Kosovo and the 1997 Landmines Treaty

Tasneem Jamal Conventional Weapons

Mary Foster

The Ploughshares Monitor September 1999 Volume 20 Issue 3

This text was excerpted from a longer paper available from Mines Action Canada (MAC), where Mary Foster is a staffperson. The views expressed in this paper are her own.

The 1997 Landmines Treaty has been celebrated as a humanitarian achievement towards ending the grievous impact of landmines. How did the treaty measure up in the first war Canada entered after it became international law?

Over the last few years, Canada has championed a new way of conducting international relations aimed at reversing regressive trends such as increased civilian casualties in war. The new emphasis, advancing under the flag of “human security,” questions the focus and the preferred methods of traditional state security policy. One of the most prominent elements of this project for Canada has been its support for the 1997 Convention on the Prohibition of the Use, Stockpiling, Production and Transfer of Anti-personnel Mines and on their Destruction. This treaty was intended to increase the security of civilians during and after armed conflict. It is therefore important to ask to what extent it succeeded in doing so in the first military conflict in which Canada engaged since the treaty entered into force, the war in Yugoslavia/Kosovo.

Use of landmines

Anti-personnel mines were heavily used by both the Serbian army and the Kosovo Liberation Army (KLA) both before and after NATO entered the war. The international media reported as early as October 1998 that the KLA was using mines to halt the progress of Yugoslav forces; by January 1999, the KLA was mining roads, blowing up a Serb police vehicle. After the war, British Brigadier John Hoskinson, faced with the task of clearing the mines, said KLA mining was the “joker in the pack.” The KLA’s tactic of digging up and reusing Serb mines – as well as planting its own – had made the records the Serbs handed over to UN forces inaccurate and the mines much more difficult to locate.

Neither entity committed itself to a ban on anti-personnel mines and apparently neither felt constrained by the international norm to any practical degree. Nor did the ban in the rest of the world seem to restrict supply for either body. The Federal Republic of Yugoslavia (FRY) has its own production capacity. It is not known where the KLA procured its landmines, other than by digging up and reusing Serb mines (but it is certainly worth trying to find out). The use of anti-personnel mines by the two parties has contributed to making Kosovo very dangerous to returning refugees and peacekeepers; by 15 July, 150 people had been injured or killed by mines and other unexploded ordnance, according to a World Health Organization press release.

Traditional anti-personnel mines were not used by NATO states, only two of which (the US and Turkey) have not signed the treaty. The US denied allegations that they were using the Gator system, a combination of anti-personnel mines and anti-vehicle mines. Although no evidence to the contrary has come to light, the Americans did assert that they reserved the right to use the weapon.1

Thus, the current situation in Kosovo speaks to the need and the difficulty of convincing intransigent states to sign and adhere to the treaty and it highlights one weakness of the treaty – the failure to address the problem of non-state actors (NSAs), such as the KLA, using landmines. (The Landmines Treaty is directed exclusively at states, although this could be addressed either in a future revision of the treaty or through a parallel process aimed at NSAs.) Furthermore, before declaring the triumph of the treaty in restraining use of anti-personnel mines by the powerful NATO alliance, it is important to look at other weapons with similar effects that NATO forces did use in their bombing campaign and at the impact these had on civilians. (It is worth noting in this context that the Pentagon linked its denial of anti-personnel mine use with an admission that it was using cluster bombs, “We have not dropped those weapons [Gators]. We have dropped … cluster bombs, but we have not dropped the ones that [Senator Leahy] was talking about [Gators].”)

The Landmines Treaty defines an anti-personnel mine as, “a mine designed to be exploded by the presence or contact of a person and that will incapacitate, injure, or kill one or more persons.” Members of the International Campaign to Ban Landmines have advanced an alternate definition based on effect instead of design. According to the effect-based definition, an anti-personnel mine includes “any device or piece of ordnance which, although its primary purpose or design may be other than [to explode by the contact, presence or proximity of a person], can be deployed in a manner to achieve this effect without modification or through a specific design feature.”2 It is clear that if the treaty is to succeed in its determination, expressed in its preamble, “to put an end to the suffering and casualties caused by anti-personnel mines,” the fundamental consideration in deciding which weapons to ban must be how they actually function when deployed. Otherwise, armed forces are free to rely on weapons with similar effects but a different marketing strategy – and we are no further ahead.

The effect-based definition already has some support from governments. Most notably, Italy’s implementing legislation for the Landmines Treaty encompasses mines defined by effect. Canadian representatives involved in the treaty negotiations asked that an interpretation not far from the effect-based definition be read into the record of the conference. During the period between the negotiation and signing of the treaty in December 1997, Canadian officials took the position in several debriefing sessions that the treaty definition was effect-based.

Canada in fact went some way towards implementing this interpretation, ordering the destruction of tilt rod fuses for anti-vehicle mines in November 1997. Tilt rods are “thin flexible rods” which usually emerge from the top of a mine and are designed to be set off by the passage of a tank. In practice, however, they can be detonated by a person – only 2 kg of pressure were required to detonate the tilt rods that Canada destroyed (Mines Advisory Group 1997). Nevertheless, some of the weapons Canada has retained do fall under the effect-based definition, including cluster bombs and anti-tank mines with anti-personnel capability.

Cluster bombs

Cluster weapons are containers filled with hundreds of small explosive devices or munitions. Delivered as bombs, the containers break open in mid-air and scatter their munitions over a wide area. In this way, the munitions are similar to air-delivered anti-personnel mines. Unlike landmines, the cluster munitions are intended to explode just before, during or just after impact.

However, not all cluster munitions do what they are designed to do. Instead of exploding on impact, anywhere between 5 and 30 per cent fail. Their failure can be due to numerous factors, including malfunction, landing in soft earth, water or a tree, or being released from the wrong altitude or at the wrong speed. The unexploded cluster munitions are prone to exploding when someone encounters or innocently tampers with them. As development and human rights experts – the people who are in a position to assess the impact of these weapons in human terms – have made clear, unexploded cluster munitions become effective landmines. Like mines, they kill indiscriminately, last after the war is officially over, and deny safe land to refugees and farmers. Like air-dispersed mines, they hide in soft earth and vegetation, or attract the attention of children and other curious passers-by.

The broader human impact of cluster bombs has been well-documented in Laos. Over a quarter of a century after the last load of cluster bombs was dropped over Laos, cluster munitions (“bombies”) continue to kill and, by denying land, contribute to desperate poverty. At a national level, clearance, mine risk education, and health care for victims continues to consume resources. Overall national productivity is diminished, increasing dependence on foreign aid. More recently, after the Gulf War, unexploded bombies from modern cluster bombs dropped by American, British and French coalition forces are estimated to have killed or injured thousands of Iraqi and Kuwaiti civilians and impeded post-war recovery. The Kuwaiti government spent $800 million clearing mines, cluster bomblets, and other unexploded munitions after the war.

The US and the UK have acknowledged using cluster bombs in Yugoslavia/Kosovo. The UK began using a cluster bomb called the BL 755 a few days into the bombing campaign. The US used the BLU 97/B submunition (delivered in a CBU-87 dispensing system). This is a “combined-effect munition,” which means it has three nasty party tricks: a bullet type effect, an explosion of fragmented metal and a shower of flame. It is capable of “disabling lightly armoured units … and killing people in the surrounding area.” An “official American document” of July 1997 is cited as saying that an unexploded bomblet from the CBU-87 can “be detonated or exploded by the presence, proximity or contact of a person. And, when detonated or exploded, produces an effect similar to a traditional anti-personnel landmine.”

The cluster bomb that Canada stocks (Rockeye II Mk 20 system with Mk 118 bomblets) was also reported to have been used in Yugoslavia/Kosovo. Each of these Rockeye cluster bombs contains 247 cylindrical submunitions which are designed to detonate on impact, but frequently – as often as 30 to 40 per cent of the time in the Gulf War – don’t. Canada has neither acknowledged nor denied using the weapon in Yugoslavia/Kosovo in response to a query from Mines Action Canada.

Not surprisingly, the impact of cluster bomblets in Yugoslavia/Kosovo has been no different than elsewhere. Because mines and cluster bomblets are so similar in effect, they are not always distinguished in reports on injuries and mine clearance. This makes it difficult to accurately assess the impact of bombies alone. However, specific accounts identifying the weapon paint a familiar picture in cameos: five young boys were killed on 24 April near Doganovic in Kosovo when they tried to pry open a cluster munition; in June, two British peacekeepers were killed while defusing an American cluster bombie in a school yard; three young men died and seven others were hospitalised near the village of Jahoc in July when they tried to pry open a cluster bombie; in April, a police officer in Tropoje in northern Albania died after picking up a Serbian fired cluster bombie. Of the approximately 150 injuries and deaths noted above, some forty per cent were caused by cluster bomblets.

NATO acknowledges using 1,500 cluster bombs in Yugoslavia/Kosovo, each containing an average of 200 bomblets. By the US State Department’s estimate, the Americans alone left 11,000 unexploded American bombies in Kosovo (Newsweek 1999). However, early reports from independent sources in Kosovo indicate that the failure rate of bomblets was high – perhaps 30 per cent – partly because NATO forces seem to have used outdated cluster bombs. If this estimate by mine clearance agency HALO Trust is accurate, it means as many as 90,000 volatile bomblets are scattered about the area.

Because cluster bombs are not designed to be victim-activated, they generally are not considered to fall under the treaty prohibition, even though they are known to act in the same way. Thus the treaty currently fails to prevent civilians from being subject to this mine-like weapon. Future reviews of landmine-related treaties may be good opportunities to address this limitation internationally but, in the meantime, states such as Canada which are committed to addressing the problems caused by this type of weapon could take unilateral action at the domestic level.3

Anti-vehicle mines

According to the International Committee of the Red Cross (ICRC), “… in many areas anti-vehicle mines placed on roadways used by civilians pose a serious threat to the civilian population.” The Landmines Treaty exempts anti-vehicle mines, defined as mines designed to be detonated by a vehicle as opposed to a person. The lack of definition for “vehicle” has raised concerns that this will result in the use of anti-vehicle mines triggered by very light vehicles. There is support within the German government for broadening the ban to include these victim-in-vehicle-activated weapons, partly as a result of this concern.

Anti-vehicle mines were used in Yugoslavia/Kosovo. Russian peacekeepers alone reported destroying ten anti-vehicle mines. Two incidents exemplify in different ways how they have affected civilians. On 18 April, during the war, five members of a family fleeing Kosovo were killed when their car drove over an anti-vehicle mine on the border. After the war, on 13 July, one civilian was killed and three passengers injured when their vehicle, driving down a mud track, hit two anti-vehicle mines placed on top of each other (AFP 1999).

While all anti-vehicle mines endanger civilians, anti-vehicle mines are increasingly equipped with devices that allow them to be set off by individuals. These anti-handling devices are intended to prevent anti-vehicle mines from being moved by the opposition. It is a matter of concern to both the ICRC and the International Campaign to Ban Landmines (ICBL) that the treaty exempts certain anti-handling devices from the ban. Canada is retaining one mine, the FFV 028 anti-vehicle mine, which appears to have anti-personnel capability in this way.

As information on the use of anti-vehicle mines with and without anti-handling devices in Yugoslavia/Kosovo emerges, it will be important to establish the impact of these mines with careful attention to the role of the treaty in changing the behaviour of militaries.

Assisting activities banned by the treaty

The Landmines Treaty prohibits the assistance, encouragement, or inducement “in any way” of anyone to engage in any activity prohibited under the treaty. Canada and a number of other NATO members have interpreted this prohibition very narrowly to apply only to “active assistance.” Activities such as carrying supplies to an ally who is using landmines would be permitted under the “Understanding” Canada appended to its instrument of ratification. The interpretation has allowed Canada to define an ally’s minefields as “natural obstacles” and to plan to make strategic use of them in battle. The narrow interpretation is highly controversial – it certainly does not correspond to the ordinary meanings of ‘encouragement,’ ‘assistance,’ and ‘in any way’ – and is clearly inconsistent with the spirit of the treaty. Germany’s implementing legislation, in contrast, would prevent German forces from participating in joint operations where anti-personnel mines are used.

In Yugoslavia/Kosovo, NATO allies cooperated on a number of fronts, including developing joint strategies, providing logistical support, sharing intelligence, and coordinating battlefield operations. When the Pentagon announced that it reserved the right to use Gator mines, Canada and other signatories gave no indication that they would cease participating in NATO joint operations if the Americans took advantage of their “right.” However, it can be assumed that the embarrassment the US would have created for its allies inhibited American use of the weapon. In this sense, the treaty, backed by the ‘new superpower’ of public opinion, prevented civilians from being subject to American Gator mines.

Nevertheless, the ICBL is taking the position that the only escape from future compromise for NATO members who have signed the treaty lies in a NATO policy of no use of landmines.4 Until NATO adopts such a policy, NATO governments are free to exercise their sovereignty and withdraw from joint operations when there is no guarantee that anti-personnel mines won’t be used by the US or Turkey.

More strikingly, NATO allies provided support to the KLA which used mines for several months prior to NATO’s entry into the war, and continued to do so during the period of NATO’s direct involvement. Assistance to the KLA by States Parties such as the UK could well constitute a violation of the treaty. The matter of States Parties giving assistance to the KLA demands serious investigation by ban protagonists and, as mandated under Article 8 of the treaty, other States Parties.

Responsibility for humanitarian mine action

One of the most promising features of the treaty is its reach beyond disarmament to demand that signatories address the problems created by mines already deployed. Mine-affected states are required to clear mines from areas under their jurisdiction and, pending clearance, to take steps to protect their citizens from minefields. Although primary responsibility for mine clearance rests with the mine-affected state, the treaty goes some way towards ensuring that the broader problem is shared among all States Parties. States are granted a general right to “seek and receive” assistance from other States Parties to fulfill their treaty obligations. All states are required to provide assistance for the rehabilitation of “mine victims,” for mine awareness programs, and for mine clearance (Article 6). None of these obligations to provide overseas assistance is unqualified – but neither are they entirely empty.

Who is responsible for repairing the damage caused by the mines used in Yugoslavia/Kosovo? There are many layers of responsibility under international law and according to moral principle: various moral and legal duties of states and “the global community” to meet the needs of the vulnerable and to fulfill international human rights law; compensation owed generally by states who developed mine technology and warfare and accepted the use of mines; moral and legal duties of compensation of producers and users in Yugoslavia/Kosovo (arising not least from human rights law and international humanitarian law). This last, the “polluter pays” principle, puts a special burden on the FRY, the KLA, and NATO states – as well as manufacturers like Alliant Tech in the US.

Under the treaty, primary responsibility for mine clearance, surveying and marking in areas under the authority of the FRY government (that is, FRY exclusive of Kosovo) would fall to it. FRY would be entitled to ask other states for assistance in this effort. This arrangement is obviously not satisfactory in the present circumstances where FRY has not signed the treaty and is, in any case, in no position to carry out or seek assistance in carrying out the work. Moreover, it leaves a question mark about responsibility for clearance in Kosovo, which is under international authority. However, the treaty also requires all States Parties (currently around 80) “in a position to do so” to assist people injured by mines, provide for mine risk education, and to provide assistance for mine clearance. These obligations apply to people and land in the entire area affected by the recent war (not just Kosovo). However, if adequate assistance is not forthcoming, attempts to secure compensation through legal action may find stronger grounds in alternate legal avenues.

Agenda for action

While the treaty succeeded in preventing NATO states – including, remarkably, the US – from using traditional anti-personnel mines and contributed to a swift mobilisation of resources for humanitarian mine action in Kosovo, the experience in Yugoslavia/Kosovo exposes several areas urgently requiring attention in efforts to eliminate the injustices caused by landmines.

We are faced with a number of challenges for improving the effectiveness of the treaty and for developing alternate strategies. Chief among the former are the possible treaty violations by States Parties: assistance provided to the KLA; failure to support mine action in affected areas outside Kosovo; and possible use of weapons which fall under the treaty, such as certain anti-vehicle mines with anti-handling devices. Among the latter, the most urgent challenge is eliminating weapons with anti-personnel mine effect which do not fall under the landmine treaty, such as cluster bombs.

Challenges for improving the effectiveness of the treaty:

• reaching governments who are not responding to the norm established by the treaty (such as Federal Republic of Yugoslavia);

• tightening international controls on transfer of mines and the ability of armed forces with no internal production capability, such as the KLA, to procure them;

• eliminating anti-vehicle mines with anti-personnel capability that fall under the treaty;

• ensuring that illegal assistance and encouragement is not provided to armed forces using anti-personnel mines; and

• ensuring that mine clearance and victim assistance is provided to all populations threatened by landmines, regardless of political objectives.

Challenges requiring alternate strategies:

• reaching Non-State Actors;

• eliminating victim-activated weapons, including cluster bombs and all anti-vehicle mines with anti-personnel capability, including those that fall outside the treaty;

• providing reparation to people injured by mines, as a matter of justice;

• ensuring funds for mine action are spent according to the priorities of affected people; and

• ensuring mines are cleared and victims assisted where governments of affected areas are not bound, able, or willing to provide humanitarian mine action or where there is no government.

1 In fact, according to information from Human Rights Watch, they prepared to do so by upgrading their B-52 bombers to deliver the Gator system.

2 The full definition, originally proposed by Rae McGrath, is given in the “Guidelines for Mine Action Programmes from a development-oriented point of view” (the Bad Honnef guidelines), drafted by international experts in June 1997 (www.landmine.de).

3 The process of responding to criticisms by ‘upgrading’ the munition to a (nearly) fail-safe, self-destruct fuse is already underway (“Cluster Bomb Dud Rates Cut, Army Says,” Ron Laurenzo, 2 June 1999). The same arguments used by the ICBL and states supporting the Ottawa process to deny an exception for self-destruct anti-personnel mines apply here.

4 Such a policy has a precedent in NATO’s declaration that, “Alliance strategy does not include a chemical or biological warfare capability. The Allies support universal adherence to the relevant disarmament regimes” (The Alliance’s Strategic Concept, NATO, 24 April 1999).


AFP 1999, “Landmines Kill, Maim 97 in Kosovo,” 14 July.

Mines Advisory Group 1997, “Definitions and Anti-Handling Devices Discussion Paper,” July.

Newsweek 1999, “A Scandal Waiting to Blow,” 2 August.

Spread the Word