright decides to weigh in on this, as the Albanians are struggling with this. Why did she go, and what was the significance of her arrival? She comes down, her plane lands, she's got the black Stetson hat on, and she comes down those steps. Tell me about the significance of that, of her landing and her coming down.

Albright's coming to Rambouillet is the signal that now the final push for a Kosovo agreement in the Rambouillet courts is really going to happen. Albright is convinced, and has been told by her aides, that just a bit of a push, and the Kosovars will sign on. That's part of the strategy--get the Kosovars to sign on, get the Serbs to renege, bomb the Serbs, get the Serbs to sign on, deal--that's the strategy. Albright gets down there, convinced by her aides that the force of her personality, that the fact that she is secretary of state of the largest, most powerful country in the world, is more than sufficient to get a deal.

In the end she succeeds. But the Kosovars tell her, "We agree with this in principle. We're willing to sign in two weeks, but we're not going to sign now." That is truly a very big blow to her prestige, because she has put her prestige on the line to get the Kosovars to sign on, and it doesn't work.

Ultimately it worked. Even if it was unflattering, it worked. Albright got her strategy.

Even though she fails in Rambouillet to get the Kosovars to sign on, the strategy works, because the Kosovars, albeit with three weeks delay, sign on the dotted line on March 18, and at that point, all the ducks are in a row. The Kosovars have signed. NATO has issued a threat to use force if the Serbs don't sign. The Serbs don't sign; in fact, the Serbs have been using the time to increase their forces to be ready for whatever it is that they're ready for at that point. NATO is ready. So the strategy by March 24, the day the bombing starts, has worked. The strategy was to get the Kosovars to sign on, and if the Serbs failed to sign, to get NATO united behind a phased air campaign.

Again, the White House calls some outside experts together in order to get briefed on what is happening at Rambouillet. You get a visit to the West Wing of the White House. Take me there.

A group of us from the outside are invited to talk to some senior administration officials in the West Wing of the White House, who are laying out what is happening in Rambouillet, and what they think needs to be done to get to a successful conclusion. Many of us are concerned about what happens if it fails. Some of us raised the issue that, that if you're going to go in and bomb, that you are also prepared to follow that up with ground troops. What happens if the bombing doesn't work, are you prepared? And the answer is actually quite interesting. On the one hand, the answer we get from the administration is that nobody thinks that a bit of bombing is likely to get Milosevic back to the table, but at the very least our bombing will severely degrade his capacity to do any harm inside of Kosovo, so we don't really have to worry. In essence, NATO bombing would become the air force of the KLA.

If the bombing doesn't work in getting him to the table, we were going to bomb his forces so that he couldn't do any damage inside Kosovo. We weren't going to send any ground troops in because Congress would not support it, or so we were told.

Our strategy is to bomb, to bomb, and to bomb, in the hope that at some point he will give up. We also, at that point, are pretty much convinced that bombing is sufficient to take care of the Serb armor, their artillery, and the Serb paramilitary forces, to prevent them from doing the kind of widescale slaughter and atrocities that they would engage in.

On March 24, the president is going to tell America that we're about to bomb Milosevic. You get a phone call that afternoon from the White House. Tell me what happens--where are you, what time is it, and what is the call about?

In the early afternoon, I get a phone call from people in the White House, telling me that the president will go on television and that this is real, and they hope that I can support the United States in its actions. I said, "Of course. I've been calling for this for a year now, so I'm glad it's finally happening. Don't worry about it. What is he going to say about ground troops?" The answer is, "We're going to say we have no plans to put in ground troops." I said, "You can't say that, because if we don't have any plans to use ground troops, we ought to fire the person who is responsible for drawing them up. So either you don't have plans and you're incompetent, or you're lying, so you can't say that. How about saying something like 'intention'--we have no intention to use ground troops?" One of my biggest mistakes is to give them the kind of words that they would, in fact, use, and I would criticize them for it within days.

Why weren't we more prepared to go to war?

There was a widespread conviction on the part of civilian and military leaders that bombing would either get Milosevic to back off, or get him to the table. That was the lesson we thought we had learned from Bosnia. It was a lesson that we had inculcated ourselves, with regard to what we believed Milosevic was all about. He was the kind of bully who, if you hit him across the head, he'll come back and do what you want him to do. That was the conviction. So the details of the military plan were less important than the fact that you were willing, and demonstrated that you were willing, to use force, because that would get the bully to come to your side. It was the assumption of the secretary of state, of the national security adviser, of the NATO commander, of the president, and of everybody else. The guiding assumption of using force was that, by doing so, you would demonstrate that you were serious, and that was sufficient to get Milosevic to back off.

We don't want to go to war. In fact, we're not at war--we're just using force in the service of diplomacy in order to change his mind. As Wes Clark would put it, what we are trying to do is to get Milosevic to back off, to get out of Kosovo, to allow the interim agreement that was negotiated at Rambouillet to be implemented--that's what it was all about.

You've said that we need to think about this war as having two phases. Explain to me briefly what you mean by that.

This is a war that has two very big phases. The first we lose, the second we win. In the first four months of this war, the Serbs have the upper hand in two very important instances. One, strategically, they know what they want. We don't. They want to get the Albanians, as many as possible, out of Kosovo. We want to get him to the table and we don't really know how we're going to get Milosevic to the table. He wins, we lose. He has achieved his objective inside Kosovo by kicking the Albanians out, by slaughtering many thousands of them.

It's only after about a month that we realize that we can't lose this war. Politically we can't lose the war, militarily we can't lose this war, and strategically we can't lose this war. If we lose this war, NATO is ended, and the credibility of American foreign policy is at an end. We put into place a strategy that was designed to reverse what has happened inside Kosovo. Our goals by April, as we said, are no longer to prevent him from conducting massive atrocities against the Albanians or to degrade his capacity to do so. Now our goal becomes to have refugees return. Those people who were still inside Kosovo when we started bombing are now gone, so our goal becomes Serbs out, NATO in, refugees back. Those are our three goals, and we put into place a strategy designed to achieve those goals, which we do by June 10.

Did the bombing create a refugee crisis?

No, Mr. Milosevic created a refugee crisis. Mr. Milosevic's forces were sent in, village by village, to kick the Albanians out, to murder, rape and torture those who remained. He was responsible for it. The bombing took the lid off. It allowed Milosevic to conduct this kind of operation without having to fear that NATO would intervene. Before March 24, he feared that NATO would intervene. Once NATO had intervened, it was his incentive to go in and do it as quickly as possible, and then present the United States and NATO with a fait accompli: "I've cleansed the entire Kosovo. Now what are you going to do about it?"

He got the first part right--he got his job done.

He got that job done. And the question then became for the United States and NATO, how do we reverse this? How do we make clear that we're not going to accept this fait accompli? We have to reverse it. A new strategy was put into place, the strategy of victory, which is designed to reverse what had happened in the first month of the war.

Let's move to the ground troops, because that's part of it, and the push for ground troops. Here we are at this critical moment, towards the end of April, and I want to know what people are thinking about ground troops. Sandy Berger--what's he thinking?

Sandy Berger is convinced, and remains so to the end, that a bombing campaign with a strong diplomatic strategy will eventually convince Milosevic that the time is up, to give in, that NATO has to go in, that the Serbs have to go out, that the refugees have to come back.

President Clinton has a conversation before the NATO summit with Prime Minister Blair. By this time, Blair is convinced that the only way the war can really be won is to prepare for a ground invasion, and if necessary, to put those troops in, to take Kosovo back. By late April, President Clinton, having listened to Blair, comes to the realization that he was to win this war, that he can't lose this thing. NATO will be in bad shape, and the credibility of American foreign policy will be completely in tatters, so has to win this war. He comes to the realization by late April or early May that he has to do what it takes, and if that means ground troops, then he's willing to go ahead.

Let me ask you to set that meeting up. It's an important meeting on the eve of the summit. Blair arrives in Washington. He goes straight to the White House for a long dinner meeting with the president. What did they talk about, and what was the significance?

Blair comes to Washington a day early for the NATO summit and talks to the president in a very small group, with one mission. The mission for him is to get the president to consider the use of ground troops if necessary, and therefore to start planning a process that would be necessary to prepare for a possible ground invasion. The president's primary concern on that evening is not to say that Tony Blair is wrong. His primary concern is to have the issue of ground troops not be part of the summit that will take place the next day.

The deal that Clinton and Blair strike in that dinner the night before the NATO summit is that Blair will not bring up the issue of ground troops during the next couple of days. NATO unity is the key, and finding an agreement to intensify the air campaign is what Clinton wants. In return, the president agrees to start very quietly planning preparations between the United States and the UK for a ground invasion.

How big a turning point is that summit? At the end, everybody seems to get what they want. For people who are worried about ground troops, it's not out there publicly. People who want ground troops, well, we're starting to work on it. It seems like everybody walks away getting a little of something.

I think the summit is the turning point in the war. It is the reaffirmation of the strategy that NATO finally puts in place only in late April--get the Serbs out, get NATO in, get the refugees back. It reaffirms that the fundamental goal is to reverse what has happened in Kosovo, and there is a unity of purpose among the 19 members to that effect. There is no real agreement about how you're going to do that, but there's agreement about the goal, and we're not going to change. The other thing that the NATO summit does is that, by the end of the summit, by the end of April, the United States is committed to a three-track strategy to achieve the goal. One is the air campaign, an intensification of strategic bombing. Second is the planning, and if necessary, moving ahead to prepare for using ground forces. Third is a diplomatic strategy that engages the Russians directly as a means towards an end, to have the Russians carry basically our water in Belgrade, so that we can get a deal.

On May 19, General Clark travels from Brussels and goes to the Pentagon. He goes where the joint chiefs of staff meet, and briefs the joint chiefs on two plans he has. One is a plan to invade and prepare for an invasion of Kosovo--175,000 troops, and what do we take on the American side, who would have to contribute on the allied side to get this done. The chiefs are highly pessimistic and dismissive of this, and worried that he is moving in this direction. The second is an idea to increase the number of troops inside of Macedonia and Albania--ostensibly to provide a rapid capacity to move in to Kosovo if there is an agreement if Milosevic gives in, and to establish what becomes the Kosovo force. But it's also to provide a basis for augmenting the force later on to prepare for an invasion, in order to cut down the time it would take to prepare for such an invasion. And while the chiefs are clearly against the notion that we should go in forcefully, they buy the logic of increasing the KFOR enabling force from about 25,000 at that point, to about 45,000-50,000 later on.

What are the chiefs so worried about here? Why are they so concerned about the idea of ground troops?

The chiefs don't want to go to war over this issue. They don't want to have a ground war. They think it's going to be bloody, and it's going to be bad. It's not an issue that they generally support the use of force over. They were reluctant participants in the entire process and they were not going to be supportive of a ground campaign, in general, and particularly in the army. The air force believed that you could do it from the air. There was a large degree of opposition to this.

On June 2, you're invited to a meeting with Sandy Berger. Set that up for me. Where are you going, and why, and who are you meeting with?

On June 2, a meeting is called by the White House to meet with Sandy Berger in the Roosevelt Room with a number of people who have been critical of the administration with regard to its strategy on Kosovo. Most of the people who were there were people who had called for using ground forces, or at least preparing for ground forces much earlier.

Berger made what I thought was an excellent presentation on where we were and where we were going to go. He said there were four points to remember. One, NATO is going to prevail. Two, prevailing means exactly what we have said it means, Serbs out, NATO in, refugees back. Three, the air campaign is working. We're inflicting a lot of damage on Serb forces and on the Serbs themselves. We have indications that things are moving in the right direction. But four, all options are on the table, go back to one, we will win. As he explained, that meant that ground forces were an issue seriously under consideration. NATO and the United States were planning for all kinds of options, and if necessary, ground forces would be part of the equation. We were going to win, no matter what it took.

That was the kind of message that nobody had said publicly up to that time in those kind of terms. It had not been clear to many of us on the outside that the administration was fully and totally and 100% committed to winning, even if that meant putting in ground forces. The message that Berger left on June 2 was that we will, if necessary, use ground forces. We don't think it's necessary yet, but we will if necessary.

Two and a half months into a war, and the security adviser is basically saying we're committed to win. That's the message he has to portray.

They never went to war. It took quite a long time for the administration to recoup both the trust and the confidence of those of us who had called for a strategy of victory, a strategy that would guarantee victory and winning, no matter what it took. It takes a long time, because they weren't committed in the beginning. They were committed to a strategy of changing Milosevic's mind. That's not a strategy of victory. They weren't committed to doing everything necessary to achieve it.

We've talked to a number of people. The allies and a number of people outside were concerned whether this president, and this administration, would hang in there.

Absolutely. There was a big concern whether the administration would remain committed to achieving the goals that NATO had set, and there were many people inside the administration arguing that they didn't think that it was going to happen. Some of the allies didn't believe that the administration would, and therefore were not willing themselves to put forward the kind of policies that would have been necessary to achieve this. And many on the outside were not convinced that the administration had the will to push this through. The signs come in May, and then by early June, that the administration is committed. That's important.

What does that say about leadership, though? Two and a half months into a war, you have to reassure your allies and those here that you are actually committed to the strategy.

The problem was that the strategy we adopted in going into this conflict was not a strategy that was designed to achieve our political objectives militarily. It was a strategy designed to somehow convince somebody that we were committed to something we were not committed to. We were not committed to winning this war. We were not even committed to war. We never engaged in what was called a war. We were engaged in the use of force in order to support our political objectives. In the end, even General Clark still calls this coercive diplomacy, rather than war. That was a fundamental misapprehension. What we were engaged in once the first bomb was dropped was a war, and you cannot lose wars. It took a while for the administration to realize that, one, we were in a war and two, and we couldn't lose it.

June 1. Why did Milosevic capitulate?

In the end, he capitulated because the strategy that was amorphously put together around late April worked. The intensification of strategic bombing forced Milosevic down a corridor with a dead end, but there were two doors through which he would hope to escape. One was a political door, in which the Russians were going to come to his rescue politically--preventing a political deal would have prevented the UN from coming aboard. That door was shut down on June 3, when the Russians joined the United States and the European Union on the same NATO demands and said, you've got to deliver. The other door was a military door--the hope that if he just sat down and waited and waited and waited, NATO would fall apart, and there wouldn't be any ground invasion. By May, and certainly by early June, the signs were all there that, if necessary, NATO was going to go in on the ground, so that door was closed too. Now he was in a corridor that had a dead end. The moment he realized that, which is the moment the Russians closed the political door, he said, "Okay, I've had enough. Whatever you want, as long as I stay in power, I'll quit," and he did.

So we're right there, at the last possible moment, and we get word that he's going to cave.

Absolutely.

The president doesn't have to make that tough political call. This guy survives scandal, survives the Republicans who end up losing the battle on the Hill, takes hits for it, and all of a sudden he gets off again. Lucky.

This guy is the luckiest president we've ever had. That said, I believe that Bill Clinton in time became convinced that he had to win this war. He would have done whatever was necessary to win the war, and he was leading his administration in a way that the commander in chief must lead his administration. In the month of May and into June, he went right down the path of making all the necessary decisions. Sure, he was trying to postpone those decisions--he didn't want to decide about ground forces for as long as possible--but in the end, he would have made that decision. He would have made the right one, and we would have won in any case.

Who won here--NATO, Milosevic or the KLA?

In essence, NATO, Milosevic and the KLA won. Milosevic stays in power, which is what he cares about most. NATO won, because it showed that if it uses force, it can do so for a sustained period of time, and can achieve the objectives that ultimately they agreed upon. And the KLA won because they basically run Kosovo today.

How dicey is all of this? We're not going to war. We said we don't plan to go to war. The next thing we know, we've got a refugee crisis that nearly topples the Macedonian government, which is what we are trying to avoid in the first place. Then we have this little problem with the Chinese embassy and those relations. The Russians go off to Pristina airport, and we're almost going to take them on. We're just days from making a decision. We're potentially months from fighting a bloody ground war. How dicey is this?

When you make a strategy of war on the fly, you are likely to have all kinds of problems. You may find that Macedonia is about to be overrun and destabilized by refugees when you only were prepared for 25,000 refugees. You may find that your relationship with China goes down the toilet, because unfortunately you happen to bomb their embassy. You may find that you have to go in with ground troops and start a major war in the middle of Europe, almost on the scale of the Gulf war. If you don't know what it is that you want, and you don't know how you're going to get what you want, you're going to make it up on the fly. You increase the risk of failure, and increase the stakes necessary for success. It's not a way to go to war.

General Short's notion is that we had the wrong strategy, and that he knew what the right strategy was, if only he had been allowed to do what he wanted to do. This ignores the fact that this was a war conducted by an alliance, and ignores the fact that the strategy that he wanted to implement would never have received approval in the White House, which was convinced that a bit of bombing was more than enough to do what needed to be done.

He may well have been wrong. . . . How does he know what makes Mr. Milosevic tick, and how does he know at what point would have caved? He doesn't even know why Milosevic caved in the end. He has the hubris to suggest that he knew the right strategy when others didn't. That's just wrong. The only way in which you could guarantee success is to have had a military strategy where you could deny militarily what Milosevic was

seeking--and you do that with ground forces--not with air power.

There's the idea that, if they bombed more heavily, they could have shortened the war. It would have happened quicker, there would have been fewer problems, less misery, and they could have at least put the pressure on. Ultimately they got there. But why not do it from the start?

If we would have bombed downtown Belgrade the way that General Short would have done, the question is, would that have changed Milosevic's mind? Because if it hadn't changed Milosevic's mind, the forces on the ground in Kosovo would have done exactly what they did. So the issue here is, how do you change Milosevic's mind with air power? My suggestion is that General Short may think he knows, but he doesn't necessarily have any evidence for knowing how you change Milosevic's mind. You can only make that case if you believe that, in the end, the bombing of the electricity was what changed Milosevic's mind. He doesn't know the center of gravity of the Serb forces any more than I do. He doesn't know that Milosevic could have done what he did inside Kosovo, even if we'd hit the bridges in downtown Belgrade on day one.