EUROPE-ASIA STUDIES Vol. 62, No.

7, September 2010, 11351152

Why Kosovar Albanians Took Up Arms against the Serbian Regime: The Genesis and Expansion of the UCK in Kosovo
KEIICHI KUBO

SINCE THE END OF THE WAR IN KOSOVO, THE SUCCESS of the Kosovo Liberation Army (Ushtria clirimtare e Kosoves, UCK) has been widely publicised both inside and outside Kosovo. Despite this wide publicity for the UCK however, its origins remain rather mysterious and have rarely received serious scholarly scrutiny. How and why did the UCK start to organise and sustain its rebellion against the Serbian regime? Given that the Serbian regime was powerful and ruthless, it is important to ask, following Petersen (2001, p. 1), how and why do individuals accept enormous risks in the process? To answer these questions, this essay analyses the genesis of the UCK and its evolution into a large-scale guerrilla army, based on interviews and materials collected in Kosovo.1 This essay consists of ve sections: the rst briey reviews theoretical arguments in the literature on ethnic conict and rebellion; the second briey examines the historical background of the conict in Kosovo, pointing out that various factors often cited in the literature on the occurrence of violent conict cannot fully explain the case of
1 The author conducted eldwork in Kosovo and Macedonia in 2006. For the emergence and evolution of the UCK, this study relies heavily on selected interviews with former protagonists, particularly an interview with Ibrahim Kelmendi in Tetovo in May 2006 on the preparatory phase of the rebellion. The author is aware of the possible pitfalls of such prioritised reliance on one selected key informant, especially given the specic post-war context of politicised memory constructions. In order to avoid such pitfalls, an attempt was made to verify the statements of the informant as much as possible by using the published materials and statements made by other people in Kosovo. Due to the clandestine nature of the early illegal movement however, such verication is often simply impossible. However, this prioritised reliance is justiable in the present study, especially because Ibrahim Kelmendi is one of the rare survivors of the founders of the clandestine organisation, while other founders such as the Gervalla brothers and Kadri Zeka were assassinated (see below). In addition, unlike those who are now politically active in Kosovo and Macedonia such as Hashim Thaci and Ali Ahmeti, Kelmendi has not spoken publicly about his activities and has never been cited (as far as the author is aware) in the literature on the ethnic conict in Kosovo, although he has been an inuential gure behind the scenes in the Albanian society in ex-Yugoslavia, working as a political advisor to Ali Ahmeti in Macedonia as of May 2006.

ISSN 0966-8136 print; ISSN 1465-3427 online/10/071135-18 2010 University of Glasgow DOI: 10.1080/09668136.2010.497022

1136

KEIICHI KUBO

Kosovo; and the subsequent three sections analyse three distinct phases in the development of the UCK, namely the preparatory phase from the 1970s up to 1996, the initial phase from 1996 up to March 1998, and the expansionist phase after March 1998. The essay concludes with a recapitulation of the key ndings and an evaluation of the theoretical arguments in the literature, pointing out some important factors that have not received adequate attention. Theoretical overview Many scholars have asked why peopleor ethnic groupsrebel. In the 1970s, Gurr (1970) asked why men rebel. Horowitz (1985) analysed ethnic groups in conict in the 1980s. Furthermore, since the end of the Cold War, there has been a remarkable development in the study of ethnic conict and civil war, and various authors have attempted to build new explanatory models2 and to identify the causes of conict based on large-N analyses. For example, while the occurrence of ethnic conict or rebellion has been traditionally explained in term of grievances held by the members of the ethnic group (Gurr 2000), some scholars argue that grievances do not matter and what really matter are greed and opportunity (Collier & Hoeer 1998, 1999, 2001), or the right conditions for insurgency such as political instability at the centre and the mountainous terrain (Fearon & Laitin 1999, 2003). While the study of ethnic conict and civil war has seen signicant development, it also has some limitations. Firstly, in many of the theoretical arguments and models cited above, it is often assumed that ethnic groups act as unitary actors and make collective decisions to take up arms. Game-theoretic accounts of ethnic conict often make this assumption: they explain the occurrence of ethnic conict by some strategic interaction between the ethnic group and the state. As some prominent scholars have already pointed out, however, this is rarely a plausible assumption (Fearon 2004, p. 407), particularly regarding ethnic groups (Laitin 1998, p. 331; Brubaker & Laitin 1998, p. 438). The second problem in the literature is the conceptualisation of the occurrence of violence as a unique event. In large-N analyses of the occurrence of civil war, for example, the dependent variable is binary. Under such a research design, the onset of civil war is inevitably conceptualised as a unique event: the peace lasts until a certain year (y 0 where t 0, 1, . . . n), and then the civil war occurs in the next period (y 1 when t n 1). This conceptualisation of the onset of large-scale violence is also seen in game-theoretic accounts of ethnic rebellion. The unitary actor assumption discussed above aects this conceptualisation. When the ethnic group is assumed to be a unitary actor and to take a collective decision to take up arms at some point, the occurrence of violence is implicitly assumed to be a unique event, which occurs when the ethnic group makes this decision. This essay proposes to take a step forward by relaxing the unitary actor assumption and conceptualising the occurrence of rebellion as a process rather than a unique event. There are radicals and moderates within the ethnic group, and the former may decide to take up arms while the latter do not. People also make a decision to take up arms at dierent times. Some members of the ethnic group become what Petersen (2001) calls
2

See, for example, Lake and Rothchild (1998), Fearon (1998), Weingast (1998) and Oberg (2002).

WHY KOSOVAR ALBANIANS TOOK UP ARMS

1137

rst-actors or entrepreneurs who initiate collective action. Others may join the rebellion later. Some people may join the rebels at one point but drop out later. The level of violence may vary as a function of the number of the rebels and the amount of resources available to them. Given that ethnic groups normally lack a formal decisionmaking body, these assumptions are far more realistic than the unitary actor assumption and the conceptualisation of the occurrence of violence as a unique event. This suggests that one should take a nuanced approach in evaluating the eect of the causes of conict. When the dependent variable is binary, scholars are compelled to take an all-or-nothing approach to this evaluation. If the ethnic group is not a unitary actor and the occurrence of rebellion is a process, some factors may aect the emergence of rst-actors without aecting the actions of people who join the rebels later. Some factors that are not signicant for the emergence of rst-actors may be signicant for the expansion of the rebel organisation and the escalation of rebellion. This nuanced approach thus requires that one analyses separately the emergence of rst-actors, their initial decision to take up arms, and the expansion (or reduction) of the rebel organisation. This essay attempts to analyse the case of Kosovo based on this nuanced approach. The next section will briey examine the historical background of the case of Kosovo. Historical background of the conict in Kosovo This section briey examines the background of the case of Kosovo, focusing on developments in Kosovo under the communist regime after World War II.3 It demonstrates that various factors often cited in the literature of ethnic conict and civil war cannot fully explain the occurrence of ethnic rebellion in Kosovo, which makes it worthwhile to take a more nuanced approach to analyse the process of ethnic rebellion. Kosovo became an autonomous region within Serbia in 1946 (and was later upgraded to an autonomous province on a par with Vojvodina in 1963). In the 1940s and 1950s however, the autonomy mostly remained on paper, as the institutions of Kosovo were dominated by Serbs and Montenegrins and were under the control of Aleksandar Rankovic, a Serb leader who dominated the Yugoslav security service as Vice President and Chief of Police.4 After the downfall of Rankovic in 1966, large demonstrations by the Kosovar Albanians occurred in 1968. The Yugoslav federal authorities started to make a series of concessions to the Albanians from then onwards, including an increase in the number of Albanian ocials in the provincial authorities, granting the right to raise the Albanian ag, and giving permission to establish a university with instruction in the Albanian language (Pipa 1989). Republics and Autonomous Provinces became almost fully equal under the 1974 Federal Constitution, according to which the two autonomous provinces gained full autonomy over their parliaments, budgets and judicial systems, and the governments of the provinces were able to veto any policy of the Republic of Serbia, while the Serbian
3 For further discussion of the history, culture, religion and society in Kosovo, see for example Malcolm (1998); Duijzings (2000); and Schwandner-Sievers and Fischer (2002). 4 For discussion of the situation of Kosovo under the communist regime and the policies towards the Kosovar Albanians, see Vickers (1998, pp. 144217) and Clark (2000, pp. 745).

1138

KEIICHI KUBO

government did not have equivalent powers over decisions made by the provincial governments (Woodward 1995, pp. 40, 65; Vickers 1998, pp. 17881). These measures satised many Albanians in Kosovo and made them loyal to the communist regime. Indeed, the 1970s are seen now by many Kosovar Albanians as a golden age (Judah 2008, pp. 5563). The Kosovar Albanians, however, took part in demonstrations again in 1981 after the death of Tito, after which radical Albanians were purged from the party or arrested by the authorities (Vickers 1998, pp. 197217). Inter-ethnic relations in Kosovo started to deteriorate seriously after the rise of Milos evic to power in the Republic of Serbia in 1987. Milos evic started implementing radical policies against the Albanians in Kosovo, such as the purge of Albanians from workplaces and the eective annulment of the autonomous status of Kosovo. Facing repressive measures taken by the Serbian authorities, the Albanian elite started making secessionist demands. In July 1990 Albanian MPs of the parliament of Kosovo unilaterally declared that Kosovo should become a republic and independent from Serbia, even though it still remained a part of the Yugoslav federation (ASRA 1993, p. 331). At the end of September 1991 a referendum was held on the sovereignty and independence of Kosovo, in which the voter turnout was 87%, and 99.87% of those who voted supported independence. In May 1992, Albanians organised their own presidential and parliamentary elections, and the Democratic League of Kosovo (Lidhja demokratike e Kosoves, LDK) won an overwhelming majority (76.44%). As a result its leader, Ibrahim Rugova, became President of Kosovo. Under this leadership Kosovar Albanians began building a parallel society (Kostoviova 1996), establishc ing their own tax system, in which not only local Albanians but also diaspora communities in Western Europe contributed. Kosovar Albanians were thus very well mobilised and organised under the leadership of the LDK by the end of 1992. This situation did not lead to violent conict in Kosovo, because the LDK adopted non-violent strategies and did not choose to organise armed rebellion. Even when Croats and later Bosnian Muslims encouraged Albanians to take up arms and open the southern front against Serbia, Albanian leaders rejected their requests (Judah 2000, pp. 11315). Why did they choose peaceful protest rather than violent means? There are several factors that aected their decision. First, there was the inuence of the ideas of democratic opposition in Eastern Europe: the idea of a parallel system or a shadow government was inuenced by the notions of autonomy and selforganisation developed among Central European intellectuals, particularly Polish Solidarity (IICK 2000, pp. 4445). Secondly, there was a move among ordinary Albanians to counter the primitive and uncivilized stereotype portrayed by the Serbs: Albanians decided to abolish the traditional practice of blood feud, establishing a Council of Reconciliation which tracked down Albanian families and brought them together for mass reconciliation (IICK 2000, pp. 4445). Thirdly, there was a pragmatic judgment as well: the Serbian armed forces were too strong to ght against. In 1992, for example, Rugova made the following remark:
we would have no chance of successfully resisting the army . . . in fact the Serbs only wait for a pretext to attack the Albanian population and wipe it out. We believe it is better to do nothing and stay alive than be massacred. (Vickers 1998, p. 264)

WHY KOSOVAR ALBANIANS TOOK UP ARMS

1139

Fourthly, the attitude of external actors was also important. Albanian President Sali Berisha did not support an openly violent course because he was not willing to jeopardise economic assistance from abroad and the security of Albania by inviting Serbian reprisals (Judah 2000, p. 115). He also openly opposed the idea of unication of Albania and Kosovo: for example, Berisha stated in 1993 that the idea of a Greater Albania is not considered in serious Albanian political circles (Vickers 1998, p. 270). Due to these factors, the non-violent movements led by the LDK were predominant among Albanians until the UCK started its violent actions in 1996. In the case of Kosovo, various factors often cited as causes of violent conict in the literature cannot fully explain the occurrence of ethnic rebellion. For example, the high level of grievances, caused by the discriminatory policies and annulment of the autonomy, did motivate the Kosovar Albanians to seek secession from Serbia, but did not necessarily motivate them to take up arms immediately. The political regime in Serbia remained strong and authoritarian, and there was no political instability at the centre which could have motivated the potential rebels to take up arms. The economic situation in Kosovo (poverty and high rate of unemployment), geographic concentration of the Albanians and the mountainous terrain may provide a partial explanation of the occurrence of ethnic rebellion in Kosovo, but these factors had been present for a long time before the onset of rebellion by the UCK in 1996, and thus they cannot explain why the Albanians did not choose rebellion in the early 1990s (and most of them did not do so until 1998) while some Albanians started to rebel in 1996. In sum, the timing of the commencement of rebellion remains unexplained by the oft-cited factors. In the sections that follow, this essay will attempt to examine other factors that aected the occurrence of rebellion in Kosovo. The preparatory phase: from the 1970s to 1996 In the case of Kosovo, the rst-actors were the leaders of the illegal anti-Yugoslav organisations established in the 1970s in Western Europe. These organisations were not the rst illegal movements by the Kosovar Albanians.5 Among those that preceded them, for example, in the period from 1945 to 1952, was the National-Democratic Albanian Movement (Levizja Nacional Demokratike Shqiptare), although it was brutally crushed by the Serbian authorities. During the period from 1958 to 1968, two organisations were established, namely the Revolutionary Party for the Unication of the Albanian Lands (Partia Revolucionare per Bashkimin e Tokave Shqiptare me Shtetin Ame) founded by Metush Krasniqi, and the Revolutionary Movement for the Unication of the Albanians (Levizja Revolucionare per Bashkimin e Shqiptareve) founded by Adem Demaci. These two parties played an important role with respect to the subsequent development of illegal parties, because they established the basis of a national programme for the Albanians and because they contributed to the escalation of student demonstrations in 1968 (Ceku 2004). These parties were, however, again

For detailed documentation of the illegal movements in Kosovo, see Ceku (2004).

1140

KEIICHI KUBO

brutally suppressed by the Serbian authorities, and they did not directly lead to the emergence of the armed rebels in Kosovo.6 It was the student demonstrations of 1968 and the state repression against the demonstrators that played a crucial role in the emergence of the radicals who later established the UCK. After the 1968 demonstrations in Kosovo, many Albanians were arrested and banned from working or studying, and these people went abroad to seek refuge. Major illegal organisations were created in Western Europe by these immigrants or refugees. For example, the Red Popular Front (Fronti i Kuq Popullor, FKP) was established in 1978 by Ibrahim Kelmendi, who had been banned from studying due to his involvement in a small demonstration and had gone into exile in Germany in 1976.7 The National-Liberation Movement of Kosovo and Albanian Lands in Yugoslavia (Levizja Nacional-Clirimtare e Kosoves dhe te Viseve tjetra Shqiptare ne Jugosllavi, LNCKVShJ) was founded by Jusuf Gervalla and his brother Bardhosh Gervalla who went into exile in Germany in 1979. Kadri Zeka, who left Yugoslavia in 1979 to seek refuge in Switzerland, became a representative of the Organisation of MarxistLeninists in Kosovo (Organizate e MarksisteLeninisteve te Kosoves, OMLK). After the failure of the 1981 demonstrations, leaders of these groups began negotiating together to unify the three organisations.8 Their negotiations failed initially, but they nally agreed on unication and the political platform of the new organisation in 1982. While the process of unication was interrupted by the assassination of Zeka and the Gervalla brothers in Germany in the same year, the three groups were merged into the Popular Movement for the Republic of Kosovo in Yugoslavia (Levizja Popullore per Repubilken e Kosoves ne Jugosllavi) later in 1982.9 In 1991, it changed its name to the Popular Movement for Kosovo (Levizja popullore per Kosoves, LPK). The leading gures of the LPK decided to establish an army, which Ali Ahmeti named the Kosovo Liberation Army. In making preparations for the uprising the LPK rst sent a four-man group to Kosovo to build a network of secret cells across Kosovo and to make contact with other armed groups, such as the Jasharis.10 Secondly, they attempted to establish contacts with Albanian ocers in the Yugoslav Peoples Army (Jugoslovenska narodna armija, JNA). These ocers were to continue their military career in the ex-Yugoslav countries and to join the rebels when the time came, in order to help professionalise
6 For example, Adem Demaci was arrested several times by the Serbian authorities and served several lengthy prison sentences, he was released from prison for the last time in April 1990. Therefore, he could not be active during the 1980s while others were founding the illegal organisations in the West. Demaci, however, became the political spokesman for the UCK in August 1998, and his embrace of the UCK lent enormous respect to the organization because he was the uncompromising symbol of Kosovar Albanian militancy (Perritt 2008, p. 34). 7 Interview with Ibrahim Kelmendi, Tetovo, 30 May 2006. 8 Interview with Ibrahim Kelmendi, Tetovo, 30 May 2006. 9 In Yugoslavia was taken from its name in 1986. Interview with Ibrahim Kelmendi, Tetovo, 30 May 2006. 10 According to Kelmendi, this group consisted of Ali Ahmeti, Xhavit Haliti, Sabri Kicmari and Emrush Xhemaili. Interview with Ibrahim Kelmendi, Tetovo, 30 May 2006. Judah cites dierent names for this four-man group: Kadri Veseli, Hashim Thaci, Xhavit Haliti and Abaz Xhuka (Ali Ahmeti); see Judah (2000, pp. 11516).

WHY KOSOVAR ALBANIANS TOOK UP ARMS

1141

their rebel army. For example, they were in close contact with Agim Ceku after he deserted from the JNA and joined the Croatian army in 1991.11 According to Kelmendi, the LPK had established contacts with roughly 100 Albanian ocers in the region by the second half of the 1990s. Thirdly, the LPK established a homeland calling fund (Fondi Vendlindja therret) in 1993 to collect donations for their armed struggle. Kelmendi explained as follows:
We attempted to agree with Bujar Bukoshi in 1992. He promised to nance our activities from his fund [established to nance the activities of the government led by Ibrahim Rugova], but actually never did so. So we established our own fund and called for the Albanian diaspora to nance us rather than Bukoshis fund.12

Why did the leaders of the LPK become rst actors and initiate preparations for the rebellion? There are at least three factors. The rst is the political repression by the Yugoslav authorities. As mentioned above, after the 1968 demonstrations, many Albanians were arrested and banned from working or studying, and these people sought refuge abroad. They often became radicalised by the experience of arrest and imprisonment. This process of radicalisation through political repression continued for many years, and it led to an inow of radicalised people to the LPK. For example, Hashim Thaci, who later became head of the UCK political directorate, was a head of the students union in 1991 and went to Switzerland, where he joined the LPK (Judah 2000, p. 117). Bardhyl Mahmuti, who later became a francophone spokesman of the UCK in Switzerland and a foreign minister of the Kosovo provisional government after the war, was imprisoned from 1981 to 1988 after the 1981 demonstrations, and then went to Switzerland (ICG 1999, p. 15). Pleurat Sejdiu was also punished for taking part in the demonstrations in 1981 and left for Bucharest as a student, and then worked as a mini-cab driver in London, where he emerged as a UCK spokesman (Judah 2000, pp. 11718; ICG 1999, p. 16). Sabri Kicmari was only 13 when he was rst arrested for writing banners in 1981 (Judah 2000, p. 117), and he later went abroad and joined the LPK. The commitment of such people to the rebellion cannot be explained without this process of radicalisation, since they could have lived a normal and far easier life as workers or students in Western Europe if they had wanted to. The second factor is the inuence of Albanian Enverist ideology which increased in Kosovo in the 1970s after the Yugoslav authorities allowed the use of textbooks from Albania in Prishtina University. It was also in this period that Albania became a supporter of illegal anti-Yugoslav organisations (Ceku 2004). The anti-Yugoslav ideology propagated by Albania thus seems to have motivated the rst-actors to organise a rebellion against Yugoslavia. Two points, however, should be noted here. Firstly, many Kosovar Albanians had never been to Albania, and the MarxistLeninists who supported an uprising had no idea what Enverism wasthey just wanted to get rid of the Serbs (Judah 2000, p. 106). Some argue that Enverism was used to wrap a nationalist cause, which could have been
Interview with Ibrahim Kelmendi, Tetovo, 30 May 2006. Ceku led a retirement request in 1998 to go to Kosovo and join the UCK, though he was allowed to do so only in 1999. 12 Interview with Ibrahim Kelmendi, Tetovo, 30 May 2006.
11

1142

KEIICHI KUBO

associated with the nationalistfascist groups of the wartime period, in a fashionable radical leftism (Judah 2000, p. 106). Secondly, the degree of ideological commitment varied among the organisers of the LPK. For example, according to Kelmendi, the OMLK was far more committed to Marxist ideology than the FKP, which was more committed to the nationalist cause; an attempt to unify the three illegal organisations initially failed because the OMLK people wanted to create a communist-type organisation, but we [FKP] were not ready to be led by communists.13 The third factor that led to the leaders of the LPK becoming rst actors relates to the start-up costs and xed costs of the rebellion. As Lichbach (1995, p. 45) pointed out, early joiners often must bear the entire initial cost of creating a dissident organisation and as the set-up costs of dissident institutions increase, collective dissent is less forthcoming. Indeed, in a communist country such as Yugoslavia, it was too costly to create an illegal anti-regime organisation: organisers had to risk their jobs and even their lives. For those who were in the West, on the other hand, the cost was relatively lower: they enjoyed freedom of speech and association and did not have to worry about being arrested; they often worked as Gastarbeiter and thus had a source of income to sustain themselves. Therefore, it is not without reason that the illegal anti-Yugoslav organisations were created in Western countries rather than inside Yugoslavia. This does not mean, however, that involvement in the anti-Yugoslav organisations was costless in the West. According to Kelmendi, it was dicult to persuade Albanian immigrants in the West to support the LPK because many feared being detected by Yugoslav secret agents.14 Indeed, Zeka and the Gervalla brothers were assassinated because of their involvement in underground organisations. Therefore, the establishment of the anti-Yugoslav organisations cannot be explained without other factors such as political repression and the inuence of ideology. The initial phase: from 1996 to 1998 The initial phase of the UCK rebellion began in 1996. The rst violent action allegedly taken by the UCK was the killing of a Serbian policeman in 1995, but it was not until 1996 that an organisation calling itself the UCK claimed responsibility for the attacks (IICK 2000, p. 51). The rst planned assaults took place on 22 April 1996, when four almost simultaneous attacks were launched in separate locations that killed two policemen (ICG 1998c, p. 2). After that there were some sporadic UCK attacks on Serbian policemen, even though the intensity of rebellion remained quite low. It was on 15 October 1997 that the rst UCK man in uniform died while attacking a police station at Kliina (Judah 2000, p. 117). On 28 November, some UCK members c appeared in public for the rst time at the funeral of Halit Gecaj, an Albanian killed in the crossre of the UCK attack on Serbian police (ICG 1998a, p. 29).15 While the number of attacks was increasing (31 in 1996, 55 in 1997 and 66 in January and
Interview with Ibrahim Kelmendi, Tetovo, 30 May 2006. Interview with Ibrahim Kelmendi, Tetovo, 30 May 2006. 15 Now, 28 November is a date observed by Albanians everywhere as Flag Day, a holiday of great patriotic signicance. See ICG (1998c, p. 3).
14 13

WHY KOSOVAR ALBANIANS TOOK UP ARMS

1143

February 1998 alone), the death toll remained fairly low in this period. According to Ljubis ka Cvetic, who was then a spokesman for the Ministry of Internal Aairs of Serbia, these attacks from 1996 to February 1998 led to the deaths of 10 Serbian policemen and 24 civilians (ICG 1998a, p. 30).16 Why did the LPK/UCK decide to begin their rebellion activities in 1996? In explaining the occurrence of rebellion in Kosovo, many often argue that the availability of arms from neighbouring Albania played a crucial role. Following the collapse of a number of pyramid investment schemes in Albania in spring 1997,17 the Albanian government lost control and the country fell into chaos. During this period hundreds of thousands of Kalashnikovs and ammunition were looted by local Albanians and sold to the UCK on the black market. It is true that the UCK bought an enormous amount of arms and, as noted in the next section, that the rapid expansion of the UCK would have been technically impossible without these arms.18 However, this factor is not sucient to explain both the initial decision making by the UCK and its expansion. Firstly, the UCK began its rebellion before the collapse of the pyramid schemes in Albania. This means that the availability of arms was not crucial for the initial decision to begin a rebellion in Kosovo. Secondly, the sudden inow of a large number of small arms did not lead to an immediate escalation of the rebellion. The intensity of the rebellion remained fairly low throughout the year of 1997 despite the availability of many small arms. This suggests that the availability of arms is not sucient to explain the initial decision by the LPK/UCK and the expansion of the UCK. As for the initial decision in 1996 to start a rebellion, a stronger argument is that a more crucial factor was the growing disillusionment among Kosovar Albanians with Rugovas policy of peaceful resistance, which led to the decisions of some non-UCK people to switch their support from Rugova to the UCK. To substantiate this argument, let us briey examine how the disillusionment with Rugovas policy grew in Kosovo. Since the beginning of the 1990s, the LDK led by Rugova enjoyed strong support from an overwhelming majority of the Albanians both inside and outside Kosovo. The LDK leaders struggled to achieve the independence of Kosovo by peaceful means, and especially they sought to achieve their goal by appealing to major powers to recognise the independence of Kosovo. Their eorts, however, did not yield any substantial results. While many international conferences were held on the Yugoslav crisis after 1991 and the Albanian leaders attempted to put the issue of Kosovo on the agenda, their
16 For a list of the dates and places of the attacks associated with the UCK during this period, both on the Serbs and Albanian collaborators, see Kresovic et al. (1998, pp. 4049). 17 Pyramid schemes are a non-sustainable form of investment with the promise of extremely high rates of return for the investors deposit and for enrolling other people into the scheme, without any product or service actually being delivered. They emerged in Albania in the 1990s and expanded rapidly until they collapsed in 1997, prompting a general uprising against the government. 18 While the exact number of arms that owed into Kosovo is unknown, it has been suggested that the amount was indeed huge. According to some reports about 750,000 weapons were stolen from military depots and many of them ended up in the hands of the UCK (ICG 1998b, p. ii). According to General Obrad Stevanovic of the Vojska Jugoslavije (VJ), the number of weapons seized by the Serbian authorities by 20 June 1999 amounted to 1,045 hand-held rocket launchers, 8,320 machine-guns, semiautomatic, automatic and similar ries, 360 pistols, 4,224 mines and other explosive devices, and 723,531 pieces of ammunition of dierent types. See, ICTY transcript of case IT-02-54 (Slobodan Milos evic), p. 39574 (2005/05/17).

1144

KEIICHI KUBO

eorts were simply ignored. Other ex-Yugoslav republics such as Slovenia, Croatia and Bosnia were recognised by the European countries on the basis of a referendum on independence in each republic, but the referendum held in Kosovo in September 1991 was simply ignored by the European countries. In 1994 and 1995, the so-called KK plan was popular among Kosovar Albanians, according to which they would seek a similar level of autonomy to that experienced by the Krajina Serbs (Simic 2000, p. 96), but the argument for this plan was weakened when the self-proclaimed state of Krajina Serbs collapsed in 1995. The Albanian leaders lobbied the US government to solve the issue of Kosovo at the Dayton peace conference,19 but their eorts failed again. In September 1996, Rugova made an agreement with Milos evic on the normalisation of education, brokered by a Catholic organisation, Comunita` di SantEgidio,20 but this agreement was never implemented. The successive failures to achieve their goals by peaceful means led to increasing frustrations among Albanians in Kosovo and their growing disillusionment with nonviolent resistance. Already by 1993, there was growing resentment at international negotiators for not according the Kosovar Albanians full status at Geneva, and this resentment had begun to shift the balance within Rugovas movement toward the radical militants who preached a military solution (Woodward 1995, p. 359). In March 1994, Bukoshi stated that the Kosova governments pacist approach was losing credibility within the population (Clark 2000, p. 118). The Dayton Agreement in 1995 was clearly the turning point in this regard. The Dayton Agreement itself was an extraordinary trauma for the Kosovo Albanians because it rewarded the violent actions taken by Serbs with the recognition of Republika Srpska, and conrmed to them in the most dramatic and humiliating way that Rugovas policy of passive resistance had failed (Judah 2000, pp. 12425). Simic (2000, p. 54) observes that the failure of the LDK to put the question of Kosovo on the agenda at the Dayton peace conference led to a division within the LDK cadres and the radicalisation of the Albanian youth and Albanian national movements. Actions taken by the international actors after Dayton furthered a sense of disillusionment among Albanians. The UN embargo imposed on Yugoslavia was lifted, and the EU states ocially recognised the Federal Republic of Yugoslavia.21 Bonn and Belgrade made an agreement and Germany returned 100,000 Albanian refugees to Yugoslavia, and it appeared as if the last international means to put pressure on Belgrade were also lost (Petritsch & Pichler 2002, p. 74). Thus the Independent International Commission on Kosovo (IICK) (2000, p. 59, emphasis in original) concluded that the international community sent a message that Kosovo was denitely o the current international agenda and this demoralised and weakened the non-violent movement in Kosovo which felt betrayed by the international community and began to doubt the eectiveness of its own tactics.
19 Interview with Edita Tahiri, Prishtina, 9 May 2006. Tahiri was the former Foreign Minister of the government in exile of the Republic of Kosova and a member of the presidency of the LDK, and is currently the president of the reformist party, the Democratic Alternative of Kosova (Alternativa Demokratike e Kosoves, ADK). 20 For the details of the agreement, see Simic (2000, pp. 99100) and Troebst (1998, pp. 8183). 21 Note, however, that the US maintained so-called outer wall sanctions and blocked Yugoslavias return to the World Bank and the International Monetary Fund, citing the issue of Kosovo as one reason.

WHY KOSOVAR ALBANIANS TOOK UP ARMS

1145

This disillusionment with Rugovas policy led to the decisions of some non-UCK people to switch their support from Rugova to the UCK. For example, according to Florin Krasniqi, an Albanian immigrant who began organising nancial assistance for the UCK from the United States: Dayton made me realize that ghting was the only way. Peaceful resistance brought us nothing. Every other disgruntled minority in Yugoslavia . . . fought for independence and got it. The Albanians took the peaceful road and were ignored (Sullivan 2004, p. 3). Seeing this growing disillusionment with Rugovas peaceful resistance policy, the LPK/UCK judged that it was the right moment to claim legitimacy for its violent option and to begin a rebellion. Kelmendi explains as follows: By the end of 1995, it became clear that peaceful resistance proposed by Rugova failed. If we fail to achieve our goals by political means, we are entitled to resort to armed resistance. Thats why we took up arms in 1996.22 Indeed, some people were persuaded by the UCKs message that armed resistance was the only option. According to a Kosovar Albanian from Prizren who later became one of the deputy zone-commanders in that region:
When I heard of the news about the UCK in November 1996, I thought that only they can solve the situation in Kosova. I tried to contact them and, when I succeeded in doing so, I went into the mountains in early 1997 without saying anything to my parents and friends.23

The number of Kosovar Albanians who joined the UCK after its appearance in 1996, however, remained rather low. According to Mahmuti, for example, the UCK had only about 150 active men up to 1997 (Judah 2000, p. 118). According to the IICK, until late 1997, active armed resistance groups in Kosovo were very small and without permanent bases in the province (IICK 2000, p. 52). Even after the collapse of the pyramid schemes in Albania, according to Judah (2000, p. 129), Jashar Salihu found that few people were prepared to take the guns he was oering them from Albania. In order to explain why the vast majority of Kosovar Albanians did not join the rebellion immediately after the appearance of the UCK, one should take into consideration not only the fact that the Serbian regime was still powerful and ruthless but also Rugovas strong rejection of violence and his tight control of the LDK structure. Whenever the UCK appeared in Kosovo, Rugova implied that its existence was fabricated by the Serbian regime to justify repression. According to Edita Tahiri, almost everybody in the presidency of the LDK supported a violent uprising by the end of 1997, but Rugova stubbornly opposed it.24 Despite Rugovas international image as a pacist leader, he governed the LDK in an authoritarian manner, not allowing any opposition to his decisions.25 His strong rejection of violence and his tight, even authoritarian, control of the LDK structure prevented most Albanians from joining the UCK.
22 23

Interview with Ibrahim Kelmendi, Tetovo, 20 May 2006. Interview with a Kosovar Albanian who wished to remain anonymous, Prishtina, 12 May 2006. 24 Interview with Edita Tahiri, Prishtina, 9 May 2006. 25 Interview with Edita Tahiri and Edi Shukriu, Prishtina, 9 May 2006. See also Clark (2000, pp. 84, 123).

1146

KEIICHI KUBO The expansionist phase: after March 1998

What, then, explains the expansion of the activities of the UCK in 1998? The crucial factor was the Serbian police operations which killed almost all members of the Jashari clan. The police operations were conducted in the Drenica region, where the UCK was reported to have a strong presence. According to Troebst (1998, p. 14), heavily armed police units were mobilised in the operations, equipped with 20 helicopter gunships and 30 armoured personnel carriers. Between 28 February and 7 March 1998, Serbian police forces attacked villages of Likoshan, Cirez and Prekaz i Poshtem in the Drenica region. One of the main targets was the home compound of Adem Jashari, a local strongman who allegedly had killed a Serbian policeman several years previously and was sentenced to 20 years imprisonment for terrorism (AI 1998, p. 78; Judah 2001, pp. 2223). In Prekaz, a special anti-terrorist unit of the Serbian police killed almost all the members of the Jashari clan, a total of 59 people.26 While the Albanians seem to have oered some armed resistance against the police, many of those killed were civilians: according to Amnesty International (1998, p. 18), it was clear that many of the victimswho included at least 12 women and 11 childrenhad no involvement in the attacks on police. In Likoshan and Cirez, the Serbian forces killed 26 Albanians, using machine guns, rocket-propelled grenades, helicopters and armoured vehicles in the operation. The eect of these police operations was not to pacify but to electrify Kosovo (Judah 2001, p. 23): after the deaths of the Jasharis, the UCK began to expand. Baton Haxhiu, who was then a journalist with the Koha Ditore newspaper in Kosovo, reported that he did not see more than 100 UCK soldiers in 1997, but following the Prekaz and Likoshan massacres, the revolt spread out all over Kosova and everyone thought it [sic] could nd solutions by taking to the mountains and they rallied around this armed group which was called UCK.27 According to Hockenos (2003, p. 247), the events of this massacre became the rallying cry that would radicalize the Albanian population in Kosovo and the diaspora. Surroi has stated that the events in Prekaz were a fundamental moment and the war in Kosovo began symbolically on 5 March 1998, while the UCK emerged publicly earlier.28 In June 1998, even a Serbian police ocial admitted that the police operation had increased support for the UCK in the villages away from the main road.29 As a result, the size of the UCK increased very rapidly after the events in Prekaz. Pleurat Sejdiu recalls, for example, that the events in Prekaz led to a big inux of volunteers and it was unstoppable (Judah 2000, p. 141). Shukri Buja reported that the UCK didnt have any problem recruiting soldiers, because there was a great ood of young people to become members of the UCK.30 It seems that the UCK themselves
26 For the list of all people killed in Prekaz during the Serbian police operations, see Tahiri (2006, pp. 8688). This event was called the Drenica Massacre by Albanians, and the site of the massacre, with ruined buildings of the former Jashari compound, was turned into a memorial complex. For pictures and descriptions of the complex, see Thaci (2004). 27 ICTY transcript of case IT-02-54 (Slobodan Milos evic), p. 5428 (2002/05/23). 28 ICTY transcript of case IT-02-54 (Slobodan Milos evic), p. 3492 (2002/04/19). 29 Observer, 14 June 1998. 30 ICTY transcript of case IT-02-54 (Slobodan Milos evic), pp. 642728 (2002/06/06).

WHY KOSOVAR ALBANIANS TOOK UP ARMS

1147

were surprised by the course of the events: everyone was shocked by what was happening, no one more so than UCK men themselves (Judah 2000, p. 140). According to Sejdiu, the plan of the UCK was to start a war in 1999 but they were forced to speed things up due to the large inux of volunteers (Judah 2000, p. 141). Kelmendi explained as follows:
as of March 1998, we only wanted to conduct partisan-type guerrilla attacks, and did not want to begin an open rebellion, because Serbia was still too powerful. After Prekaz, however, a general uprising started against our will and we had to coordinate it.31

Volunteers also came from abroad. For example, in April 1999, according to a political representative of the UCK in Tirana: Until now, the number of people coming from the West, mostly from Germany and Switzerland, has reached eight thousand.32 Serbian scholars suggest that the UCK had about 1,200 members in May 1998, growing to 25,000 members by July 1998 (Mijalkovski & Damjanov 2002, p. 128). One should note that the expansion of the UCK took place in a rather chaotic manner. Besides the original UCK which began the rebellion, a new structure under Bukoshi called the Armed Forces of the Republic of Kosova (FARK) appeared on the ground, as well as local armed ghters organised by youths on a village-by-village basis, calling themselves Rugovas UCK, who fought until the end of the war under the impression that the UCK was under Rugovas control (Kola 2003, pp. 33637). This rather chaotic expansion was reected in the fact that the UCK had a decentralised and localised structure and that there was no functioning general headquarters. According to Kelmendi, for example,
the General Headquarters of the UCK was not functioning until very late . . . it was during the Rambouillet conference that the zone commanders sat down and elected Sylejman Selimi as the UCK general commander, whom Agim Ceku replaced in May 1999 when President Tudman nally allowed him to leave his post in Croatia and go to Kosovo.33 

Ramush Haradinaj, who was a zone commander of the Dukagjini area during the conict, also said that there was no general commander until the selection of Sylejman Selimi (Hamzaj 2000, p. 128).34 Shemsi Syla, who was then a deputy zonecommander of the Karadak area, explained as follows:
The General Headquarters of the UCK did not give orders . . . they just gave advice and worked on coordination between zone commanders because we were waging a guerrilla war and decisions had to be made by zone commanders according to the specics of Operative Zones.35
Interview with Ibrahim Kelmendi, Tetovo, 30 May 2006. AIM Press, 20 April 1999. 33 Interview with Ibrahim Kelmendi, Tetovo, 30 May 2006. 34 Haradinaj also explains how he was elected as a zone commander in a meeting of representatives from all headquarters of villages in Gllogjan. See Hamzaj (2000, pp. 7172). 35 Interview with Shemsi Syla, Prishtina, 10 May 2006. According to him, the UCK had seven operative zones (OZs) in Kosovo, each of which had its zone commander. These were as follows: OZ
32 31

1148

KEIICHI KUBO

In any case, the rapid expansion of the UCK naturally led to an escalation in the intensity of armed conict in Kosovo. For example, General Delic of the Yugoslav Army testied that there was a sharp increase in the number of attacks by the UCK after March 1998.36 In January and February 1998 the situation in Kosovo remained as usual with only a few attacks, but then from the month of March onwards, there was a particularly large number of attacks against civilians and members of the police, and as a result, there were only about 70 attacks in 1997 whereas in 1998 there were 1,470.37 The UCK had taken control of three major routes in Kosovo by the summer of 1998.38 In July 1998, the Serbian authorities began counterinsurgency operations against the UCK, in which the Yugoslav military forces (Vojska Jugoslavije, VJ) were mobilised. Artillery of various calibres, battle tanks and armoured personnel carriers of the VJ were deployed to provide heavy indirect and direct re support for the police forces.39 Tens of thousands of Albanians ed and hid in the hills and woods, their houses were looted and burned, and crops and cattle were destroyed. By the beginning of August 1998, reports estimated that between 200,000 and 300,000 Albanians had been displaced from their homes as a result of the military operations (IICK 2000, p. 74). As a result of the Serbian military campaign, the UCK simply melted into the woods, realising that it could not take on the far more heavily armed Serbs (Judah 2001, p. 23). But as more and more people ed their homes to become displaced persons, which was extensively covered by the international media, international pressures on Milos evic to halt military operations mounted.40 On 23 September 1998 the Security Council adopted Resolution 1199 calling for a ceasere, the withdrawal of security forces, and cooperation with the international monitoring eorts; NATO approved an activation warning for air campaigns against Yugoslavia.41 The US special envoy Richard Holbrooke was dispatched to Belgrade to negotiate with Milos evic. Threatened with the prospect of NATO air strikes, Milos evic made an agreement with Holbrooke on 13 October 1998. By the end of October, some 4,000 Serbian special police forces had been withdrawn and the OSCE Kosovo Verication Mission (KVM), a team of 2,000 observers who would monitor enforcement of the agreement, was deployed.
Drenica (zone commander: Sami Lushtaku), OZ Pashtrik (Musa Jashari, Ekrem Rexha and Tahir Sinani), OZ Dukagjini (Ramush Haradinaj), OZ Shala (Rahman Rama), OZ Llap (Rustem Mustafa), OZ Nerodime (Shukri Buja) and OZ Karadak (Ahmet Isu). See also Republika Srbija (2003, map 1), for the territorial division of the operative zones of the UCK. 36 He did so, of course, without admitting that the sharp increase in the number of UCK attacks was a result of the excessive use of force by the Serbian security forces. Indeed, he denied any excessive use of force by the Serbian authorities. See ICTY transcript of case IT-02-54 (Slobodan Milos evic), pp. 932729 (2005/06/22). 37 ICTY transcript of case IT-02-54 (Slobodan Milos evic), pp. 4124243 (2005/06/21). 38 ICTY transcript of case IT-02-54 (Slobodan Milos evic), p. 7926 (2002/07/10). 39 ICTY transcript of case IT-02-54 (Slobodan Milos evic), p. 7927 (2002/07/10). 40 Hockenos argues that this was a result of the UCKs strategy. He argues that part of the UCK strategy entailed provoking the Serbs, getting them to lash back with predictable ferocity and thus forcing a Western military response because the more cruel the repression, the more vivid the message that Albanians could not live under Serb domination (Hockenos 2003, p. 250). 41 ICTY transcript of case IT-02-54 (Slobodan Milos evic), pp. 696667 (2002/06/13).

WHY KOSOVAR ALBANIANS TOOK UP ARMS

1149

This agreement did not lead to peace in Kosovo, since the UCK took advantage of the withdrawal of the Serbian forces and renewed its military actions. Haradinaj admitted that the Serbian oensive in the summer had inicted heavy losses on the UCK and called the Holbrooke agreement life saving for the UCK (Hamzaj 2000, p. 115). Agim Ceku, who became the UCK Chief of Sta during the 1999 war, later said that the ceasere was very useful for them (Sell 2002, p. 293). During the period in which there was a reduced level of ghting, UCK members were able to return to burnt-out villages since the Serbs did not have enough men to hold on to the territories they had seized from the UCK, and the UCK had the time to train seriously and to consolidate a rather chaotic command structure (Judah 2001, p. 24). While the UN, NATO and OSCE were alarmed and called upon the UCK to cease provocative actions, it continued its military activities. Facing the UCK military activities, the Serbian army again moved into Kosovo in large numbers, with tanks and other heavy military equipment, establishing permanent positions in various areas in Kosovo. When Serbian forces assaulted Raak village and executed 45 Kosovar Albanians, c the OSCE-KVM investigated the site on the next day and concluded that they found evidence of arbitrary detentions, extra-judicial killings and mutilation of unarmed civilians (IICK 2000, p. 81), although the Serbian authorities denied that any civilians had been killed. The Contact Group composed of six countries, namely the US, Russia, Britain, France, Germany and Italy, organised peace negotiations to be held in Rambouillet, France. When the peace talks failed due to the refusal of the Yugoslav delegation to sign an agreement, NATO started air strikes on 24 March, which lasted for 78 days. When Milos evic nally accepted the G8 principles, Serbian forces withdrew from Kosovo and the UN administration in Kosovo was established, which eectively ended insurrections by the UCK. Conclusion Analysing the genesis of the LPK and UCK, this essay has argued that the LPK/UCK dates back to the 1970s when those who went abroad after the 1968 demonstrations organised a number of illegal organisations. As for the factors that motivated the rst-actors to form these organisations, three factors have been emphasised, namely political repression under the non-democratic regime (communist Yugoslavia), the inuence of Enverist ideology and the set-up costs of rebellion. These factors correspond to ideas put forward by some scholars in the theoretical literature. For example, Gurr (2000) argued that non-democratic regimes tend to cause rebellions whereas democratic regimes tend to give rise to peaceful protests. Lichbach (1995, pp. 3637, 4546) stressed the importance of political discontent, direct personal experience and what was described earlier as set-up costs, to explain the actions of rebels. The case of Kosovo can be used as an example that supports their arguments. The essay then analysed the start of rebellion, pointing out that increasing disappointment and frustration among Albanians, both within and outside Kosovo, led to the UCKs decision to take up arms. At the beginning of the 1990s, almost all Albanians supported the peaceful resistance proposed by the LDK, but as time went on, people started to become frustrated by the lack of achievement and subsequently

1150

KEIICHI KUBO

questioned the real eectiveness of peaceful resistance. The analysis here emphasises the importance of the time dimension in explaining the occurrence and dynamics of violence. When peaceful activities fail to lead to any visible achievements, more and more people start to question their eectiveness and look for other, more violent means to achieve their goals. In analysing when collective violence is used in general, Lichbach (1995, p. 60) has pointed out that lengthy campaigns tend to be associated with few concessions by government and a strong dissident group. Hence, as the struggle wears on, dissidents turn to more violent tactics. However this time factor has yet to receive adequate attention in the theoretical literature on ethnic rebellion and civil war. The case of Kosovo suggests that this factor is theoretically important for the explanation of the occurrence of ethnic rebellion or civil war. As for the expansion of the UCK, this essay has pointed to the importance of the excessive reaction of the Serbian state authorities, which led to the rapid radicalisation of Albanian society in Kosovo and the rapid expansion of the UCK. Many scholars have stressed the generally counterproductive eect of state repression. For example, analysing the sources of uprising by Kashmiri people against the Indian State, Bose (2003, p. 116) points out that the regime of repression had the eects of further radicalising public opinion and of convincing thousands of Kashmiri youths to take up arms to ght the Indian state. In relation to the Kurdish question in Turkey, it has been pointed out that security operations and the practice of burning villages had the eect of fuelling Kurdish nationalism and forcing especially young people to join the ranks of the rebels (Kiris ci & Winrow 1997, p. 131). In Sri Lanka, the police and military responded brutally when the Liberation Tigers of Tamil Eelam (LTTE) started its low-intensity rebellion, and the military terror and repression directed against the Tamil population played a vital catalytic role and contributed to the ascendancy of the LTTE (Bose 1994, p. 91).42 Also, the counterproductive nature of coercion was pointed out almost 30 years ago by Hibbs (1973, p. 182), who concluded that the nearly instantaneous response to repression is most often more mass violence. This argument suggests the importance of bringing the state back into the explanatory framework. For example, OLeary and Tirman (2007, p. 12) argued that econometric analyses of civil war are analytically one-sided, instead asking why are statesrather than just insurgentsnot modelled as suppliers of violent exploitation or predation, a perspective that would seem more consistent with the genre of economic theorising from which their work ows? Oberg proposed to bring the governments calculations back into the equation, and make the argument strategic (Oberg 2002, p. 18). Indeed, in the 1960s, Eckstein (1965, p. 145) pointed out that it is important to examine the role of the state authorities in explaining internal war:
One crucial choice that needs to be made is whether to put emphasis upon characteristics of the insurgents or incumbents, upon the side that rebels or the side that is rebelled against. Not surprisingly, the existing literature concentrates very largely on the rebels . . . This would seem
42

Bose described this dynamic as the dialectic of state repression and nationalist resistance (1994, p. 89). For the further detail of this dialectic, see Bose (1994, pp. 92116). LTTE membership grew from a mere 30-odd individuals in July 1983 to some 4,000 in July 1987 (Bose 1994, p. 87).

WHY KOSOVAR ALBANIANS TOOK UP ARMS

1151

to be only natural: after all, it is the rebels who rebel. At least some writings suggest, however, that characteristics of the incumbents . . . must be considered jointly with characteristics of the insurgents, indeed perhaps even emphasised more strongly.

Not all state authorities resort to excessive repression. Therefore it would be important and interesting to ask why the state authorities in some countries resort to repressive measures against a small rebel group while others do not. This question requires a systematic inquiry that compares the actions made by the state authorities against a small rebel group. This however, is a task for future research. Waseda University

References
Academy of Sciences of the Republic of Albania (ASRA) (ed.) (1993) The Truth on Kosova (Tirana, Encyclopaedia Publishing House). Amnesty International (AI) (1998) Kosovo: The Evidence (London, Amnesty International). Bose, S. (1994) States, Nations, Sovereignty: Sri Lanka, India and the Tamil Eelam Movement (Thousand Oaks, CA, Sage Publications). Bose, S. (2003) Kashmir: Roots of Conict, Paths to Peace (Cambridge, MA, Harvard University Press). Brubaker, R. & Laitin, D. (1998) Ethnic and Nationalist Violence, Annual Review of Sociology, 24. Ceku, E. (2004) Shekulli i Ilegales: Proceset gjuqesore kunder ilegales ne Kosove (Prishtina, Brezi81). Clark, H. (2000) Civil Resistance in Kosovo (London, Pluto). Collier, P. & Hoeer, A. (1998) On Economic Causes of Civil War, Oxford Economic Papers, 50, 4. Collier, P. & Hoeer, A. (1999) Justice-Seeking and Loot-Seeking in Civil War, World Bank Research Paper, available at: http://econ.worldbank.org/, accessed 6 March 2010. Collier, P. & Hoeer, A. (2001) Greed and Grievance in Civil War, World Bank Research Paper, available at: http://econ.worldbank.org/, accessed 6 March 2010. Duijzings, G. (2000) Religion and the Politics of Identity in Kosovo (London, Hurst & Co). Eckstein, H. (1965) On the Etiology of Internal Wars, History and Theory, 4, 2. Fearon, J. (1998) Commitment Problems and the Spread of Ethnic Conict, in Lake, D. & Rothchild, D. (eds) (1998). Fearon, J. (2004) Separatist Wars, Partition, and World Order, Security Studies, 13, 4. Fearon, J. & Laitin, D. (1999) Weak States, Rough Terrain, and Large-Scale Ethnic Violence since 1945, Paper presented at the Annual Meetings of the American Political Science Association, available at: http://www.stanford.edu/group/ethnic/workingpapers/insurg1.pdf, accessed 6 March 2010. Fearon, J. & Laitin, D. (2003) Ethnicity, Insurgency, and Civil War, American Political Science Review, 97, 1. Gurr, T. (1970) Why Men Rebel (Princeton, NJ, Princeton University Press). Gurr, T. (2000) Peoples versus States: Minorities at Risk in the New Century (Washington, DC, US Institute of Peace). Hamzaj, B. (2000) A Narrative about War and Freedom (Dialog with the Commander Ramush Haradinaj) (Prishtina, Zeri). Hibbs, D. (1973) Mass Political Violence: A Cross-national Causal Analysis (New York, WileyInterscience). Hockenos, P. (2003) Homeland Calling: Exile Patriotism and the Balkan Wars (Ithaca, NY, Cornell University Press). Horowitz, D. (1985) Ethnic Groups in Conict (Berkeley, CA, University of California Press). Independent International Commission on Kosovo (IICK) (2000) Kosovo Report: Conict, International Response, Lessons Learned (Oxford, Oxford University Press). International Crisis Group (ICG) (1998a) Kosovo Spring, ICG Report, available at: http:// www.crisisgroup.org/home/index.cfm?id1601&l1, accessed 6 March 2010. International Crisis Group (ICG) (1998b) The View from Tirana: The Albanian Dimension of the Kosovo Crisis, ICG Balkans Report, No. 36, available at: http://www.crisisgroup.org/home/ index.cfm?id1602&l1, accessed 6 March 2010.

1152

KEIICHI KUBO

International Crisis Group (ICG) (1998c) Kosovos Long Hot Summer: Brieng on Military, Humanitarian and Political Developments in Kosovo, ICG Balkans Report, No. 41, available at: http://www.crisisgroup.org/home/index.cfm?id1597&l1, accessed 6 March 2010. International Crisis Group (ICG) (1999) Whos Who in Kosovo, ICG Balkans Report, No. 76, available at: http://www.crisisgroup.org/home/index.cfm?id1590&l1, accessed 6 March 2010. Judah, T. (2000) Kosovo: War and Revenge (New Haven, CT, Yale University Press). Judah, T. (2001) The Growing Pains of the Kosovo Liberation Army, in Waller, M., Drezov, K. & Gokay, B. (eds) (2001) Kosovo: The Politics of Delusion (London, Frank Cass). Judah, T. (2008) Kosovo: What Everyone Needs to Know (Oxford, Oxford University Press). Kiris ci, K. & Winrow, G. (1997) The Kurdish Question and Turkey: An Example of a Trans-state Ethnic Conict (London, Frank Cass). Kola, P. (2003) The Search for Greater Albania (London, Hurst). Kostoviova, D. (1996) Parallel Worlds: Response of Kosovo Albanians to the Loss of Autonomy in c Serbia 19891996 (Keele, Keele European Research Centre). Kresovic, T., Obradovic-Savuljesko, J., Dinic, M. &  okic, G. (1998) Kosovo Dossier: Through D Terrorism to Independence (Beograd, BINA). Laitin, D. (1998) Identity in Formation: The Russian-Speaking Populations in the Near Abroad (Ithaca, NY, Cornell University Press). Lake, D. & Rothchild, D. (eds) (1998) The International Spread of Ethnic Conict: Fear, Diusion and Escalation (Princeton, NJ, Princeton University Press). Lampe, J. R. (1996) Yugoslavia as History: Twice There Was a Country (Cambridge, Cambridge University Press). Lichbach, M. (1995) The Rebels Dilemma (Ann Arbor, MI, University of Michigan Press). Malcolm, N. (1998) Kosovo: A Short History (London, Macmillan). Mijalkovski, M. & Damjanov, P. (2002) Terorizam Albanskih Ekstremista (Beograd, Vojska). Oberg, M. (2002) The Onset of Ethnic War as a Bargaining Process: Testing a Costly Signaling Model (Uppsala, Department of Peace and Conict Research, Uppsala University). OLeary, B. & Tirman, J. (2007) Thinking about Durable Political Violence, in Heiberg, M., OLeary, B. & Tirman, J. (eds) (2007) Terror, Insurgency, and the State: Ending Protracted Conicts (Philadelphia, PA, University of Pennsylvania Press). Perritt Jr, H. (2008) Kosovo Liberation Army: The Inside Story of an Insurgency (Urbana, IL, University of Illinois Press). Petersen, R. (2001) Resistance and Rebellion: Lessons from Eastern Europe (Cambridge, Cambridge University Press). Petritsch, W. & Pichler, R. (2002) Dugi Put u Rat: Kosovo i Medunarodna Zajednica, 19891999 [Der  lange Weg in den Krieg. Kosovo und die Internationale Gemeinschaft 19891999] (Beograd, Samizdat B92). Pipa, A. (1989) The Political Situation of the Albanians in Yugoslavia, with Particular Attention to the Kosovo Problem: A Critical Approach, East European Quarterly, 23, 2. Republika Srbija (RS) (2003) Albanski Terorizam i Organizovani Kriminal na Kosovu i Metohiji (Beograd, Vlada Republike Srbije). Schwandner-Sievers, S. & Fischer, B. J. (eds) (2002) Albanian Identities: Myth and History (London, Hurst & Co.). Sell, L. (2002) Slobodan Milosevic and the Destruction of Yugoslavia (Durham, NC, Duke University Press). Simic, P. (2000) Put u Rambuje: Kosovska Kriza 19952000 (Beograd, Nea). Sullivan, S. (2004) Be Not Afraid, For You Have Sons in America: How a Brooklyn Roofer Helped Lure the US into the Kosovo War (New York, St. Martins). Tahiri, B. (2006) Adem Jashari: Legjende e Legjendave (Prishtina, Rilindja). Thaci, T. (2004) Memorial Complex: Adem Jashari, Prekaz (Prishtine, Shtypshkronja Dukagjini). Troebst, S. (1998) Conict in Kosovo: Failure of Prevention? An Analytical Documentation, 19921998 (Flensburg, European Centre for Minority Issues). Vickers, M. (1998) Between Serb and Albanian: A History of Kosovo (London, Hurst & Co.). Weingast, B. (1998) Constructing Trust: The Political and Economic Roots of Ethnic and Regional Conicts, in Soltan, K., Uslaner, E. & Hauer, V. (eds) (1998) Institutions and Social Order (Ann Arbor, MI, University of Michigan Press). Woodward, S. (1995) Balkan Tragedy: Chaos and Dissolution after the Cold War (Washington, DC, The Brookings Institution).

Copyright of Europe-Asia Studies is the property of Routledge and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.