FROM BAGHDAD TO KABUL:

THE IMPACT OF INTERNATIONAL HUMANITARIAN LAW ON COALITION AIRPOWER AND THEIR CONSEQUENCES FOR HUMANITARIAN ACTION

The author wishes to thank Professor Rex Brynen for his much appreciated academic guidance.

The author has worked as a civilian for two United Nations peacekeeping missions (Cambodia and Haiti), as an unexploded ordnance survey consultant for Handicap International in Laos, and as a landmine education coordinator and volunteer workshop organizer in International Humanitarian Law for the Canadian Red Cross, Ottawa-Carleton branch.

The last decade has witnessed a substantial increase in the use of airpower for peace enforcement. Coalition airstrikes in the 1991 Gulf War, the use of NATO airpower against Yugoslavia in 1999, and the use of US-led airpower in the conflict in Afghanistan, are examples of this trend.

The use of airpower presents important implications for the laws of armed conflict while having consequences for the internationally-sanctioned delivery of humanitarian relief to war victims. What have been the consequences of coalition air campaigns upon civilian populations and civilian objects? Are humanitarian operations possible during coalition air campaigns?

The last decade has witnessed a substantial increase in the use of military airpower for peace enforcement and other political objectives, such as United Nations Security Council (UNSC) enforcement of no-fly zones, safe havens and sanctions. Coalition airstrikes in the 1991 Gulf War, the use of airpower by the North Atlantic Treaty Organization (NATO) and especially the United States (US) against Bosnian-Serbs in 1995 and against Yugoslavia in 1999, and the US reliance on airpower in the 2001-02 conflict in Afghanistan, are all examples of this trend.

During civil and international wars, specialized humanitarian agencies have for decades delivered emergency relief to civilian victims of armed conflict. Although the use of modern airpower has proven largely effective for achieving political objectives, increasing means of coercion from the air raise important questions about state compliance to the laws of armed conflict and about the consequences of such airpower to internationally-sanctioned relief operations to war victims. This study will examine both of these questions, identifying legal gaps and humanitarian tensions.

In the first major section, the paper will examine the behavior and results of coalition airpower against Iraq, Yugoslavia and Afghanistan in order to determine the extent of coalition compliance with International Humanitarian Law (IHL). Politico-military tensions about targeting decisions and differences in national interpretation of IHL compliance will be studied, including the targeting of dual-use civilian-military facilities. Attention will be paid to the American interpretation of and coalition compliance with Additional Protocol I of the Geneva Conventions.

Do US-led airpower coalitions meet the legal obligations of IHL? Has the use of airpower increasingly limited civilian casualties since the Gulf War? While research focuses on the means and methods of (air) warfare, jus in bello, no examination will be made of the legality of using force or of the right to militarily intervene, jus ad bellum.

In the second major section, examination will be made of the implications of the above three air campaigns upon the operations of international relief agencies. Airstrikes present consequences for the safety of civilians, whether internally-displaced persons (IDPs) or local and international aid workers. A brief analysis of the humanitarian value of US military aid drops over Afghanistan in 2001 will be made. While reviewing the work of specialized UN relief agencies, particular attention will be paid to the operations of the International Committee of the Red Cross (ICRC) as both the guardian of IHL and as a body globally-mandated to provide relief and protection to victims of armed conflict.

Can relief agencies effectively and impartially operate during airstrikes sanctioned by the UN or NATO? Does humanitarian work remain possible within the constraints of today’s use of airpower technology? Have coalitions increasingly learned to heed an operational ability to relief agencies since the Gulf war?

This paper also challenges the notion among some proponents of airpower that airpower alone can provide solutions to political problems at a lesser political, civilian and humanitarian cost. Air campaigns without ground-based intelligence or ground forces increase the vulnerability of civilians and aid workers to targeting failures.

While much research has focused on assistance to victims of conflict who succeed as refugees in escaping areas of conflict, there has been insufficient research examining the access of emergency relief operations during air campaigns. Research will conclude that, despite the legal rights of impartial aid agencies to access victims of armed conflict, ambiguities in the laws of war permit decisions by military lawyers and commanders to argue the military necessity of certain actions which consequently obstruct effective humanitarian efforts. It has not however meant that humanitarian relief operations during airstrikes are impossible, as will be revealed. The involvement of political scrutiny, coalition pressures, legal concerns, technological advances and media exposure of the conduct of air warfare have combined to increasingly limit incidents of collateral damage.

This paper forms part of a thesis on coalition airpower, its legal compliance and effects on relief operations, submitted during graduate research at McGill University, Montreal, Canada.

LAW SECTION: THE CONSEQUENCES OF AIRPOWER UPON INTERNATIONAL HUMANITARIAN LAW

The Geneva Conventions and their Additional Protocols seek to limit military action to what is militarily necessary during battle, guided by universally-accepted principles of proportionality, discrimination and humanity. Although it signed Protocol I in 1977, the US announced in 1987 that it would not ratify but would consider itself legally bound to those rules which reflected customary international law. In its arguments against ratification of Protocol I, the US recognized that, in lowering previous standards by which combatants were required to distinguish themselves from civilians, the ambiguity favoured protection of terrorists and their struggles. [2]

Since April of 2001, when France deposited its documents of ratification to the Swiss government, the US has become the only permanent Security Council member not party to Protocol I. Both Britain in 1998 and Germany in 1991 had indicated reservations during their ratification pertaining to the ambiguity of what in Article 52.2 constitutes “effective contribution” and “definite military advantage” in deciding an attack. [3]

The 1976 US Air Force Pamphlet and the 1991 US Rules of Engagement Pocket Card during Operation Desert Storm reflected provisions of Protocol I in regards to distinction, proportionality and military necessity. The 1976 US instructions on the legal limits of applying force would later repeat almost verbatim the provisions of Article 57 of Protocol I. [4]

While IHL remains static, a belligerent’s Rules of Engagement (ROE), which comprise military, legal and political objectives, [5] can shift and adapt to changing circumstances during conflict, thus politically altering what is militarily necessary. IHL compliance becomes positively affected by a state’s military ROE.

During Vietnam, ROE and targeting decisions were complicated by politicization from the White House and the Pentagon, “a case of airpower being undermined by civilian control of air operations, with images of President Johnson and Secretary of Defense McNamara on their knees on the Oval Office selecting targets.” [6]

US ROE were reviewed during Operation Desert Storm, after Iraq began the illegal practice of human shielding, [7] placing anti-aircraft guns on civilian property like apartment buildings, schools and hospitals, [8] and commingling military assets with civilian people, [9] in violation of Articles 51.7 and 58 (b) of Protocol I.

ROE for NATO pilots during Operation Allied Force in 1999 only later evolved to include the striking of north Yugoslav dual-use targets: bridges, television transmitters and the Belgrade power grid, later discussed.

During the Afghan air campaign, ROE changed to reflect ground realities dictated by the advance of Northern Alliance Forces southward. ROE also acknowledged the presence of relief convoys operating along certain roads. [10]

Case study examination will analyze and compare coalition airpower behavior and results, statistics on civilian casualties, levels of use of Precision Guided Munitions (PGMs) and collateral damage from targeting errors. Legal, media and NGO critiques on the effects of airpower will be examined.

It must be noted that no consistent, independent and transparent sources exist for the compilation of figures on civilian casualties. Periods of airpower, available ground facts, and types and numbers of bombs used varied between case study, complicating effective comparisons.

The Gulf War witnessed the heaviest aerial bombardments since World War II, in a campaign which received greater attention to Protocol I than any previous conflict. [11] The introduction of stealth fighters, satellite imagery and precision-guided munitions over Baghdad in 1991 was popularized by sensationalist CNN coverage, welcomed and promoted by the Pentagon. Cruise missiles said to be entering the front door of Iraqi command and control centers, or descending air vents to explode on a desired floor, have done much to artificially and unrealistically elevate the public’s expectations of surgical aerial bombing in general [12] and of zero-casualty warfare in particular.

During the 43-day UN-authorized Operation Desert Storm, between 3000 [13] and 3200 Iraqi civilians were reported killed, [14] compared to some 146 US military combat deaths [15] and 17 British military combat deaths. [16]

During 72,000 sorties, 8.8 percent of the 84,700 tons of bombs dropped were PGMs. [17] Human Rights Watch (HRW) stated without citing its source that the estimated accuracy rate of the remaining dumb bombs was 25 percent. [18] According to official US reports, even among precision munitions dropped by F-117 Nighthawks, 20 percent of their precision bombs did not hit their targets: “about 400 tons of bombs from the F-117s alone may have caused collateral damage.” [19]

Iraqi capacities to fill its skies with metal of all kinds soon prevented the coalition’s initial strategy of radar-evading low-flight attack tactics. After the Royal Air Force had lost five aircraft in seven days, a switch of ROE resulted in high-altitude flights dropping 1000-pound from 20,000 feet, offering pilot safety while increasing the chances of collateral damage from less accurate ordnance drops. [20]

Despite the optimism of Gulf commanders estimating target accuracy at 80 percent, a congressional report later stated that “the real hit rate may have been as low as 41 percent.” [21] In a 1997 US General Accounting Office evaluation of the effectiveness of Operation Desert Storm airpower, conclusions included the following: “weapon system performance were overstated, misleading, inconsistent with the best available data, or unverifiable.” [22]

Unlike during the Vietnam war, military commanders in the Gulf theatre had the necessary freedom to plan and execute the campaign: “not once did Pentagon officials reverse decisions from the Black Hole [23] about what weapons to use, what targets to strike, and how and when to attack them.” [24] However, after the bombing of the Amiriya bunker in week 4, the Pentagon took back some targeting control from field commanders. [25] Approximately 25 percent of combat sorties returned without having dropped their ordnance, [26] an indication of ROE complementing both military economy of force and IHL efforts to minimize civilian casualties.

The coalition had no ground spotters in Iraqi urban centres to permit better targeting accuracy, although some small teams of British and US Special Forces operated inside western Iraq on scud-busting operations. [27] Among the most criticized coalition strikes were those against the baby milk factory in late January and against the Amiriya command bunker/civilian shelter on 13 February, the latter killing between 300 and 400 people. [28] Spain called for an international inquiry into the bunker attack and an end to blanket bombing. [29] In the case of the former, military intelligence claimed that the target was a biological weapons factory. [30]

Military intelligence regarding the Amiriya site revealed that the facility, whether used for civilian or military purposes, was covered in camouflage, protected by barbed wire and armed guards at the door, and that it had been converted to a command and control centre during the Iran-Iraq war. [31] Additionally, intercepted command communications originating from the building indicated its military use. [32] Iraqi President Saddam Hussein was known throughout the war to commingle civilian and military objects and people in order to provide some deterrence. [33] Also, after a month-long air campaign, Iraq had begun the illegal practice of human shielding, hiding weapons in schools, hospitals and mosques, while placing anti-aircraft guns on the rooftops of public buildings. [34]

Another highly criticized behavior of coalition airpower regarded the unnecessary level of military retribution used against fleeing Iraqi troops leaving Kuwait with the spoils of war along the Basra highway, later dubbed the “Highway of Death.” Although it is not illegal in war to attack retreating troops, a disproportionate level of force was used against the Iraqi army, the results of which were globally publicized by the media.

According to a Greenpeace study, Iraqi water pumping stations, treatment plants and distribution facilities were however not directly targeted by coalition forces. [35] Despite UN cries of epidemics [36] caused by the intensity of coalition airstrikes upon dual-use facilities, “the ICRC team itself did not find any evidence confirming pessimistic media reports of widespread epidemics”. [37]

Following the withdrawal of OSCE monitors [38] from Kosovo and a failed attempt at mediation with Yugoslav President Slobodan Milosevic by US envoy Richard Holbrooke on 23 March, NATO Secretary General Javier Solana authorized the launch of airstrikes the next day. During the 78-days of Operation Allied Force against Yugoslavia in 1999, the US-led coalition, involving aircraft from 14 of the 19 NATO members, operating from 24 European bases and three aircraft carriers, conducted 37,225 sorties, one-third of which were strike and air defence suppression sorties, non-US sorties constituting about 39 percent of this total. [39] In what was expected to be a short, [40] expensive cruise-missile oriented campaign of possibly a dozen days, [41] NATO had but 50 pre-approved targets. [42]

Collateral damage included the deaths of some 500 civilians killed in 90 separate incidents of which 78 were investigated by HRW. [43] Some 820 civilians were wounded. [44] Some 600 Serbian military and special police were killed, according to Yugoslavia government officials. [45] HRW noted that Yugoslav civilian death figures were three times higher while those of the US government were lower than the HRW result. [46]

Although the beginning of the campaign saw a 90 percent count in use of PGMs, clear weather began to permit greater use of dumb bombs, culminating with a 34 to 37 percent PGM-use over the campaign period. [47] By the end of the air campaign’s first month, poor weather had caused the cancellation of over 50 percent of strike sorties during 20 days. [48] Weather limited the number of sorties in a considerable way, forcing “NATO to cancel at least half of its total number of planned sorties on 39 days of the 78-day campaign, and allowed unimpeded air strikes on only 24 of 78 days.” [49] Two-thirds of US strikes occurred at night. [50]

Two NATO planes were shot down, one US F-117A stealth fighter on Day 4 and one US F-16 fighter jet; both pilots were recovered. [51] NATO bombs erroneously hit three embassies (China, Sweden, Switzerland) and two foreign countries (Albania and Bulgaria) during the campaign. [52] In May of 1999, NATO strikes mistakenly hit a KLA outpost, killing one rebel. [53] Other NATO bombing errors and the NATO targeting and striking of dual-use civilian-military targets are addressed in detail in later sections.

To ensure the safety of foreign dignitaries in Belgrade during mediation efforts, NATO limited its airstrikes on the city during the visits of Russian and Finnish envoys. [54] “NATO pilots were forced to drop millions of dollars of ordnance in the Adriatic and on open countryside because they could not find their targets or engage them properly due to bad weather and the aerial rules of engagement imposed by politicians.” [55]

Airpower technology without sufficient ground intelligence made targeting vulnerable to human error. Dependence upon airpower alone to achieve NATO’s objectives proved more difficult than expected. [56] After the first six weeks, there were more Serb forces in Kosovo than before the campaign. [57] This military difficulty caused a rift in NATO planning policies, with Americans preferring the punishing of northern Yugoslav command and control centers, including Milosevic’s Presidential palace [58] in Belgrade, while the Europeans wanted bombing concentrated on southern Yugoslavia, upon Serb forces in Kosovo. [59] The Europeans were far short of such advanced US technological airpower capacities and were more fearful of collateral damage [60] and its political consequences upon continued refugee flows and regional economic instability.

After the 78-day campaign and the negotiated withdrawal of Serb forces, the number of retreating tanks and troops [61] proved NATO’s inability to win a campaign purely by aerial means in Kosovo without a ground option. NATO ultimately decided to repeat the World War II practice of hitting targets that would maximize the adversary’s political vulnerability to public pain. The change in NATO ROE, to strike more dual-use targets, attempted to increasingly deteriorate the moral fabric of Yugoslav society to support its own leadership, hopefully forcing Milosevic to the table, later discussed. Such a strategy could only work against democratic societies, not Iraqi or Taliban dictatorships with little accountability to and regard for their citizens.

Ultimate military victory was never permitted against Yugoslavia, NATO politicians, Britain notwithstanding, and the US government constantly having to pacify Clark’s desires as an army soldier to send in ground troops to engage Serb forces. The NATO objective was strictly to use sufficient force to create the political conditions, unavailable during the Rambouillet talks of 1998, that would complement diplomatic efforts, achieved by Finnish President Maarti Ahtisaari and former Russian Prime Minister Victor Chernomyrdin in June.

Amnesty’s June 2000 report criticized NATO’s conduct of airstrikes from 15,000 feet, causing risk of incidental loss of civilian life. [62] While AI accused NATO of violating international law, the June 8, 2000, Final Report by the ICTY found no basis for further investigation, [63] nor for charging NATO or its leaders with war crimes: NATO had not violated the laws of armed conflict since “there was no deliberate targeting of civilians or unlawful military targets.” [64] The ICTY Final Report did remark that NATO, in answering specific questions about specific incidents, replied only in “general terms and failed to address the specific incidents.” [65]

It should also be noted that the ICTY was created in 1993 by resolution of the UNSC, a political body comprising veto-wielding states which also happened to be the leading users of airpower in 1999. As the ICTY reports to the UNSC and not to a separate UN ombudsman or independent world body, the court’s absolute independence remains in question.

A US-UK coalition with NATO support began bombing the Taliban regime on 07 October, 2001. The objective of Operation Enduring Freedom differed from the other three air campaigns which had publically sought the withdrawal of forces from a distinct territory. Enduring Freedom sought the annihilation and replacement of a regime, the Taliban and its largely foreign-based Al-Qaeda allies.

The air campaign, executed from above 10,000 feet and supported by three US aircraft carriers, [66] went through several phases of action, depending on the advance of Northern Alliance forces, a loose military grouping comprising various decentralized, ethnically-based opposition clans, most of which were formed in the 1980s as anti-Soviet Mujahedin forces. The Northern Alliance was comprised largely of non-Pashtun ethnicities like Uzbeks, Tadjiks, Hazaras. Once the Northern Alliance had taken control of Kabul on 13 November, US bombing diminished. While concentrating on isolated pockets of resistance in Kunduz, Kandahar, Tora Bora and Gardez, the air campaign entered a period more sensitive to the need for accuracy given the increased presence of foreign troops, foreign media and expatriate relief workers. For the purposes of comparison, given that a prime reason for Operation Enduring Freedom was the termination of power of the Taliban regime, largely achieved by the New Year, [67] analysis will be limited to the bombing period ending in early February of 2002, although the defeat of the Taliban can be estimated to have taken place by December 8-9, when the regime’s politico-spiritual leader, Mullah Omar, fled his Kandahar base.

According to HRW, at least 1,000 civilians were killed by coalition bombing, while a Reuters news agency estimate counted 982 civilian deaths in 14 incidents. [68] The Associated Press review of bombings estimated the civilian death toll at between 500 and 600. [69] According to estimates by a senior MSF worker based on reports from hospitals and field workers around the country, some 2000 to 3000 civilians were killed. [70] HRW is expected to publish findings from field investigations in September, 2002; until then its claims are inconclusive. [71]

In an exhaustive 18 January study carried out by the Project on Defence Alternatives (PDA) at the Commonwealth Institute, a public policy research organization dealing with defence issues in Massachussetts, it was estimated that between 1,000 and 1,300 civilians may have been victims of collateral damage. [72] The rate of civilians killed per bomb dropped was four times higher than during the NATO-led Kosovo campaign, a result of a switch to less precise GPS-guided from laser-guided bombs. [73] There was also a greater reliance on the dropping of bombs by bombers than by fighter jets; in Operation Allied Force only 2 percent of bombs dropped were launched from bombers while 11 percent of bombs dropped over Afghanistan through until 10 December originated from high altitude bombers. Bombing by fighter jet and by laser-guided targeting thus seems to afford the greatest likelihood of bombing accuracy and the least number of civilian casualties. While fighter jets carry fewer bombs, they also cost less than bombers.

In a more contested compilation of civilian casualties, Economics professor Marc Herold, of the University of New Hampshire, concluded on 10 December that 3,767 civilians had been killed, an average of 62 civilian deaths per bombing day. [74] Regarding methodology, sources for the PDA study were chosen from non-Asian, Western media (BBC, Reuters, Agence France-Presse, Associated Press, The Independent, The Times, The Guardian) [75] while the Herold study depended on reports from Indian, Pakistani and European media [76] including internet surfing practices [77] and other reports such as the Afghan Islamic Press, a pro-Taliban news agency based in Pakistan. [78]

Herold charged the Pentagon with waging a war without witnesses. [79] On 11 October, the Pentagon was said to have purchased exclusive rights to all satellite images from Space Imaging, a US company with information capable of surveying and corroborating claims of collateral damage from coalition bombings. [80] It may also be possible that such a measure was meant to prevent enemy knowledge of the position of coalition allies, coalition ground forces and laser-wielding forward air controllers.

William M. Arkin, a columnist for the Washington Post and former army intelligence analyst for MSNBC, working as military advisor to HRW while teaching at the US Air Force School of Advanced Airpower Studies, noted some 300 “incidents” during the campaign, a third of which it was estimated needed further attention because of civilian casualties. [81] HRW had two researchers on the Pakistani-Afghan border for 11 weeks compiling a general picture of targeting failures and errors. [82]

According to one American reporter, 2 out of three bombs dropped during the air campaign were PGMs. [83] In the first six weeks of airstrikes, over 60 percent of bombs were precision-guided munitions. [84] An analysis of the first five days of bombing determined a 43 percent use of smart bombs, which includes satellite-guided and laser-guided bombs. [85] The first two months of bombing saw 72 percent of bombs dropped by the US Air Force as being PGMs, and 60 percent, by the US Navy. [86] During the first week of February, US General Tommy Franks, the commander of Operation Enduring Freedom in Tampa, claimed that among the 18,000 bombs dropped, 10,000 were precision munitions. [87]

The PDA study estimated that 56 percent of bombs dropped by 10 December were considered to be PGMs. [88] While the Kosovo action breakdown of used bombs witnessed two-thirds being laser-guided and one-third being unguided, the Afghanistan action, until 10 December, saw 20 percent of bombs being laser-guided, 40 percent being GPS-guided and 40 percent being unguided. [89] The Pentagon announced that two-thirds of Navy fighter jets returned to their carriers without having dropped their bombs, [90] another example, as during the Gulf War, of the complimentarity of concerns about military economy of force (accuracy vs wasted valuable, limited resources) and inadvertent IHL concerns.

One of the reasons for civilian casualty rates may have been the difficulty by bombers and ground spotters of distinguishing soldiers from civilians in an impoverished population of largely non-uniformed or untraditionally uniformed combatants dressed similar to their civilian counterparts. It is not certain that US satellites can differentiate armed combatants in civilian dress from unarmed civilians.

During the first night of US airstrikes against the Taliban regime in Afghanistan, much to the discomfort of Defence Secretary Donald Rumsfeld, efforts to minimize incidental loss of civilian casualties were noted when US military lawyers of the Central Command vetoed Central Intelligence Agency (CIA) permission [91] to allow a missile strike by an unmanned Predator drone against a suspected convoy alleged to include Taliban leader Mullah Omar. [92] Tests for the carrying and firing of missiles by the remote-controlled aircraft had only been completed the previous February. [93]

As in the Gulf War, [94] the US decided against a compilation of military and civilian casualties, [95] although “the Air Force created a special assessment team at its air operations center in Riyadh to look at cases of possible civilian losses and other unintended damage.” [96] While investigations regarding technological limits, malfunctions and weather problems are one concern, intentionally bombing a target later deemed a mistake due to human intelligence presents a more embarrassing concern. Neither the US nor Afghan governments intend to attempt a tally; the UN and ICRC have not compiled figures. [97]

The Pentagon has investigated some bombing errors and targeting failures, but only two bombing mishaps involved civilian deaths, according to a 29 March Pentagon report. [98] The often repeated Pentagon claim has been that “any civilian deaths were the result of unavoidable ‘collateral damage’ from attacks on military targets or were people killed by bombs that went astray.” [99] Among US military enquiries was the deliberate striking of an ICRC warehouse compound on two occasions. [100] A CNN interview claimed that the Taliban had placed an anti-aircraft battery next to the ICRC warehouse. [101] Falsely blaming the ICRC for not having notified them of the warehouse location, which it had, [102] a Pentagon investigation later revealed that a US Air Force General had exceeded his authority in ordering the intentional second strike, a change in its original claim of human error. [103]

Despite claims by US Defence Secretary Donald Rumsfeld that American weaponry was “probably 85-90 percent reliable,” [104] any ability to confirm such estimates is very difficult, requiring further investigation results from independent reporters and NGOs. General Franks called the campaign the most accurate war in US history. [105] Any compilation of war casualties would need to account for the Afghan habit of immediately collecting and burying the dead.

US targeting information depended heavily on ground-based cooperation with and intelligence from regional warlords, many with political agendas unknown or misunderstood by the coalition. In a society born out of war, Afghan military principles of discrimination, proportionality and restraint did not compare to those of their Western allies. To collect intelligence, 12-member US Special Operations Forces, CIA teams and British SAS began co-mingling themselves into friendly Afghan forces in October. By mid-November, ten such teams were deployed, and by 08 December, 17 teams existed. [106] By early November, 80 percent of US combat sorties were devoted to directly supporting opposition forces in the field. [107] While targeting information may have been accurate, the reasons for targeting certain groups, convoys, caves, buildings and installations may have been detrimental to civilians. This manner of proxy belligerence came at a high price to civilians, making the US increasingly vulnerable to intelligence error or deliberate warlord misinformation, as seen in the following examples.

On 20 December, a convoy of suspected terrorists in Paktia province was struck by US bombs, killing some 52 people. [108] While the Kabul regime quickly claimed all victims to be Taliban supporters, residents countered that 107 innocent people had died, including tribal leaders who had been heading to the inauguration of the new interim Afghan leader, Hamid Karzai. [109]

In another case of mistaken intelligence investigated by the Washington Post, US airstrikes killed another 52 people from the village of Qalai Niazi on 29 December where people had congregated for a wedding. [110] The incident, killing 25 children as well as the bride and groom, was being investigated by both the ICRC and the UN. [111] While it was the first time that the UN had demonstrated a public concern about the bombing, the Pentagon stated that the presence of ammunition in the village made the village a valid military target. [112] The IHL principle of proportionality was violated, as stated in Protocol I, articles 51.5 (b) and 57.2 (a) iii. No advance warning was known to have been issued by the coalition, a violation of Article 57.2 (c).

Targeting decisions: national interpretation of IHL, collateral damage, coalition cohesion and the media

This section will expose some differing national interpretations to IHL compliance by key coalition states during air campaigns, highlighting the most controversial incidents and coalition repercussions of collateral damage in the four cases. Of particular interest to this study is the effect of political involvement in targeting selection upon IHL compliance and civilian casualty rates. It is also important to make note of the influence which coalition cohesion and the media may have had on IHL compliance in the cases.

The requirement of taking all feasible precautions to minimize civilian casualties (Article 57.2) must be weighed against the military necessity to safeguard a belligerent’s capacity to fight and ensure its pilots’ survival. Even though weapons failures and mistaken target identification were admitted by the Gulf coalition in 1991, state practice during the conflict revealed that noncombatant immunity had become respected as a customary norm, strengthening Articles 48 and 51.2. [113]

According to the legal interpretation of one American military lawyer however, the “law of war is not a suicide pact. It does not require that an attacker employ the most discriminate force available to him.” [114] Use of the words “incidental loss of civilian life” in Article 51.5 (b) of Protocol I signifies that death may be a consequence of belligerency and that such force could not be deemed illegal if the rule of proportionality is respected: “the law recognises that an attack upon a military objective may entail ‘collateral’ civilian casualties and damage.” [115]

A political link exists between a state’s interpretation of IHL obligations, collateral damage and the need to maintain coalition cohesion in the face of a common adversary, particularly when the global media has access to the adversary’s terrain during bombing campaigns. Targeting failures accentuate political interest in airpower management given the public accountability of democratically-elected coalition leaders.

The 13 February 1991 airstrike upon the Amiriya civilian bunker/command center resulted in a coalition decision to “sharply reduce the bombing of targets in central Baghdad; in the two weeks after the incident, only 5 were struck, compared with 25 in the preceding two weeks.” [116] Three weeks into the NATO air campaign against Yugoslavia, after the 12 April airstrike against a rail bridge in which a passenger train was hit, and the 14 April airstrike upon a refugee convoy, forward air controllers previously restricted to 15,000 feet, received NATO clearance to descend to as low as 5,000 feet if necessary to ensure positive identification of ground targets in Kosovo. [117]

The Gulf War and anti-Taliban campaigns however witnessed the relative freedoms of military lawyers and commanders to interpret the laws of war largely without political interference. [118] The Kosovo air campaign almost choked as politically-approved targets were quickly exhausted by the third night, [119] and delays in the alliance targeting mechanism forced NATO pilots to strike some targets repeatedly. [120] The Kosovo campaign required commanders to oversee air campaigns while negotiating the complexities of coalition politics. [121]

Unlike the large degree of civilian involvement in targeting decisions against Yugoslavia in 1999, [122] only a minimum of political restraint was exercised against Afghanistan, particularly during the opening weeks of the war, to limit strikes north of Kabul. [123] Otherwise, ROE were comparatively loose and flexible, remaining almost entirely within military circles. According to a British parliamentary report on the coalition targeting process in October, American pilots operated in what were called “engagement zones,” with pilots given general types of targets, such as moving military vehicles, and gaining rapid approval to hit specific emerging targets from forward air controllers. [124] The Taliban also moved in non-military commercial trucks and foreign aid agency vehicles which they commandeered or stole. [125]

Coalition behavior and intentions differed during the air campaigns. While the 2001 coalition wanted nothing less than the end of the Taliban hold on Afghan power, the other air campaigns sought the withdrawal of enemy forces from a region as an operation’s end state, not the replacement of a regime. Different degrees of freedom in commander decision-making were used to achieve these goals.

Varying degrees of technology, intelligence and national interpretation of obligations to IHL weakened the unity of Balkan airpower coalitions. Politicians more than generals wanted to publically defeat Milosevic. But “making war by accepting political constraints that impede sound military preparations can be a prescription for defeat – and nearly was.” [126] The Kosovo campaign revealed the highest example of learn-as-you-go political involvement in targeting decisions. Although airstrike targeting plans by NATO and a separate US cell had begun in the Spring of 1998, “NATO began the war without having achieved any consensus on what the alliance would do if the hostilities extended beyond 48 hours.” [127]

US Generals Norman Schwarzkopf and Tommy Franks were comparatively not required to daily and hourly sell their evolving targeting plans to coalition leaders. [128] Gen. Clark’s US Air Force subordinate, Lt Gen Michael Short often found his targeting proposals temporarily or permanently vetoed by Clark, Washington or NATO.

In 1999, the US, British and French interpreted the Geneva Conventions through their own political lense. By early 1999, France, Turkey and the US were the only NATO members not parties to Protocol I. All members of the European Union except France were party to Protocol I. Against Yugoslavia, “each instructed their air-crews to stay on the ground when missions they considered legally dubious were taking place.” [129] The Europeans were more fearful of consequences upon regional values and legal obligations. While many nations, especially NATO member Greece, [130] had called for a pause in bombing in March, [131] the idea was not supported by the NATO commander, and had been rejected by the US State Department. [132] A pause was also not tolerated by the Finnish mediator, President Ahtisaari, a non-NATO state, during June mediation efforts. [133]

General Clark learned the difficulties of waging a war with 14 veto-wielding force-contributing countries arguing over the vulnerability of target selection to political repercussions from possible civilian casualties. Describing the multi-state and multi-level targeting approval process as regularly “ricocheting around NATO headquarters”, [134] Clark admitted that the management of such a sensitive campaign lead more than once to NATO almost crumbling. [135] Even within the US position, the avoiding of cracks in public unity [136] proved a tiring exercise for an American commander in a European civil war: the “stress of the relationship with the Pentagon had been the worst part.” [137]

The overwhelming number of strike packages sent over Yugoslavia were led either by Americans or Canadians. [138] But even Canada, whose air force interoperability [139] is closest to that of the US Air Force, presented a legal challenge to American plans. “Canada’s different legal approach was an issue during the Kosovo conflict in 1999, when the U.S. Air Force would not pair its pilots with Canadian fighters who, under a protocol of the Geneva Convention, were required to be more selective of targets.” [140]

France, a former World War I ally of Serbia, tempered US ambitions while the pro-interventionist British balanced the French and Americans. The newly-elected German government impressively learned fast from developments while rekindling a traditional open door with Russian counterparts. French President Jacques Chirac remarked that it was thanks to him that any bridges were spared during these political wrangles. [141] No British pilots participated in attacks on the TV station and electric power grid because of fears of potential IHL violation. [142] British and French governments prevented their aircrews from striking ground forces if these were within 500 meters of a village. [143] It was after the NATO bombing of Belgrade’s Interior Ministry in early April, broadcast on global television, that calls for member consultations led to the creation of a “bombing by committee” approach, resulting in foreign ministers from the US, UK, France, Germany and Italy conferring daily on targeting decisions. [144] As during the Cold War, [145] the Europeans continued to exercise a pacifying affect upon American foreign policy behavior.

Although Operation Allied Force was seen to be a coalition effort, an absolute political necessity given the 50^th anniversary of the Alliance in April, the Americans by far conducted most of the fighting, based on their own separate operations centre: flying “over 80 percent of the strike sorties, over 90 percent of the advanced intelligence and reconnaissance missions, over 90 percent of the electronic warfare missions and fired over 80 percent of the precision guided weapons and over 95 percent of the Cruise missiles.” [146]

General Clark ensured that the coalition’s varying standards would not inhibit Washington’s desire to punish Milosevic, “by keeping NATO out of missions using American planes. At Vicenza (the NATO Combined Air Operations Centre in Italy) there were two completely separate targeting teams, called cells, one for NATO warplanes, the other strictly American assets.” [147]

The US also views its obligations to IHL differently because of its far superior level of airpower technology in comparison to its NATO allies and air campaign adversaries. Such airpower technology enables greater discrimination in targeting efficiency and thus a greater capacity for IHL compliance. But some military analysts argue that American technological superiority should not become a legal restriction on the behavior of all its airpower assets. “As the undisputed leader in military technology, the United States has every incentive to ensure that its technological supremacy doesn’t evolve into a legal liability.” [148]

If the capacity to discriminate is there, will states place legal expectations on PGM use in place of available stocks of dumb bombs? If allies know that US technology exists to pre-detonate a camera-loaded cruise missile accidentally heading for a civilian train, embassy or bomb shelter, and such an option is ignored, what would happen to coalition cohesion and to IHL compliance? IHL may not be a suicide pact for pilots, but a commander’s requirement to use all means necessary to avoid civilian objects and populations, Article 57.2, applies, regardless of technology and of questions about available bomb inventories.

Perhaps complementary to legal, technological and humanitarian efforts to limit human suffering caused by air campaigns was the role of the media and its impact upon the conduct of airpower and selection of targets, an argument meriting much further research beyond the scope of this study. One could argue that the presence of foreign media in war zones also weighed heavily on any civilian motivations to oversee targeting decisions. In 1991, the “mass media carried every detail of the campaign into the homes of a domestic public that, although supportive of their governments, proved to be acutely sensitive not only to the losses of their own armed forces but even to those inflicted on the enemy.” [149] Public televising of the scorched remains of fleeing Iraqi military convoys along the US-bombed “Highway of Death” in late February 1991 is also said to have influenced coalition policy to conclude the war. [150]

There is an absolute correlation between media exposure of civilian casualties, public opinion, political involvement in targeting decisions and subsequent IHL compliance. The relative absence of foreign commercial media in Iraq (except Baghdad [151] ) and Afghanistan (except Kabul [152] ) mirrored an absence of political interference in air campaigns and thus proportionally higher civilian casualties. Unlike during the Kosovo campaign, [153] foreign journalists and analysts were not capable of providing visual proof of civilian suffering which could have pressured the public and coalition leaders into greater IHL compliance. In October, the Pentagon is said to have bought the entire output of satellite photographs on ground movements in Afghanistan. [154] In mid-November of 2001, a US missile hit the Kabul office of the Al-Jazeera television station. [155]

Milosevic and Hussein both made strategic use of Western media to publicize the mistakes of American bombings [156] in Baghdad and Serbia: “stories of civilians carbonized in bombed trains and media workers incinerated by strikes on television stations.” [157] In Kosovo, even though only 20 bombs went astray, out of a total of 23,000 launched by NATO against Yugoslavia, their impact was known worldwide, thus underlining the importance of media superiority, “that the PGMs which matter the most are precision-guided messages.” [158]

Such worldwide publicity immediately increased the involvement of civilians in targeting decisions which thus curtailed bombing. Vietnam had already proven to be a dire lesson for the Pentagon, where the media became the conscience of the White House. Besides reasons of safety, it is possibly this foreign presence in the form of relief agency workers which is not highly preferred by the Pentagon during air campaigns, a presence examined in section two of this study.

Among the most debated targets struck by NATO during the Kosovo campaign were, in chronological order: 1) a passenger train on 12 April, 2) a civilian convoy on 14 April, 3) the RTS TV station on 23 April, 4) the Chinese embassy on 07 May, and 5) the electric power grid in Belgrade at various times. General Clark stated that the three most contentious targeting issues for the coalition were the TV station, the bridges and the electrical system. [159]

The targeting of a civilian passenger train was not intentional as the military target was the bridge being used at the time. [160] The moment at which the train became visible would have required, if technically possible, the option of avoiding civilian casualties by auto-detonation of the Maverick missile which was about to strike the bridge. It is known that the pilot or co-pilot did not have such an option. [161] Incidental loss included over 10 civilians.

One is left to wonder when technology will allow humans to abort strike missions or pre-detonate a munition once it is deemed to jeopardize civilians. The speed at which camera-wielding munitions travel does not necessarily permit missions to be so handily aborted in order to avoid collateral damage.

Over 70 people were killed when an F-16 fighter jet struck a convoy of internally-displaced civilians in mid-April. The practice of human shielding, or commingling of Serb forces with civilians, a violation of Article 51.7 of Protocol I, had been observed by NATO since the previous September. “This included the practice of military convoys joining columns of refugees.” [162] The ICTY judgment admitted the possibility of human error in determining the protected status of persons from 15,000 feet: “While there is nothing unlawful about operating at a height above Yugoslav air defences, it is difficult for any aircrew operating an aircraft flying at several hundred miles an hour and at a substantial height to distinguish between military and civilian vehicles in a convoy.” [163]

The reasons for attacking the Serbian television and radio station (RTS) would not have been legal if they were simply for purposes of halting propaganda. [164] But the coordinated hitting of the radio relay buildings and towers with electrical power transformer stations intended to stop their dual civilian-military function, thus denying communications to Serb troops and affording NATO military advantage. Two previous warnings of the airstrike did not prevent the deaths of between 10 and 17 people using the building. [165] Prior to the 2AM attack, advance notice had been given by the Pentagon to the Yugoslav government and to foreign war correspondents using the station, on April 12 and 18. [166] Following one Pentagon news briefing, “the Serbs had ordered all international journalists to report to the Serb television building”, [167] a violation of IHL articles forbidding human shielding. [168]

On 21 June 2002, the former chief of the television network was sentenced to 10 years in prison for his role in the deaths of 16 television employees killed by a NATO missile. He had failed to order the evacuation of the station during NATO air raids, instead “ordering the employees to remain at work so their deaths could be used as a propaganda tool.” [169]

The intentional striking of the location later revealed to be the new Chinese embassy and not the Federal Procurement Directorate was the result of an intelligence failure attributed to old maps used by the CIA. [170] The Rendulic rule applies at this point. The US apologized for the error and paid $32.5 million in compensation to the Chinese government and families of the three killed journalists and 15 injured civilians. [171]

The targeting of the Belgrade electric power grid was originally desired in the first week of the campaign, as was the case during the Gulf War. [172] The power grid was not hit until Day 40, the point at which NATO had agreed it was time to alter ROE and begin targeting the morale of the population in order to inflict pain sufficient to pressure a change in the actions of political leadership, a strategy used in the two World Wars. NATO was able to cut off power to 70 percent of Yugoslavia, although most of it was back on within a day. [173]

While the targeting of the electric power grid took place six weeks into the NATO campaign against Yugoslavia, [174] the Iraqi national power grid had ceased to function by the seventh day of the campaign. [175] HRW comments on a 30 January briefing by US coalition commander General Norman Schwarzkopf, indicating some recognition of attacks being excessive in relation to their concrete and direct military advantage: “allies had rendered 25 percent of Iraq’s electrical-generating facilities ‘completely inoperative’ and an additional 50 percent ‘degraded’.” [176] According to one legal commentator, “(t)he enormous devastation that did result from the massive aerial attacks suggests that the legal standards of distinction and proportionality did not have much practical effect.” [177] The hitting of dual-use electrical generating plants in Iraq seriously affected food supply, water purification, refrigeration capacities, sewage disposal systems and medical facilities, promoting the spread of disease and contamination. [178] The coalition continued to bomb power stations until the last day, the Al-Hartha power plant, which provided water flow and sewage pumps, being hit twelve more times. [179]

The targeting of dual-use civilian-military infrastructures is not illegal except in the case of those works and installations containing dangerous forces like nuclear stations, dams and dikes (Article 56).

One retired US military lawyer, claimed loosely that once a civilian object is put to military use, it loses its protected status. [180] This is not the wording of Protocol I, which uses the word “strictly military objectives”. However, any loose interpretation of the Protocol renders too easily most any object used by the military vulnerable to attack. “Were this argument to be taken to its logical conclusion, every civilian object that could possibly be used by the military would become a military objective.” [181] Protocol I defines a military object as affording an “effective contribution to definite military advantage.”

In both the Gulf and Kosovo conflicts, coalition leaders pledged that the campaign was not aimed at the civilian population. Yet the foregoing of strikes against fielded forces in favour of direct attacks upon industrial infrastructure in 1991 and 1999, as during the two world wars, served the dual purpose of destroying both the enemy’s ability to fight and his morale. [182] Once all leaders saw the difficulties inherent in combating Serb forces from the air through bad weather and lack of ground intelligence, NATO sorties were given permission to shift ROE from targets in Kosovo to the nerve centre of the war, Belgrade.

Having reviewed the laws of war and the behavior of airpower in the three case studies, US-led air force coalitions have largely met their legal wartime obligations. The measuring of progress in estimating any success at avoiding civilian casualties remains difficult given the different campaign durations, population densities, access to information and number and type of bombs dropped. The studies revealed roughly the following number of civilian casualties: 3000 Iraqi civilians killed over 44 days, 500 Yugoslav civilians killed over 78 days and some 1150 Afghan civilians killed over 103 days ending 18 January, [183] the later still contested and awaiting HRW’s April 2002 ground investigation findings. Until the air campaign against the Taliban, coalition efforts to increasingly avoid civilians were more obvious. [184]

An analysis of coalition behavior reveals a general increase in the use of PGMs by airpower coalitions, an expensive investment in the military principles of economy of force and the humanitarian obligation for distinguishing civilian and military objectives and populations. Use of PGMs reached 9 percent in the Gulf War, increasing to 35 percent in Yugoslavia and rising again to 60 percent in Afghanistan. [185] According to the PDA study, Operation Enduring Freedom “failed to set a new standard for precision in one important respect: the rate of civilians killed per bomb dropped.” [186]

The involvement of political scrutiny, coalition pressures, media exposure and legal concerns about the conduct of air warfare combined to greatly limit incidents of collateral damage in 1999, as proven by comparing military conduct and the non-involvement of civilians in targeting decisions during the 1991 and 2001 campaigns.

Until the Afghan campaign, coalition respect of IHL had demonstrated increased compliance. In 2001, civilian controls were largely waived by the new Bush administration for at least two reasons: the measure of public American resolve for the war after a first attack upon US territory since the Japanese attack on Pearl Harbour in 1941, and the appreciated lessons learned from a largely military-managed Gulf War by the same Presidential entourage as in 1991: Secretary of State Colin Powell and Vice President Dick Cheney, both veteran leaders of Gulf War operations, as the former Chief of Defence Staff and the former Secretary of Defence to President George Bush Sr. These circumstances encouraged the coalition to revert to Gulf War-style targeting decisions largely made by the Pentagon’s Central Command in Florida. Given the attack on American lives on September 11th, an interpretation of military necessity would not match that of any of the other three cases.

The ground element to enable effective airpower employment played a positive role in all cases. British and then American Special Forces were deployed to west Iraq in early 1991 as SCUD busters because airpower was limited by weather and especially by the mobility of SCUD missiles hiding under highway underpasses. [187] Throughout the Kosovo campaign, General Clark never relented in pushing for use of ground forces supported by the 5000-strong Joint Task Force Hawk and its 24 Apache helicopters already stationed in Albania. PGMs could not have engaged the Serbs or Taliban without a ground force component, the KLA and the Northern Alliance, respectively.

In recent years, military commanders, lawyers, politicians and scientific acumen have combined to make the waging of air warfare increasingly survivable. This statement is qualified by American airpower superiority and the low intensity of the 1999 and 2001 conflicts. Although no pilots have been killed in air campaigns since the early days of the Gulf War, coalition practices toward minimizing collateral damage had until 2001 changed for the better. The promise of surgical airpower, hailed by politicians wary of body-bags, but in the absence of ground spotters, remains however a non-conclusive means of achieving objectives free of error, a practice which negatively pushes the limits of publically-acceptable military necessity.

HUMANITARIAN SECTION: THE CONSEQUENCES OF AIRPOWER UPON THE “HUMANITARIAN SPACE”

Because it is important to understand the point at which the use of airpower can conflict with and inhibit, if not prevent, the legal and practical access of impartial humanitarian players on the ground, this section will examine the quality of the “humanitarian space” in each case study, a space [188] created by the actions and legal rights of international humanitarian agencies involved in assisting victims of complex political emergencies: the ICRC, the UN Children’s Fund (UNICEF), the UN High Commissioner for Refugees (UNHCR) and the World Food Program (WFP). The right of initiative and of access of emergency relief organizations will be tested. This section will argue the necessity by state-oriented and non-civilian bodies to respect the rights of humanitarian agencies entitled to work in conflict zones, whether from the air or not. The humanitarian problematic of US air-dropped relief to Afghanistan will be addressed.

In comparing relief actions in the three cases, the following issues, concerns and obstacles will be analyzed: pre-air campaign agency status, presence of foreign aid workers, level of access to aid recipient populations, communication capacities of local aid staff and partners, security risks/casualties of staff and associated partners, impartiality of aid agency, relations of aid agency with airforce coalition.

It should be noted that the socio-economic development and humanitarian needs of states prior to coalition bombings varied greatly. In 1988, the UN had brokered a peace in the eight-year war between the semi-industrialized, oil-exporting states of Iraq and Iran. In early 1999, despite political oppression and unemployment, Kosovars enjoyed relative health in an educated and technologically-advanced society where IDPS and refugees had access to cell phones. [189] In the Fall of 2001, the five-year old Taliban regime had come to power in an impoverished, landmine-contaminated state ravished by war since the Soviet invasion of 1979. The UN human development index for Afghanistan was far below that of Iraq and Kosovo.

The US-led military build-up to Operation Desert Storm gave itself five months between Iraq’s invasion of Kuwait in August of 1990 and the ultimatum for withdrawal of Iraqi forces from Kuwait on pain of force, beginning on 16 January, 1991. Although aid agencies had ample time to organize their operations, only the ICRC remained in Iraq. [190] Having previously consolidated relations with belligerents during the 1980s Iran-Iraq war, 9 delegates remained in Baghdad on 16 January, restricted to their premises for security reasons. [191]

UNSC resolutions in 1990 failed to remind member states of their obligations to IHL, including the right of impartial and unimpeded access of relief under Article 23: “one of the early sanctions resolutions violated the rule that requires free passage for many sorts of humanitarian aid intended for civilians, even civilians of an adversary.” [192] The UN Sanctions Committee established by the UNSC strictly limited and controlled all UN and other relief efforts into Iraq. The ICRC informed the UN Sanctions Committee and the Iraqi government that it would notify the committee of its humanitarian shipments but would not seek its approval to do so. [193]

The UNSC measure transformed UN agencies into complicit parties favouring a belligerent. When Iraq asked the UN to send a humanitarian mission to Baghdad to evaluate food shortages and medical needs, these agencies refused to act until the cessation of hostilities. [194] The image of UN relief agencies suffered greatly from association with the UNSC decision to use force. Attempts by UN humanitarian staff to distance themselves from the political aspects of the UNSC were unsuccessful, causing protests in front of the UNDP offices in Jordan. [195]

ICRC mobility was hampered by low fuel availability and communications problems. The only means of communicating with ICRC delegates in Baghdad since Day 2 of bombardments was during their few visits to the Iranian border, a 150-mile drive. [196] The ICRC was however able to launch the war’s first relief operation on 31 January with the support of both national Red Crescent societies of Iran and Iraq, convoys entering from Iran with nineteen tonnes of emergency medical supplies for Iraqi civilians. [197] A week later, another Red Cross convoy was organized. [198]

On 12 February, a visit to Baghdad via Tehran by Angelo Gnaedinger, the ICRC Director of Operations from Geneva, was made to discuss the prisoner situation and relief access. [199] Unable to gain access to the Foreign Ministry, he was however able to meet the Iraqi Health Minister and the Iraqi Red Crescent Society to win formal agreement for relief. On 17 February, for the ninth time since the UN trade embargo issued in August, 1990, a convoy of Jordanian Red Crescent trucks left for Baghdad, carrying medicine, doctors and infant milk. [200] On 23 February, once the ICRC Baghdad team became more mobile, a sanitary engineer went to Baghdad to begin an assessment, after which a programme was launched. [201]

In an effort to distance itself from the US-led coalition sanctioned by the UN, American UNICEF Director James Grant, in an effort to repeat their successful cooperation during the Cambodian crisis, visited the ICRC on 26 January for talks. [202] In early February, the UNICEF head officially petitioned the coalition and Iraq to allow safe passage to a convoy of medical supplies through a “corridor of peace” or “fire-free zone” from the Iranian border to Baghdad. [203] The roads and bridges from the Jordan border, a much greater distance, had been severely damaged by coalition strikes. [204] Grant referred to the pledges made in 1990 by states at a World Summit for Children conference, at which article 20, paragraph 8 of a summit declaration called for recognition of “periods of tranquility and special relief corridors.” [205] On 16 February, UNICEF and the World Health Organization jointly organized a convoy of medical supplies intended for women and children. [206] Medecins Sans Frontieres (MSF) was cooperating with Oxfam in Syria in preparing refugee camps along the border, seeing Syria for the first time welcome voluntary relief agencies into its country. [207]

Out of concern for its citizens held in Baghdad, the British government publicly jeopardized the neutrality of the ICRC by linking the organization to ill-informed British bribing statements: “any response to its humanitarian needs depends on Iraq living up to its obligations under the Geneva Conventions and allowing the Red Cross to visit PoWs.” [208] The ICRC survived Thatcher’s attempt to make it a pawn of political leverage.

The post-conflict situation revealed Iraqi and coalition cooperation with the Red Cross over prisoner exchanges, with coalition and Iraqi prisoners being repatriated in Red Cross planes. [209] Still the ICRC went to great effort in distancing its operations from UN and coalition policies toward Baghdad and from the militarization of humanitarian assistance to the displaced Kurdish people of northern Iraq [210] where it conducted its own separate civilian-needs assessments. [211]

Following the failure of mediation attempts and the decision to withdraw international monitors, NATO recommended that UNHCR and other UN personnel leave Kosovo. [212] Until 23 March, the WFP had been dropping and distributing food to some 210,000 Kosovars “facing food shortages because of Serb repression.” [213] The WFP had stockpiled food in the border regions of Kosovo’s neighbours. [214]

Prior to NATO action, the ICRC had 70 staff in Kosovo while UNHCR provided assistance to some 400,000 people in Kosovo with relative freedom. [215] With the initiation of airstrikes on 24 March, the ICRC office in Pristina closed 5 days later, citing security reasons. [216] On 26 April, the ICRC President visited Belgrade, its delegation office never having closed, where he met with the Yugoslav President and received his personal assurances for the organization’s freedom of mobility and safety in Kosovo. [217] Re-opening on 24 May, 17 days before the end of NATO airstrikes, delegates returned to Kosovo after having received Yugoslav assurances and encouragement from NATO. [218] In May, joint ICRC and Yugoslav Red Cross (YRC) activities provided relief to airstrike victims in six towns throughout the territory. [219] While not accusing NATO of violations for targeting dual-use facilities, the ICRC stated that it was rebuilding the water supply network of a northern city where the water supply was cut to 90,000 people.

The ICRC found it difficult to work in the extremely hostile environment that Kosovo had become during the last three weeks of joint ICRC-YRC operations. [220] The operations were able to avoid being struck by airstrikes by keeping open, transparent channels with all sides in the conflict. The ICRC and NATO, having signed a memorandum of understanding in 1996, [221] had reached consensus on communication channels and procedures. Before the launch of NATO airstrikes, the ICRC had communicated warehouse and office locations in Yugoslavia. During the bombing, the ICRC head office in Belgrade would provide the full-time ICRC delegate in Brussels with information on intended relief convoy operations with 48-hours transmission notice to NATO. NATO would be informed of field information on convoy times of departure, intended routes, number and nature of vehicles, all in clearly-marked Red Cross vehicles. The actual distribution of relief stocks was carried out by the Yugoslav Red Cross, under ICRC monitoring, throughout the territory of Yugoslavia, in both Serbia and Montenegro, while responding to the effects of bombings on the population. Security concerns within Kosovo itself were compounded by combat between the KLA and Yugoslav/Serbian armed forces.

To underline the need for relief operations beyond border areas, the Pope called on the Belgrade government to create a humanitarian corridor to improve relief to IDPS and refugees. [222] The UN High Commissioner for Refugees however stated that the option of calling upon the services of a neutral third-party for delivery of relief within Kosovo was not viable in such an intense conflict, and that any UNHCR aid workers would only return once Yugoslav forces had withdrawn and international peacekeepers had been deployed. [223] Her political pronouncement immediately disqualified UNHCR from Article 59 status as an impartial relief agency because it made the agency sound like it was working for NATO. In the face of Serb-led minority oppression, KLA activities and NATO threats, MSF, [224] UN agencies [225] and other NGOs found themselves unable to maintain a sense of independence from all parties [226] during the civil war, the result of which was the abandoning of the internally displaced inside Kosovo and the rest of Serbia.

However, third-party intervention was attempted in a limited capacity. A joint humanitarian assistance operation called “Operation Focus” was organized and conducted by four countries during the bombing: NATO member Greece, Switzerland, Russia and later Austria. [227] On 28 April, the Greeks, having distanced themselves from NATO bombings, launched relief operations into Kosovo while the Russians were based in Belgrade, the Swiss in Podgorica, Montenegro, and Skopje, Macedonia, and the Austrians in Nis, Serbia. [228] Non-ICRC Swiss operations were coordinated from Bern. [229]

Five relief trucks entered Kosovo from the Greek border on 06 May, followed by four more trucks on 28 May. [230] Attempts to then distribute aid stored in Gracanica, 12 kms from Pristina, were blocked by Serb authorities who arrested the Greek aid coordinator, putting him under house arrest for nine days. [231] Because the Greek relief convoy had not been cleared by Serb customs, their distribution was halted until eventually given to local branches of the Yugoslav Red Cross. [232] Under Article 59 of the 1949 Conventions, belligerents are allowed to search relief convoys.

On 05 May, another Greek-based relief effort by international NGO Doctors of the World [233] (Medecins du Monde or MDM) entered Kosovo with three trucks and a jeep carrying medicines and a Greek neurosurgeon. [234] Although having received Yugoslav approval, and bearing clearly-visible signs of its humanitarian nature, the convoy was almost directly struck by a munition, exploding some 100 meters away, while nearing the town of Urosevac, 25 miles from Pristina. [235] Nobody was injured. While Serbia’s Tanjug state media blamed NATO for the “attack,” claiming 2 of the 3 trucks to be damaged, [236] NATO denied hitting the convoy, adding that the NGO had publically stated that it did not know the origin of the explosion, whether a munition had been dropped from the air or fired from the ground. The NGO’s Greek spokesperson in Athens did state that the convoy had suffered damage, that it “had been hit.” [237]

Operation Focus relief trucks were able to avoid NATO bombs by their communications with NATO coordinators in Skopje, informed of the times of travel and routes employed by the Greek trucks. [238] Operation Focus trucks also avoided potential NATO targeting by placing distinctive orange textile sheets on their convoys. [239] The Greek humanitarian relief coordinator acknowledged that Operation Focus succeeded in its political intention of distancing Greece from its NATO members by adopting a humanitarian approach, not unlike the public relations objective of US military aid drops over Afghanistan, later discussed.

During these relief operations, humanitarian-military liaising took place at US CENTCOM in Florida in order to share information intended to avoid the bombing of these relief operations. [240] NATO wanted to know in advance the movements of these relief operations. [241] General Clark mentioned his concern about hitting the relief convoys arriving from Greece, thus complicating his airstrike plans against Serb forces. [242] Clark also admitted two near misses upon ICRC convoys during NATO airstrikes in mid-May. [243]

On 16 May, a UN humanitarian assessment team, including UNHCR, WFP, UNICEF and six other specialized agencies, arrived in Yugoslavia, spending ten days visiting areas of Yugoslavia, including various locations in Kosovo. [244] The mission excluded the ICRC, never having left Yugoslavia, but included Save the Children, said to represent all NGOs. [245] The mission succeeded in convincing Yugoslav authorities to allow UN convoys to re-enter the country. [246] But no such activity took place until Yugoslavia signed an agreement on 09 June committing the withdrawal of its forces from Kosovo. [247]

In an unprecedented example of humanitarian resolve during the final days of the air campaign, and much to the dislike of some US Air Force personnel, [248] the US-based International Rescue Committee (IRC), funded by the US Agency for International Development (USAID), dropped food aid over Kosovo on 30 May in coordination with NATO. [249] The IRC [250] had contracted a Canadian company, Skylink Air, hiring Moldovan pilots to fly Russian Antonov 26 cargo planes from Moldova which had been painted a distinctive white with orange stripes. [251] Previous to the aid drops, the IRC had dropped multi-lingual leaflets to inform people of the pending drops, in order to avoid injuries from falling items. General Clark required NATO aircraft to divert action from the aid drop area. [252] After the halt of NATO airstrikes ten days later, the IRC had planned to step up drops in order to supplement the problematic rush of aid convoys into Kosovo. [253]

As in 1998, the Taliban decreed the departure of all humanitarian aid agencies from Afghanistan following the events of September 11, 2001. All expatriate aid workers left, mostly for Pakistan, including 34 expatriates working for the WFP. [254] However, many local Afghan staff of UN agencies, MSF [255] and over 1,000 from the ICRC [256] continued to work. [257] Given the continued mobility of WFP food convoys by local partners and their privately-contracted truckers, the ICRC made its food stocks available to WFP distribution plans while its own medical stocks in Kabul continued to be replenished by means of ICRC convoys from Pakistan. [258] The WFP had almost no communications with local staff at UN offices in Afghanistan but were able to resume daily trucking operations after a brief halt in September. [259]

US-led airstrikes began on 07 October, directed from the US Central Command (CENTCOM) in Tampa, Florida. On the second day, a UN-funded demining office was hit, killing four demining technicians [260] and injuring two guards.

On 09 October, the WFP and UNICEF began sending convoys of trucks to Kabul, Herat and the North West, carrying food and other supplies into Afghanistan on a test basis. [261] But on 15 October, shrapnel from a US bomb wounded an aid worker loading grain from a WFP warehouse for an Oxfam delivery at the Kabul airport. [262] The coalition target was said to be a military parking lot near the WFP warehouse in Kabul. [263] A part of the dual-use problematique, airports are always among the first military targets of conflict, but also a familiar and convenient location for the storage of humanitarian supplies.

On 16 October, a clearly-marked ICRC warehouse 1.5 kilometers from the Kabul airport was struck by a US Navy jet, injuring one guard. [264] Previous to the air campaign, the US had been notified of all ICRC facilities, clearly marked by the Red Cross emblem. [265] The Pentagon claimed that the Taliban had been storing military equipment there. [266] Although the warehouse was then included on a no-strike list, [267] a pilot was apparently not informed of this list and targeted the same ICRC warehouse compound on 26 October, despite previous ICRC transparency. [268]

A lack of ground intelligence about the humanitarian situation, a result of Taliban decrees against the presence of foreign aid workers and journalists [269] and the possession of satellite phones, challenged the ability of aid agencies to coordinate relief operations from outside the country. Oxfam called for a pause in airstrikes, at least in some areas, to permit the replenishing of pre-winter food stocks, [270] while CARE USA called for greater sharing of information with the Coalition Humanitarian Liaison Center (CHLC) in order to safeguard its facilities, people and operations. [271]

A CHLC had opened in Islamabad in late October [272] with the intention of providing aid agencies with security information on safe routes inside Afghanistan. Although denying that the creation of the CHLC had any relation to the US bombing of the ICRC warehouse, the office was staffed with British and American soldiers with civil and special operations experience from previous multinational operations in Kosovo, Sierra Leone and East Timor. [273] The UN Regional Humanitarian Coordinator met regularly with the coalition in Islamabad. [274]

Although humanitarian workers faced communications problems and coordination challenges from across the Pakistani border, small UNICEF-funded teams within the country still managed to implement a country-wide polio eradication program from 21 to 25 September. [275] Involving thousands of teams, traveling by hired car, donkey or foot, visiting house to house, [276] oral vaccines were successfully administered to millions children under 5 years old. [277] While not making specific requests for security arrangements with CENTCOM, UNICEF communicated its intention to conduct the scheduled campaign by liaising with the coalition through OCHA. [278] There were some difficulties for the coalition in considering the risks of such a geographically-expansive program given UNICEF’s inability to inform CENTCOM of all the teams’ movements. [279] No special precaution was used to identify traveling polio vaccination teams. [280]

In late October, the WFP continued food distribution operations with the cooperation of local Afghan staff from 19 international NGOs in Afghanistan. [281] UNICEF operations also continued, a convoy entering from Iran reaching the Western city of Herat. [282] All WFP food was distributed by commercial means, hired truckers of the same ethnicity as the intended aid recipients, paid by the WFP to reach warehouses where local Afghan NGO partners distributed the food on behalf of the WFP. [283] Aside from a temporary two-week halt in distributions following the events of September 11, largely a result of lorry firm concerns for safety under the Taliban, WFP convoys were conducting more operations in October than in the month before the 07 October coalition attacks. [284] During the months of October and November, while relief may have reached storage locations in urban centers, few local NGO partners were available to enable redistribution because insecurity had become prohibitive. [285] Nevertheless, the month of November witnessed a doubling of food tonnage delivered in October by the WFP, with December roughly doubling November figures. [286]

In a late October interview, the WFP stated that US bombing was not significantly inhibiting its ground relief operations. [287] But on 10 November, a 22-truck WFP convoy en route to Bamian was struck by two US bombs while parked along a highway. [288] While nobody was hurt, two vehicles were destroyed while others were damaged; 80 percent of the food was said to be no longer useable. [289] Traveling in unmarked trucks, in an area of central Afghanistan then under Taliban control, the convoy had not coordinated its travel with the coalition, according to CENTCOM. [290] But a WFP spokesperson stated that aid agencies had been careful to inform the US military of their travel times and routes. [291] On 16 November, a WFP warehouse in Kandahar was also struck by US bombs. [292]

After the unexpected Northern Alliance sweep of central Afghanistan, including Kabul on 13 November, expatriate aid workers began to return to Afghanistan, including those from the ICRC, WFP, UNHCR, MSF. [293] A mid-November bulletin by the UNJLC reported that more than 200 locally-hired trucks operated in the country every day delivering relief. [294] Relief convoys made use of “literally all the roads in Afghanistan” before snow arrived. [295] In addition to the access of relief convoys, the first significant land shipment of aid arrived by barge across a river from Uzbekistan on 15 November. [296] MSF was the first NGO on location in many instances including Kandahar as soon as the Taliban was absent. [297]

Because the air campaign went through several phases of ground action and airforce ROE, relief operations did likewise. The first UN flight into Afghanistan in 65 days included the Regional Humanitarian Coordinator appointed by OCHA in June, Michael Sackett, a former WFP country director in Afghanistan. [298] Since late October, the WFP had chartered a C-130 aircraft capable of airdropping food as a contingency plan. [299] The WFP only began dropping food for the first time in remote areas on 23 November. [300]

The first ICRC flights into Afghanistan since September landed in Chaghcharan, Ghor province, arriving from Iran on 15 December, followed by regularly airlifted food resuming to the remote, mountainous, snowy region from 20 January onward. [301] These relief flights into Afghan airspace were cleared “for military traffic deconfliction” with all parties: coalition offices in Islamabad and CENTCOM, as well as the Regional Air Movement Control Center (RAMCC) based in Qatar since January and the Afghan Department of Civil Aviation when it opened. [302]

As bombing circumstances dictated, and as some cities and warlord domains varied in their pro-Taliban hostilility, with the Taliban seizing WFP warehouses in Kabul and Kandahar, [303] WFP convoys began bypassing urban areas with partner warehouses, heading directly to recipient villages. [304] Commercial truckers for the WFP felt it safer not to be visibly identifiable by any markings because they did not want to attract attention from either Taliban or coalition forces. [305] The presence of US Special Operations ground spotters played a role in validating which convoys were known to be Taliban or genuine relief convoys. [306]

On 11 October, liaising had taken place in Florida between CENTCOM and UN emergency relief agencies: OCHA, WFP, UNICEF and UNHCR. [307] Pentagon and US State Department officials met with relief agencies to discuss airlifts and aid drops. [308] Humanitarian-military discussions enabled an assessment of operational requirements and attempted to create a methodology for CENTCOM and WFP to conduct their respective operations. UN Joint Logistics Centres (UNJLC) had been activated in Rome and Islamabad on 25 September. [309] Liaising between the coalition and UN agencies, through a UNJLC liaison officer in Florida, [310] benefited the coordinated use of military assets for the airlifting of relief to the region. [311]

As the air campaign progressed, greater freedoms in humanitarian-military information-sharing took place, including coalition information on locating the dropping of cluster bomb units (CBUs) which permitted OCHA to deploy its explosive ordnance experts for clearance projects. [312] In November, such cooperation included the sharing of coalition satellite imagery free of charge for the detection of camps for the internally-displaced. [313] After US explosive ordnance experts repaired and cleared the bombed Mazar-I-Sharif airfield, the first ICRC relief flight landed there on 10 December. [314]

Although the coalition had established a mechanism by which to maintain open channels with relief organizations, not all UN agencies responded in the same way to the US initiative. [315] The ICRC did not station a representative at CENTCOM because of its liaison offices already in Washington and New York. [316] While the coalition wanted to know in advance all the movements of relief operations in Afghanistan, this was not entirely possible, given the many actors on the ground in Afghanistan. [317] Even if movements were reported, the coalition gave no assurances of safety. [318] When the coalition representative in Islamabad asked NGOs and UN agencies to provide the precise GPS grid references of their compounds so as to avoid targeting mishaps, MSF did not comply since this measure had obviously not benefited ICRC warehouses. [319]

According to the terms of reference of the UNJLC representative at CENTCOM, the UNJLC was mandated to coordinate the requesting (from UN agencies) and seconding (from the coalition) of all military and civil defence assets (MCDA), all the while keeping in mind the delicate issue of the perceived independence of the humanitarian community. [320] The UNJLC was also charged with informing the coalition of relief activities which may be affected by military operations, sharing UN agency information with the coalition to improve the safety of personnel working in conflict zones, and updating UNJLC on developments within the coalition pertaining to security of personnel and the storing and moving of relief. [321]

While the UNJLC in Islamabad coordinated relief agency activities into Afghanistan, the CHLC in Islamabad kept open communications with the UNJLC, each visiting each other’s offices to share information, although the CHLC never physically stationed a liaison officer at the UNJLC. [322] When possible, the CHLC would record all pertinent information on expected relief convoy activities, such as route to be taken, time of departure, destination, expected return. [323] This information was shared with CENTCOM who in turn notified pilots who were sometimes given strict ROE to not strike any moving targets because WFP convoys were using commercial trucks as was the Taliban. [324] At times, the Taliban and Northern Alliance commandeered commercial and UN agency trucks. [325]

While the CHLC began to gain improved visibility on WFP activities, the CHLC had less success in gaining information on the movement of some NGOs like MSF which kept an arms length with the military. [326] MSF stated that this liaising was more of a one-way channel, for the military to know where any aid trucks were headed than for relief agencies to learn anything about military operations. [327]

When pilot ROE changed due to evolving ground circumstances in the conflict, the CHLC would notify the UNJLC in Islamabad. [328] CENTCOM targeting decisions largely spared the main entry roads recognized for accessing major population centers, these being the Peshawar-Jalalabad-Kabul route from the East, the Western entry from Iran to Herat and the Northern access route toward Mazar-I-Sharif. [329]

Although the Russians had established another CHLC in Kabul in mid-December, [330] the new interim Afghan government voiced its displeasure in January that the US still had no airstrike liaison officer stationed in Kabul’s Ministry of Defence, especially given repeated reports of US bombs going astray in December. [331]

In light of the growing militarization of humanitarian action, and consequent blurring of humanitarian-military roles during complex political emergencies, this section briefly addresses the issue of US aid drops in 2001.

In Afghanistan, on 07 October, 2001, the US military began a simultaneous dropping of yellow-wrapped food rations [332] and bombs, including CBUs containing hundreds of yellow bomblets. [333] The HDR initiative was undertaken with advice and data from USAID which maintained a representative at CENTCOM. [334] With 19 sites chosen, the USAID representative at CENTCOM said that all but three HDR drop sites were in areas under Northern Alliance control. WFP maps and figures on the locations of hunger-stricken people were already in the public domain. [335]

On 01 November, the Pentagon announced a colour change in HDR drops from yellow to blue as a result of confusion with yellow-coloured bomblets used against Taliban forces. [336] On 29 November, UN officials reported that 2 children were killed and three wounded when they ran into a minefield to collect rations. [337]

In addition to encouraging starving people into dangerous or mined areas, regional warlords fought each other over the food and clothing dropped by US planes. [338] A report by retired American Special Forces officers stated that warlords sold collected rations for profit. [339] The report was based on a two-week survey in northern Afghanistan during November.

The US military actually claimed that their aid drops played a “major role”, even a “critical role,” in averting a humanitarian disaster in Afghanistan. [340] USAID chief Andrew Natsios however stated that US aid drops actually represented a mere one-quarter of one percent of the food needed to feed the Afghan people. [341]

Legitimizing the NGO argument that relief must be properly distributed to have effect and avoid endangering people, an Afghan woman was killed when a US relief pallet crushed her house on 29 November, after which the US suspended aid drops. [342] Military aid drops were ended on 13 December. [343] Between 07 October and 22 December, the US Department of Defence flew 162 missions, dropping 2.5 million daily rations, 816 tons of wheat and 73,000 blankets. [344]

Having examined the operations of some key humanitarian relief agencies during the air campaigns, this study has shown that air force coalition respect for the legal, operational presence of a “humanitarian space” in the midst of war has made progress since the Gulf War but remains a formidable challenge. The national interests of coalition powers continue to take precedence over the operational rights of such agencies as well as the rights of war victims to receive relief. There is no doubt that airpower presents dangers to any human activity, including the accessing of victims of armed conflict.

All air campaigns revealed some degree of humanitarian relief taking place. The extent of agency access to the field can be measured by the amount of transparency the humanitarian and military communities are willing to sacrifice to a UNJLC-type or OCHA-type information-sharing body. Given that Operation Focus was led by Greece, a NATO member with close historical and commercial ties to Yugoslavs sharing a common religion, the safety of its convoys was an important consideration for General Clark.

No foreign aid workers were killed during any of the air campaigns because they had largely left targeted regions. Although not totally operational, foreign ICRC delegates remained in major cities targeted by coalition airstrikes (Baghdad and Belgrade) during two of the air campaigns, a testament to the value of its impartiality, diplomatic approach and transparent relations with all warring parties. Although it publically reminded belligerents of their legal obligations as signatories of the Geneva Conventions, the ICRC never criticized coalition targeting practices in any of the campaigns, except in 2001 when US bombs twice hit its warehouses near Kabul. Because it maintains confidential relations with warring parties in order to preserve its access to warzones, it has seldom been ICRC practice to publically list violations of IHL. Given such relations with all sides during the Gulf War and Yugoslavia campaign, its operations did manage to claim a droit d’ ingerence, resuming before the end of the air campaigns and before those of any other major relief agency.

In Kosovo, except for the ICRC, all international aid agencies had left the region before bombing began. [345] In the last three weeks of the 1999 campaign, the ICRC no longer remained confined to their premises, becoming actively involved in distributing relief throughout Yugoslavia. In 2001, the ICRC left because of a Taliban decree against the presence of all relief agencies in the country.

Until the Afghan campaign, the ICRC had proven to be the most capable relief agency during air campaigns. In Afghanistan, after the use of airports by relief agencies became impossible, the WFP and its international and Afghan NGO partners proved the most capable relief operation at reaching recipients by road. Being the most impressive example of humanitarian capabilities during any of the air campaigns, WFP operations were possibly due to a decades-old Afghan habit of working amidst warring factions. WFP impartiality among belligerents was not a limiting factor because it delivered food by commercial means in unmarked trucks, a manner which has also at times been used by the ICRC, although the ICRC never delegates actual distribution. [346] Security threats to WFP convoys [347] were more related to warlord activities [348] than to risks associated with stray bombs.

Can relief agencies effectively operate during airstrikes? They can only be effective when they are in contact with all belligerents. The transparency of relief efforts to all sides in a conflict can best maintain an agency’s impartial nature and thus potential for accessing aid recipients. For relief agencies operating during air campaigns, remaining steadfastly aloof of available coalition information-sharing avenues for the sake of distancing association with and identification to coalition practices, as did MSF in Afghanistan, meant that valuable opportunities for accessing war victims and for safeguarding aid worker activities may have been lost.

In Afghanistan, in addition to the bombing of ICRC and WFP warehouses and of a demining office, causing the deaths of four UN-paid clearance specialists, there was one reported incident of a relief convoy being struck by coalition bombs.

The example of relief access in Afghanistan, facilitated by humanitarian-military liaising at Tampa and Islamabad, did not take place in 1999 and thus in part explained the resulting mass exodus of Kosovars to neighbouring states which forced NATO troops to organize refugee camps. One could argue that this NATO lesson in 1999 stimulated CENTCOM’s aid drop program in 2001. While one military analyst erroneously claimed that the air campaign exacerbated the humanitarian crisis in Afghanistan, [349] an unprecedented degree of humanitarian-military information-sharing contributed to the averting of a massive post-conflict humanitarian crisis.

Humanitarian-military information-sharing during Operation Enduring Freedom offers some hope to future relief work during airstrikes. That NATO and the US CENTCOM wanted to know in advance the movements of relief operations in Yugoslavia and Afghanistan, respectively, something which proved difficult during the Afghanistan campaign given the much greater number of relief operations, [350] indicates increased coalition acknowledgement of the right of such humanitarian initiative, dictated by Articles 23 and 59 of the Geneva Conventions.

The direct use of military logistics cannot substitute for effective, impartial, long-term, culturally-sensitive, ground-coordinated and capacity-building relief operations by experienced agencies coordinating with other regional efforts. Short of this, such unguided military aid drops are but band-aids on massive wounds in a country like Afghanistan. Similar to unguided bombs, the usefulness of expensive aid drops without the requisite ground intelligence for reaching intended recipients, is questionable, largely wasteful and in some cases fatal.

The first major conclusion of this thesis is that, until Operation Enduring Freedom in Afghanistan, the use of airpower by US-led coalitions had witnessed a proportional, general progress in diminishing numbers of civilian casualties since the Gulf War. [351] Efforts to comply with IHL were legally met during the three air campaigns, with increasing percentages of PGM-use in proportion to campaign duration. Independent public investigations of bombing mistakes have not revealed evidence of a deliberate targeting of civilian populations or strictly civilian objectives, although HRW findings of coalition airpower conduct in Afghanistan will be published in the Fall. The anti-Taliban campaign, although it used a greater percentage of PGMs, resulted in proportionally greater numbers of civilian casualties, an indication that there does not exist a trend in air campaign coalitions meeting higher degrees of IHL compliance.

There remains a legal obligation by states to “take all reasonable precautions” at their disposal, during sea and air warfare, to minimize incidental loss of life, Article 57.4 of Protocol I. Article 57.2 (a)(iii) raises the humanitarian threshold of what is militarily necessary to remain within acceptable incidental losses. The scale of proportional damage permissible is heightened from what are “reasonable precautions” to arguably refraining from what later could be judged as “excessive in relation to the concrete and direct military advantage anticipated”. After an airstrike, it is not possible for a commander to predict whether, a week later, hostilities will end, thus accentuating any possibly unwarranted escalation, or whether the conflict will last for years, thus rendering actions relative to overall military gains. A loophole to this obligation is the Rendulic rule, developed as a result of the acquitted trial of a German general for action in a WWII campaign in Norway. The rule “holds that a commander in the field is not to be judged by knowledge gained in hindsight” but on the information at hand. [352] A legal grey zone herein acknowledges the complexities of waging war, called the heat of battle or fog of war.

Although the Geneva Conventions, last updated in 1977, do not explicitly specify an obligation by states to use PGMs, enhanced by GPS, or even better, to deploy aircraft with laser-wielding forward air controllers, ground-based if possible, these means, when available, could be argued to be legally required depending on the circumstances at the time.

The counter argument also requires consideration of the military necessity to only use PGMs instead of dumb, gravity-driven bombs when there is no possible way of knowing if PGMs, in limited stock, should also be reserved or restricted to the most important combat missions, possibly yet to come, during the whole of a conflict’s duration. NATO’s PGM reserves, largely concentrated in the first five weeks of the 1999 campaign were said to have become increasingly exhausted. IHL also does not require that pilots and ground spotters sacrifice their lives and a state’s full inventory of aircraft and materials to preserve every civilian life endangered in the midst of conflict.

As a second major conclusion, research has also identified three areas which merit legal clarification and emphasis in future state deliberations to update the laws of armed conflict, at least pertaining to air warfare.

1) There is a need to clarify the legality of targeting and striking certain dual-use civilian-military facilities and infrastructure, referred to in Article 52.2 of Protocol I, as they present the greatest challenge to the protection of non-combatants and to the effective work of humanitarian organizations. In Kosovo, research showed that the targeting of dual-use facilities was highly unpopular. In Iraq and Afghanistan, such political concerns were largely unimportant or unreported.

The counter argument would reveal that it is militarily impossible to codify the innumerable circumstances under which the targeting of such infrastructure would be deemed illegal, given the fog of war and the vulnerability of all law to state re-interpretation. While roads, bridges, ports, airstrips, energy sources and communications networks remain vital to relief agency operations, use of such facilities by belligerents makes them a legal target for destruction.

2) There is also a need to strengthen airpower coalition member obligations to provide, as per Article 57.2 (c) of Protocol I, clear, public, repeated and accessible warnings of imminent bombing missions and their areas of concentration, for the benefit not only of civilians in those areas but also to permit the access of impartial humanitarian agencies, NGOs and their local partners. Coalitions must be much more forthcoming in making such IHL compliance evident. Unlike in 1999, the US-UK coalition failed numerous times to issue such warnings. Afghan civilians and aid workers were killed or injured as a result. The coalition also allowed itself to fall victim to high dependence on local warlords with political agendas void of IHL concerns at the expense of civilian lives.

3) As particularly evident in Afghanistan, there remains a need for legal clarification in Article 44.3 of Protocol I on the obligations of identification of non-uniformed combatants. During the 1970s state deliberations which culminated with Protocol I, the demand for a uniform or recognizable sign was dropped, on condition that such out-of-uniform combatants openly carried their weapons. [353] By legitimizing out-of-uniform, non-state guided violence at the expense of clear codification, articles 44.3 and 51.3 accentuated the difficulties for combatants to fulfill the obligations of Article 57 on precautions during attack. It is unrealistic to expect states to increase IHL compliance while non-state or extra-state entities like Al-Qaeda find comfort in legal ambiguities. For the sake of protections owed to all civilians, there must be no room for turnstile “civilian combatants”.

There is a fine legal line between the operations of military coalitions to combat armed, non-uniformed, non-state, international terrorists and international police efforts (CIA, Interpol) to apprehend criminals. The new International Criminal Court could possibly address the issue of IHL obligations for such civil police actions.

As a third general conclusion, research has determined a number of reasons for which, until the Afghanistan air campaign, civilian casualties had proportionally fallen and PGM-use risen.

a) The principal reason: increased political involvement in targeting decisions. Coalition politicians have limited a commander’s freedom to determine what is militarily necessary to accomplish military objectives. This joint civilian-military management of air campaigns greatly restrained airpower use from ideas of military victory to more political objectives enabling diplomatic success while salvaging vital coalition cohesion and minimizing further political embarrassments from targeting failures and human error. In 1999, the political survival of coalition leaders and of coalition unity was as important as any military success. IHL compliance benefited immensely from the input of political concerns regarding targeting. The cases revealed that, regardless of technology, there were fewer civilian casualties when Rules of Engagement became politicized.

It must be acknowledged that not all case studies are comparable. While in 1999, politicians sought the use of airpower for reaching conflict resolution and international accommodations with a belligerent, in 2001 airpower sought unconditional regime change. This was not the case against the regimes of Iraq and Yugoslavia, whose preservation was a political objective. ROE and decisions about military necessity were not the same in each case.

b) The second reason: technological advances in airpower. Bombing and aircraft technology have exponentially enabled a simultaneous and complementary economy of force and a minimizing of civilian casualties, accomplishing with fewer bombs the destruction of targets previously requiring, as during the Gulf War, more numerous strike sorties and more risks to pilot and civilian. PGMs have become cheaper over the last decade. No US or coalition pilots have been lost to enemy fire since the Gulf War although the Serbs downed two US planes in 1999.

The human element in managing such technology however remains paramount. The deployment of coalition ground spotters enabled greater targeting accuracy in Iraq and Afghanistan. In populated areas of Iraq and Yugoslavia, an absence of ground-based intelligence, accentuated by weather obstructions, weakened the ability of such technology to totally avoid collateral damage to civilians and non-military objectives. Despite incredible advances in science, communications, bombing equipment, aviation, naval base support and space technologies, military capabilities have still proven fallible. Given NATO’s deliberate striking of a government procurement facility later revealed to be the Chinese embassy in Belgrade, even the use of PGMs absent requisite human intelligence has its limits. Mistakes will always be possible, such as the deaths of 4 Canadian soldiers killed by a US-dropped PGM in Afghanistan in April 2002.

However, Afghan air defences, made irrelevant within days, were a far cry from Iraqi and Serb anti-aircraft systems which had forced coalition pilots higher and which increased the susceptibility of civilians to bombs. Given a weaker enemy in the Taliban, coalition planes capable of flying at lower altitudes could have resulted in fewer civilian casualties. The presence of ground spotters to facilitate accurate targeting should also have meant fewer civilian casualties. Although fighter jets carry fewer bombs, bombing by means of fighter jet, assisted by laser-pointing ground spotters, or pilot-calibrated lasers, could have afforded the greatest likelihood of bombing accuracy and reduced vulnerability to civilians at a fraction of the cost of $28 billion dollar B-1 bombers.

c) The third reason: increased legal obligations by principal airpower coalition members. Both Britain and France have ratified Protocol I since the Gulf War. The US and Turkey remain the only non-ratifying NATO members.

d) The fourth reason: the importance of coalition cohesion. Regardless of the varying legal obligations of coalition members, the case studies revealed the influence of incidents of collateral damage upon coalition cohesion and IHL compliance. Coalition cohesion positively affected the targeting decisions of commanders and politicians, resulting in greater concerns about civilian casualties. After coalition strikes upon the Amiriya site in Baghdad in 1991 and the Chinese embassy in Belgrade in 1999, the repercussions upon coalition unity from public reports of collateral damage weighed heavily upon the minds of politicians and commanders, thus increasing the employ of PGMs in order to minimize embarrassing airpower mistakes.

Civilian coalition leaders emphasized the importance of using PGMs instead of dumb bombs not only because of concerns about civilian casualties and IHL compliance but also due to the importance of hitting particular targets in order to maintain coalition cohesion. PGMs can best avoid the potential repercussions, political and strategic, of missing key objectives. For instance, during SCUD-busting missions in west Iraq, PGMs were employed in non-urban areas due to the political importance of preventing further Iraqi attacks upon coalition allies Saudi Arabia and Israel.

Maintenance of coalition cohesion was paramount in 1991 as it was in 1999. Larger coalitions have resulted in greater IHL compliance by playing a pacifying role upon the military conduct of each other. The small US-UK coalition failed to meet higher expectations of IHL compliance, meaning proportionally fewer civilian casualties, earlier reached in 1999. While the US refused to compile statistics on numbers of civilian casualties or incidents with collateral damage, coalitions eventually became sensitive to such public figures.

e) Increased communications and formal relations between the ICRC and coalitions, particularly with NATO members since the Gulf War. These measures expanded state understanding of the purposes and obligations of the Geneva Conventions, of the right of war victims to receive provisions and protection, and of the right of access of humanitarian organizations during conflict.

f) Increased presence and influence of global media during air campaigns, especially upon the targeting decisions of politicians given public opinion about airpower mistakes committed by pilots. There has also been an increased vigilance by humanitarian and human rights organizations of coalition conduct during air campaigns. Decisions and results on airpower targeting and the consequences thereof have never been so accessible to the general public. The role and impact of the media during air campaigns remains a vast area deserving of further examination.

As a fourth general conclusion, research on humanitarian operations during air campaigns has proven that humanitarian action by international relief organisations is not impossible but has only positively evolved most recently in the last case study. Such relief action was briefly attempted in Kosovo and was greatly increased during the bombing of Afghanistan. Although neither conflict offered the same ground realities, the increased presence of specialized relief agencies during air campaigns revealed that some progress had been made by the humanitarian community to convince coalitions of the importance of their access.

However, until the bombing of Afghanistan, the legal access of impartial relief agencies to areas under bombardment and the right of victims to receive provisions were largely not respected and remained highly misunderstood by air force coalitions, their military commanders and political leaders. Thousands of civilian lives affected by air campaigns depended on this legal understanding. The US CENTCOM seems to have changed in this regard, much to the benefit of WFP operations in Afghanistan in 2001.

The creation and respect of neutral, demilitarized zones for civilians, as outlined in Article 15 of the 1949 Geneva Conventions, did not take place during any of the air campaigns, evidence that humanitarian-military relations require greater mutual understanding of each other’s weaknesses and opportunities. The militarization of aid drops in 2001 proved useless, wasteful and detrimental in the end. It is coalition leaders more than soldiers who remain most responsible for the blurring of humanitarianism.

It will also be necessary for the international community to better address the implications of military aid drops for humanitarian policy during future air campaigns. The harnessing of military zeal for aid drops into greater coordination with relief agencies, as evident in Kosovo in 1999, could make coalitions a much lesser threat to the humanitarian space.

A combination of pressures and capacities, from legal obligations, political involvement in targeting decisions and technology to humanitarian-military relations, concerns for coalition unity and media access to air campaigns, have contributed to making the conduct and effects of airpower safer for civilians and aid workers since the Gulf War, although unfortunately less so in the case of Afghan civilians during Operation Enduring Freedom.

[1] Wesley K. Clark, Waging Modern War: Bosnia, Kosovo and the Future of Combat, (New York: Public Affairs, 2001): 259. General Clark failed to acknowledge that, as per Article 2 of the 1949 Geneva Conventions, conflict need not be officially declared for IHL to apply.

[2] Judith G. Gardam, “Noncombatant Immunity and the Gulf Conflict”, 826, citing Abraham D. Sofaer, “The Rational for the United States Decision,” in American Journal of International Law, 82 (1988): 784.

[3] Peter Rowe, “Kosovo 1999: The air campaign – Have the provisions of Additional Protocol I withstood the test?” in International Review of the Red Cross (herein IRRC), 82, 837 (31 March 2000): 1-2.

[4] Matthew C. Waxman, International Law and the Politics of Urban Air Operations, (Santa Monica: RAND, 2000): 12, referring to US Air Force Pamphlet (AFP) 110-31, International Law – The Conduct of Armed Conflict and Air Operations, (1976): 5-9.

[5] Garth J. Cartledge, “Legal Constraints on Military Personnel Deployed on Peacekeeping Operations,” in The Changing Face of Conflict and the Efficacy of International Humanitarian Law, Helen Durham and Timothy L. H. McCormack, eds. (The Hague: Martinus Nijhoff, 1999): 123.

[8] Lt Col John G. Humphries, “Operations Law and the Rules of Engagement in Operations Desert Shield and Desert Storm,” in Airpower Journal, 6, 3 (fall 1992): 9, off website www.airpower.maxwell.af.mil/airchronicles/apj/hump.html

[9] Maj. Ariane L. DeSaussure, “The Role of the Law of Armed Conflict During the Persian Gulf War: An Overview,” in Air Force Law Review, 37 (1994): 65.

[10] Col. David N. Blackledge, CENTCOM Public Affairs office, Tampa, Florida, phone conversation, 14 March 2002.

[11] Christopher Greenwood, “Customary international law and the First Geneva Protocol of 1977 in the Gulf conflict,” in The Gulf War 1990-91 in International and English Law, Peter Rowe, ed., (London: Sweet and Maxwell, 1993): 65.

[12] Grant T. Hammond, “Myths of the Gulf War,” in Airpower Journal, 12, 3 (fall 1998): 6, www.airpower.maxwell.af.mil/airchronicles/apj/apj98/fal98/hammond.html

[14] Ian Traynor, “The unfinished war – Afghans are still dying as air strikes go on,” in The Guardian, 12 February 2002, 4. Because the US refused to estimate the number of Iraqi military and civilian casualties, figures on deaths differ greatly, with Greenpeace estimating between 5,000 and 15,000 civilian deaths. See Anthony H. Cordesman, The Lessons of Modern War, volume IV: The Gulf War, (Boulder: Westview, 1996): 27-28 and 342-3.

[15] Grant T. Hammond, “Myths of the Gulf War,” 2. On 25 February, 28 US military were killed and over 100 wounded by debris from an Iraqi SCUD missile fallen near US barracks at Dhahran, Saudi Arabia. See David Travers, “A Chronology of Events” in The Gulf War 1990-91 in International and English Law, Peter Rowe, ed. (London: Routledge, 1993): 25. The US declared 24 troops missing-in-action, including under ten pilots. See Tracy Wilkinson and Nick B. Williams, “Iraq to return remains of 14 Allied dead,” in Los Angeles Times, 12 March 1991, A1. The US lost 145 troops due to non-combat deaths. See Juan Tamayo, “Low Death Count Reflect U.S.‘s Technological Might, Taliban’s Weaknesses,” in Miami Herald, 01 March 2002.

[17] John G. Humphries, “Operations Law and the Rules of Engagement in Operations Desert Shield and Desert Storm.”, 7, citing briefing by Gen. Merrill A. McPeak, Air Force Chief of Staff, 15 March 1991, in US Department of Defence, Conduct of the Persian Gulf Conflict: An Interim Report to Congress, July, 1991, p. 4-5.

[19] Alberto Bin, Richard Hill and Archer Jones, Desert Storm: A Forgotten War, (Westport: Praeger, 1998): 103.

[20] A.P.V. Rogers, “Zero-casualty warfare,” 2, citing Allen, Charles, ed. Thunder and Lightening, HMSO, 1991, pp. 74, 80.

[22] US Government Accounting Office, Operation Desert Storm: Evaluation of the Air Campaign, GAO/NSIAD-97-134, (06/12/1997) www.fas.org/man/gao/nsiad97134/index.html

[24] John G. Humphries, “Operations Law and the Rules of Engagement in Operations Desert Shield and Desert Storm,”6.

[25] Ariane DeSaussure, “The Role of the Law of Armed Conflict During the Persian Gulf War,” 58, citing Thomas Matthews, “The Secret History of the War,” in Newsweek, (March 18 1991): 28, 36.

[26] John G. Humphries, “Operations Law and the Rules of Engagement in Operations Desert Shield and Desert Storm.”, 12, citing W. Hays Parks, Chief, International Law Branch, Office of the Army Judge Advocate General, Memorandum, Review of Coalition Air Operations, 10 December 1991.

[28] Iraq had announced the recovery of some 288 bodies, including 91 children, according to Martin Walker, David Fairhall and John Hooper, “The Gulf War: Bush faces backlash over bunker deaths,” in The Guardian, 15 February 1991. The wreckage has become a shrine dedicated to the 408 civilians killed, according to Timothy Appleby, “Inside the ‘axis of evil’,” in Globe and Mail, 09 March 2002, F8.

[29] Martin Walker, David Fairhall and John Hooper, “The Gulf War: Bush faces backlash over bunker deaths,” in The Guardian, 15 February 1991.

[30] Ariane DeSaussure, “The Role of the Law of Armed Conflict During the Persian Gulf War,” 64, citing Chairman of the Joint Chiefs of Staff, General Colin Powell, briefing of 23 January 1991.

[31] Steven Keeva, “Lawyers in the War Room,” in American Bar Association Journal, 77 (December 1991): 58, citing W. Hays Parks without reference. During the Iran-Iraq war, most Western states supported Iraq, including the building of such bunkers.

[33] Hella Pick, “PoW’s death in allied raid raises fears of Saddam’s human shield ploy,” in Guardian, 30 January 1991.

[34] John G. Humphries, “Operations Law and the Rules of Engagement in Operations Desert Shield and Desert Storm.”, 8.

[35] Yves Etienne and P. Giorgio Nembrini, “Establishing water and sanitation programmes in conflict situations: The case of Iraq during the Gulf War,” in Social and Preventive Medicine, 40, 1 (01 January 1995): 18-26. www.icrc.org/icrceng.nsf/4dc394db5b54f3fa4125673900241f2f/dee32cbe2370bf1f412, citing WM Arkin, D. Durrant and M. Cherni, On impact: modern warfare and the environment: a case study of the Gulf War, (Washington DC: Greenpeace, 1991).

[36] Susan Okie, “Health Official Sees Threat of Epidemics in Iraq; Allied bombing Leaves Shortage of Clean Water and Sewage Treatment Facilities,” in Washington Post, 26 February 1991, A10.

[37] Yves Etienne and P. Giorgio Nembrini, “Establishing water and sanitation programmes in conflict situations: The case of Iraq during the Gulf War,” 18-26.

[39] Anthony Cordesman, The Lessons and Non-Lessons of the Air and Missile Campaign in Kosovo, (Westport: Praeger, 2001): 43, citing US Department of Defence, Report to Congress: Kosovo/Operation Allied Forces After Action Report, Washington, 31 January 2000, 32, 67-69, 87-88, 104.

[40] NATO commander Gen. Clark predicted a 40 percent chance of a 4-day war. See Dana Priest, “United NATO Front Was Divided Within,” in Washington Post, 21 September 1999, http://www.washingtonpost.com/wp-srv/inatl/daily/sept99/airwar21.htm NATO Secretary General Javier Solana saw the campaign as lasting “days, not months.” See Wesley Clark, Waging Modern War, 208.

[41] Anthony Cordesman, The Lessons and Non-Lessons of the Air and Missile Campaign in Kosovo, 21, citing Steven Erlanger, “NATO was closer to ground war than is widely believed,” in New York Times, 7 November 1999, A-6.

[42] Anthony Cordesman, “The Lessons and Non-Lessons of the Air and Missile Campaign in Kosovo,” citing Rebecca Grant, The Kosovo Campaign: Aerospace Power Made It Work, (Arlington: Air Force Association, September 1999): 6.

[43] Human Rights Watch, Civilian Deaths in the NATO Air Campaign, (New York: Human Rights Watch, February 2000), 12 (1D), 4, www.hrw.org/reports/2000/nato/Natbm200.htm

[44] International Criminal Tribunal for the Former Yugoslavia (ICTY), “Final Report to the Prosecutor by the Committee Established to Review the NATO Bombing Campaign Against the Federal Republic of Yugoslavia,” (herein “ICTY Final Report”) in International Legal Materials, 39 (08 June 2000): 1283.

[45] Dana Priest, “France played skeptic on Kosovo attacks,” Washington Post, 20 September, 1999. http://www.washingtonpost.com/wp-srv/national/daily/sept99/airwar20.htm. Three US prisoners of war were released during the campaign, after ICRC reminders of IHL obligations and the visit of US diplomat Jesse Jackson to Belgrade.

[46] Kenneth Roth, executive director of Human Rights Watch - New York, letter to The Guardian, 15 February 2002, 23.

[50] William M. Arkin, “Operation Allied Force: ‘The Most Precise Application of Air Power in History’,” in War Over Kosovo: Politics and Strategy in a Global Age, Andrew J. Bacevich and Eliot A. Cohen, eds. (New York: Columbia University, 2001): 21.

[52] Bjorn Moller, “Kosovo and the Just War Tradition,” in Copenhagen Peace Research Institute, working papers, (August 2000): 8, Columbia International Affairs Online, www.cc.columbia.edu/sec/dlc/ciao/wps/mob11/mob11.html See also Wesley Clark, Waging Modern War, 214 and 314.

[53] Raf Casert, “NATO admits hitting Kosovar rebels at border post,” from Nando Media, 22 May 1999, http://archive.nandotimes.com/Kosovo/story/general/

[55] Timothy L. Thomas, “Kosovo and the Current Myth of Information Superiority,” in Parameters, (spring 2000) http://carlisle-www.army.mil/usawc/parameters/00spring/thomas.htm

[56] Clinton considered the ground option for the first time on 19 May. See William Shawcross, Deliver Us from Evil: Peacekeepers, Warlords and a World of Endless Conflict, (New York: Simon and Schuster, 2000): 382.

[60] A 1999 ICRC worldwide survey on modern war determined that about two-thirds of the public in France and the UK accepted an absolute standard on distinguishing combatants and civilians, compared to a majority in the US: “Those that accept less than an absolute standard are more likely to accept military practices that put civilians at risk… Fully 63 per cent of the American public say there are no laws to stop attacks on populated areas, but so do 46 per cent of the British and 30 per cent of the French. See Greenberg Research, People on War: country report France, United Kingdom, United States, ICRC worldwide consultation in the rules of war, (Geneva: October 1999): iii-iv http://www.icrc.org/eng/onwar_reports

[62] Steven Erlanger, “Amnesty Slams NATO bombing as violation of international law,” of New York Times, in Globe and Mail, 8 June 2000, A10.

[67] David Rohde and Eric Schmitt, “Taliban Give Way in Final Province Where They Ruled,” in New York Times, 10 December 2001.

[68] Murray Campbell, “Thousands of Afghans likely killed in bombings,” in Globe and Mail, 03 January 2002, A1.

[69] Laura King/Associated Press, “Civilian toll in Afghan war likely lower: A review suggests hundreds fell to U.S. bombs, not the thousands the Taliban said,” in Philadelphia Inquirer, 12 February 2002, A1.

[70] Ian Traynor, “The unfinished war – Afghans are still dying as air strikes go on,” in The Guardian, 12 February 2002, 4.

[71] After a one-month investigative field trip in Afghanistan, the HRW team returned to New York in early April. Kenneth Roth, executive director of Human Rights Watch - New York, email to author, 05 April 2002 and on 25 June 2002.

[72] Carl Conetta, Operation Enduring Freedom: Why a higher rate of civilian casualties, Briefing Paper #11, Project on Defense Alternatives, 18 January 2002, www.comw.org/pda/0201oef.html#appendix1.

[74] Murray Campbell, “Thousands of Afghans likely killed in bombings,” in Globe and Mail, 03 January 2002, A1. Also Michael Massing, “Grief Without Portraits,” in The Nation, 04 February 2002.

[75] These sources seemed more attuned to the issue of civilian casualties than were US newspapers. See Carl Conetta, Operation Enduring Freedom: Why a higher rate of civilian casualties.

[77] Kenneth Roth, executive director of Human Rights Watch - New York, letter to The Guardian, 15 February 2002, 23.

[78] Juan Tamayo, “Low Death Count Reflect U.S.‘s Technological Might, Taliban’s Weaknesses,” in Miami Herald, 01 March 2002.

[79] Murray Campbell, “Thousands of Afghans likely killed in bombings,” in Globe and Mail, 03 January 2002, A1.

[80] Roberto J. Gonzales, “Pentagon veils true toll of war,” in San Francisco Chronicle/Deseret News, 06 January 2002, A7.

[81] Barry Bearak, Eric Schmitt and Craig S. Smith, “Uncertain Toll on the Fog of War: Civilian Deaths in Afghanistan,” in New York Times, 10 February 2002, 1.

[82] Ian Traynor, “The unfinished war – Afghans are still dying as air strikes go on,” in The Guardian, 12 February 2002, 4.

[83] John Hendren, “Afghanistan yields lessons for Pentagon’s next targets,” in Los Angeles Times, 21 January 2002, A1. Article emailed to author by Emily Clark of the Center for Defence Information (USA) on 22 January 2002.

[84] James Dao and Thom Shanker, “US ready to send additional troops to hunt bin Laden,” in New York Times, 21 November 2001, www.nytimes.com/2001/11/21/international/asia/21MILI.html The same figure of 60 percent was used on 24 December. See Eric Schmitt and James Dao, “Use of Pinpoint Air Power Comes of Age in New War,” in New York Times, 24 December 2001. www.nytimes.com/2001/12/24/international/24WEAP.html

[85] Daniel Green of www.danshistory.com Email to author on 12 January 2002, pursuant to an enquiry by the author to Globe and Mail global issues correspondent Marcus Gee, who closely monitored the campaign and stated that 9 out of 10 bombs dropped by the US had been PGMs. See Marcus Gee, “Mostly the Right Stuff: How the Americans make war,” in Globe and Mail, 22 December 2001, A21.

[86] John A. Tirpak, “Enduring Freedom,” in Air Force Magazine, 85, 2 (February 2002): 4, www.afa.org/magazine/Feb2002/0202airwar.asp

[87] Laura King/Associated Press, “Civilian toll in Afghan war likely lower: A review suggests hundreds fell to U.S. bombs, not the thousands the Taliban said,” in Philadelphia Inquirer, 12 February 2002, A1.

[88] Carl Conetta, Operation Enduring Freedom: Why a higher rate of civilian casualties, citing Richard Newman, “How the war in Afghanistan is being run from Florida,” in US News and World Report, 17 December 2001, 18.

[91] Rebecca Grant, “The War Nobody Expected,” in Air Force Magazine, 85, 4 (April 2002) www.afa.org/magazine/April2002/0402airwar.asp

[92] Reuters, “US had Omar in sights, magazine says,” in Globe and Mail, 15 October, 2001, A2. See also Seymour M. Hersh, “King’s Ransom,” in The New Yorker, 22 October 2001 www.zmag.org/hirshsaudi.cfm

[93] Thomas E. Ricks, “U.S. arms unmanned aircraft,” in Washington Post, 18 October 2001 www.s-t.com/daily/10-01/10-18-01/a02wn013.htm

[95] Juan Tamayo, “Low Death Count Reflect U.S.‘s Technological Might, Taliban’s Weaknesses,” in Miami Herald, 01 March 2002. In five and a half months, the US lost 1 CIA agent on 28 November, 3 soldiers due to friendly fire from US bombs on 05 December, 1 soldier in combat on 04 January, and a total of 26 troops during accidents, 21 of which were involved in aircraft crashes.

[96] The team acknowledged but a handful of bombing malfunctions. See Barry Bearak, Eric Schmitt and Craig S. Smith, “Uncertain Toll on the Fog of War: Civilian Deaths in Afghanistan,” in New York Times, (10 February 2002): 1.

[97] Laura King/Associated Press, “Civilian toll in Afghan war likely lower: A review suggests hundreds fell to U.S. bombs, not the thousands the Taliban said,” in Philadelphia Inquirer, 12 February 2002, A1.

[98] Charles J. Hanley, “Bombing victims gather in Kabul, ‘angry and sad,’ to press U.S. Embassy for compensation,” from Associated Press, 06 April 2002.

[100] Barry Bearak, Eric Schmitt and Craig S. Smith, “Uncertain Toll on the Fog of War: Civilian Deaths in Afghanistan,” in New York Times, (10 February 2002): 1.

[101] Jim Clancy, Martin Savidge and David Grange, “Civilian Casualties in Fog of War,” on CNN International, 11 February 2002.

[102] ICRC Press Release 01/48, Bombing and occupation of ICRC facilities in Afghanistan, 26 October 2001. Corroborated by author’s interview of field worker.

[103] Jamie McIntyre, “Pentagon probes bombing of Kabul Red Cross,” in CNN, 19 March 2002. www.cnn.com/2002/03/19/ret.pentagon.redcross/index.html

[104] Barry Bearak, Eric Schmitt and Craig S. Smith, “Uncertain Toll on the Fog of War: Civilian Deaths in Afghanistan,” in New York Times, 10 February 2002, 1.

[105] John Donnelly and Anthony Shahid, “Civilian Toll in US raids put at 1,000 bombing flaws, manhunt cited,” in Boston Globe, 17 February 2002, A1.

[109] John Donnelly and Anthony Shahid, “Civilian Toll in US raids put at 1,000 bombing flaws, manhunt cited,” in Boston Globe, 17 February 2002, A1.

[110] Edward Cody, “Villagers, U.S. At Odds Over Lethal Bombing: Residents Say Al Qaeda, Taliban Never There,” in Washington Post, 10 January 2002, A1.

[111] Ian Traynor, “The unfinished war – Afghans are still dying as air strikes go on,” in The Guardian, 12 February 2002, 4.

[112] Anthony Shahid, “Victims of Circumstance,” in Middle East Report, 222 (spring 2002) www.merip.org/mer/mer222/222_shahid_main_article.html

[115] This was the legal argument, in para. 29, used by Christopher Greenwood in his memorandum to the House of Commons of the United Kingdom on 22 November, 1999, to defend the use of means employed by NATO against the Federal Republic of Yugoslavia, the Legality of the Use of Force case before the International Court of Justice. See website: http://www.parliament.the-stationery-office.co.uk/pa/cm199900/cmselect/cmfaff/28/0020802.htm

[116] Thomas A. Keaney and Eliot A. Cohen, Summary Report: Gulf War Air Power Survey, (Washington: Office of the Secretary of the Air Force, 1993): 22, 69, 219, cited in Alberto Bin, Desert Storm: A Forgotten War, 108.

[117] Benjamin S. Lambeth, NATO’s Airwar for Kosovo: A Strategic and Operational Assessment, (RAND, 2001): 49.

[119] US General Mike Short, Air Force commander in charge of the NATO air campaign and operating under US General Clark, cited by David Halberstam, War in a Time of Peace: Bush, Clinton and the Generals (New York: Scribner, 2001): 444.

[121] Don D. Chipman, “The Balkan Wars: Diplomacy, Politics and Coalition Warfare,” in Strategic Review, (winter 2000): 30.

[122] A large degree of politicized targeting was also evident during the 16-day NATO air campaign against Bosnian Serbs in 1995. See Paul C. Forage, “Bombs for Peace: A Comparative Study of the Use of Air Power in the Balkans,” in Armed Forces and Society, 28, 2 (winter 2002): 216.

[123] Thomas E. Ricks and Alan Sipress, “Attacks Restrained by Political Goals,” in Washington Post, 23 October 2001, A1, www.washingtonpost.com/wp-dyn/articles/A36257-2001Oct22.html

[124] Operation Enduring Freedom and the Conflict in Afghanistan: an Update, Research Paper 01/81, International Affairs and Defence Section, House of Commons Library, (London: 31 October 2001) www.parliament.uk/commons/lib/research/rp2001/rp01-081.pdf

[125] Col. David N. Blackledge, CENTCOM Civil Military Operations, Tampa, Florida, phone conversation, 14 March 2002. See also Alissa J. Rubin, “Humanitarian Aid a War Victim,” in Los Angeles Times, 13 November 2001.

[126] Ivo H. Daalder and Michael E. O’Hanlon, Winning Ugly: NATO’s War to Save Kosovo, (Washington: Brookings Institution, 2000): 20.

[127] William M. Arkin, “Operation Allied Force: ‘The Most Precise Application of Air Power in History’,” in War Over Kosovo: Politics and Strategy in a Global Age, Andrew J. Bacevich and Eliot A. Cohen, eds. (New York: Columbia University, 2001): 3-5.

[128] It would be useful to know if Gen. Franks was stationed at Tampa (while Gen. Schwarzkopf was in Saudi Arabia in 1991) because of its time zone with Washington, thus enabling easier political access to targeting performance, something that proved frustrating to Washington in 1999 while Clark was based in Belgium.

[130] For Czech, Hungarian and Greek reservations on NATO airpower use, see Serge Schmemann, “Storm Front: A New Collision of East and West,” in New York Times, 04 April 1999, cited in Simon Chesterman, Just War or Just Peace? Humanitarian intervention and international law, (Oxford: Oxford University Press, 2000): 220.

[136] During the 50^th anniversary of NATO, Clark was not originally welcome by the US Defence Secretary. The 1986 Goldwater-Nichols Act had intended to correct the Vietnam lessons of decentralized command in executing wars by ensuring more joint operations between the US military services as well as greater Pentagon support to field commander decisions. While US General Schwarzkopf as Gulf War commander benefited tremendously from the Act, it had not envisioned the complexities of NATO disunity and Clark’s position within this vortex. See David Halberstam, War in a Time of Peace, 455.

[138] David G. Haglund and Allen Sens, “Kosovo and the case of the (not so) free riders: Portugal, Belgium, Canada and Spain,” in Kosovo and the Challenge of Humanitarian Intervention: Selective Indignation, Collective Action and International Citizenship, Albrecht Schnabel and Ramesh Thakur, eds. (Tokyo: United Nations University Press, 2000): 195.

[139] Canada’s 18 CF-18 fighter jets were PGM-capable, unlike most of the other allied air forces. Although accounting for 2 percent of the 912 allied aircraft in the operation, because of superior training experience, Canadians flew 10 percent of all strike sorties, often led by Americans or Canadians. See David G. Haglund and Allen Sens, “Kosovo and the case of the (not so) free riders,” 195.

[140] Alan Freeman, “U.K. to hand suspects to Afghans,” in Globe and Mail, 18 January 2002, A4, citing comments made by Michael Byers, Professor of Law, Duke University, who stated that his remarks were learned during discussions with a “very high ranking” US Air Force officer during an academic conference. Email communication of 11 February 2002.

[145] See Thomas Risse-Kappen, Cooperation Among Democracies: the European Influence on U.S. Foreign Policy, (Princeton: Princeton University Press, 1995).

[146] Michael Ignatieff, Virtual War, 206, citing Anthony H. Cordesman, The Lessons and Non-Lessons of the Air and Missile Campaign in Kosovo.

[150] Charles J. Dunlap, “Technology: Recomplicating Moral Life for the Nation’s Defenders,” in Parameters, (autumn 1999): 3, http://carlisle-www.army.mil/usawc/parameters/99autumn/dunlap.htm

[151] Especially monopolized by CNN. Founded in 1980 as the world’s first 24-hour all news channel, CNN’s coverage of the war from Baghdad led to the post-Cold War coining of the term “CNN-factor,” a phenomenon explaining media influence upon foreign policy.

[152] Although the Taliban had decreed the departure of all foreign journalists, two Al-Jazeera correspondents and three Afghan reporters working for Reuters, Agence France Presse and Associated Press had been allowed to stay. See BBC, “Al-Jazeera goes it alone,” 08 October 2001, http://news.bbc.co.uk/hi/english/world/monitoring/media_reports/

[153] An Italian journalist was killed during a NATO airstrike in late May of 1999. See Wesley Clark, Waging Modern War, 334.

[154] Roxanne Farmanfarmaian, “The Media and the War on Terrorism: Where Does the Truth Lie?” in Cambridge Review of International Affairs, 15, 1 (2002): 160.

[155] Roberto J. Gonzales, “Pentagon veils true toll of war,” in San Francisco Chronicle/Deseret News, 06 January 2002, A7. Viewed by some US officials as propagating anti-Western views, Al-Jazeera, the Qatar-based station founded in 1996, broadcast Afghan demonstrators attacking the US embassy in Kabul on 26 September and aired video tapes of Osama Bin Laden statements in October. See BBC, “Al-Jazeera Kabul offices hit in US raid,”13 November 2001.

[156] US intelligence tracked a senior Yugoslav air defence team to Baghdad in February of 1999 for two days. From International Herald Tribune, 31 March 1999, cited in Lawrence Freedman, “The split-screen war: Kosovo and changing concepts of the use of force,” in Kosovo and the Challenge of Humanitarian Intervention: Selective Indignation, Collective Action and International Citizenship, Albrecht Schnabel and Ramesh Thakur, eds. (Tokyo: United Nations University Press, 2000): 431.

[158] Grant T. Hammond, “Myths of the Air War over Serbia: Some ‘Lessons’ Not to Learn,” in Aerospace Power Journal, 14, 4 (winter 2000), www.airpower.maxwell.af.mil/airchronicles/apj/apj00/win00/hammond.htm

[160] Hilaire McCoubrey, “Kosovo, NATO and International Law,” in International Relations, 14, 5 (August 1999): 41.

[166] Dana Priest, “France played skeptic on Kosovo attacks,” in Washington Post, 20 September 1999, 5, www.washingtonpost.com/wp-srv/national/daily/sept99/airwar20.htm

[169] Associated Press, “Yugoslav TV ex-official sentenced to 10 years,” in Globe and Mail, 22 June 2002, A13.

[170] Wesley Clark, Waging Modern War, 296-7. Another claim has been made that the Chinese were assisting Milosevic. See http://www.guardian.co.uk/Kosovo/Story/0,2763,203214,00.html

[172] Don D. Chipman, “The Balkan Wars: Diplomacy, Politics and Coalition Warfare,” 30 General Short, Clark’s Air Force subordinate, had flown in the Gulf War.

[177] Oscar Schachter, “United Nations Law in the Gulf Conflict,” in American Journal of International Law. 85, 3 (July 1991): 466.

[178] Ariane DeSaussure, “The Role of the Law of Armed Conflict During the Persian Gulf War,” 62, citing Harvard Study Team Report: Public Health in Iraq after the Gulf War, May 1991.

[179] Ariane DeSaussure, “The Role of the Law of Armed Conflict During the Persian Gulf War,” 63, citing Bernd Debusmann, “Allied Motives Queried in Raids on Iraqi Plant,” in Washington Post, 28 January 1992, A14.

[180] James A. Burger, “International humanitarian law and the Kosovo crisis: Lessons learned or to be learned,” in IRRC, 82, 837 (March 2000): 133.

[181] Peter Rowe, “Kosovo 1999: The air campaign – Have the provisions of Additional Protocol I withstood the test?” in IRRC, 82, 837 (31 March 2000): 148.

[182] Kenneth R. Rizer, “Bombing Dual-Use Targets: Legal, Ethical and Doctrinal Perspectives,” in Air Power Chronicles, (01 May 2001): 12. www.airpower.maxwell.af.mil/airchronicles/cc/Rizer.html

[183] Average of PDA study estimate by Carl Conetta, Operation Enduring Freedom: Why a higher rate of civilian casualties, 18 January 2002.

[184] According to an ICRC assessment, the 16-day NATO air campaign over Bosnia resulted in the loss of 27 civilians, attributing deaths as a direct result of NATO bombings or of UN artillery fire from Rapid Reaction Forces. See Tim Ripley, Operation Deliberate Force: the UN and NATO campaign in Bosnia 1995, Centre for Defence and International Security Studies, (Lancaster: Lancaster University, 1999): 316. See also the claim that 26 people were killed during the NATO airstrikes, in Don C. Chipman, “The Balkan Wars: Diplomacy, Politics and Coalition Warfare,” 26, citing Col. Robert C. Owen, “Summary of Operation Deliberate Force,” unpublished manuscript, Air University School of Advanced Airpower Studies, Montgomery AL, 64.

[185] During NATO’s 1995 air campaign, 69 percent of released bombs were PGMs, hitting 97 percent of planned targets. See John A. Tirpak, “Deliberate Force,” in Air Force Magazine, 80, 10 (October 1997): 3, www.afa.org/magazine/1097deli.html

[186] Ian Traynor, “The unfinished war – Afghans are still dying as air strikes go on,” in The Guardian, 12 February 2002, 4.

[187] Peter de la Billiere (Sir Gen.), Storm Command: A Personal Account of the Gulf War, (London: Harper Collins, 1992): 224.

[188] For a definition of the humanitarian space, see Beatrice Megevand Roggo, “After the Kosovo conflict, a genuine humanitarian space: A utopian concept of an essential requirement?” IRRC, 82, 837 (March 2000): 31-46.

[189] Unlike the average humanitarian crisis in Asia and Africa, Kosovar Albanians, having lived in the most advanced Communist society in Europe, “did not conform to clichés of destitution”. See Michael Ignatieff, Virtual War, 43. The majority of such refugees stayed with relatives, not in camps. See Ian Christoplos, “Reviewing International Experience and Lessons from Kosovo,” in Kosovo and the Changing Face of Humanitarian Action, Uppsala University conference, 08 May 2001, www.kus.uu.se

[190] Leonard Doyle, “Crisis in the Gulf: Stance of UN stops it helping Iraqi civilians,” in The Independent, 26 January 1991.

[191] Paul Watson, “Iraq blocking aid, Red Cross says ‘political problems’ are cited,” in Toronto Star, 18 February 1991, A12.

[193] Larry Minear, U.B.P. Chelliah, Jeff Crisp, John MacKinlay and Thomas G. Weiss, United Nations Coordination of the International Humanitarian Response to the Gulf Crisis 1990-1992, Occasional Paper 13, (Providence: Brown University, 1992): 23.

[195] Larry Minear, United Nations Coordination of the International Humanitarian Response to the Gulf Crisis 1990-1992, 22.

[196] Rone Tempest, “Red Cross says Iraq blocking efforts on POWs,” in Los Angeles Times, 04 February 1991, A5.

[197] David Travers, “A Chronology of Events,” in The Gulf War 1990-91 in International and English Law, Peter Rowe, ed. (London: Routledge, 1993): 20.

[199] Christophe Girod, Tempete sur le desert: Le comite de la Croix-rouge et la guerre du Golfe 1990-1991, (Paris: Bruylant-Bruxelles, 1995): 168.

[201] Yves Etienne and P. Giorgio Nembrini, “Establishing water and sanitation programmes in conflict situations: The case of Iraq during the Gulf War,” in Social and Preventive Medicine, 40, 1 (01 January 1995): 18-26. www.icrc.org/icrceng.nsf/4dc394db5b54f3fa4125673900241f2f/dee32cbe2370bf1f412

[202] Christophe Girod, Tempete sur le desert: Le comite de la Croix-rouge et la guerre du Golfe 1990-1991, 166.

[203] Olivia Ward, “U.N. seeks ‘peace corridor’ to bring in supplies,” in Toronto Star, 09 February 1991, A12.

[204] Paul Lewis, “War in the Gulf: Relief; Allies Asked to Guarantee Safe Transit on Medical Aid,” in New York Times, 09 February 1991, 7.

[205] Leonard Doyle, “Crisis in the Gulf: UNICEF wants Iraq aid corridor,” in The Independent, 01 February 1991, 2.

[207] Frances Williams, “Crisis in the Gulf: ‘Respect Geneva Conventions’,” in The Independent, 18 January 1991.

[209] Rick Atkinson, “15 Americans Among Second Group of POWs Flying to Freedom,” in Washington Post, 07 March 1991, A25.

[210] In the wake of a defeated Iraq, coalition forces arrived in northern Iraq where returnees and IDPs were to be protected from Iraqi Republican Guards by the threat of airpower. See Nicholas Wheeler, Saving Strangers: Humanitarian Intervention in International Society, Oxford: Oxford University Press, 2000): 151. Over 23,000 US-commanded troops from NATO states, plus Australia, arrived in northern Iraq. See Cpt. Michael J Harrington, “Operation Provide Comfort: A Perspective in International Law,” in Connecticut Journal of International Law, 8, 2 (spring 1993): 650.

[211] David P. Forsythe, “The International Committee of the Red Cross – A policy analysis,” in IRRC, 314 (31 October 1996): 5 of web version.

[213] Tracey Lawson, “Trapped Albanian Villagers Face Starvation in the Hills,” in The Scotsman, 15 April 1999, 10.

[214] Paul Vallely and Christopher Brading, “War in the Balkans: Famine Stalks the Fleeing Hordes as Relief Agencies Struggle to Keep Up,” in The Independent, 01 April 1999, 3.

[215] Michael Pugh, “Civil-Military Relations in the Kosovo Crisis: An Emerging Hegemony?” in Security Dialogue, 31, 2 (June, 2000): 233, citing Nicholas Morris, “UNHCR and Kosovo: A Personal View From Within UNHCR”, in Forced Migration Review, 5 (August 1999): 15.

[217] ICRC Press Release 99/23, Crisis in the Balkans: Yugoslav authorities give ICRC President assurances for a return to Kosovo, 26 April 1999.

[218] Cornelio Sommaruga, “Kosovo: All Sides Must Let the Red Cross Work in Kosovo,” Extract from International Herald Tribune, 26 May 1999.

[221] Cornelio Sommaruga, Visit to the North Atlantic Treaty Organization: Statement by the International Committee of the Red Cross, (Brussels, 22 December 1999), in IRRC, 837, 82 (March 2000): 259.

[223] Douglas Hamilton, “Brussels – No air drop, no convoys for Kosovo stranded,” from Reuters, 14 April 1999.

[224] Michael Pugh, “Civil-Military Relations in the Kosovo Crisis: An Emerging Hegemony?” 236, from Elizabeth Becker, “With NATO in Charge, Relief looks Less Neutral”, in New York Times, 10 April 1999, citing Joelle Tanguy of MSF.

[225] In September of 1998, a UNHCR convoy was turned back by a KLA checkpoint. See Wendy Lubetkin, “UN Appeals for Funds to Avert Catastrophe in Kosovo this Winter,” 08 September 1998, USIA European Correspondent, from Human Rights Watch, Humanitarian Law Violations in Kosovo, (New York: Human Rights Watch, 1998): 86.

[226] In July of 1998, KLA forces temporarily confiscated an MSF vehicle. See “Kosovo Rebels Confiscate an MSF Vehicle,” Agence France Presse, 24 July 1998, from Human Rights Watch, Humanitarian Law Violations in Kosovo, (New York: Human Rights Watch, 1998): 86.

[227] Christoph I. Lang, Swiss Ministry of Foreign Affairs, Department of Humanitarian Aid and SHA, Bern, email communication of 28 February 2002.

[228] Konstantinos Georgiou, head of Operation Focus Greek mission into Kosovo, Greek Ministry of Foreign Affairs, Athens, email to author, 18 March 2002, with collaboration from the Greek embassy in Ottawa.

[230] Konstantinos Georgiou, 18 March 2002. Hellenic Radio (ERA) announced on 26 April that 6 Greek relief trucks arrived from Thessaloniki in Pristina that day accompanied by Greek Ambassador to Belgrade Panayotis Vlasopoulos. Operation Focus was not mentioned. See www.hri.org/news/greek/eraen/1999/99-04-26.erean.html#07

[233] After co-founding MSF, French doctor Bernard Kouchner helped found MDM along similar principles of droit d’ ingerence: the right of humanitarian initiative being more important than state sovereignty.

[234] Associated Press, “Relief Convoy struck by bomb in Kosovo,” from USA Today, 05 May 1999, www.usatoday.com/news/index/kosovo/koso520.htm.

[240] Lang acted as Swiss government liaison with NATO for the security of humanitarian relief operations of the Greek-led Operation Focus during the Kosovo air campaign.

[244] Agence France Presse, “UN Mission to Meet Milosevic as Yugoslavia Gets Respite From Bombing,” 17 May 1999.

[245] OCHA, Report of the Inter-Agency Needs Assessment Mission Mission dispatched by the Secretary General of the United Nations to the Federal Republic of Yugoslavia, 10 June 1999.

[247] There was no available information on NATO coordinating its airstrikes to avoid the traveling ten-day UN mission.

[248] David Ensor and Kevin Bohn, “U.S. group to begin Kosovo food drops Monday,” from CNN, 28 May 1999. Corroborated in phone interview with Col. David Blackledge, 14 March 2002.

[249] William Branigin, “Airdrops Planned for Starving in Kosovo,” in Washington Post, 02 June 1999, A16.

[256] ICRC Press Release 01/32, Afghanistan: ICRC expatriates on standby in Pakistan, 16 September 2001. See also ICRC Press Release 01/56, Afghanistan: ICRC reinforces team, 19 November 2001.

[259] Peter Popham, “Alarm over aid drop in ‘world’s biggest minefield’,” in The Independent, 09 October 2001, www.independent.co.uk/story.jsp?story=98478

[260] Human Rights Watch, Afghanistan and Refugees: Need for Humanitarian Action, Statement to the Senate Foreign Relations Committee, New York, 10 October 2001, www.hrw.org/press/2001/10/afghan-testi-1010.html

[261] DFID Afghanistan Crisis: Situation Report No. 8, October, 2001, cited in Operation Enduring Freedom and the Conflict in Afghanistan: an Update, Research Paper 01/81, International Affairs and Defence Section, House of Commons Library, (London: 31 October 2001) www.parliament.uk/commons/lib/research/rp2001/rp01-081.pdf

[262] Human Rights Watch, Afghanistan: Attacks on Aid Increasing. New York, 18 October 2001, www.hrw.org/press/2001/10/aid1018.htm

[263] Khaled Mansour, WFP Public Affairs Officer in Islamabad, email communication to author, 15 March 2002. The casual worker/porter was released from hospital a day later.

[266] Joe Havely, “The Pentagon has admitted that its bomb had ‘inadvertently’ targeted the warehouse,” in CNN Hong Kong, 17 October 2002.

[267] Jamie McIntyre, “Pentagon probes bombing of Kabul Red Cross,” in CNN, 19 March 2002. www.cnn.com/2002/03/19/ret.pentagon.redcross/index.html

[269] Unlike in Iraq and Yugoslavia, where some journalists were present in some of the targeted areas. See Paul Koring, “First strikes aim to cripple Taliban,” in Globe and Mail, 08 October 2001, A4.

[270] Oxfam Briefing Note, Food has now run out for many Afghan people. 01 November 2001, www.oxfam.org/news/docs/011101_2.htm

[271] CARE USA, Afghanistan: Options for humanitarian access, 25 October 2001, www.care.org/info_center/afghanistan/afghanistanpolicy2.asp

[273] Pakistan Link, Humanitarian liaison center opens in Islamabad, 06 December 2001, www.pakistanlink.com/headlines/Dec/06/15.htm

[275] Mohammad Jalloh, Polio campaign spokesperson, UNICEF office New York, email communication with author, 02 April 2002.

[277] Mohammad Jalloh, 02 April 2002. The campaign, in partnership with the UN World Health Organization, Rotary International, the US Centers for Disease Control and Prevention and many other agencies, had been pre-authorized by the Kabul government at the time and was not restricted to Taliban-held territory.

[279] Col. David Blackledge, CENTCOM Public Affairs office, Tampa, Florida, phone conversation, 14 March 2002.

[282] Cahal Milmo, “Air Strikes On Afghanistan: Humanitarian Operation – Aid Agencies Race Against Time to Feed 7.5 Million,” in The Independent, 13 October 2001.

[289] Alissa J. Rubin, “Humanitarian Aid a War Victim.” See also Afghanistan - OCHA Situation Report No. 20, 12 November 2001, www.pcpafg.org/news/Situation_rep/Afghanistan_Crisis_OCHA_Situation_Report…

[290] Thom Shanker, “A Nation Challenged: Inquiries; Convoy Struck In November Found to Lack U.N. Markings,” in New York Times, 11 April 2002.

[293] David R. Sands, “US, global groups plan mass humanitarian relief effort,” in Washington Times, 16 November 2001, A17.

[296] David R. Sands, “US, global groups plan mass humanitarian relief effort,” in Washington Times, 16 November 2001, A17.

[301] Email communication to author by foreign aid worker, 05 June 2002. See also ICRC News 02/22, “Afghanistan – Ghor airlift ends,” 30 May 2002.

[308] Elizabeth Becker and Eric Schmitt, “US Planes Bomb a Red Cross Site for Second Time,” in New York Times, 26 October 2001.

[309] Christoph Lang, 28 February 2002. Lang took part in UNJLC liaising with the military in Islamabad in October before being posted to the same effort at CENTCOM in Florida in November. See also www.unjlc.org/Bulletins/01/Bulletin01.htm

[311] The UK, through a DFID staff secondment to UNJLC Rome, coordinated the use of military assets. See www.unjlc.org/Bulletin/02/Bulletin02.htm

[314] Sgt William Patterson, “Civil Affairs teams re-establish Afghan relief,” in Army Link News, 21 December 2001, www.dtic.mil/armylink/news/Dec2001/a20011221cateams.html

[325] UNICEF spokesperson Chulho Hyun stated that the Northern Alliance commandeered 10 of their trucks in mid-November, while the Taliban did so in Mazar-I-Sharif. See Alissa J. Rubin, “Humanitarian Aid a War Victim.” Corroborated by Col. David Blackledge, 14 March 2002.

[330] USAID, Central Asia Region – Complex Emergency Fact Sheet #35, 12 December 2001, www.usaid.gov/hum_response/ofda/situation.html

[333] US-dropped CBUs in Indo-china also contained yellow bomblets. The CBU opens during free fall, spreading its bomblets over targeted enemy terrain. The author did previous research for Handicap International in Laos in the mid-1990s where stories abounded of rural children being injured by curiosity to the colour.

[334] Communication to author from Oliver Ulich of UN OCHA New York citing information received from Steve Catlin, USAID representative at CENTCOM during the HDR drop program, 22 April 2002.

[337] Elizabeth A. Neuffer, “Food drops found to do little good,” in Boston Globe, 26 March 2002, A1. www.nytimes.com/2002/01/06/weekinreview/06BECK.hmtl

[338] David Filipov, “Idle Combatants Now Wage Battle for US Aid Drops,” in Boston Globe, 29 November 2000, A1.

[340] Jim Garamone, “U.S. aid helps avert famine in Afghanistan,” in Defense Link: American Forces Press Service, 4 January 2002, www.defenselink.mil/news/Jan2002/n01042002_200201043.html

[341] Elizabeth Becker, “The Danger of Doing Good Deeds.” Relief agencies also claimed the US drops to be worth under one percent of the Afhan need. See Marc Kaufman, “Taliban Seizes Relief Food, Two Main UN Warehouses,” in Washington Post, 18 October 2001, A23.

[342] Associated Press, “Airdropped relief kit kills Afghan civilian,” in Globe and Mail, 30 November 2001.

[343] Agence France Presse, “US ending food, supply drops as relief comes in by road, rail: Pentagon,” 13 December 2001.

[344] Jim Gamarone, “Humanitarian mission averts Afghan starvation,” in Defense Link: American Forces Press Service, 24 January 2002, www.defenselink.mil/news/Jan2002/n01242002_200201242.html

[345] Paul Vallely and Christopher Brading, “War in the Balkans: Famine Stalks the Fleeing Hordes as Relief Agencies Struggle to Keep Up,” in The Independent, 01 April 1999, 3.

[347] Even before the commencement of coalition airstrikes against the Taliban regime, the WFP had initiated air drops due to security problems on roads. See Donald Urquhart, “UN seeks funds to charter aid ships,” in Business Times Singapore, 10 October 2001.

[349] Carl Conetta, Project on Defense Alternatives. See Ian Traynor, “The unfinished war – Afghans are still dying as air strikes go on,” in The Guardian, 12 February 2002, 4.

[351] See summary of statistics on air campaign civilian casualties and PGM use in section conclusion on pages 66-7.

[353] Jacques Moreillon, “Humanitarian Law, The ICRC, and Promoting the Geneva Conventions,” in American University Law Review, 31, (1982): 824-5. Speaking as Director of the ICRC Department of Principles and Law, Moreillon questioned the place of politics in humanitarian law and the mixing of war objectives and military conduct. He stated that wars of national liberation or any other motives for fighting, jus ad bellum, had no place in the codification of laws intended to regulate the means and methods of war, jus in bello. The reasons for warring were not to be the concern of IHL or the ICRC.