Why Did We Bomb Belgrade?

Why did we bomb Belgrade?

MICHAEL MCCGWIRE

In principle, the question was answered by our political leaders in the days following the start of NATO’s air campaign against Serbia. In practice, the answers raised new questions as to why these intelligent men were saying such things. Using similar words in various formulations, George Robertson and Robin Cook explained (repeatedly) that the political objective was to avert a humanitarian disaster in Kosovo and/or to prevent a crisis from becoming a catastrophe. This was to be achieved by strategic and precision bombing of military targets (initially in Serbia) in order to reduce the capability of Serb forces to: continue with their violence; repress the Kosovar Albanians; order ethnic cleansing. There was an obvious disjunction between the stated military objective of degrading Serbia’s military capability (a slow process) and the immediate political objective of halting the forced expulsions and associated killings in Kosovo. Indeed, bombing Belgrade seemed likely to inflame the situation, and made sense only if one believed that this demonstration of NATO resolve would cause Slobodan Milosevic to halt the process himself. That did not happen. We were then told—despite events in Bosnia and Croatia—that no one could have foreseen that Milosevic could have been so wicked. The continued bombing was justified by describing what was happen- ing in Kosovo (which was terrible enough) using exaggerated and emotive lan- guage, including talk of genocide which, in common parlance, clearly did not apply. Meanwhile, our leaders continued to demonize Milosevic. In the past, demonizing has been used to justify offensive military action that in other circumstances might be questionable. Abdul Nasser (whom Britain likened to Hitler) at the time of the Suez crisis is one such example; Muammar Qadhafi as ruler of Libya is another. So what exactly was afoot? Was this a punishment beating in the Balkans, where NATO, dissatisfied over UN ineffectiveness, was taking the law into its own hands? Or was there something more to the whole affair? This article is divided into two parts. The first is descriptive, reviewing the situation through to the end of 1998 and then summarizing events during the first six months of 1999. The second part revisits the evidence, following up

International Affairs 76,  () ‒

76.1/01.Mccgwire 1 20/12/99, 1:30 pm Michael MccGwire

various anomalies to construct a story that is more plausible (and in many ways more laudable) than what we were told at the time.

What happened? Disintegration Yugoslavia was one of the states most severely affected by the foreign debt crisis in the 1980s, but its special role in NATO strategy had ensured continuing access to Western credits. However, the winding down of the Cold War removed this leverage and exposed Yugoslavia to the full rigours of IMF conditionality. At the same time, Yugoslavia lost its niche position in the Cold War global economy and found itself in competition with the newly inde- pendent states of central Europe for Western trade and favours. The resultant economic austerity and budgetary conflict was hugely divisive and placed heavy strains on the complex political and socio-economic structure of the federal state, accentuating nationalistic tendencies which, for different reasons, were on the rise in Serbia, Slovenia and Croatia. These strains and the economic attractions of independence for Croatia and Slovenia led inexorably to the disintegration of the federal state. Yugoslavia was a federation of the Southern Slav peoples, and its political structure combined aspects of a union between sovereign territories with established borders, and a union between sovereign peoples, whose members could be living anywhere within the Federation. The right to secede was unclear and contested. The West (particularly the United States) took the view that the former republics were states-in-waiting and treated claims on their territory as international aggression. For Serbia and Croatia, which had significant minorities living outside their borders, such claims involved legitimate disputes about how to divide up a failed state. This applied particularly to Bosnia-Herzegovina, which comprised three national communities. None had an overall majority, and the Serb minority strongly opposed the move to declare independence from the Federation. In other words, the West was oversimplifying when it identified Serb aggression as the root cause of the Bosnian conflict; even the label ‘multi-ethnic civil war’ missed the full complexity of what followed the failure of the Yugoslav state, about which the West had ample warning. And, whatever the merits of individual cases, there was little to choose between the different factions and their leaders. Each was adept at manipulating the aid process and the media, and made whatever gains that opportunity allowed, the level of brutality being largely determined by relative capabilities and the situation on the ground. Territories and their people were both the pieces and the board for this ruthless game. Although all sides were guilty of atrocities, up to July 1995 (and including events at Srebrenica), the Bosnian Serbs had committed the over- whelming majority. However, in August 1995, the US-supported Croatian army, on the direct authority of the Croatian President Franjo Tudjman, drove

76.1/01.Mccgwire 2 20/12/99, 1:30 pm Why did we bomb Belgrade?

Serb forces from the Krajina and ‘cleansed’ the area of its 200,000 inhabitants. This was by far the largest expulsion up to that point, and was achieved in four days. The Croats started by shelling villages and towns to create panic and a disorganized mass exit; this was followed up by assault troops looting shops and dwellings, and then by the brutal paramilitaries. The Croatian modus operandi was so effective that the Serbs adopted the same approach when they occupied Kosovo in March 1999. By the time of the assault on the Krajina, the war had turned against the Bosnian Serbs and NATO was able to bomb with impunity. In 1993, Milosevic had been unsuccessful in pressurizing them to accept UN-brokered peace pro- posals; the Bosnian Serbs had felt there was a better deal to be had by continuing the war. In 1995, they finally accepted that it was time to negotiate, and Milosevic once again acted as the influential intermediary between his fellow Serbs and the West, leading ultimately to the Dayton Accord.

Kosovo By all accounts, Milosevic is a callous, ruthless, politically adept, power-hugging man. He can be socially engaging, but must never be trusted. The West has alternated between vilifying him and finding him indispensable. He did not create the Kosovo problem, but he did exploit it. The Albanians had resisted incorporation into Yugoslavia in 1918 and again in 1944. There was an inherent conflict between their long-established wish to unite Kosovo with Albania, and the emotional attachment by Serbs throughout Yugoslavia to Kosovo and its holy places. By 1961, Albanians already comprised 67 per cent of Kosovo’s population and the Serbs only 24 per cent; in 1991, the ethnic imbalance would be 90:10. In 1974 the federal constitution had been amended to make Kosovo an autonomous province of the Serbian republic. This went a long way towards satisfying the aspirations of the Albanian majority, but there were still those who pressed for full republican status for Kosovo and ultimate union in a Greater Albania. Following Tito’s death, there were widespread student-led riots and a period of federally imposed martial law. Meanwhile, there was growing political protest among the 200,000 Serbs about their subordinate status and the Albanians’ oppressive discrimination. Milosevic seized on this issue as a means of gaining the leadership of the Serbian Communist Party in 1987, which allowed him to appeal to a much wider constituency in Serbia, including anti-communists and right-wing nationalists. By the time of the ‘million-man’ celebration of the anniversary of the Battle of Kosovo in 1389, Milosevic had been elected President of the Serb Republic. More significant, in the multi-party elections held in December 1990 (the first in half a century), Milosevic’s newly formed Serbian Socialist Party won the support of 65 per cent of those voting (47 per cent of the full electorate). New laws disadvantaging the Albanians were introduced in 1989 and a con-

76.1/01.Mccgwire 3 20/12/99, 1:30 pm Michael MccGwire

stitutional amendment revoking Kosovo’s autonomy was passed in September 1990. The Kosovo Albanians responded by declaring independence and holding a clandestine referendum, which elected Ibrahim Rugova, leader of the Demo- cratic League of Kosovo (DLK) and a convinced Gandhian, as President. As part of their response to the ‘Serbianization’ of the province and what was seen as colonial rule from Belgrade, they set about organizing an alternative society with an improvised political structure, schools and a medical system. The Kosovo Liberation Army (KLA) emerged fully on the scene only in the wake of the Dayton Accord, when it argued that Rugova’s strategy of passive resistance had failed. Apart from some sporadic terrorist attacks, it was mainly occupied establishing the organizational structure to support a guerrilla war, but the situation was radically transformed when rioting in Albania led to the looting of state armouries. This resulted in the ready availability of a large number of Kalashnikovs at $10 each. In February 1998, following a particularly brutal Serbian reprisal, the Albanians in the Drenica area took up arms, leading to a major KLA offensive; by July they controlled more than 30 per cent of the province. Emboldened by its success, the KLA declared its goal to be a Greater Albania, banned all political parties in the territory it controlled, repressed various minorities (particularly Serbs and Romanies) and denounced Rugova, his DLK party and the Kosovo shadow parliament. No love was lost between the two factions. The KLA leadership came mainly from those who had been imprisoned for their political views and moved abroad on release. Rugova’s DLK included those who had worked for and with the provincial authorities, prior to autonomy being revoked. At the end of July 1998 the Serbs launched a counter-offensive, driving back the KLA and burning rebellious villages. The human cost of this operation was some 800 dead and perhaps 200,000 persons displaced. This cursory review indicates that while Milosevic exacerbated the situation in Kosovo, inflaming ethnic animosities and introducing discr- iminatory and repressive policies, the underlying problem was not of his making. There is a strong argument that the one possibility of defusing the situation in Kosovo lay with the Albanian majority after provincial autonomy was granted in 1974. They had the singular opportunity at that time of adopting policies that would reassure the dwindling Serb minority that its rights and interests were being fully observed and would continue to be respected in the future. But, as we know from Northern Ireland, that is not the way majorities behave. By 1987 the oppression of the Serb minority in Kosovo was a political issue ripe for picking, if not by Milosevic, then by one of the right-wing nationalist parties. It is relevant that Kosovo was first placed on the political agenda by Serbian academics; that the revocation of Kosovo’s autonomy had widespread support among the political class in Serbia; that the necessary constitutional amendment was made possible by a deal struck with Slovenia; and that Serb repression in Kosovo was not an issue during the anti- Milosevic protests in winter 1996–7.

76.1/01.Mccgwire 4 20/12/99, 1:30 pm Why did we bomb Belgrade?

Promoting Milosevic to arch-demon of the Balkans may have eased the West’s collective conscience about its inept performance in 1985–95, but he was just one player in the Balkan tragedy, a play in four acts with a cast of corrupt and ruthless demons (‘gangsters wearing suits and ties’) spawned by the disintegration of Yugoslavia. It is fanciful to think that General Ratko Mladic and his nominal political boss Radovan Karadzic were merely puppets dancing to Milosevic’s strings. As for the President of Croatia, the arch-nationalist Franjo Tudjman, he was an equally, if not more, unpleasant character—as witnessed by the ‘cleansing’ of Krajina on his orders, described above. Demonizing Milosevic diverted attention from the intractable nature of the Kosovo problem and Serbia’s legitimate interests as the sovereign power. Kosovo comprised elements of anti-colonial nationalism (Kenya, Southern Rhodesia), gathering in the ‘greater’ Albanian nation (Serbia, Croatia, Ireland) and, in terms of sheer animosity, the Arab–Israeli conflict, with its contested historical claims for the same territory. These elements were interlaced with the competing demands of human rights, internal security and the security of the Serbian state, which faced the classic conundrum of how to suppress the insurgents without acting as their recruiting sergeant. During the first half of the 1990s the threat could be easily contained, thanks to Rugova’s policy of passive resistance. But following Dayton and the sub- sequent availability of Kalashnikovs, there emerged a qualitatively new threat of Albanian-based guerrillas enjoying local support inside Kosovo, which was inflamed by the brutality of the Serb response.

The road to Rambouillet On 27 December 1992 George Bush warned Milosevic that if Serbia began a war in Kosovo, the United States would consider it a direct threat to US national interests and would be obliged to act. This warning was reiterated by Bill Clinton on taking office in January and amplified in February by his Secre- tary of State, who expressed the fear that if Serbian influence extended into Kosovo and Macedonia, it would draw into the fray countries such as Albania, Greece and Turkey. The ‘Christmas message’ reflected the view in Washington that the root cause of the Yugoslav conflict was territorial aggression by Serbia, which was assumed to be an old-fashioned expansionist state led by a dictator. This made it easy to assume that, having been prevented from taking over Slovenia, Croatia and Bosnia, Milosevic would turn his eyes to Macedonia. At that time, human rights were not the primary issue. The failure to consider Kosovo’s future status at Dayton is partly explained by the fact that any viable solution to the Kosovo problem would create a dan- gerous precedent that could destabilize the fragile inter-ethnic modus vivendi in the southern Balkans. That failure was bound to discredit Rugova’s policy of non-violent resistance and embolden those who advocated liberating Kosovo

76.1/01.Mccgwire 5 20/12/99, 1:30 pm Michael MccGwire

by resort to arms; but on their past showing, and particularly after the implosion of Albania in 1997, they were not deemed to pose a serious threat to political stability in the province. This assessment was overturned by the success of the KLA’s impromptu offensive, and its taking control of more than one-third of the province. Northern Ireland is comparable in size to Kosovo; this was equivalent to the IRA seizing military control of Counties Fermanagh and Tyrone. It is widely believed that the United States gave Milosevic the go-ahead to launch a counter-offensive to put the KLA back in its box. Whether this is true or not, it makes absolute sense in terms of the policy concerns that prompted the ‘Christmas message’. In 1998 the KLA was still classified by the State Department as a terrorist organization, and its publicly stated goal was a Greater Albania which, if achieved, would unstitch the complex patchwork of the southern Balkans. In May 1998, at the height of the KLA offensive, US envoys persuaded Milo- sevic to meet Rugova, and by the beginning of September the two sides had agreed to talk about a three-year ‘interim’ solution involving a degree of Kosovar autonomy. Meanwhile, a new player had joined the cast: the Armed Force of the Republic of Kosovo (FARK). This was unsurprising; not only did the KLA have irredentist aims, but it was difficult to know who was in charge. In contrast, FARK (which was said to rely on Saudi money and Turkish logistic support) was directly responsible to Rugova. The KLA rubbished Rugova’s proposal and fighting continued. Later that September, in the centre of Tirana, the KLA assassinated the man who had been tasked by Rugova to set up FARK. International concern about the conflict in Kosovo, continuing Serb repression, and the effect of winter on the Albanians who had taken to the hills to escape the counter-offensive, was expressed in UN Resolution 1199, and early in October the United States persuaded Milosevic to agree to a ceasefire. Under the threat of NATO bombing, he also agreed to accept that the conflict was now an international issue, to withdraw heavy weapons and equipment from Kosovo, to allow air reconnaissance over the territory, and to permit 2,000 unarmed international observers to monitor the ceasefire. Serb forces in Kosovo were to revert to their level prior to February, namely 22,000 (10,000 Ministry of Interior Police [MUP] and 12,000 army). As the Serb forces withdrew, so the KLA moved forward, often occupying the Serbs’ former positions. Prior to the KLA’s spring offensive, the great majority of Kosovars had supported Rugova’s conflict-avoidance strategy and were sceptical of the KLA’s pretensions. But having tasted ‘victory’ that summer (followed swiftly by ‘defeat’), the local populace switched to actively supporting the guerillas. In acknowledgement of this new reality, the United States started talking to the KLA in November, with two meetings in Kosovo and one in Switzerland; early in December, a further meeting took place in Washington. Meanwhile, renewed fighting had broken out. The KLA were applying a lesson learned from the Bosnian conflict: to gain their independence, they would need to draw the United States in on their side. The Serbs played their part to perfection.

76.1/01.Mccgwire 6 20/12/99, 1:30 pm Why did we bomb Belgrade?

On 15 January, in a brutal reprisal for some KLA violence, Interior Ministry troops killed 45 people in the village of Racak and some 5,000 fled to the hills. Meeting on 29 January 1999, the six-nation ‘Contact Group’ summoned the warring parties to attend at Rambouillet on 6–14 February, to agree an interim political settlement to the Kosovo conflict. The Serbs were warned that if they refused to attend or to comply with that settlement, they would be liable to military reprisals.

Rambouillet It was one of the oddities of Rambouillet that although Russia was a member of the Contact Group (which originated in 1992), it was only ‘involved in the outskirts’ of the proximity talks, which were strictly confined to NATO members. In preparation for this meeting, the Contact Group had drawn up a set of ten principles that would structure the final agreement; these were non- negotiable. Key points were that interim arrangements were to be agreed for a period of three years, during which the final settlement would be negotiated; that the territorial integrity of Yugoslavia was to be respected, implying continued Serbian sovereignty over Kosovo; that the rights (particularly cultural) of all communities were to be respected, implying far-reaching autonomy for Kosovo; and that these interim arrangements were to be implemented through international participation. As the Serb parliament had endorsed the restoration of Kosovo autonomy when it considered the original US interim solution late in 1998, Milosevic had no difficulty in accepting these principles as the basis for a settlement. Despite having met with the United States on 27 January, the KLA (which now had the upper hand in the Kosovar delegation) was unwilling to accept the draft, as there was no commitment to an ultimate option of independence. Although the US Secretary of State arrived to take personal charge of the talks, the KLA still refused to agree to the terms. For their part, the Serbs had few problems with the existing political provisions (including restoring Kosovo autonomy) and the schedule for withdrawing Serb forces from Kosovo; but they could not stomach certain aspects of the implementation provisions, most importantly the insistence that the 28,000 strong implementation force (K-FOR) would be an arm of NATO. Their initial position was that they would agree only to an unarmed, non-NATO force, but as the talks neared their deadline, Milan Milutinovic (Serbia’s prime minister and chief negotiator) appears to have accepted the need for an international military presence in Kosovo. He was adamant that Belgrade could not agree to NATO forces policing the agreement, but is said to have indicated that participation by certain European members of NATO (and of course Russia) would be acceptable, as long as it was not an explicitly NATO force. However, as the nature and composition of the implementation force were not open to negotiation, this feeler was not followed up.

76.1/01.Mccgwire 7 20/12/99, 1:30 pm Michael MccGwire

Even if the Serbs had been willing to accept the presence of NATO forces, the scope of the ‘status of forces’ set out in an appendix to the Agreement created an insurmountable obstacle. The terms were standard UN wording and were acceptable in respect to Kosovo. However, the appendix explicitly referred to the Federal Republic of Yugoslavia (FRY), which meant the terms applied equally to Serbia and Montenegro. NATO forces (men, weapons, equipment) were to be allowed unimpeded access throughout the FRY and its associated air and sea space; ‘access’ included the right to bivouac and manoeuvre, and the use of any areas or facilities as required for support, training and operations. NATO personnel (in general; not just K-FOR) had the right to enter/exit the FRY on production of an identity document. On 23 February, faced by both parties’ refusal to accept the interim agreement, the negotiations were adjourned until 15 March. To persuade the KLA to sign, the United States is reported to have committed itself to early elections, to retention by the militias of their personal weapons, to preventing any future Yugoslav challenges to the interim or final political status of Kosovo, and to considering the issue of independence if regional and international circumstances permitted. Besides the incentive that NATO would be unable to bomb Serbia if the KLA did not sign, it seems that the clincher was an informal promise by Madeleine Albright that a referendum on self-deter- mination would be held after three years. There were no such sweeteners for the Serbs, and Belgrade faced up to the near-certainty of NATO military action. NATO already had military units deployed on Kosovo’s southern border, which had originally been intended as an ‘extraction force’ should the safety of the OSCE observers be endan- gered. This force was now some 9,000 strong, posing an immediate threat. On 13 March, fighting resumed in the north and south of Kosovo, probably initiated by the KLA, to ensure that the United States remained on side. The meeting reconvened on 15 March, but neither NATO nor the Serbs were prepared to compromise on the major points at issue. Negotiations were terminated on 19 March, on the 20th OSCE observers were ordered to withdraw, on the 22nd the Serbs launched their offensive in Kosovo, and on the 24th NATO bombing began.

Sub-text to Rambouillet: NATO’s 50th birthday The celebratory NATO summit and 50th birthday party were scheduled for 23–5 April in Washington. For at least the previous nine months, this long- planned anniversary had been an increasingly important consideration in US policy regarding Kosovo, and it introduced time constraints that ranged from the trivial (Rambouillet could not reassemble until the foreign ministers had returned from attending the accession ceremony for new NATO members on 12 March), to the strategic. There was growing pressure to get Kosovo sorted out before the birthday

76.1/01.Mccgwire 8 20/12/99, 1:30 pm Why did we bomb Belgrade?

summit, and preferably before the Easter celebrations at the beginning of April. The Contact Group’s rapid response to the killings at Racak and the original schedule for Rambouillet (Serb agreement to its terms by 14 February—or else) had allowed six weeks for military action before April. KLA prevarication meant five of those weeks had been wasted. The timetable was now uncomfortably tight. The birthday celebrations, moreover, were much more than a deadline; the forthcoming summit was intended as a powerful affirmation of NATO’s continuing relevance in the post-Cold War world. NATO (or rather its leading members) had not come well out of the Bosnian imbroglio. The confrontation with Milosevic would be a crucial test of NATO’s credibility. Military action against Serbia would also provide useful precedents for the new strategic concept that was to be approved at the April summit. The US favoured a policy of NATO willingness to take enforcement action in the absence of specific UN authorization. In the case of Kosovo it could be argued that NATO was acting within the spirit of two earlier UN Resolutions (1160 and 1199), which identified Serbian misdemeanours and implied UN action, should corrective measures not be taken. Furthermore, some members (primarily the United States) wanted NATO to extend its role beyond Alliance borders. The Balkans were technically ‘out of area’, but they were also located on NATO’s side of the old ‘iron curtain’ and clearly of concern to western Europe. The anniversary would be the occasion for welcoming Poland, Hungary and the Czech Republic as new members of NATO. For understandable reasons, Russia viewed NATO enlargement with concern, particularly as it breached the bargain struck in 1990 that allowed a unified Germany to be part of NATO. While membership for the ‘Visegrad Three’ was accepted as reflecting a certain political, cultural and geographical symmetry, Russia was not reconciled to NATO enlargement as an ongoing process and was adamant in vetoing membership for the former Soviet republics. The expansion of NATO had been seen by the Clinton administration as an evolutionary process that could possibly include a democratic Russia, which would in all circumstances participate as partner in a larger security regime embracing greater Europe. Unfortunately, the Republican-dominated Congress was against partnership with Russia. It stressed rivalry, even enmity, and favoured incorporating as many countries west of Russia’s borders as possible, particularly the Baltic republics. These mutually exclusive objectives explain contradictory NATO pronouncements that advocated a cooperative security regime in Europe while insisting that the door to NATO membership must remain open. In May 1997 the Founding Act between NATO and Russia established a mechanism ‘for consultations, coordination and … where appropriate, for joint decisions and joint action’. A year later, the Senate resolution ratifying the treaty on NATO expansion laid down conditions that would exclude Russia from NATO’s decision-making process and limit mutual consultations to only those topics on which NATO already had an agreed position.

76.1/01.Mccgwire 9 20/12/99, 1:30 pm Michael MccGwire

War and peace Once the Serbs had seen the ‘Implementation’ chapter of the Rambouillet agreement, they would have realized that war with NATO was inevitable. When it came, aerial bombardment was certain, but the Yugoslav military had to cover the full range of threats. This included NATO’s 9,000 strong ‘extraction’ force in Macedonia (which could be reinforced) and the likelihood of a major Kosovar insurrection, led by a rapidly expanding KLA that would be supplied and reinforced from Albania. The adjournment on 23 February seems to have been taken as a signal that the die was cast. By November 1998, Serb force levels in Kosovo had been brought back to 22,000 (as agreed), but at the end of February 1999 they stood at 25,000; by mid-March they had risen to 29,000 and the first tanks were arriving. When NATO air strikes started on 24 March, there were 36,000 troops already in the province (of which 16,000 were MUP), with another 8,000, armoured, in transit. The order on 20 March to withdraw OSCE observers provided final confirmation, and on 22 March the Serbs launched their pre-emptive offensive. The Serb military requirement was relatively straightforward: rapidly clear the southern and western parts of Kosovo that bordered Macedonia and Albania, driving the KLA across the border in the process; at the same time, impose ruthless internal security throughout the province. And that is what seems to have happened. According to the UN’s disaster assessment (as reported in July), the northern and eastern parts of the province suffered little, whereas the villages stretching to the Albanian mountains in the west were charred rubble and whole sections of the western cities of Pec and Djakovica were blackened shells. The military rationale for the operation drew on numerous precedents from the Second World War, through Vietnam, to the present day. The Croatian cleansing of the Krajina provided the most recent template for an operation that had to be well under way by the time NATO air assets became available to deny Serbia its use of heavy weapons. The operation resulted in more than 800,000 Albanians fleeing the province, while tens of thousands left their homesteads for the hills. It is a moot point whether this mass expulsion and the associated killings were also an example of premeditated ‘ethnic cleansing’ or just a particularly brutal demonstration of the MUP at work and the vicious behaviour of paramilitaries. The figures for refugees imply that something approaching half the Albanian population of Kosovo remained in place. The argument that Belgrade saw this as a manageable number which could be balanced by the resettlement of the 600,000 Serbs expelled from Bosnia and Croatia is plausible—except that those refugees had consistently chosen to settle elsewhere. In any case, the military argument was compelling. On 24 March, NATO’s air campaign got under way, By 10 June it had flown over 27,000 sorties and used more than 23,000 bombs and missiles, with no casualties to NATO personnel. There was extensive damage to Serbia’s military

76.1/01.Mccgwire 10 20/12/99, 1:30 pm Why did we bomb Belgrade?

infrastructure, generously defined to encompass industrial installations, road and rail transport, and telecommunications facilities (including TV stations). A quarter of the SAM radars were destroyed, but NATO was less successful in respect of Serb armour and heavy weapons, with exact numbers in dispute. The Serbs put their losses in the low tens and admitted to only 500 men killed. There is reported to have been significant collateral damage to civilian facilities, including two hospitals and several schools. It appears to be accepted that some 1,500 civilians were killed, and presumably several times that number wounded. The immediate effect of the air strikes was to rally Serb opinion against NATO and to increase Milosevic’s intransigence over the deployment of an international military force. However, by the end of April the principle of an international force had been largely accepted. With Viktor Chernomyrdin acting as mediator, the focus was on the composition of the force and who would be in charge, with Russia seeking to emphasize the role of the UN. The final formula, spelt out in the peace plan agreed on 3 June and amplified a week later in UN Resolution 1244, was for an international security presence with substantial NATO participation. Milosevic’s acceptance of this formula may have been influenced by evidence that Britain and France were actively contemplating a ground-force assault on Kosovo, and growing Russian exasperation was certainly a factor. But it was more important that significant changes had been made to the formula Milosevic rejected in March. In particular, three points of principle had been conceded: • The deployment of an international security and civil presence in Kosovo was to be under UN rather than NATO auspices. • There was no mention of K-FOR’s right to ‘unrestricted passage and unimpeded access’ throughout the rest of the Federal Republic (i.e. Serbia). • While the Rambouillet accords were to be ‘taken into account’ when ‘facilitating a political process designed to determine Kosovo’s final status’, there was no mention of a three-year transition period, nor of Madeleine Albright’s assurance to the KLA that the political process would involve a referendum.

NATO could point to the fact that the timetable for withdrawing all Yugoslav forces from Kosovo had been cut from months to a matter of days, and that only a very small number would be allowed back for border control and guard duties. More important was the specification of substantial NATO participation in the international security presence and the need for unified command and control. While the latter implied NATO C3 in Kosovo, this was significantly less than Rambouillet, which had specified that K-FOR would operate ‘under the authority and subject to the direction of and the political control of the North Atlantic Council through the NATO chain of command’.

76.1/01.Mccgwire 11 20/12/99, 1:30 pm Michael MccGwire

What was it all about? Reviewing the outcome of the Kosovo imbroglio, two questions cry out for answers. Why did NATO go to war with Belgrade? And, having done so, why did it make such a mess of it? I refer to the situation on the ground in Kosovo and not to the relatively few targeting mishaps of the air campaign. In strictly military terms, what NATO air forces set out to do they did well and with remarkable precision, achieving their assigned objective of degrading Yugoslavia’s military capability. Whether it was the optimal objective is another matter. To answer these questions, we need to revisit Rambouillet and the run-up to the talks to see if we can arrive at a clearer understanding of what was actually taking place.

Why did NATO go to war? In their statement of 29 January, the Contact Group noted that the critical situation in Kosovo raised ‘the prospect of a humanitarian catastrophe’. The letter from the NATO Secretary-General to Milosevic declared that the Alliance’s objective was ‘to avert a humanitarian catastrophe’. This would be done by ‘compelling compliance’ with the demands of the international community, and ‘acceptance’ of an interim political settlement. The Contact Group’s statement made it clear that the terms of that settlement had largely been decided, with only a ‘limited number of points’ needing final negotiation between the parties, and required Yugoslavia to ‘accept the level and nature of international presence deemed appropriate by the international community’ (i.e. the Contact Group). ‘The demands’ of the international community included a ceasefire, an end to the disproportionate use of force, and the reduction of Yugoslav force levels in Kosovo to those pertaining in February 1998. The means of compellance would be NATO air strikes on Yugoslavia. These ultimata were addressed to both sides of the dispute, but as NATO had no means of ‘compelling’ the KLA to do anything, this procedural impartiality did not correct the impression that NATO was already parti pris. This was reinforced by reports of US consternation when the KLA resisted signing the settlement, and that one of the inducements to sign was that only when they did so would NATO be able to bomb Belgrade. As a reason for going to war, ‘averting a humanitarian catastrophe’ certainly resonated, but what was the definition of a catastrophe? According to NATO, the conflict in 1998 (including the KLA spring offensive and the summer counter-offensive by the Serbs) resulted in the deaths of something over 1,500 Kosovar Albanians, while the KLA (according to George Robertson) killed a slightly larger number of Serbs during the same period. Was this why we went to war in 1999, killing another 2,000 Serbs (1,500 of them civilians) and indirectly causing the deaths of an even larger number of Albanians?

76.1/01.Mccgwire 12 20/12/99, 1:30 pm Why did we bomb Belgrade?

The great bulk of the deaths in 1998 occurred before the deployment of the Kosovo [ceasefire] Verification Mission, established by the OSCE in October. According to Robertson, the KVM had allowed tens of thousands of Kosovars to return to their homes, and this despite the lengthy delay in getting the process up and running, with the result that only 1,300 observers (out of 2,000 authorized) were in place by 20 March. Since the KVM seems to have been achieving positive results, notwithstanding start-up problems and the shortage of monitors, one might have expected the ‘international community’ to have persisted with this peaceable and potentially cost-effective approach, perhaps tripling the number of observers to allow 24-hour cover and a more effective presence. But, despite the doubts of some members, there seems to have been a willingness to war among NATO political leaders. This has been explained as reflecting the opinion that Milosevic was the root cause of the Balkan tragedy, coupled with a sense of revulsion—and guilt—over Srebrenica. In 1993, when the Security Council had declared six ‘safe areas’ in Bosnia, the Secretary- General had asked for a force of 70,000 troops under NATO control to prevent attacks launched from inside these areas and to deter attacks from the outside. This had been turned down by the United States and other members, with the result that in July 1995, Srebrenica’s only protection was a few hundred UN peacekeepers, who were taken hostage by the Serbs. Whether or not Srebrenica was the reason, the available evidence supports the widespread impression that Rambouillet was set up to fail. The summons to the meeting was explicit that the lengthy Interim Agreement was not open to negotiation except on a few points, and those did not include the nature and size of the ‘implementation force’. The studied avoidance of any reference to the UN and the insistence that the operation would be solely under NATO auspices, NATO-led and NATO-manned, was a known obstacle to acceptance by Belgrade. To make matters worse, as noted above, the appendix specifying the ‘status of forces’ allowed K-FOR the free run of Serbia and Montenegro. Others have argued that the appendix merely replicated the standard agreement for UN peacekeeping forces; this is true, but misleading. The issue was one of geographical scope and, by UN procedure, the appendix should have applied to Kosovo only and not the whole Federal Republic. If it was a drafting error, this could have been easily corrected, so it is fair to assume that the inclusion of this obviously unacceptable condition was deliberate. Why the belt and braces? Perhaps to cover the possibility that Milosevic might unex- pectedly soften his stance on a NATO implementation force. What, then, was the purpose of Rambouillet? There were five possible outcomes: (1) both sides refuse to sign; (2) both sides sign; (3) Belgrade refuses until the brink of bombing; (4) Belgrade gives in after NATO’s military resolve has been tested; (5) Belgrade embarks on a contest of political will and cohesion. One would have expected (2) to have been the ‘purpose’ of the meeting, but the evidence argues otherwise, which leaves (3) or (4) as the preferred outcomes. This brings us back to the sub-text of NATO’s birthday, the importance of

76.1/01.Mccgwire 13 20/12/99, 1:30 pm Michael MccGwire

demonstrating the continuing relevance of the alliance on its 50th anniversary, and the opportunity presented by the Kosovo crisis to further the out-of-area issue and to establish NATO’s right to act without specific UN endorsement. There was also an urge to punish and humiliate Milosevic. Although the plan (if such it was) went horribly wrong, it provides a plausible explanation for much that is puzzling about Rambouillet, particularly the inclusion of a ‘killer clause’ in the text. Lord Carrington, former Secretary- General of NATO and subsequently Chairman of the Hague Peace Conference on Yugoslavia, has observed that Rambouillet required Milosevic to ‘allow NATO to use Serbia as a part of the NATO organization’, a loss of sovereignty that was clearly unacceptable. There remains, however, an anomaly that cannot be explained by the politics of NATO’s anniversary and its new strategic concept, namely the ceiling imposed on Serb force levels in Kosovo. This seems to be of some significance to NATO, since the fact that the ceiling was breached was always included in the list of indictments, however truncated. But the permitted force levels did not make military sense. In October 1998, NATO insisted that Milosevic agree to limit Serb forces in Kosovo to the numbers deployed at the beginning of February; roughly 20,000 in all. True, a ceasefire was in place and the deployment of a verification mission was planned. Nevertheless, NATO’s insistence that the Serbs should revert to the force levels that pertained before the emergence of a guerrilla movement enjoying substantial local support was hard to justify, in absolute as well as relative terms. For comparison, British army strength in Northern Ireland during the first half of the 1970s peaked at 22,000; there were another 4,000–5,000 men in the Ulster Defence Regiment, and the Royal Ulster Constabulary was 13,000 strong, making roughly 40,000 in all, with the UDR and RUC having substantial reserves living locally. Although approximately the same size, Northern Ireland had a significantly smaller population than Kosovo, only 35–40 per cent of whom were potential nationalists; 90 per cent of Kosovars were potential nationalists and the borderlands with Albania were mountainous. The force-level anomaly, the seeming lack of urgency in establishing the verification mission, the unsubstantial nature of the ‘humanitarian catastrophe’ used in January to justify summoning the parties to Rambouillet, the absolute insistence that K-FOR should comprise NATO forces ‘subject to the direction of and the political control of the North Atlantic Council’, and the close involvement of US Special Envoys Richard Holbrooke and Christopher Hill since (at least) May 1998: all these can be explained by reference to George Bush’s 1992 ‘Christmas message’ to Milosevic. To recap, President Bush had warned Milosevic that if war were brought to Kosovo, the United States would be obliged to act; the message was reiterated by Clinton on taking office in January 1993, and the fear that such a war would serve as a fuse train for a major conflict in the southern Balkans was spelt out in February. Five years down the road, the story would have run something like this.

76.1/01.Mccgwire 14 20/12/99, 1:30 pm Why did we bomb Belgrade?

Following an unexpected KLA offensive in the spring, the Serbs had re- established the status quo ante by September 1998, resulting in a large number of refugees from the fighting. Holbrooke and Hill had meanwhile been negoti- ating the outline of a long-term peace settlement involving the restoration of Kosovo autonomy, which was accepted in principle by Milosevic and Rugova, but rejected by the KLA. Failure to create an alternative military force (FARK) under Rugova’s political control, and the rapidly growing political and military strength of the KLA (bolstered by funds from the diaspora), brought the larger problem into focus. If events were allowed to continue along their present tra- jectory, the inevitable result would be major civil war in Kosovo (or, more properly, a war of national liberation), with foreseeably dire consequences. It would seem that, drawing on the experience of 1991–5, the US decided it was no good intervening once such a war had started; that it was unrealistic to expect the Security Council to authorize a pre-emptive intervention; and that even if it did, the UN would inevitably complicate matters, if not render such an operation ineffective. Only NATO could meet the requirement. Bosnia provided a persuasive precedent. NATO troops had been deployed there for nearly three years, implementing an agreement signed by the warring parties at Dayton. A similar approach could be adopted towards Kosovo. Using the authority of existing UN Resolutions and under the threat of air strikes for non-compliance, at a suitable moment the warring parties would be summoned to attend proximity talks. There, they would be required to sign an agreement drawn up by NATO negotiators, setting out the terms of a peace settlement and providing for the deployment of a NATO implementation force in Kosovo. The size of the force would be determined in part by the possibility that Serb units would ignore orders to allow NATO a peaceable entry. Obviously, the fewer Serb forces the better, but to insist on lower numbers than were present in February 1998 would arouse suspicion. On 14 October, under the threat of NATO air strikes, Milosevic agreed to comply with the terms of UN Resolution 1199, including withdrawing his counter-offensive forces. This augured well for air power, and on 24 October the adoption of the ‘Dayton model’ as NATO policy was signalled by the visit of SACEUR and the chairman of NATO’s Military Committee to Belgrade, where they secured Milosevic’s personal agreement on force ceilings in Kosovo. The agreement would also provide a legalistic justification for NATO air strikes should Belgrade, faced by the certainty of war, reinforce its troops in Kosovo. By early November, NATO national authorities had started work on planning the deployment of the implementation force and SACEUR had been told to prepare for an air force war with zero losses. It is unclear when the development of this policy became enmeshed with the political build-up to NATO’s anniversary. It does, however, seem likely that the timetable for the celebrations was what prompted the decision to use the killings at Racak to justify setting the Rambouillet process in train. Taking action to forestall a civil war would seem to be a commendable

76.1/01.Mccgwire 15 20/12/99, 1:30 pm Michael MccGwire

initiative. Why, then, was NATO so reluctant to admit what was afoot? One can think of two kinds of reason, one relating to undesirable precedents in the region. The more important kind relates to the terms of the UN Charter and the North Atlantic Treaty; even if it relied heavily on the Universal Declaration on Human Rights, NATO would be sailing perilously close to the legal wind. Hence George Robertson’s insistence (after the bombing was under way) that it was ‘not a war’. He had in fact let slip the forbidden term ‘civil war’ in February, but from September 1998 the codeword for that calamity had been ‘humanitarian catastrophe’. As this euphemism first surfaced in Resolution 1199, when the immediate humanitarian problem was the effect of winter on the 50,000 who had taken refuge in the mountains, NATO’s real objective was well concealed from everybody, including its electorates. So, why did NATO go to war? In essence, to compel Milosevic to accept the deployment of 28,000 NATO troops in Kosovo, whose presence was deemed necessary to avert a civil war. While keeping the peace, their role would be to oversee the implementation of a political settlement that would restore substantial autonomy to Kosovo, something the Serb parliament had already agreed in principle in late 1998. Accepting the objective of averting a civil war, could it have been achieved without bombing Serbia? One can only speculate, but the formula that underlay the final peace agreement clearly had potential and a variant is said to have been advanced by the Serbs at Rambouillet. A central feature would have been the emphasis on the UN, fudging the question of the NATO core, and enlarging the force if need be so as to ensure a reasonably balanced presence in order to reassure Belgrade. This approach would have required involving Russia as a full partner in the negotiation and implementation processes, rather than slighting its opinion until things went wrong. A possible alternative would have been to try the OSCE route, building on the verification mission to secure Belgrade’s acceptance of a sizeable and somewhat more capable body. Britain’s acknowledgment in October of the merits of a European Security and Defence Identity made it possible to think in terms of EU participation, with an Anglo-French combined joint task force at its core. On the supposition that such a body would not be opposed by Serb forces in Kosovo, it could be smaller (and hence less threatening) than the NATO implementation force, but would have the possibility of reinforcement, should the need arise. Practical or not, neither of these approaches would have been acceptable to the United States, whose taste for multilateralism had been soured by Somalia and whose relations with the UN were frequently hostile. As regards the OSCE, the United States was determined to prevent the emergence of an alternative Europe-wide security structure that could challenge its authority; NATO was a known quantity, whose collective decisions could be shaped through bilateral negotiations with member states. Meanwhile, the Republican-dominated Congress accepted only grudgingly Russia’s involvement in NATO affairs and

76.1/01.Mccgwire 16 20/12/99, 1:30 pm Why did we bomb Belgrade?

forbade any discussion on policy issues before NATO’s own position had been decided. At the celebratory summit in April, Clinton spoke of US policy being based on diplomacy backed by force. At Rambouillet the threat of force took centre stage, but of diplomacy there was precious little. Except for a very few points, the terms of the ‘agreement’ were presented as an ultimatum: ‘Sign—or else’. Nominally chaired by the foreign ministers of Britain and France, Rambouillet was Madeleine Albright’s personal project, her ‘Dayton’. But her personal credibility was also tied up with NATO’s, which was very much on the line. There was considerable pressure in official circles and in the press to demonstrate that NATO at 50 still had a role to play.

In what way did we make a mess of the war? It must first be acknowledged that NATO did achieve its objective of deploying a NATO force in Kosovo, whose mission was to prevent a civil war. And, although it was achieved at significant cost in lives and homes of the Kosovar Albanians, the latter felt it was a price worth paying to have brought NATO in on their side against Serbia, which was severely punished. That was not, however, what NATO intended. So what went wrong and why? Of the many unintended consequences, four were particularly damaging. The most immediate was the flood of refugees from Kosovo and the death and destruction from which they fled. Despite NATO claims to the contrary, there was a common-sense link between the bombing of Serbia and the behaviour of its forces in Kosovo, and the destruction did in fact peak during the first two days. To those who had been following the Kosovo crisis more closely, while Serb forces were clearly the instrument of the unfolding ‘humanitarian disaster’, NATO’s long-trailered urge to war was undoubtedly a primary cause. Second, Rambouillet bestowed a legitimacy on the KLA, which emerged from the war with an enhanced military capability, a greatly improved organizational structure, and leaders who saw themselves (and were seen by many) as Kosovo’s government-in-waiting, overshadowing Rugova. Concessions made to the KLA both before and after the war ‘because they had signed’ and despite their terroristic behaviour and aim of a Greater Albania, further enhanced the KLA’s status and their military capability, storing up trouble for the future. Third, eleven weeks of bombing not only did extensive damage to Serbia’s industrial infrastructure, it disrupted the trade and communications of its neighbours as well as more distant areas. Serbia lies at the heart of the Balkans and the fact that it will take five years or more to recover from this pounding will have a strongly negative effect throughout the region, while the cost of restoring the infrastructure will be over $20 billion. The fourth and most damaging unintended consequence, whose long-term costs have yet to be fully realized, is the effect that the prolonged war has had on NATO’s image. It was, of course, hugely embarrassing for the anniversary cele-

76.1/01.Mccgwire 17 20/12/99, 1:30 pm Michael MccGwire

bration, which had carried such high political expectations. But far more serious was the damage done to the carefully cultivated image of a redesigned NATO, intended primarily to ‘project stability’ throughout Europe, helping to bring good governance and market economics to the eastern parts of the region. Instead, the world at large saw a political–military alliance that took unto itself the role of judge, jury and executioner. Despite Russian objections to NATO military action, the alliance claimed to be acting on behalf of the international community and was ready to slight the UN and skirt international law in order to enforce its collective judgement. The world saw an organization given to moralistic rhetoric, one no less economical with the truth than others of its kind; a grouping of Western states with an unmatched technical capacity to kill, maim and destroy, that was limited only by their unwillingness to put their ‘warriors’ at risk.

Why did we make a mess of the war? Having adopted the ‘Dayton model’ as the way to solve the problem of Kosovo, NATO then failed to make the major adjustments needed to accom- modate the important differences between the two situations. After four years of on-and-off conflict, Dayton presented the warring parties with a territorial settlement that largely reflected the situation on the ground. Croatia having regained the Krajina, the settlement was between the three main ethnic groups living in the former republic of Bosnia-Herzegovina. Given the continuous involvement of the international community, the role of NATO air power in bringing the sides into rough balance, and the contested political status of the warring entities and their leaders, it was realistic to bring the three sides together in proximity talks and insist that they agree on a settlement, drawn up by mediators. It was also reasonable to insist on the deployment of a NATO-led implementation force in the former republic, which was already awash with UN officials, international agencies and UNPROFOR soldiers. The differences from the situation in 1999 are obvious, but three are worth highlighting: Serb sovereignty over Kosovo was not in dispute; Serbia was not some newly formed statelet, nor was it war-weary; and the international presence in Kosovo, agreed by Belgrade, was limited to 2,000 unarmed observers, subordin- ated to the OSCE. In such circumstances, while ‘take it or else’ proximity talks may have been suitable for agreeing a political settlement restoring substantial autonomy to the Kosovo Albanians (which had already been agreed in principle by the Serbs), Milosevic had grounds for arguing that the composition of an implementation force was a matter for negotiation between Belgrade and the UN. It appears that NATO’s political leaders assumed that Milosevic, who was seen as a dictator and the sole decision-maker, would cave in when threatened directly with air strikes; and if not, then surely after a short, sharp, bombing campaign to show that NATO meant business. By any measure, it was a high- risk strategy.

76.1/01.Mccgwire 18 20/12/99, 1:30 pm Why did we bomb Belgrade?

The gamble failed because of three common errors: overlapping objectives; a mismatch between objectives and resources; and misreading one’s opponents actions/reactions. In the Kosovo crisis, the last of these was crucial.

Know your enemy NATO insisted that the enemy was Milosevic. By demonizing one man, ignoring the complexities of Belgrade politics and absolving the Serbian people of any blame for the policy in Kosovo, NATO political leaders seem to have blinded themselves to how the Serbs would react to the threat and actuality of a bombing campaign. The Serbs have always been proud of their ability to fight in defence of their interests and their capacity to absorb punishment. There was ample evidence of popular support for Serb policy in Kosovo, and as late as February 1999 an opinion poll in the opposition paper Dnevni Telegraf showed that 37 per cent were willing to defend Kosovo by force ‘under any circumstances’. In July, a third of the Serb parliament voted against ratifying the peace agreement. As the result of events in 1992–5, NATO was seen by the (redefined) Yugo- slav army as an instrument of US policy and a declared enemy. By 1992, the United States was siding with the Bosnian Muslims against the Serbs. By 1994, retired US service personnel were helping to rebuild, re-equip and train the Croatian army. In 1995, NATO air assets essentially worked in support of Croatian and Muslim forces when they drove the Serbs back in Bosnia. And now, in 1999, the United States had aligned itself with the KLA. NATO looked as if it were preparing to do serious harm to Serbia. It was demanding it be allowed to deploy a heavily armed force in Kosovo, giving it a second beach- head in the Balkans under the political control of the North Atlantic Council. NATO already had troops in Macedonia and Bosnia, Hungary was about to join the Alliance and Romania was a favoured applicant for membership. NATO controlled the Adriatic, Montenegro was being encouraged to secede, and Yugoslav intelligence had reported US military construction in Bulgaria. It could be called encirclement. Besides misreading how seriously Serbia saw the threat to its national survival, NATO was guilty of wishful thinking about the Serb response to air attack. The situation on the ground and the sequence of events leading up to Dayton in 1995 were completely different from the circumstances that pertained in 1999. Similarly, what Milosevic conceded in October 1998 under the threat of air strikes bore no relation to what Serbia was being asked to concede at Rambouillet. Lastly, NATO seems to have pulled the wool over its own eyes by insisting that the military action it was threatening ‘was not a war’. Hence the shocked surprise of its political leaders when Serbia, facing certain attack by the remaining superpower and its allies, took preventive military action: action that SACEUR described as entirely predictable.

Overlapping objectives One can identify two long-term political objectives. One was to avert civil war in Kosovo; the other was to demonstrate NATO’s

76.1/01.Mccgwire 19 20/12/99, 1:30 pm Michael MccGwire

continuing utility and future potential. An ancillary objective—to punish Milosevic—affected attitudes, if not plans. Although there was a need for early and concerted action to achieve the first objective, an undesirable urgency was introduced by the time-bound schedule for NATO’s anniversary. The second (and less important) objective seems to have persuaded NATO to use the Dayton approach of tight deadlines and diktat, rather than concentrate on finding a way (other than the use of force) of persuading Milosevic to accept an effective international force to implement the political settlement. The fusion of the two objectives meant that a full analysis of alternative ways of averting civil war in Kosovo was sacrificed to showing off NATO’s political muscle and demonstrating its commitment ‘to meeting a wide range of threats to our shared interests and values’. There were a number of different ways of organizing (or ‘growing’) an effective international presence in Kosovo. Carrots as well as sticks could have been used to persuade Belgrade to cooperate, and more effort could have been invested in boosting Rugova and diminishing the KLA. But in the real world of politics, the longer-term and more important objective of avoiding civil war had to yield precedence to the immediate requirement to bolster NATO credibility by punishing Milosevic for taking issue with its demands.

Matching objectives to resources The resources allocated to SACEUR by NATO’s political leaders were long-range cruise missiles and manned aircraft, the latter limited to operating above 15,000 feet; i.e. a capability for strategic bombardment. While there can be argument about the tempo, intensity and targeting philosophy of the actual campaign, the capability was, in theory, sufficient to achieve the immediate political objective of persuading Belgrade to sign up to Rambouillet. In practice, if the threat itself was insufficient to make Milosevic sign, history should have warned that strategic bombing would at best take time and (in terms of the initial objective) was unlikely to be fully successful. In the process, it was bound to cause significant civilian casualties, which would cause serious damage to NATO’s image and most likely bring its public relations into disrepute. It is true that bombing degraded Serbia’s military capability, but that applied more to infrastructure than to fighting forces, which were practised in operating under enemy aerial domination. Meanwhile, bombing could do nothing about preventing the clearing of Kosovo; nor, in practical terms, could anything else. In theory, NATO could have run the operational plan in reverse, starting with the military envelopment of Kosovo by air, meanwhile deterring interference from Serbia’s main forces by threatening the aerial devastation of their country (deter- rence being more effective than compellance). In practice, the military complexity and political unacceptability of such an operation meant it was a non-starter. Some argue that NATO should have been poised to launch a land invasion of Kosovo to reinforce the threat of bombing, bringing additional pressure on Milosevic to sign. That is an exercise in hindsight; besides which, such an the

76.1/01.Mccgwire 20 20/12/99, 1:30 pm Why did we bomb Belgrade?

invasion would have had to wait until the main exodus was over. However, as the weeks of bombing went by, the growing burden of the refugee population on Macedonia’s fragile political economy gave a new importance to the immediate objective of enabling the Kosovars to return to their homes. This changed the terms of the debate about ground-force involvement.

To conclude Similarities between the Suez and Kosovo crises extend beyond the demonizing of Nasser and Milosevic and the claim to be acting on behalf of the ‘international community’. They include the failure to achieve the declared objective of either operation (safe passage of shipping/avoiding a humanitarian disaster), and the need for considerable international assistance to rectify the damage they wrought. The list goes on; but it is of interest that Suez turned out to be a watershed in the postwar world, both in terms of the policies and pretensions of the two imperial powers and in a more general sense concerning the political acceptability (and utility) of ‘great power’ military intervention. In many ways Suez can be seen as the real beginning of post-colonialism and the superpower era, along with peacekeeping and the explosion of UN membership. Kosovo may well prove to be a comparable watershed in the post-Cold War world, but it is far from clear what lies on the other side. One reads of concern around the world at NATO’s departure from long-established policies, at the slighting of the UN, and at the resort to coercive diplomacy under the guise of humanitarian intervention. And we have seen NATO’s air campaign used as a precedent by Russia, which, unfortunately for the Chechens, lacks precision- guided weapons. NATO enlargement had been put on hold by the end of 1998, but sup- portive behaviour during and after the air campaign reopened the question of Romania’s candidacy and added Bulgaria to the favoured list. NATO’s cloak now extends to the Black Sea, adding to Russia’s concerns. In the United States, rather than bolstering support for NATO, the war aroused significant opposition, not least because US forces did almost all the work and took most of the risks. This conformed to the pattern in the Gulf war, the air strikes in Bosnia and the continuing campaign against Iraq, and reinforced the ‘unilateralist’ trend in national security policy. Contrary to what was hoped of the 50th anniversary, the transatlantic relationship has re-entered the US debate. There is talk in Europe of a ‘Marshall Plan’ for the Balkans, although little has been done and Serbia, the key to the region, will be excluded until it gets rid of Milosevic. EU members are reluctant to provide the funds and the United States, having borne the major cost of the war, is unwilling to foot the bill— which reminds us that it would have been so much cheaper if, in the mid-1980s and subsequently, the West had avoided doctrinaire solutions and taken the long view. If we had continued to prop up the economy of its Yugoslav client, enabling it to weather the debt crisis and the loss of its niche position, the

76.1/01.Mccgwire 21 20/12/99, 1:30 pm Michael MccGwire

Federation could have made a peaceful transition to democracy and a market economy, rather than succumb to populist nationalism and inevitable disintegration, with incalculable costs in social disruption and human suffering. There are other lessons to be drawn from the Balkan tragedy, including the tendency to define complex conflicts in oversimplified and moralistic terms, labels which then drove policy and shaped public expectations. In part, this reflected our damaging ignorance of the convoluted politics of the situation, a debility compounded by our slow learning curve. Reacting to each conflict in turn, the West applied undigested lessons from the previous one, whereas the underlying causes of the ‘four successive wars’ were qualitatively different, each requiring a particular set of responses. The application of the ‘Dayton model’ to Kosovo is a case in point, the effect being accentuated by NATO’s simultaneous pursuit of two objectives that were not fully compatible. It is symptomatic that the substantive political objective of averting civil war had to yield precedence to the time-bound public relations objective, with its deadline of NATO’s birthday. This ruled out consideration of potentially more fruitful ways of achieving the substantive objective within a different time-frame. This was a classic example of image taking precedence over substance, which is not uncommon in today’s political world. It is often associated with a rhetorical style that is more concerned with effect than with accuracy; in Animal Farm-speak, words come to mean what our leaders say they do. When commenting on Chechnya, a spokesman for Number 10 can still describe the mass expulsions from Kosovo as ‘a programme of racial genocide’, although the facts on the ground say otherwise. Robin Cook speaks of ‘the liberation of Kosovo’ although he knows (as co-chairman of Rambouillet) that Serbia remains the sovereign power. But the really grotesque example is describing eleven weeks of bombing as ‘humanitarian intervention’. We are also told that Kosovo was a new kind of war—one designed to pro- tect values, not interests. But is that factually correct? Although our concern for human rights in Kosovo had been registered by a UN Resolution in March 1998, the evidence suggests that the need for NATO military intervention was not finally decided until October. It was prompted by the KLA’s brutal rejection of the US-brokered political solution to the problem, which was based on restoring provincial autonomy with Rugova as president. The purpose of NATO’s intervention was to pre-empt a civil war, a war that would trigger George Bush’s ‘Christmas warning’ about US action to defend vital interests in the Balkans. The second objective, of demonstrating NATO’s continuing utility and future potential, was clearly concerned with public relations, not values. As final refutation, the resources allocated to SACEUR were intended to force Milosevic to agree to the deployment of a NATO force in Kosovo, and were not designed to avert a humanitarian disaster. It was hoped that the threat of bombing would cause Milosevic to cave in. If he did not, as SACEUR himself has said, the forced expulsion of the Kosovar Albanians was utterly predictable, and NATO would be unable to do anything about it.

76.1/01.Mccgwire 22 20/12/99, 1:30 pm Why did we bomb Belgrade?

No one questions the underlying good intentions, but one suspects that much of the moralistic rhetoric, the demonizing, the claim to be pioneering a foreign policy based on values as well as interests, was a form of denial. It served to conceal from all of us the unpalatable fact that leaders and their people have to accept their share of the blame for unintended consequences—in this case the humanitarian disaster in Kosovo and the civilian casualties in Serbia.

*** In November it was refreshing to learn on the evening news that a senior Russian general had officially acknowledged civilian casualties in Chechnya, for which Russia ‘accepted moral responsibility.’ NATO, please copy.

76.1/01.Mccgwire 23 20/12/99, 1:30 pm