A PHASED CAMPAIGN

97 The air campaign was divided conceptually into three phases,[215] though these were very compressed and overlapped. Phase 1 was the suppression of enemy air defences; Phase 2 was strikes against Serb forces on the ground in Kosovo; Phase 3 was strikes against "strategic" targets in Serbia.[216] The Chief of Joint Operations told us—

I suspect that NATO hoped that through phase one of the campaign it could achieve its objectives, although we all understood that we might be in it for the long haul.[217]

98 We were told that the Secretary General authorised the transition from Phase 1 to Phase 2 on 27th March. To have moved to Phase 3 would have required the full consent of the NAC. There is some ambiguity about the nature of the post-Phase 2 stage of operations. Sir John Goulden told us that the Council took—

... the final decision on targetting ... as early as 29 or 30 March. By 29 March we had authorised all the powers that the military needed for the campaign ...[218]

However, our informal discussions would suggest that the formal decision to move to strategic bombing of Serbia was never put directly, in quite those terms, to the NAC. Rather, an extension of the delegation to the Secretary General was made on or around 30 March. This interpretation is confirmed by a comment of General Naumann's in a BBC documentary broadcast in March 2000. He said—

Phase three could have been seen as an all-out air war against Yugoslavia and the NATO nations, well not all NATO nations, were prepared to go as far ... and for that reason we never took the risk to ask the question knowing that we may run into some problems.[219]

The remaining 72 days of bombing instead fell into a rather uneasy definition of 'Phase 2A' of the strategy. This may have been another example of a pragmatic political compromise—however, it also indicates to us a certain amount of sustained ambivalence amongst the politicians directing the operation towards the reality of the military means they needed to deploy to achieve their ends. This ambivalence about authorising the strikes against 'strategic' targets was, we conclude, unhelpful to military planning and contributed to the mixed signals that were sent to Milosevic, the NATO military and the public of the allied nations (and Serbia). That mixed message may have further undermined NATO's capacity to signal its determination in the early stages of the bombing campaign to see through to the end what it had begun.

99 As the air campaign moved into 'Phase 2A,' from the end of March, UK and NATO targeteers made some effort to identify and strike targets that were not just of military value to Serbian air defences, military command and control and the fielded forces in Kosovo, but would also influence perceptions. There appear to have been two target audiences for this. The first was the Serbian people as a whole. There was a belief—or hope—in the UK and in the wider Alliance that Serbian morale would 'crack' and that the Serbian population would be encouraged by the air campaign to protest against the policies of the Milosevic government. The second was the Serbian decision-makers. Some attacks appear to have been directly on assets and symbols believed to be of high value to Milosevic and the civil and military élite.

100 However, air strikes targetted at the Serbian population in general were never at the heart of Alliance strategy. Indeed, the UK and the Alliance went to great lengths to avoid mobilising Serb opinion against NATO by minimising collateral damage when striking targets in Serbia. Lord Gilbert gave a hostile account of this policy—

It was after it was over that President Chirac said publicly that it was him, that M. Chirac had vetoed the destruction of any bridges across the rivers in Belgrade ... I congratulated him on his statesmanship for having the courage to take responsibility for the fact that the hostilities went on far longer than they need have done, that the poor people in Kosovo suffered consequently far more than they need have done, and that his own pilots were put at risk far more than was necessary ... The whole story of the targetting ... is one of political timidity ...[220]

This restraint was deliberate, and did enable the Alliance to emphasise that it was fighting the Serbian regime and not the populace as a whole—an important political message in this political war. The more important strategic targets were assets and symbols that it was hoped would directly coerce the regime's decision makers. These included political symbols such as the Belgrade HQ of the Socialist Party (21 April), totems of state authority such as Belgrade's main TV station (23 April), and industrial assets in which key régime figures had an economic interest. While this approach to targetting reflected the political dimensions of the coercion strategy, in retrospect it appears that neither the MoD nor the Alliance had the doctrine and intelligence support to mount a coordinated, systematic attack on the régime's psychological centres of gravity. As General Klaus Naumann has stressed since the campaign, in the future—

... we should aim at the ¼ goal of enforcing our will on our opponent¼. [this would involve concentrating] in Phase 2 on those targets which constitute the opponent's pillars of power.[221]

We conclude that in the event, from the outset of the bombing campaign, both aspects of the Alliance's strategy—coercion and denial—may have suffered from a politically engendered uncertainty over what effect the campaign was supposed to have on whom.

101 We now examine separately the effectiveness of each of the three phases of the air campaign .

PHASE 1: SUPPRESSION OF ENEMY AIR DEFENCES

102 The Alliance's first task in the air campaign was to suppress or destroy the air defence system of the Federal Republic of Yugoslavia, so that allied aircraft could operate unimpeded over the territory. There is some disagreement about how successful this phase of the campaign was in achieving its objectives. When asked about the extent to which the Alliance disabled the Serbian integrated air defence (IAD) system, the Chief of Defence Intelligence responded—

It depends how you define disable. I think we did disable their defence ... we had a very good understanding of their IAD system. We knew what we needed to do in terms of our suppression of enemy air defence package. Those packages did work. We managed to suppress it and we managed to conduct those air operations. I think we did do it very successfully ...[222]

Certainly, elements of the Serbian IAD were successfully neutralised. There was a general belief at the time of the operation that no sorties were flown below 15,000 feet. CDI told us, however, that—

... as the war went on ... for specific targets, aircraft were allowed to go down below that and go down to the 6,000 feet level to conduct certain operations against targets.[223]

Nonetheless, the MoD's own Lessons from the Crisis reports that—

The decision by NATO that operations should be flown at medium altitude (above 15,000 feet) was taken in recognition of the threat from Yugoslav air defences, which continued throughout the campaign ... The 15,000 feet minimum operating altitude was introduced to ensure that NATO aircraft operated at an acceptable level of risk but, as the operation progressed, some NATO aircraft operated at lower altitudes when necessary to acquire and identify targets.[224]

Despite flying over 1600 sorties, including over a thousand strike sorties, and despite over 700 missiles being fired by Yugoslav air defences against NATO aircraft, no UK aircraft were shot down during Operation Allied Force and no UK lives were lost.[225] However, the US DoD report to Congress admits that—

While NATO prevailed in delivering a punishing air offensive with virtually no loss to its forces, we must acknowledge some concerns for the future. Although among the most capable that the United States has faced in combat, the Yugoslav air defence systems do not represent the state of the art. Much more capable systems are currently available for sale in the international arms market.[226]

103 As both the Gulf War, and now the Balkans have shown, Electronic Warfare (EW) support is an essential feature if operations are to be conducted without heavy losses against a modern air defence system. The Suppression of Enemy Air Defences (SEAD) can be achieved with either hard-kill systems such as bombing or anti-radar missiles,[227] or by soft-kill systems that rely on electronic attack on enemy IADs. Electronic counter-measures (ECM), or jamming, can be directed against air defence surveillance systems (so that aircraft are not detected in the first instance) or against SAM missile guidance systems (either the ground-based element, or in the missile itself).

104 Even against relatively unsophisticated air defences a comprehensive jamming capability is essential if losses are to be restricted, particularly if aircraft are to penetrate at low level. In the case of the US EA-6B Prowler the aircraft stands-off in such a way that the geometry of the total attack allows its jamming to cover the penetrating bombers. In the Balkans almost all of the stand-off ECM capability was provided by the EA-6B, a force of only a little over 90 aircraft world-wide, already stretched, and one providing the bulk of the US capability in this field. As a result, providing aircraft for the Balkans resulted in withdrawing units from the Middle East and even reinforcing from the Far East. The increasing age of the EA-6B force is a concern for the Pentagon and there is now much debate about a follow-on system, but that will be expensive and certainly not easy to bring into service quickly.[228]

105 No European Ally deployed airborne electronic warfare aircraft with capabilities to match the American EA-6B Prowler. This aircraft and system proved critical in suppressing enemy air defences and the need for this capability is being examined by the MoD with other NATO European allies.[229] We discussed European shortfalls in this and other areas in our recent Report on European Security and Defence.[230] Remedying these shortfalls is part of NATO's 'Defence Capability Initiative'. We taxed the Secretary of State on whether Europe was ever likely actually to acquire such capabilities. He told us—

There is little doubt, at the moment, given what we know about the Kosovo campaign and the kind of equipment assets that were required, that even then [in 2003] we would not be in a position, given the capabilities we are setting out in the headline goal, to be able to conduct precisely this kind of operation. This is because very many of the assets, particularly in the air campaign, are simply not assets that European nations for the moment have available.

We asked, therefore, if he would agree that if we are going to be dependent for some considerable time on the United States for the ability to suppress enemy air defences. He responded—

It depends on the scale. There are undoubtedly air campaigns that we could conduct as a country on our own. There are other campaigns that we could conduct with appropriate European allies. If you had said to me, as I have already conceded, "Could we conduct tomorrow, or even in the near term, a Kosovo style air campaign?" I would say we could not without involving the United States.[231]

106 When asked if we or our European Allies would be acquiring EA-6B Prowlers or their equivalents, the Secretary of State commented—

There is very considerable work underway on looking at this but ... we have to make a judgment at what point we have the right technology, and the best technology so that it is not superseded by developments elsewhere. Work is underway in a vigorous way to get this right.[232]

However, the MoD's Lessons from the Crisis report says—

Electronic Warfare (EW) and Suppression of Enemy Air Defence (SEAD) capabilities were vital force enablers during the air campaign ... The bulk of this effort was provided by the US ... We are looking at SEAD capability, and whether a significantly increased capability could be achieved ... [without] buying more SEAD-capable aircraft ...[233]

107 There are other approaches to SEAD. When possible, the best means of avoiding enemy air defences is to operate in such a way as to be outside the capabilities of the defensive systems. For many years NATO (and the RAF) planned to penetrate below the radar cover of the Warsaw Pact defences. The RAF's expertise in low-level flying and the terrain-following radar in the Tornado both resulted from that doctrine. A different approach was taken by the US U-2 reconnaissance aircraft that operates at a height above the capability of most air defence missile systems. Another means was taken by the SR-71, a very high-speed aircraft that could out-run defensive missiles. A further strategy is to use stealth technology that so delays the detection of target aircraft that the defensive system has insufficient time to execute an attack. In Kosovo, the US used B-2 stealth bombers, operating from air bases in the USA.[234] Modern cruise missiles, some incorporating stealth features together with other counter-measures, can also penetrate air defence systems.

108 Where air defence systems cannot be avoided they have to be engaged, with physical attacks on airfields and aircraft, Surface-to-Air Missile (SAM) systems and/or Command, Control and Communications facilities. Once destroyed, equipment is out of action for the duration of a conflict, and there is a superficial attraction to that. However, some air defence systems can have many separate elements and built-in redundancy, and could take a long time to degrade sufficiently to allow for unimpeded operations. Soft-kill jamming, on the other hand, can degrade much of an air defence system at the same time or, alternatively, can degrade only those parts that pose a threat for any specific operation. In practice, a combination of both hard and soft-kill methods needs to be used.

109 Lack of sophistication in some defensive systems themselves can sometimes foil more sophisticated countermeasures. Anti-aircraft artillery (AAA), once fired, cannot be jammed or decoyed. AAA mixed with tracer rounds can appear very threatening and despite the poor probability of a kill most pilots treat AAA with a great deal of respect. Shoulder-launched SAM systems are widely deployed, and the latest Russian (SA-18) and US (Stinger) missiles are very effective. It is the combination of light AAA and the shoulder-launched SAM, both highly mobile systems, that place the unknown into the equation and drove the decision to operate strike aircraft at medium-level in the Balkans (and, eight years earlier, in the Gulf War).

110 The battle between the penetrating aircraft and the air defence system will continue with the advantage swinging from one to the other. It is essential to engage in this contest if air power is to be relied upon in the future. More particularly, the risk-averse nature of politically sensitive operations makes SEAD even more critical to the effectiveness of operations like that over Kosovo and Serbia. To concede the battlefield to the defence will result in losses, at what most probably will be politically unacceptable rates. Part of the offensive air capability, therefore, must include the capability to engage in SEAD. Currently, the non-US members of NATO, including the UK, do not have SEAD capabilities sufficient to allow operations such as those in Kosovo without US support. There is a risk of divergence in the Alliance if the US decides to pursue stealth technology as its main technique for defeating air defence systems, because if as a result it allows its electronic counter measures systems to become obsolescent, the protection they currently afford to other Allies' aircraft will wither. As the expense of stealth technology is probably beyond the reach of many European Allies' pockets or political will, they may have to choose between developing their own systems to protect their more vulnerable aircraft, or relinquishing the strike role to the US Air Force.

111 Although the UK and Germany appears to be discussing ways of harmonising and sharing their existing SEAD capabilities,[235] the lack of a wider European SEAD capability raises serious questions about whether the Europeans could risk engagment in a mission such as Operation Allied Force in the future. The alarming deficit in European capabilities for suppressing and destroying even relatively unsophisticated air defences suggests that Europe must either accept that its scope for action independent of the US is very limited indeed, or face up to the requirement of improving its capabilities sufficiently for it to act independently. A sea change in political and military will, along with the required financial provision, will be needed to provide such a European capability and, currently, there is no evidence of either being present in the UK or amongst our European Allies.

PHASE 2: STRIKES AGAINST FIELDED FORCES

112 Following the partial success of the first phase of the air campaign, a conventional strategy of denial was attempted against VJ and MUP forces operating in Kosovo. It was hoped that they could be progressively destroyed as the air campaign went on. The Chief of Joint Operations told us—

From a military perspective the bombing campaign achieved three things: it degraded the Serbian air defence system; it disrupted the Serbian military's command and control systems; and it disrupted the efforts of Milosevic's forces in Kosovo to use their weapons against the Kosovo Albanians.[236]

However, there has again been considerable controversy over the effectiveness of the Alliance attacks on Serb ground forces. We examined the evidence at some length with our witnesses and in informal sessions in NATO HQ and at SHAPE. The MoD's Lessons from the Crisis reports—

General Clark (the Supreme Allied Commander Europe or SACEUR) made public NATO's Battle Damage Assessment of attacks against mobile targets on 16 September 1999 ... In short, he reported validated strikes on 93 tanks, 153 armoured personnel carriers, 389 artillery pieces and mortars and 339 military vehicles. It is not always possible to ascertain whether a target has been totally destroyed or what degree of damage it has sustained.[237]

113 BDA figures for mobile targets that have been accepted by the MoD and the US Department of Defense were brought together by a NATO assessment team that conducted—

... a comprehensive day-by-day , mission-report-by-mission-report reconstruction of the operation to determine the actual number of mobile targets struck with high confidence.

The BDA assessment covered all 78 days of the operation but focussed only on strikes in the area of Kosovo and the Presevo Valley. The team took into account a number of contributory factors such as, for example, whether the KLA had destroyed or damaged tanks or armoured personnel carriers, and whether some NATO missions had struck the same targets on more than one occasion. The assessors drew from multiple sources such as cockpit videos, image intelligence, signals intelligence, interviews with forward air controllers and on-scene witnesses.[238] The US Report, unlike that of the MoD, goes on to provide further information on unconfirmed, decoy and multiple strikes, drawing attention to the limitations of the assessment, and stressing that no claims were made on the proportion of the total mobile targets that were hit, nor the level of damage inflicted on the targets that were struck.

114 The range of possible figures can be illustrated by the basis of assessment for strikes against tanks. 26 tank 'carcasses' were found in Kosovo after the Serbs withdrew. The USAF/NATO assessment considers that a reasonable level of confidence can be felt about a figure of 93 destroyed. Mission reports from pilots suggested 181 successful strikes (some of which were almost certainly against decoys and others against targets previously hit). There will never be any conclusive evidence, though better figures may exist somewhere in Belgrade. NATO's final battle damage assessments are lower than originally claimed but not excessively cautious. Indeed, the Chief of Defence Intelligence conceded that NATO's final figures were "probably ... optimistic",[239] and Sir John Goulden granted that "There is a fairly wide margin of uncertainty ...".[240] The Chief of Joint Operations also considered that—

... in retrospect we probably thought we were having rather more success at the time than in hindsight we thought we achieved.[241]

General Sir Mike Jackson was similarly sceptical.[242]

115 Some of the problems encountered in operations against fielded forces appear to have arisen because of the failure to eliminate the Serbian air defence network, which made it very difficult to get to grips with Serbian military forces which were on the ground in Kosovo. MoD witnesses emphasised, however, that the air campaign against mobile targets forced the VJ and MUP to hide equipment thereby denying its use against the Kosovo Albanians. Of course, had the VJ had to prepare to counter a ground attack, such concealment would have been much more difficult, and more of their armour would have been vulnerable to air attack. The Chief of Joint Operations told us that Serbian commanders knew that if they brought weapons out into the open they would be hit,[243] and asserted that NATO did a good deal to restrict Milosevic's operations and reduce his capability to prosecute his campaign of ethnic cleansing.[244] Sir John Goulden argued that while—

We accept that there is a margin of uncertainty ... The key bottom line is that we bottled up the equipment that was in Kosovo.[245]

The MoD's report takes the same tack—

Attacks against tactical targets in Kosovo proved to be a significant challenge to the Alliance, given the difficulties in locating and positively identifying targets. The Yugoslav/Serbian security forces concealed their assets to a considerable extent, and were adept at deception techniques, including the use of decoys. But we were successful in that our efforts in Kosovo forced the security forces to conceal their tanks and heavy weapons from NATO attack, and thus limited their ability to use these weapons against Kosovo Albanians. Through attacks, and the threat of attack, NATO aircraft influenced the situation on the ground.[246]

Nevertheless, whatever the level of military constraint, it did not stop Serbian forces from forcing civilians from their homes and manipulating the refugee flow to preoccupy allied military forces and create dangers of destabilisation in Albania, and more especially in Macedonia. Moreover, though constrained, VJ and MUP forces in Kosovo were able to withdraw from the province after 10 June 1999 with their flags flying, plausibly claiming to have remained undefeated.[247] This gave Milosevic an important (if temporary) psychological victory when the Allies were doing all they could to undermine his domestic political credibility.

116 The attacks also had the consequences of boosting KLA military operations against Serbian forces in Kosovo. The Allies claimed emphatically that their military operation was not undertaken on behalf of the KLA or in support of its desire for independence.[248] Nevertheless, the partial success of the Allies in constraining VJ and MUP forces in Kosovo allowed the KLA to mount a successful ground operation under the cover of the NATO air offensive. This had the effect of creating some of the air/ground synergy for which military planners would normally hope, and NATO attacks in Kosovo became more effective. All our witnesses denied that there were any formal links between NATO and the KLA over strategy. As the KLA advanced, Serbian units were forced out into the open, thus giving NATO a better hit rate as the air campaign went on.

117 Despite some success in bottling-up Serbian forces, the strikes against fielded forces in Kosovo unarguably failed in their declared primary objective of averting a humanitarian disaster. The limitations of airpower in pursuit of such humanitarian goals were clearly demonstrated, and this lesson must be learned.

118 In relation to the effectiveness of these strikes as part of the coercive strategy, the evidence of the relatively poor kill rate against Serbian armour can only lead us to the conclusion that the contribution of this axis of the bombing campaign to achieving the Alliance's overall objectives was, at best, marginal.

216  Q 136 Back

217  Q 134 Back

218  Q 871 Back

219  Moral Combat: NATO at War, transcript Back

220  Q 1046 Back

221   K. Naumann, "Democracies Fighting a War," World Defence Systems 2000, Royal United Services Institute Back

222  QQ 370-371 Back

223  Q 381 Back

224  Cm 4724, para 7.12 Back

225   Although three Serviceman were killed in follow-on operations over the summer and autumn Back

226  DoD, p 70 Back

227  Eg the US HARM or UK ALARM systems Back

228  DoD, p 1 Back

229   QQ 245 and 496-7 Back

230  Eighth Report, Session 1999-2000, European Security and Defence, HC 264 Back

231  QQ 1047-1049 Back

232  Q 1049 Back

233  Cm 4724, para 7.43 Back

234  DoD, p 91 Back

235  As part of wider discussions "with like-minded allies to develop, bi-laterally, improvements in our capabilities" (Secretary of State's interview on the BBC Radio 4's Today programme, 9 October 2000) Back

236  Q 107 Back

237  Cm 4724, para 7.17 Back

238  DoD, p 84 Back

239  Q 384 Back

240  Q 860 Back

241  Q 227 Back

242  QQ 667-8 Back

243  Q226 Back

244  Q228 Back

245  Q 861 Back

246  Cm 4724, paras 7.5 and 7.6 Back

247  Q 734 Back

248  Q 305 Back