News | Oct. 14, 2021

Read the Manual: Reversing the Trends of Failure in NATO Humanitarian Interventions with Airpower

By Michael Clark, Erik Jorgensen, and Gordon M. Schriver Joint Force Quarterly 103

Download PDF

Lieutenant Colonel Michael Clark, USMC, is a Defense Attaché at the U.S. Embassy in Monrovia, Liberia. Lieutenant Colonel Erik Jorgensen, USA, is a Joint Action Officer at the Air Land Sea Application Center. Commander Gordon M. Schriver, USN, is the Executive Officer at Airborne Command & Control Squadron (VAW) 117.

Girls wave and flash victory signs at passing helicopter during military parade in western city of Zawiya, Libya, held to mark anniversary of uprising last year that cleared way for anti-Qadhafi forces’ march on Tripoli, June 11, 2012 (United Nations/Iason Foounten)
Girls wave and flash victory signs at passing helicopter during military parade in western city of Zawiya, Libya, held to mark anniversary of uprising last year that cleared way for anti-Qadhafi forces’ march on Tripoli, June 11, 2012 (United Nations/Iason Foounten)
Girls wave and flash victory signs at passing helicopter during military parade in western city of Zawiya, Libya, held to mark anniversary of uprising last year that cleared way for anti-Qadhafi forces’ march on Tripoli, June 11, 2012 (United Nations/Iason Foounten)
Girls wave and flash victory signs at passing helicopter during military parade in western city of Zawiya, Libya, held to mark anniversary of uprising last year that cleared way for anti-Qadhafi forces’ march on Tripoli, June 11, 2012
Girls wave and flash victory signs at passing helicopter during military parade in western city of Zawiya, Libya, held to mark anniversary of uprising last year that cleared way for anti-Qadhafi forces’ march on Tripoli, June 11, 2012 (United Nations/Iason Foounten)
Photo By: Iason Foounten
VIRIN: 211013-D-BD104-012

In 1999 and again in 2011, North Atlantic Treaty Organization (NATO) leaders employed unique military means to try to achieve similarly unique strategic ends. During Operation Allied Force (OAF) in Kosovo and the military intervention in Libya, which included Operations Odyssey Dawn and Unified Protector, NATO leaders attempted to use airpower without a conventional ground force in place to protect civilians and set conditions for a lasting, self-sustaining peace.1 Neither intervention achieved these objectives, and, based on its poorer performance in Libya compared with that in Kosovo, the Alliance appears to have become less capable of using airpower for humanitarian purposes. Significant literature written about both interventions cites ways that NATO leaders can reverse this negative trend; however, none of the literature examines the military doctrine already available to Alliance leaders that, with increasing wisdom, sought to guide them to conduct these operations more effectively.2

To reverse this trend of NATO failures, Alliance leaders should more heavily weigh insights from their own military doctrine when deliberating if and how to embark on another humanitarian intervention using airpower without a conventional ground force. At a minimum, such consideration should give NATO leaders a better sense of what is realistically possible with airpower. With this better sense, they should be able to make more effective decisions on, if, and how to use the military instrument to achieve humanitarian objectives if airpower is the most robust military means available to them.

The Kosovo intervention proved ineffective at protecting civilians, but it effectively contributed to a lasting peace. The Libya intervention proved unsuccessful at both protecting civilians and contributing to a lasting peace, even though substantially more refined doctrinal recommendations were available to NATO leaders before conducting it. Had the leaders not chosen to deviate from these recommendations as wildly as they did, they likely could have come closer to achieving their strategic objectives in Libya; reversed NATO’s trend of failure in conducting humanitarian interventions primarily via airpower; and made it harder for strategic competitors, particularly the Russian and Chinese governments, to stymie future efforts at conducting similar interventions. Because modern doctrine is even more robust, well researched, and accessibly written compared with what NATO leaders had available to them before the Libya intervention, it should provide an even better guide for contemporary leaders—if they heed its advice more carefully than they appear to have done in the past.

Operation Allied Force

On the night of March 24, 1999, NATO military forces initiated Operation Allied Force, primarily employing airpower to stop an ethnic-cleansing campaign in the Federal Republic of Yugoslavia’s (FRY’s) Kosovo province. The operation initially failed to protect Kosovo’s civilians and likely incentivized those perpetrating the genocide. Months later, OAF finally helped bring about the end of this campaign, and since the operation’s conclusion, NATO and other international partners have deployed conventional ground forces as peacekeepers. As a result, Kosovo’s civilians have seen far less violence.

In the 1990s, the FRY was a Serb-majority country led by a Serbian nationalist, Slobodan Milosevic. By 1998, levels of violence in Kosovo had increased between FRY and Kosovo Serb security forces and ethnic Albanian militant groups, most notably the Kosovo Liberation Army (KLA). The FRY and Serb forces responded by initiating an ethnic-cleansing campaign in the province against its ethnic Albanian majority. KLA forces also committed war crimes during this period, but neither with the same intensity nor body count as the FRY and Serb offensive.3

To address the deteriorating situation in Kosovo, the UN Security Council (UNSC) passed UN Security Council Resolution (UNSCR) 1199 on September 23, 1998. It called for an immediate ceasefire, internationally mediated negotiations between FRY and ethnic Albanian leaders, and Milosevic to withdraw FRY security forces from Kosovo. The resolution did not explicitly authorize force if Milosevic did not comply with its demands, largely due to Russian support of his regime and Chinese obstruction.4 After repeated ceasefire agreement violations by FRY and Serb forces, NATO leaders decided to enforce UNSCR 1199 themselves with military action by initiating OAF. They claimed legitimacy for the operation by interpreting UNSCR 1199 more liberally than did Russian, Chinese, and other international leaders. In the 2 years before the NATO air campaign began, FRY and Serb forces killed more than 2,500 ethnic Albanian Kosovars.5

In April 1999, the U.S. military deployed an Army task force to neighboring Albania to assist in the operation, but diplomatic constraints kept these forces from being employed. Later that month, bolstered by growing domestic political support, NATO leaders expanded the air campaign’s targets in an attempt to reduce Serbian domestic support for Milosevic. NATO domestic support increased in part due to the UN War Crimes indictment of Milosevic and other Serbian leaders for crimes against humanity on May 27.6

Although NATO leaders were confident that OAF would compel Milosevic to capitulate quickly, he resisted. The operation also failed to protect Kosovo’s ethnic Albanian population and likely incentivized Milosevic to intensify the ethnic-cleansing campaign.7 Several factors led to OAF’s failure to protect civilians: bad weather; FRY and Serb forces effectively camouflaging themselves; FRY and Serb forces dispersing and hunkering down during airstrikes; NATO leaders conducting only limited coordination with KLA militants on the ground; and coalition airstrikes initially targeting command and control nodes, which had minimal control over the decentralized field units conducting the atrocities.8 Such futility likely incentivized Milosevic to intensify the ethnic-cleansing campaign in the hope FRY and Serb forces could remove as many ethnic Albanians as possible from Kosovo until either he was content with the level of genocide or NATO leaders finally employed a more successful strategy.9 Over the course of OAF, FRY and Serb forces killed approximately 7,000 ethnic Albanians.10

Despite OAF’s operational failures, in early June 1999, NATO finally compelled Milosevic to cease the ethnic-cleansing campaign and begin FRY’s withdrawal from Kosovo; however, OAF’s air campaign was not the only contributor to this outcome. The negotiators who delivered the terms that Milosevic ultimately accepted were representatives from the Russian and Finnish governments, the latter representing the European Union. The choice of these non-NATO negotiators was the result of robust Alliance diplomatic efforts during the operation. Ultimately, Milosevic never offered a concrete explanation for why he accepted the terms when he did. His rationale likely factored in a loss of Russian diplomatic patronage, personified by the Russian envoy sent to deliver the terms; his seeing the U.S. deployment of Army forces to Albania as a precursor to a ground invasion; a loss of Serbian national support, stemming from the expanded bombing campaign; fear of a more expansive bombing campaign if he continued to reject the June ceasefire agreement; and a belief that the June terms were likely the best he could get.11

When OAF ended on June 10, 1999, the UNSC adopted UNSCR 1244, authorizing the creation of an international peacekeeping force dubbed the Kosovo International Security Force (KFOR). Its primary mission was—and remains to this day—to foster a “safe and secure environment” to enable political efforts that promote a stable and peaceful Kosovo.12 The period immediately after OAF saw a rise in violence. By the end of 2000, this fighting had largely stopped: It was forcing an internal displacement of Serbs that effectively partitioned Kosovo along ethnic lines, and KFOR had enough troops, including Russian ones, to offer adequate security for the province. In the 5-year period after OAF ended, just over 200 fatalities were reported in Kosovo.13 Low-level violence has remained there after OAF through 2019 largely due to its unresolved political status.14

The Libya Intervention

Almost 12 years to the day after OAF began, on the night of March 19, 2011, NATO forces initiated strikes against the Libyan regime. The U.S. component was called Operation Odyssey Dawn. At the end of the month, the NATO-led Operation Unified Protector began, and just like during OAF, the combined NATO intervention in Libya primarily used airpower. Also like OAF, the Libya intervention failed to adequately protect civilians, but unlike after OAF, NATO leaders did not dispatch conventional ground forces to a peacekeeping force, leaving behind a violent and unstable environment in Libya. There, the Alliance ultimately failed to achieve any of its strategic objectives.15

The “Arab Spring” came to Libya in February 2011, when protests broke out in the city of Benghazi. These quickly became a rebellion and spread nationwide, with the Benghazi-based National Transitional Council (NTC) emerging as the voice of the rebels. Libyan security forces, led by Muammar Qadhafi, responded to the rebellion harshly, and by early March these forces initiated a counteroffensive, rapidly reasserting control over the country as they approached Benghazi.16

In response to the crisis in Libya and the growing threat to Benghazi’s citizens, the UNSC adopted UNSCR 1973 on March 17, 2011. The resolution authorized member states to take all necessary measures “to protect civilians and civilian populated areas under threat of attack in the Libyan Arab Jamahiriya.” It explicitly banned foreign troops from the country and called for the establishment of a no-fly zone over Libya “to help protect civilians.” The resolution’s primary strategic objective was finding a “peaceful and sustainable solution” to the political crisis that precipitated the violence.17 Between the 5 weeks when the protests began and the day before the UNSC approved the resolution, approximately 1,350 people had been killed in the fighting between regime forces and rebels.18

Two days later, on March 19, French forces initiated the first airstrikes to enforce UNSCR 1973. They were followed later that night by airstrikes and naval cruise missile attacks from U.S. and British militaries. Shortly thereafter the Alliance established a naval blockade to embargo arms and fighters from entering Libya, enforcing an earlier UNSCR. Within 3 days, the coalition achieved its initial objectives by stopping the regime’s armored advance on Benghazi, protecting the city’s inhabitants, and establishing a no-fly zone over Libya. Aircrew over Libya had greater success than did their Alliance counterparts over Kosovo because they experienced better weather and could find and fix regime targets more easily by using improved aircraft sensors over more open terrain.19 After achieving their initial objectives, coalition aircraft quickly shifted from targeting regime forces that were directly threatening civilians to regime forces that were threatening the overall rebel offensive.

On March 31, NATO assumed responsibility for the campaign, and Unified Protector began. Weeks later, French President Nicolas Sarkozy, British Prime Minister David Cameron, and U.S. President Barack Obama expanded their support to the rebels by supplying them with arms, essentially violating their own blockade and the UNSCR that authorized it. They did this while making seemingly paradoxical public statements to deny the coalition was supporting regime defeat. They declared, “Our duty and our mandate under UN Security Council Resolution 1973 is to protect civilians, and we are doing that. It is not to remove Qadhafi by force. But it is impossible to imagine a future for Libya with Qadhafi in power.”20

Rather than help protect civilians in Libya, such statements, along with NATO actions to impede a negotiated solution, likely did the opposite by extending the violent conflict. NATO leaders denied multiple African Union leaders’ requests to mediate negotiations between the Qadhafi regime and NTC representatives to end the fighting, even during periods of no immediate regime threat to Libyan civilians. By providing direct support and arms to the rebels, NATO likely disincentivized them from engaging in a negotiated solution. If rebel leaders thought they could defeat Qadhafi outright, then they would have little to gain by negotiating with him.21 In April 2011, NTC leaders rejected a ceasefire agreement that Qadhafi had accepted.22

Also likely extending the violent conflict were NATO leaders’ increasingly aggressive statements, NATO support for the rebels, and the International Criminal Court indictment of Qadhafi, his son, and the Libyan intelligence chief on June 27. All these actions could have made it clear to Qadhafi the only way to ensure an outcome that did not result in his incarceration or unnatural death was a regime military victory, even an increasingly unlikely one.23 Qadhafi likely felt more threatened by his International Criminal Court indictment than Milosevic did of his own. Milosevic was not living in Kosovo when he was indicted; Qadhafi lived in Libya. With all the impediments to finding a negotiated solution, the conflict persisted for more than 7 months with sustained levels of violence against Libya’s civilians. Over the course of NATO’s combined intervention, approximately 4,140 people were killed in the country by all sides of the conflict.24 According to a 2012 UN Human Rights Council report, both Qadhafi’s and rebel forces committed war crimes and broke international human rights law “in a climate of impunity.”25

In August 2011, Tripoli fell from regime control, and on October 20, NTC forces captured and killed Qadhafi after a NATO airstrike stopped his convoy. Alliance leaders denied knowing Qadhafi was in the convoy, reiterating the Alliance’s policy not to target individuals. The very next day, NATO leaders began deliberations to end the air campaign. On October 31, they declared the Libya intervention complete. The timing of the campaign’s end—so soon after Qadhafi’s death—makes their denials that the intervention did not seek regime change ring hollow.

As mentioned above, no international peacekeepers deployed to Libya when the Libya intervention ended. The near total disintegration of a well-armed Libyan state meant that almost every type of conventional weapons system, including attack aircraft and tanks, became available to Libya’s various militant groups, exacerbating the violence. Since the end of the air campaign in 2011 through 2019, more than 15,500 people have died in Libya—an average of about 1,900 fatalities a year.26

Doctrine and Decisionmaking

Since the 1990s, NATO’s understanding of how to employ the military instrument of national power, including airpower, to protect civilians and set conditions for a lasting peace has significantly evolved. One measure of this understanding is Alliance and U.S. joint military doctrine. Before OAF, minimal doctrinal guidance was available to inform NATO leaders on how to use military force to achieve these objectives; however, before the Libya intervention, considerably more insightful doctrine existed. Alliance leaders deviated from this more astute doctrine in Libya and suffered strategic failures as a result, but doctrine is not the only influence that shapes NATO decisions about employing military power. Perceptions on the use of force and domestic political concerns also inform these choices.

Prior to March 1999, there was no NATO and minimal U.S. joint doctrine to shed light on how to conduct military operations that protect civilians and enable a lasting peace. Published in 1995, U.S. Joint Publication (JP) 3-07, Joint Doctrine for Military Operations Other Than War, broadly discusses achieving these objectives. It defines one of the military operations other than war, peace enforcement operations (PEOs), as operations that use or threaten to use military force “to compel compliance with resolutions or sanctions designed to maintain or restore peace and order.” JP 3-07 does not specify if using airpower is sufficient to conduct PEOs, but the two examples of successful PEOs that it provides had extensive conventional ground intervention forces.27

Because of its objectives, OAF was a PEO. The operation ultimately furnished enough military force, alongside other instruments of national power, to compel Milosevic to comply with UNSCR 1199. The operation, however, proved insufficient to protect civilians, in large part due to a lack of extensive conventional ground forces. Only after OAF did Kosovo human security markedly improve with the introduction of the UN peacekeeping force.

As mentioned, considerably more doctrine was available to NATO leaders prior to the Libya intervention. Published in 2001, NATO Allied Joint Publication (AJP) 3.4.1, Peace Support Operations, defines success for peace support operations (PSOs) as being “related to the daily circumstances of the local populace in the former conflict area and the realization of a situation in which ‘conflicts are no longer solved using force.’” It warns that “the achievement of the military objectives and the creation of a secure environment do not guarantee the establishment of a self-sustaining peace.” It also argues that PEOs specifically, one type of PSO, should not be designed to defeat or completely destroy belligerent parties “but rather to compel, coerce, and persuade the parties to comply with a particular course of action.” One primary PEO task might be protecting civilians.28

AJP 3.4.1 also specifically addresses airpower, which includes maritime strike assets. It presents enforcing no-fly zones as an example of a stabilizing measure that “may represent the first step towards . . . negotiations for a political settlement.” The publication also promotes offensive airpower’s ability to apply “the appropriate force in any kind of conflict and to rapidly escalate or de-escalate according to the situation.” AJP 3.4.1 adds that “diplomatic activities should continue to run in parallel with military operations, and every pause in the operation should be viewed as an opportunity for further diplomatic initiatives.”29

JP 3-07.3, Peace Operations, published in 2007, does not deviate significantly from the 1995 version of JP 3-07 or AJP 3.4.1. Like its predecessor, JP 3-07.3 does not directly discuss airpower’s role in conducting PEOs, but it does state that the objective of joint fire support, including aviation fires, “is to compel or coerce the belligerents to disengage, withdraw, and comply with the mandate.”30

The Libya intervention was also a PEO, using military power to compel Qadhafi to comply with UNSCR 1973 and with a primary task of protecting civilians in immediate danger. During the intervention’s opening days, NATO military power proved sufficient to achieve this primary task. The Alliance employed airpower according to doctrine, including cruise missile strikes from naval vessels, by taking advantage of its ability to rapidly escalate to neutralize Qadhafi’s forces around Benghazi, effectively protect the city’s inhabitants, and establish a no-fly zone across the country in just 3 days.

Despite initial operational successes, NATO’s intervention in Libya failed to effectively protect civilians throughout the rest of the intervention and achieve UNSCR 1973’s objective of finding a peaceful and sustainable solution to the Libyan conflict. The intervention also did not achieve strategic success as defined in AJP 3.4.1 and JP 3-07.3. The daily circumstances of Libyans are no better—and may be worse—than they were before the NATO air campaign.31 Libyan conflicts are still solved using significant force, and the intervention did not restore peace and order. NATO leaders’ failure to achieve these objectives also validated AJP 3.4.1’s guidance that mere attainment of military operational objectives may be insufficient to guarantee a self-sustaining peace.

These failures were likely not due to purely insufficient military means but to an overall imbalance of the means NATO leaders had available to them, the ways they used their forces, and the strategic ends they eventually sought to achieve. All decisions deviated from doctrinal recommendations. NATO leaders did not build off the operation’s initial successes as a first step to initiate a negotiated solution. When Alliance airpower ensured there were no immediate threats to civilians, NATO leaders did not execute operational pauses to facilitate diplomatic efforts, and they did not rapidly activate or deactivate airpower in concert with diplomatic progress. Instead, NATO leaders expanded the scope of the air campaign to assist the rebels with destroying Qadhafi’s military capability and defeating his regime, while arming the rebels and actively impeding diplomatic efforts.

Prior to OAF and the Libya intervention, NATO leaders did receive and consider advice consistent with doctrine, but doctrine is not the only factor they evaluate when reaching a consensus on if and how to employ military force. They are influenced by many factors that likely weigh more heavily than military doctrine, such as each member state’s leaders’ and senior advisors’ individual perceptions on the use of force and domestic political concerns. Prior to OAF, the chairman of NATO’s Military Committee, German General Klaus Naumann, had been warning NATO leaders for months that they should be prepared to deploy ground troops to achieve the Alliance’s objectives.32 NATO leaders believed, however, an air campaign was the most aggressive form of intervention for which they had the domestic political support. They expected NATO citizens would find unacceptable the casualties that would likely accompany a ground intervention force.

Alliance leaders at the time also overestimated the ease with which they could employ airpower as the primary tool to compel Milosevic into compliance with a UNSCR. Four years prior, a weeks-long NATO air campaign over Bosnia did just that. This earlier victory, however, included the participation of a sizable UN conventional ground force, which was bolstered by 4,000 NATO ground troops and local military and paramilitary forces far more capable than those of the KLA.33 After OAF, the Alliance had domestic support to deploy peacekeepers.

During the Libya intervention, NATO leaders also felt they lacked the domestic support for employing a conventional ground force into Libya, under the same assumption that NATO citizens would be unwilling to accept casualties to protect Libyans. Prior to the intervention, there were intense internal debates in the White House over whether U.S. forces should intervene. Those who conceded that political realities would prohibit deploying a conventional ground intervention force and doubted the ability of Libya’s rebels to protect Libyan civilians advised against military intervention at all, particularly if the intervention led to regime change. Secretary of Defense Robert Gates led the faction advising against intervention.34 President Obama ultimately sided with Sarkozy and Cameron who, again, overestimated the ability of airpower to achieve humanitarian objectives.35 After the intervention, there was no domestic support for deploying NATO peacekeepers to Libya.

Current Doctrine

As of May 2020, the most recent NATO and joint doctrinal publications addressing the use of force to protect civilians and enable a lasting peace continue to evolve by incorporating several lessons from OAF and the Libya intervention, among other PEOs. Well researched and written, these publications are often released prior to peer-reviewed academic literature that identifies similar lessons and recommendations. Consequently, modern doctrine offers new insights on how NATO leaders ineffectively balanced military ways and means with the Alliance’s strategic ends in 1999 and 2011.

The updated title of NATO’s 2014 AJP 3.4.1, Allied Joint Doctrine for the Military Contribution to Peace Support, emphasizes the military instrument being only a contributor—and not always the most important one—to restoring lasting peace. AJP 3.4.1 dedicates an entire paragraph to enforcing no-fly zones, warning they have a limited effectiveness and could even be counterproductive when used alone to conduct PEOs. It also advises that no fly-zones “should only be adopted as part of a wider strategy.”36 OAF was part of a wider strategy that included efforts across the instruments of national power, including diplomacy, to convince Milosevic to submit to UN demands. In 2011, by contrast, NATO leaders eschewed and disincentivized diplomacy, and they failed to ever convince Qadhafi to submit to UN demands.

With respect to protecting civilians during PSOs, AJP 3.4.1 advises that the “likely consequences of military activity should not be worse than the likely consequences of inaction.”37 In Kosovo, the outcomes of military action were likely better than the outcomes of inaction. Even though OAF failed to effectively protect civilians during the operation, more civilians likely would have died had NATO never intervened at all. Milosevic probably did not concede when he did because he had completed his ethnic-cleansing campaign. After OAF, KFOR was in place to effectively protect the civilians who survived.38

Whether Libyan lives would have been better had NATO not intervened is more difficult to determine.39 What is easier to conclude is that Alliance leaders knowingly ceded the ability to have a meaningful impact on the consequences of the Libya intervention. They did this when they deliberately expanded NATO’s mission to include regime change—knowing full well that they and the UNSC lacked the political will to deploy peacekeepers to substantially influence events on the ground after Qadhafi’s regime fell.

Deposing Qadhafi and Alliance inaction, however, were not the only choices available to NATO leaders. They could have decided to strictly employ airpower within UNSCR 1973’s mandate and, in accordance with doctrine, helped facilitate a political solution by using military force to incentivize all sides to negotiate, rather than disincentivize and actively impede it.40 NATO’s interventions in Libya and Kosovo arguably exceeded the UN mandates and had other negative consequences. For example, Russian and Chinese leaders now had diplomatic cover to justify vetoing future humanitarian military interventions, most recently in Syria, on the grounds that such interventions invariably exceed their mandates.41

U.S. doctrine has also progressed since the Libya intervention ended. The 2018 version of JP 3-07.3 echoes many of the recommendations of prior U.S. and NATO publications on how to effectively conduct PEOs. It also features an eight-page annex that lists seven general operational approaches to conduct military operations with the primary purpose of protecting civilians. These approaches are taken directly from Mass Atrocity Response Operations: A Military Planning Handbook, to which JP 3-07.3 (2018) directs the reader for more information.42 The Harvard Kennedy School’s Carr Center for Human Rights Policy and the U.S. Army Peacekeeping and Stability Operations Institute published this handbook in 2010 to promote a “common military approach” that addresses the unique challenges of conducting mass atrocity response operations (MAROs) to stop genocides, crimes against humanity, war crimes, and ethnic cleansing.43 These organizations developed the handbook with representatives from U.S. European and Africa Commands, the UN, and other international organizations. Although technically available to the Libya intervention’s planners, the handbook was first incorporated into U.S. doctrine after the intervention, in the 2012 version of JP 3-07.3.44

The MARO handbook provides the clearest guidance yet on how to optimally balance military means with different operational approaches to protect civilians and enable a lasting peace. Containment is one of these approaches, and it seeks to isolate mass atrocity perpetrators with blockades and no-fly zones, making it an appropriate approach to use with airpower without a conventional ground force. Disadvantages to containment include its ineffectiveness if regime targets are not clearly identifiable from afar, its futility if regime leaders do not directly control the mass atrocity perpetrators, and the possibility it incentivizes perpetrators to accelerate mass atrocities.45 NATO leaders employed this approach during OAF, but OAF aircrew encountered all three of the listed disadvantages. NATO leaders initially employed containment during the Libya intervention, and it initially worked.

Partner-enabling is another approach that seeks to enhance local actors’ capabilities to conduct ground combat operations. This support can include advising, equipping, and providing supporting fires, and partner-enabling can be conducted with airpower without a conventional ground force. This approach’s primary disadvantages include giving enhanced capabilities and ceding operational and strategic decisionmaking to the local actors on the ground who might commit their own atrocities, have strategic interests at odds with NATO interests, or are incapable of adequately protecting civilians even with NATO assistance.46 NATO forces encountered varying degrees of these disadvantages with the KLA and Libyan rebels.

Defeating perpetrators is a third approach that directly targets perpetrator leaders and seeks to render them completely incapable of committing mass atrocities. To be effective, defeating perpetrators requires an extensive conventional ground force during the operation, and this approach may lead to regime collapse, which would require a similarly extensive conventional ground force in the postconflict phase to ultimately enable a lasting peace. The MARO handbook warns that defeating perpetrators is costly and could result in “increased levels of conflict and chaos in the country” if done improperly.47 NATO leaders did not employ this approach during OAF. Early on during the Libya intervention, Alliance leaders appeared to shift their approach from containment to a combination of partner-enabling and defeating perpetrators. NATO lacked a conventional ground force to defeat the regime during or after the intervention, and the disadvantageous qualities in the Libyan rebel groups meant that NATO lacked the requisite military strength to successfully facilitate a lasting peace after killing Qadhafi. Just as the handbook predicted a year before the Libya intervention, these deficiencies led to increased levels of conflict and chaos that remain in Libya to this day.


Operation Allied Force failed to protect civilians but contributed to Milosevic eventually complying with UNSCR 1199. The operation set conditions for the lasting peace to which NATO still contributes military forces. The Libya intervention neither effectively protected civilians nor set conditions for a lasting peace, and the violent disorder the intervention left in its wake still affects regional and Alliance security. Prior to OAF, NATO understanding of how to use force to achieve humanitarian objectives, as captured in NATO and U.S. joint doctrine, was immature and did not offer much direction to Alliance leadership. Prior to the Libya intervention, however, NATO understanding was far more refined.

It is often said that military publications, including doctrinal ones, are written in the blood of those servicemembers who made fatal mistakes. The doctrine that covers how to protect civilians, however, is written in the blood of those innocents the servicemembers failed to protect. NATO leaders should more carefully consider this doctrine when forming their opinions on the use of force and weighing their political options about using the military instrument of national power to protect civilians. When political and other factors constrain the military means that NATO leaders have available to them, such as employing airpower without a conventional ground force, greater familiarity with the doctrine of protecting civilians would likely enable leaders to balance these means more effectively with ways and strategic ends. Greater doctrinal familiarity would likely also allow leaders to more thoughtfully assess if the balance among these factors is too askew, perhaps driving them to prioritize using other instruments of national power to achieve their objectives. NATO leaders disregarding this doctrine as fervently as they did during the Libya intervention could risk the Alliance repeating the mistakes of the past, resulting in even more robust and refined doctrine borne from the blood of those whom NATO, yet again, failed to protect.

Furthermore, as the U.S. and other NATO governments’ defense establishments reorient themselves toward strategic competition with the Russian and Chinese governments, they should recognize the role these operations play. Seemingly flaunting UN limitations on military power, including limits that agree with NATO and joint doctrine, makes it easier for competitors to deny any future humanitarian interventions the legitimacy that comes with the authorization of a UN Security Council Resolution. Similarly, when NATO fails to protect vulnerable populations and enable lasting peace, particularly when it has a better understanding of how to do both, it makes a significant unforced error in the global competition for ideas and influence. JFQ


1 We do not consider the North Atlantic Treaty Organization’s (NATO’s) Operations Deny Flight and Deliberate Force in Bosnia and Herzegovina from 1993 to 1995 to meet these criteria. Although NATO as an alliance did not employ conventional ground forces to complement the air campaign, NATO member countries did employ conventional ground forces as part of the United Nations (UN) Protection Force to create safe areas to protect civilians before and during the NATO air campaigns. The UN Protection Force also coordinated with NATO aircraft during the operations.

2 See Nicholas J. Wheeler, Saving Strangers: Humanitarian Intervention in International Society (Oxford: Oxford University Press, 2000), 242–284; Alan J. Kuperman, “A Model Humanitarian Intervention? Reassessing NATO’s Libya Campaign,” International Security 38, no. 1 (2013), 105–136; Alan J. Kuperman, “Obama’s Libya Debacle: How a Well-Meaning Intervention Ended in Failure,” Foreign Affairs 94, no. 2 (2015), 66–77; Arash Heydarian Pashakhanlou, “Decapitation in Libya: Winning the Conflict and Losing the Peace,” The Washington Quarterly 40, no. 4 (2017), 135–149; Astrid Stuth Cevallos and Bryan Frederick, “Implementing Safe Areas: Lessons from History,” Survival: Global Politics and Strategy 60, no. 6 (2018), 159–180; Arash Heydarian Pashakhanlou, “Air Power in Humanitarian Intervention: Kosovo and Libya in Comparative Perspective,” Defence Studies 18, no. 1 (2018), 39–57; David Brown, “Kosovo and Libya: Lessons Learned for Limited Humanitarianism?” Comparative Strategy 38, no. 5 (2019), 467–482; Martin A. Smith, “Taking Stock After Twenty Years: The Mixed Legacy of Kosovo,” Comparative Strategy 38, no. 5 (2019), 483–496; Anthony M. Schinella, Bombs Without Boots: The Limits of Airpower (Washington, DC: Brookings Institution Press, 2019), 43–95, 223–286.

3 Human Rights Watch, “Human Rights Watch Investigation Finds: Yugoslav Forces Guilty of War Crimes in Racak, Kosovo,” January 29, 1999, available at <>.

4 UN Security Council Resolution 1199, “Situation in Kosovo, Federal Republic of Yugoslavia,” September 23, 1998, available at <>; Jason E. Fritz, Stability Operations in Kosovo 1999–2000: A Case Study (Carlisle, PA: U.S. Army Peacekeeping and Stability Operations Institute, 2018), 18, available at <>.

5 Steven Erlanger and Christopher S. Wren, “Early Count Hints at Fewer Kosovo Deaths,” New York Times, November 11, 1999, available at <>.

6 Benjamin S. Lambeth, NATO’s Air War for Kosovo: A Strategic and Operational Assessment (Santa Monica, CA: RAND, 2001), 38, available at <>; Fritz, Stability Operations in Kosovo, 31–32.

7 Elaine M. Grossman, “‘What If . . . We Were Still Fighting in Winter Weather?’: For U.S. Commander in Kosovo, Luck Played Role in Wartime Success,” Inside the Pentagon 15, no. 36 (1999), 12–13; Fritz, Stability Operations in Kosovo, 31; Phil Haun, “Attacking Fielded Forces: An Airman’s Perspective from Kosovo,” Joint Force Quarterly 95 (4th Quarter 2019), 74, available at <>.

8 Pashakhanlou, “Air Power in Humanitarian Intervention,” 43; Haun, “Attacking Fielded Forces,” 71.

9 Wheeler, Saving Strangers, 269; Brown, “Kosovo and Libya,” 473; Lambeth, NATO’s Air War for Kosovo, 28; Julie Kim, Kosovo, Review and Analysis of Policy Objectives, 1998–June 1999, RL30265 (Washington, DC: Congressional Research Service, July 21, 1999), 7–8; Grossman, “What If . . . We Were Still Fighting in Winter Weather?” 13; Tony Mason, “Operation Allied Force, 1999,” in A History of Air Warfare, ed. John Andreas Olsen (Washington, DC: Potomac Books, 2010), 234–236.

10 Erlanger and Wren, “Early Count Hints at Fewer Kosovo Deaths.”

11 Lambeth, NATO’s Air War for Kosovo, 67–86; Fritz, Stability Operations in Kosovo, 32–33; Wheeler, Saving Strangers, 283–284; Michael J. Boyle, Violence After War: Explaining Instability in Post-Conflict States (Baltimore: Johns Hopkins University Press, 2014), 175; Benjamin S. Lambeth, “Lessons from Modern Warfare: What the Conflicts of the Post–Cold War Years Should Have Taught Us,” Strategic Studies Quarterly 7, no. 3 (2013), 39; Stephen T. Hosmer, The Conflict over Kosovo: Why Milosevic Decided to Settle When He Did (Santa Monica, CA: RAND, 2001), available at <>; Mason, “Operation Allied Force,” 244.

12 JFC Naples–Allied Joint Force Command Naples, “Mission,” available at <>.

13 Boyle, Violence After War, 175–178; Brown, “Kosovo and Libya,” 477–478.

14 Heidelberg Institute for International Conflict Research (HIIK), Conflict Barometer 1999 through Conflict Barometer 2001 (Heidelberg, Germany: HIIK, Department of Political Science, University of Heidelberg), available at <>; HIIK, Conflict Barometer 2002 through Conflict Barometer 2019 (Heidelberg, Germany: HIIK, Department of Political Science, University of Heidelberg).

15 Kuperman, “Obama’s Libya Debacle,” 67; Dianne Pfundstein Chamberlain, Cheap Threats: Why the United States Struggles to Coerce Weak States (Washington, DC: Georgetown University Press, 2016), 143.

16 Human Rights Watch, “Libya: Governments Should Demand End to Unlawful Killings,” February 20, 2011, available at <>.

17 UN Security Council Resolution 1973, “On Establishment of a Ban on Flights in the Libyan Arab Jamahiriya Airspace,” March 17, 2011, available at <>.

18 Clionadh Raleigh et al., “Introducing ACLED: An Armed Conflict Location and Event Dataset: Special Data Feature,” Journal of Peace Research 47, no. 5 (2010), 651–660.

19 Pashakhanlou, “Air Power in Humanitarian Intervention,” 51.

20 Barack Obama, David Cameron, and Nicolas Sarkozy, “Libya’s Pathway to Peace,” New York Times, April 14, 2011.

21 Kuperman, “A Model Humanitarian Intervention?” 113–115; Sarah Brockmeier, Oliver Stuenkel, and Marcos Tourinho, “The Impact of the Libya Intervention Debates on Norms of Protection,” Global Society 30, no. 1 (2016), 122.

22 Peter Kenyon, “Libyan Rebels Reject AU Cease-Fire Plan,” National Public Radio, April 11, 2011, available at <>.

23 Schinella, Bombs Without Boots, 293.

24 Raleigh et al., “Introducing ACLED.”

25 Human Rights Council, Report of the International Commission of Inquiry on Libya, UN doc. A/HRC/19/68 (Geneva: United Nations, March 8, 2012), 21, available at <>.

26 Raleigh et al., “Introducing ACLED.”

27 Joint Publication (JP) 3-07, Joint Doctrine for Military Operations Other Than War (Washington, DC: The Joint Staff, June 16, 1995), III-13.

28 Allied Joint Publication (AJP) 3.4.1, Peace Support Operations (Brussels: NATO, July 2001), 2-2, 2-4, 4-5.

29 Ibid., 5-4, 6-7, 6-15.

30 JP 3-07.3, Peace Operations (Washington, DC: The Joint Staff, October 17, 2007), III-5.

31 Kuperman, “A Model Humanitarian Intervention?”; Christopher S. Chivvis, Toppling Qaddafi: Libya and the Limits of Liberal Intervention (New York: Cambridge University Press, 2014), 175.

32 Elaine M. Grossman, “Naumann: Be Ready to Use Ground Troops: Senior General Warned NATO Repeatedly of Prolonged Kosovo Conflict,” Inside the Pentagon 15, no. 16 (1999), 1, 16–17; Elaine M. Grossman, “Naumann: Coercing Miloseyic Initially ‘Did Not Work’: NATO General Says Alliance Is Prepared for Long Haul in Kosovo,” Inside the Pentagon 15, no. 17 (1999), 1, 15–16.

33 Wheeler, Saving Strangers, 256, 268; Grossman, “Naumann: Be Ready to Use Ground Troops,” 17; Grossman, “What If . . . We Were Still Fighting in Winter Weather?” 12.

34 Michael Hastings, “Inside Obama’s War Room,” Rolling Stone, October 13, 2011, available at <>; David E. Sanger and Thom Shanker, “Gates Warns of Risks of a No-Flight Zone,” New York Times, March 2, 2011, available at <>.

35 Brockmeier, Stuenkel, and Tourinho, “The Impact of the Libya Intervention Debates,” 119–120, 132; Benedetta Berti, “Forcible Intervention in Libya: Revamping the ‘Politics of Human Protection?’” Global Change, Peace, and Security 26, no. 1 (2014), 39; Hastings, “Inside Obama’s War Room.”

36 AJP-3.4.1, 2-1, 4-4.

37 Ibid., 4-5.

38 Andrew Cottey, “The Kosovo War in Perspective,” International Affairs 85, no. 3 (2009), 606–607; Wheeler, Saving Strangers, 269–270, 283–284.

39 Kuperman argues in “A Model Humanitarian Intervention?” 116–121, that far fewer Libyans would likely have died had NATO not intervened as it did. Chivvis argues in Toppling Qaddafi, 175, that Libya would be better off with Qadhafi gone only if long-term violence was not the result, which it has been in the 7 years since he published the book.

40 Kuperman, “A Model Humanitarian Intervention?”; Pashakhanlou, “Decapitation in Libya”; Chamberlain, Cheap Threats, 143; Jack Goldsmith and Matthew Waxman, “The Legal Legacy of Light-Footprint Warfare,” The Washington Quarterly 39, no. 2 (2016), 9; Cevallos and Frederick, “Implementing Safe Areas,” 169; Chivvis, Toppling Qaddafi, 90; Sarah Brockmeier, “Lessons in Statecraft Still to Be Learned 5 Years After the Libya Intervention,” War on the Rocks, March 16, 2016, available at <>.

41 Berti, “Forcible Intervention in Libya,” 39; Brockmeier, Stuenkel, and Tourinho, “The Impact of the Libya Intervention Debates,” 131; Brown, “Kosovo and Libya,” 475; Sui-Lee Wee, “Russia, China Oppose ‘Forced Regime Change’ in Syria,” Reuters, January 31, 2012, available at <>; Brockmeier, “Lessons in Statecraft.”

42 JP 3-07.3, B-1–B-8.

43 Sarah Sewall, Dwight Raymond, and Sally Chin, Mass Atrocity Response Operations: A Military Planning Handbook (Cambridge, MA: Carr Center for Human Rights Policy, Harvard Kennedy School, and U.S. Army Peacekeeping and Stability Operations Institute, 2010), 5–9, 103–105, available at <>.

44 JP 3-07.3, B-1–B-10.

45 Sewall, Raymond, and Chin, Mass Atrocity Response Operations, 55–56, 82–84.

46 Ibid., 80–81.

47 Ibid., 84–87.