Kosovo, 1944–1981: The Rise and the Fall of a Communist ‘Nested Homeland’

Click to increase image sizeClick to decrease image size

Notes

The authors contributed equally to the research and writing of this essay. This research was made possible by a postdoctoral fellowship received from the Social Sciences and Humanities Research Council of Canada, as well as by the fellowship and conference grant received from the Nuffield College, Oxford University, and by the conference grant from the British Academy. Many thanks to the Archives of Yugoslavia, Belgrade for help with access to the primary sources. The authors are also grateful to Francesc Amat, Ioannis Armakolas, Jovo Bakić, Florian Bieber, Ljubodrag Dimić, Slobodan Drakulić, Neophytos Loizides, Helder De Schutter and Lea Ypi, as well as to the anonymous reviewers for their constructive criticism of previous drafts of this essay.

For a systematic critique of this argument, see Brubaker (1998).

We analysed several thousand pages of previously unresearched internal Party materials (primarily transcriptions of meetings and final reports) produced by the Commission for Socio-Political System and Multi-National Relations of the Central Committee of the Communist Party of Yugoslavia, from 1958 to 1977. The sources were accessed in The State Archives of Serbia and Montenegro (previously Yugoslav Archives) in Belgrade, from August 2008 to February 2009.

For example, in the Soviet case, once Gorbachev's reforms reduced the threat of direct repression of protest activity, the deep institutionalisation of ethnicity determined the form and expression of nationalist demands (Gorenburg 2003, pp. 9, 24; Zaslavsky 1997, p. 89).

In North American terminology, ‘racism’ is normally preferred to ‘chauvinism’. However, in this case the group difference was generally not perceived as ‘racial’ but ‘cultural’. Hence, to call this form of ideology and policy of discrimination ‘racism’ would be misleading.

In addition, as Donald Horowitz (1998) explains, in the post-communist period, international law has evolved in the direction of the recognition of the right of secession of territorial autonomies from multi-national federal states.

While the overall communist experience with ethno-federalism can hardly be seen as positive (Emizet & Hesli 1995, p. 504; Gorenburg 2003, p. 25; Vujačić & Zaslavsky 1991, p. 137), it cannot be argued that ethno-federalism has always produced state fragmentation and disintegration. A number of democratic ethnic federations—such as Switzerland, Canada and India—have relatively successfully weathered secessionist challenges over long periods. Over-generalising from the dramatically negative experience of communist federations might make us downplay the more successful records of their democratic counterparts. Two leading scholars and advocates of ethno-federalism (or ‘multi-national democracy’) persuasively argue that studies of federalism should focus on specific conditions that make the success of an ethno-federal institution more or less likely (McGarry & O'Leary 2009). This is our goal in this essay.

For example, Nagorno-Karabakh was an Armenian ethno territorial autonomy embedded in Azerbaijan, Azeri ethno-territorial autonomy within the USSR. Unlike communist Yugoslavia, which had only one ‘nested homeland’, the USSR operated a complex hierarchy of nested autonomies. In 1939 it had 21 autonomous republics, nine autonomous oblasti, and 10 national okrugs (Suny & Martin 2001a, p. 2).

From 1944 to 1968, the territory was officially known as Kosovo i Metohija. Metohija (from Greek metoh, meaning Church land or possession). It is a region with a significant concentration of Serbian religious monuments that directly borders Albania. Of course, in focusing on the concept of the nested homeland in the case of Kosovo, it is not intended to argue that a nested homeland is either a necessary or a sufficient condition for an explanation of the rise of ethnic tensions and conflicts in the late communist period in general. As Brubaker's (1996) analysis of triadic relationships shows, communist ethno-federal institutions and policies led to ethnic conflicts in Bosnia, Croatia and Moldova, although no nested homeland institutions were implemented in these cases. Moreover, in several cases, such as North Ossetia, the nested homeland institution was implemented but did not seem to have a negative effect on ethnic relations. There were several causal paths towards ethno-national conflict at the end of the communist period, and the nested homelands effect was just one of them.

However, the nested homeland model cannot be seen as the ‘root cause’ of this conflict. The clash between Albanian and Serbian nationalist projects over Kosovo territory, dating at least to 1878 (Stefanović2005), precedes communist attempts to ‘solve the national question’ in the Balkans. Yugoslav communists cannot be blamed for the existence of this ethnic conflict, but they do have major responsibility for the failure to channel the conflict in less destructive directions.

Aleksandar Ranković was a ruthless Chief of Yugoslav State Security in the post-war period. The 1966 purge of Ranković and his close associates was followed by a rapid replacement of ethnic Montenegrins and ethnic Serbs with ethnic Albanians in the Kosovo State Security apparatus, and then, more gradually, in the Party and in other state agencies.

For a detailed example of the numerical over-representation of Serbs in Kosovo's administrative, state and Party apparatus from this period, see National Composition of Institutions, a report produced in 1956 (Arhiv Jugoslavije, Arhiv Centralnog Komiteta Saveza Komunista Jugoslavije[hereafter AJ, A CK SKJ] XVIII—Kutija 11/21). Serbs and Montenegrins represented a quarter of the local population but filled almost half of the most important administrative and executive positions.

The word Metohija (from Greek metoh, meaning Church land or possession) has negative connotations for Kosovo Albanian ultra-nationalists because of its association with the Serbian Orthodox Church (Ramet 1992, p. 191).

As Serbian communists repeatedly argued, while other Yugoslav republics also contained regions with non-titular concentrations (such as Krajina in Croatia or western Macedonia), no other Yugoslav ‘national’ republic had to cope with autonomous provinces in its territory.

Victorious Yugoslav communists ethnically cleansed Italian and German minorities and replaced them with Slavic settlers in Istria and Vojvodina. The low-scale Četnik (Serb pro-Western nationalists) guerrilla activity continued in parts of Serbia and Bosnia for several years after the war.

On the inter-war colonisation policy of the Serbian authorities, see Stefanović (2005).

For more information on the 21st Volunteer Mountain SS Division ‘Skenderbeg’ (comprised of Kosovo ethnic Albanians) and their activities, see Tomasevich (2001, pp. 154, 499).

In 1931, 33% of the inhabitants of Kosovo reported a Slavonic mother tongue and about 60% reported their mother tongue was Albanian (SZS 1998). The 1931 census reported on mother tongue and religion, not self-declared ethnic identity. It lumped together various Slavonic languages, and it makes no sense therefore to assume that all Slavonic language speakers thought of themselves as ‘Serbs’.

For example, on 2 December 1943, the Central Committee of the KPJ wrote to the Central Committee of the Communist Party of Albania that ‘every nation has the right of self-determination, including secession’, quickly adding that ‘we do not support all national movements but only those that prove themselves in the struggle against imperialism’ (Horvat 1987, p. 56). Similarly, in the winter of 1944, the Regional Secretary of the KPJ for Kosovo promised that ‘the Albanians are guaranteed the right to self-determination, and they need to earn that right in the struggle against the [German] invaders’ (Horvat 1987, p. 58).

The Kosovo Albanians' lack of resistance, collaboration, or even active support for the fascist regime needs to be interpreted in the context of the highly discriminatory and repressive policies that Royal Yugoslavia applied against this minority (see Stefanović2005). Italian and German occupiers allowed schools in the Albanian language and unification with Albania proper. Thus, for many Kosovo Albanians, the fascist occupation was a significant improvement over their position in the Yugoslav Kingdom.

AJ, A CK SKJ, XVIII-K4/1 (4–7).

In 1945, there were about 1,000 Party members in Kosovo, and only about 300 were Albanian. In December 1946, there were 5,505 Party members in Kosovo; 2,490 Serbs, 1,069 Montenegrins and 1,771 Albanians (Pavlović2001, p. 146).

The ‘Bujan Conference’ was held between 31 December 1943 and 2 January 1944, in the village of Bunjaj (Bujan) in northern Albania. The conference elected 42 delegates: 36 ethnic Albanians (of whom 15 were from Albania and 21 from the Yugoslav territory), five Serbs and Montenegrins, and one Slavic Muslim. The collective presidency consisted of seven Albanians, one Serb and one Montenegrin. The controversy comes from the last paragraph of the first point of the adopted resolution which states that Kosovo and Dukagjin [Dukađin] were regions populated predominantly by Albanians whose clear wish was to unite with Albania. The resolution goes on to state that the only way for the Kosovo (and Dukagjin) Albanians to unite with Albania was by engaging in a common struggle with the other nations of Yugoslavia against the Axis powers and their domestic collaborators. Once freedom had been won, all nations in the region (including Albanians) would have a chance to decide their own destinies by invoking the right of self-determination and seceding from Yugoslavia. The Yugoslav and Albanian communist parties are named as the guarantors of this promise. For detailed information, see Narodni odbor Autonomne Kosovsko-metohijske oblasti (AKMO) 1943–1953, Osnovni materijali sa zasedanja (Priština 1955).

During the meeting, Edvard Kardelj, the Slovenian-Yugoslav communist leader, informed Stalin: ‘On the territory of Kosovo and Metohija we now have more Albanians than Serbs. We are planning to hand over these territories to the Albanians, once we establish better relations with them’. Stalin replied: ‘Very good. That is a correct approach’ (‘A Report on Kardelj–Stalin Meeting in Moscow’, in Popović2001, p. 264).

In this period, Šiptar was a standard Yugoslav term for Kosovo Albanians. After the 1960s, the term acquired negative connotations, and the officially accepted term became ‘Albanian’ (Albanac).

Austro-Hungarian occupational authorities (1916–1918) and Italian/German occupational authorities (1941–1944) had set up Albanian-language schools during the period of their control. There were no Albanian-language schools in Royal Yugoslavia (Đekić1990, p. 178).

Schafer (2001, p. 166) shows that the Bolsheviks faced the same problem of ‘selective recognition’ during their Revolution. Like their Yugoslav students, the Bolsheviks were guided not only by overarching ideological principles, but also by tactical calculations, especially the need to recruit ethnic minorities for the Revolution by promising them recognition and improved collective status.

The understanding of the war as an opportunity for the ethnic groups to prove their worth in the eyes of the Party is well expressed in this proclamation of the KPJ Serbian HQ to Kosovo Albanians in August 1944: ‘Together with the fascist aggressors you fought against the neighbouring peoples and thus dishonoured yourselves. Because of such actions you still have not gained the right to live together with other peoples of Yugoslavia in brotherhood and unity. Now is your chance to rectify your mistakes and erase your shame’ (Marković1991, p. 216). Not surprisingly, this rhetoric did not prove terribly effective, as the Kosovo Albanian anti-communist uprising started a few months later.

While inter-war and wartime Party proclamations promised the end of the Greater Serbian oppression and ‘a brotherly, free, equal community of all nations of Yugoslavia’ (Connor 1984, p. 158), ethnic Serbs and Montenegrins were heavily over-represented in the Party, especially in the early post-war period. As late as 1971, Serbs were over-represented in Croatia, Bosnia, Vojvodina and Kosovo. This over-representation was clear for the wartime (Partisan guerrilla) generation but less clear for younger Party members (Cohen 1989, pp. 302, 420). In the early 1970s, Serbs and Montenegrins represented 43% of the Yugoslav population and 85% of all army officers (Cohen 1989, p. 298). Researchers generally explain the enduring Serbian over-representation in the Party and the Army as a consequence of ‘militaristic meritocracy’. During World War II, ethnic Serbs (especially in Bosnia and Croatia) joined communist resistance forces early and in considerable numbers. Other ethnic groups joined later, and some—such as Kosovo Albanians—had no significant participation. In the post-war distribution of commanding social positions, being ‘red’ was usually more important than being an expert. Thus, whether and how early one joined the communist resistance was frequently the most important criterion (Denich 1973, p. 102). Understandably, those who joined early, when the outcome of the war was far from certain, were seen as more genuinely committed to the cause than those who joined later, when victory was no longer in question and when, presumably, many simply wanted to be on the winning side. Since Serbs generally joined earlier, they were over-represented in the Party after the war, especially in Croatia and Kosovo (Horvat 1987, pp. 288–89). While Serbs and Montenegrins were over-represented, non-Slavic groups (such as Albanians and Hungarians) were under-represented. Furthermore, villages and regions known for their support of the communists during the war were rewarded with new roads and factories after the war (Lampe 2000, pp. 282, 294). This kind of ethnic distribution pattern was hardly the initial intention of the revolutionary leadership (which was genuinely multi-ethnic); indeed, it became a source of major ideological embarrassment for the KPJ, one that they attempted to ‘rectify’ by affirmative action hiring and the promotion of non-Serbs and non-Montenegrins, especially in the Party and the Army.

Between 1945 and 1947, 420 ethnic Albanians who held various positions in Kosovo but allegedly secretly worked for Albanian Sigurimi (state security) were arrested; a few supposedly managed to infiltrate high positions in the local branch of Yugoslav state security. For more detailed discussion of these issues see: AJ, A CK SKJ, XVIII-K4/1 (4–7).

In 1948, Serbs and Montenegrins made up about two-thirds of the active white collar workforce in Kosovo (Cohen 2001, p. 18). In 1956, Serbs were only 27.4% of the population of the province but comprised 86.6% of the state security personnel (Horvat 1987, p. 62). For a detailed example of numerical over-representation of Serbs in Kosovo's administrative, state and party apparatus from this period, see National Composition of Institutions produced in 1956 (AJ, XVIII-K11/21). Before 1966, Serbs and Montenegrins were 15 times more likely than Albanians to work in the Yugoslav State Security Kosovo branch (Popovićet al. 1990, p. 19).

A report titled ‘Campaign of Eastern European Countries and Reaction of Ethnic Minorities' (AJ, A CK SKJ, XVIII-K3/25), produced in 1958, elaborated on the propaganda activities originating from Albania and directed at Yugoslav Albanians. With regards to Enver Hoxha's Albania, the report notes that the neighbouring country publicly stated its territorial aspirations on Yugoslav territory (speech by Enver Hoxha, Second Congress of the Communist Party of Albania, April 1952; see AJ, A CK SKJ, XVIII-K3/25 (2)). Albania openly proclaimed its territorial aspirations towards Kosovo-Metohija and western Macedonia, spreading anti-Yugoslav propaganda and calling on ethnic Albanians in Yugoslavia to fight the Yugoslav regime (AJ, A CK SKJ, XVIII-K3/25: 5). The document also stated that Albania was the only member of the Eastern Bloc that continued to act against Yugoslavia after the period of normalisation (after 1955; AJ, A CK SKJ, XVIII-K3/25: 8–9). The author(s) note that the nature of the Albanian campaign changed after 1955: after initially laying claim to Yugoslav territory, Tirana argued that the status of Kosovo was ‘unresolved’.

During this period, 12,000 rifles, 1,000 machine guns, 100 heavy machine guns, 100 mortars, a few thousands handguns, a large amount of ammunition, and even one buried artillery piece were taken from local villagers in Kosovo. This particular issue was discussed by the local Communist Party leadership on 22 September 1966. Details of this discussion (and the analysis of the discussion) were produced in Sekulić (1989, pp. 340–41).

A textbook used in the Yugoslav State Security training schools specified that national minorities (such as Yugoslav Albanians) ‘have strong national feelings and thus are often ready to work for [the] intelligence agency of their home country. As they are our [Yugoslav] citizens and so have all rights of our citizens—to serve in the Army, to be state officials and people's deputies, they can use these positions to collect information [i.e. spy for neighbouring countries]’ (cited in Sekulić1989, p. 154).

In 1959, Kosovo's borders were expanded in the north to include Serb-populated regions. According to Krstić, parts of central Serbia, Ibarski Kolašin, Zvečan and Leposavić, were included in the Kosovo region to increase the number of municipalities with a Serbian population (Krstić2004, pp. 39, 50).

In 1958 at the meeting of KPJ's Committee for Interethnic Relations, it was pointed out that ethnic Albanians in Kosovo and Macedonia spoke a different dialect from Albanians in neighbouring Albania, and that the existence of two different dialects was a problem in terms of local education and the overall cost of obtaining proper print-related material (AJ, A CK SKJ, XVIII-K1/16 (37)).

We uncovered no further evidence of such an (aborted) plan by the Yugoslav communists; if it existed, the plan would have raised many additional issues, including the status of Yugoslav Albanians outside Kosovo, most importantly in western Macedonia, and questioned whether the new nation should be an equal Yugoslav one. In the latter case, Kosovo would need to become a socialist republic, and Yugoslavism would have to move away from the concept of South Slavic unity.

Some Albanians were allowed to move from Albania proper into Kosovo in the post-war period, but their numbers were small. According to Serbian demographer Ruža Petrović, in 1981 there were about 4,800 inhabitants of Kosovo who were born in Albania (Petrović 1991a, p. 175).

Albanian politician Adem Demaçi has argued that the 1950 Yugoslav–Turkish agreement activated the notorious plan of Serb ultra-nationalist Vaso Čubrilović for the expulsion of Albanians, first formulated in the 1930s (Radončić1990, p. 10). Demaçi does not provide any specific evidence to back up this claim. On the Čubrilović plan, see Stefanović (2005, pp. 481–82).

A Macedonian communist delegate at the meeting of the KPJ's Committee for Interethnic Relations (held on 13 October 1958) mentions that his Republic received 18,000 individuals (no ethnicity is specified) from Kosovo who expressed a desire to immigrate to Turkey (selling all their property in Kosovo in preparation), but they later changed their minds and decided to stay in Macedonia (AJ, A CK SKJ, XVIII K1/16: 37–38). He also says that 130,000 Turks emigrated to Turkey from Macedonia, but 9,000 were ethnic Albanians claiming to be Turks (AJ, A CK SKJ, XVIII K1/16: 38). According to another document analysed by the Committee, by the end of 1956, 86,380 individuals (both ethnic Turks and Albanians) had left Macedonia for Turkey; with regards to Kosovo, the figure is not given, but it is noted that the number is ‘much less’ due to prohibition (AJ, A CK SKJ, XVIII-K4/13: 20). Stenographic notes from the KPJ's Central Committee for Interethnic and Inter-republican meeting (held on 21 September 1967) give a better idea of the number who emigrated from Kosovo to Turkey during this period (1951–1967). Prominent ethnic Albanian Party member Fadilj Hodza states that 14,000 individuals moved from Kosovo to Turkey (AJ, A CK SKJ, XXIIIA-K3/1: 93). There is no indication how many were ethnic Turks and how many were ethnic Albanians. According to Yugoslav census data, the number of self-declared Turks inexplicably jumped from 97,954 in 1953 to 259,536 in 1961 (Hoxha 1984, p. 68; Petrović1987, p. 23). Kosovo Albanian politician Hajredin Hoxha estimates that from 1953 to 1966, 230,716 Yugoslav citizens moved to Turkey and that ‘at least’ 80,000 of them were ethnic Albanians (Hoxha 1984, p. 70), but he fails to explain how he arrived at this number. Reconstructing timeline and statistical information about the migration of Kosovo Albanians to Turkey from 1953 to 1966 is challenging. First, a number of Slavic Muslims and ‘real’ Turks were also migrating from Yugoslavia to Turkey, and the official figures do not appear to make a clear distinction between these three groups. Second, migration was going on not only from Kosovo, but from neighbouring Macedonia, which too had a large Muslim population. Third, several reports indicate that the policy on migration was more ‘liberal’ in Macedonia than in Kosovo, which reportedly led significant numbers of Kosovo Albanians to move to Macedonia so that they could migrate from there (AJ, A CK SKJ, XVIII K1/16: 37–38; AJ, A CK SKJ, XVIII-K4/13: 20). Finally, it seems plausible (and several Party documents make this point) that the ethnic boundary between Muslim ‘Albanians’ and ‘real Turks’ was blurred in this period (AJ, A CK SKJ, XVIII K1/16: 38), largely due to the lasting identification of many Balkan Muslims—including Muslim Albanians—with the Ottoman Empire. Similarly, ethnic boundaries between ‘Serbs’ and ‘Montenegrins’ are frequently blurred today, with one sister or brother self-declaring as ‘Serb’ and another as ‘Montenegrin’.

The 4th Plenum (the ‘Brioni Plenum’) was not about Kosovo per se; rather, it was the occasion for a communist-style purge through which the Party adopted a major global policy shift, distancing itself from its previous policies (calling former policies ‘deviations’), blaming them on certain individuals, and sacking those individuals. This ritual allowed the Party to claim to be perpetually ideologically ‘correct and consistent’, regardless of dramatic policy changes.

Ranković, who was a ruthless police chief, loyal to the regime, and personally devoted to Tito, was sacked as a potent symbol of the centralist model which was now abandoned by the Party (Sekulić1989, pp. 17–18).

For a detailed reproduction of stenographic records of discussions and conclusions of the 6th Plenum of the Central Committee of the League of Communists of Serbia see the Section Šesta Sednica CK SK Srbije (14–15. septembra 1966) in Popović (1999, pp. 791–840).

Slogans during the Albanian demonstrations in 1968 included: ‘We want a republic’; ‘We want a constitution’; ‘We are Albanians, not Yugoslavs’; ‘We want a national flag’; ‘We want Preševo in the province’; ‘Long live Albania’; ‘One people, one state, one party’ (Horvat 1987, pp. 100–1). Preševo Valley is a small Albanian majority region immediately east of Kosovo. According to a letter from the KPJ's Serbian provincial committee addressed to the KPJ's Kosovo regional committee dated 6 November 1944, Preševo Valley was temporarily included in the Vranje district until it could be handed over to the Kosovo regional committee. This territorial transfer never happened, and the valley remained part of central Serbia; see Petrović (1988, p. 138).

In an Internal Party Report of 1967, Asllan Fazlija noted ‘some ideological–political problems with regards to position and exercise of equality of nationalities in the SFRY’ (‘Materials for Debate’, AJ, A CK SKJ, XXIIIA-K3/1).

See AJ, A CK SKJ, XXIIIA-K3/1:16. In socialist Yugoslav state theory, nations and nationalities were fully equal, but separate legal categories continued to exist. Discussions about this issue inevitably led to questions regarding the nature of Yugoslavia itself—was the country essentially a South Slavic union, or was it purely a socialist entity regardless of the ethnic affiliation of its citizens?

AJ, A CK SKJ, XXIIIA-K3/1: 16.

AJ, A CK SKJ, XXIIIA-K3/1: 17.

AJ, A CK SKJ, XXIIIA-K3/1: 17.

AJ, A CK SKJ, XXIIIA-K3/1: 18.

AJ, A CK SKJ, XXIIIA-K3/1: 36.

These meetings included the Sixth and Eighth Plenums of the Central Committee of the League of Communists of Kosovo.

For example, see the speech by Veli Deva (Kosovo Albanian delegate) at the Sixth Plenum of the Committee of the League of Communists of Serbia, in which he argued that the Ranković-led State Security employed ‘special measures’ against members of the Kosovo Albanian intelligentsia—especially teachers of Albanian language, literature and history (Popović1999, pp. 802–3). Veli Deva used the term ‘bureaucratic-chauvinistic dreg’ (Popović1999, pp. 803–4) to describe the Ranković faction's operations in Kosovo, implying that they were pursuing Soviet-style centralism and Serb nationalism at the same time. Deva's views found their way into the official conclusions of the Sixth Plenum: Point II of the final resolution accused Ranković's faction of both ‘unitarism and nationalism’, binding—and consequently equating—the two. Transgressions of the State Security in Kosovo against ethnic Albanians were labelled as ‘the most drastic (although not the only) form of anti-constitutional and chauvinistic practice’ (Popović1999, p. 840).

Marković listed a number of Kosovo Serb communists who supported republican status for the Province, and claimed that most of Vojvodina's oligarchy did so to create a precedent for the creation of the ‘republic of Vojvodina’.

The Official Gazette of SFRY, 55/68 (Beograd, Novinska ustanova ‘Službeni list SFRJ’).

The Amendment reads: ‘Through the common struggle of the peoples and nationalities of Yugoslavia during the National Liberation War and Socialist Revolution [the provinces had joined Serbia] on the basis of the freely expressed will of the population—the peoples and national minorities of the Provinces and of Federal Serbia’ (The Official Gazette of SFRY, 55/68, Beograd, Novinska ustanova ‘Službeni list SFRJ’).

Another Serbian Communist, Đuro Trbović, argued that ‘the thesis [according to which] nationality has a right to self-determination is unsustainable. Nationalities do not have that right and KPJ's position on that is clear’ (Mišović1987, pp. 63–65).

‘Ostvarivanje Politike Ravnopravnosti’ (A CK SKJ XXIIIE-K2/5–6 Ostvarivanje politike ravnopravnosti naroda i narodnosti u SR Srbiji 1976, pp. 5, 16, 85).

The completed fertility rate refers to the average number of children per woman by the time she completes her reproductive cycle. For example, if in population A there are 10 women in this age group (45–49, usual age of completed fertility) who have 63 children, then the completed fertility rate for population A is 6.3 children per woman.

Albanian demographer Islami notes that the completed fertility of Kosovo Albanian women who lived in urban areas and were active in the labour force was only 2.74 (Islami 1989, p. 45). Serbian demographer Petrović agrees with Islami that the increased participation of Albanian women in education and employment lowered overall fertility rates, and points to the high level of patriarchical traditionalism in many Albanian families. Islami, however, blames the position of Albanian women on Kosovo's lack of economic development (Islami 1989, p. 45; Petrović1991b, p. 189). Kosovo Albanian women were five times less likely to have paid employment than Kosovo Serbian women (Popovićet al. 1990, p. 13). Thus, it appears that the affirmative action hiring policy in Kosovo only focused on ethnicity, ignoring gender, thereby channelling policy benefits to Albanian men and failing to contribute to gender equality within the Albanian community. Lack of employment opportunities for Albanian women, in turn, made very high fertility levels more likely and delayed the demographic transition of the Albanian population. Only 9.55% of working age women in Kosovo had paid employment, while in Yugoslavia as a whole the number was 32.71% (Popovićet al. 1990, p. 43).

In the late 1980s, Albanian nationalist Adem Demaçi proudly stated that ‘the [Yugoslav] Albanians succeed in doubling their numbers’ in 25 years (Radončić1990, p. 66).

In June 1977, Kardelj noted in his address to the Kosovo Albanian communists in Priština that they must modify their education policy, ‘because if you keep going without a sense of limit or focus on the expansion of the highly educated cadre, you will not be able to employ them and they will not be able to find a job anywhere in Yugoslavia, so in the end you have to expect that these people will react in a nationalist way’ (Hasani 1986, p. 320). The 1981 Albanian demonstrations proved Kardelj right.

Lazar Koliševski, a leading Yugoslav Macedonian communist, observed in April 1981 that the ‘tremendously large university in Priština keeps producing a profile of graduates who cannot find jobs and thus, may become nationalist activists’ (Popović2001, pp. 269–70).

While the 1986 survey of Serbs and Montenegrin families who left Kosovo contains a number of important findings regarding these people's experiences, it has several limitations. As the authors acknowledge, they did not collect data on the opinions and experiences of Kosovo Albanians or of Serbs remaining in Kosovo (Petrović & Blagojević1989, pp. 5, 23). The sample of 500 families was not random, which leads to problems of representativeness (Petrović & Blagojević1989, p. 5). The authors did not analyse some of the reasons that might have prevented significant Kosovo Albanian outmigration, such as Slavic discrimination (Popovićet al. 1990, pp. 5, 14). Serbian migrations from Kosovo were not systematically compared with Serbian migrations from less developed parts of central Serbia (Agani 1989, p. 119) that had no presence of ethnic others. Finally, the study occasionally uses alarmist language and tends to systematically downplay economic factors when explaining migratory decisions.

Quotas were based on the percentage of Albanians in the total population, not their share of the working age population. For example, in 1971, in response to the complaints of Serbian communists that Albanians were monopolising job opportunities, a senior Albanian communist said that the percentage of Albanians in the whole population was still higher than the percentage of Albanians in the employed population (Mišović1987, p. 274). This argument ignored the fact that, because of the much larger proportion of young Albanians, many were not old enough to be employed.

See National Composition (1956) and A CK SKJ XXIIIE-K2/5-6 Ostvarivanje politike ravnopravnosti naroda i narodnosti u SR Srbiji, 1976. While this was a rather drastic change in the Party composition, Albanians were still under-represented compared to their share in the total population of the province, which was about 74% in 1971.

As a 1976 internal Party report explains, Kosovo ‘is the only region of the country in which there is an absolute growth of rural population, which causes rural over-population [prenaseljenost] with all of its negative implications’ (A CK SKJ XXIIIE-K2/5-6 Ostvarivanje politike ravnopravnosti naroda i narodnosti u SR Srbiji, 1976, p. 18).

For example, in the early 1970s, two Kosovo Serbian communists complained about the ethnic harassment of Serb peasants, and a Kosovo Turkish communist complained about Albanian assimilationist pressure on ethnic Turks. All three were purged by the Albanian-dominated Provincial Communist Committee (Pavlovićet al. 2004, p. 26).

A CK SKJ XXIIIE-K2 /5-6 Ostvarivanje politike ravnopravnosti naroda i narodnosti u SR Srbiji 1976, p. 5.

From 1953 to 1981, the rural population in Kosovo increased by 55% (Popovićet al. 1990, p. 7).

The migration of Serbs from Kosovo provoked a heated discussion in the 1980s, with Yugoslav Albanian scholars and politicians explaining the movement as a result of economic causes (a lack of jobs in Kosovo and lower land prices in central Serbia) or Serbian unwillingness to live in equal conditions with Albanians (Dragović-Soso 2002, p. 118; Islami 1989, p. 53). Serbian scholars stressed ethnic harassment and discrimination (Dragović-Soso 2002, p. 119; Blagojević2006, pp. 151, 155). The findings of the 1986 survey indicate a combination of economic and non-economic motivations. For example, while a significant percent (15.6%) of Kosovo Serbs who moved out indicated that they would have stayed in Kosovo if they could have obtained employment, the vast majority (76.6%) indicated that they never wanted to return. The majority (57.8%) stated that the greatest improvement associated with migration was freedom from harassment and assault (Petrović & Blagojević1989, pp. 223, 306–9). However, subsequent claims by Serbian ultra-nationalists that Kosovo Serbs were victims of a ‘genocide’ committed by Albanians in the 1970s were meaningless, inflammatory exaggerations (Dragović-Soso 2002, pp. 126–27; Bracewell 2000). Instead, the flight of Kosovo Serbs should be seen in the context of the overall trends of Yugoslav migration in the post-1966 period, characterised by movement from a minority situation to a majority situation. Due to unintended effects of Yugoslav ethno-federalism, many Yugoslavs faced the dilemma of why they should be a vulnerable minority in someone else's republic, when they could be titulars in their own republic. Many resolved this dilemma by moving to the safety of their home republics. While it could be argued that the migration from Kosovo was more intensive, similar processes were happening in other parts of the country with other ethnic groups.

The Official Gazette of the SFRY, 9/74 (Beograd, Novinska ustanova ‘Službeni list SFRJ’), articles 2, 3, 4, 5, 245, 300.

The Blue Book is reproduced in its entirety in the document section of Đekić (1990, pp. 123–74).

The 1981 demonstrations, which involved widespread street violence and massive use of the security forces, marked a qualitatively new period of Yugoslav communist history. A detailed analysis is well beyond the scope of this essay, nor for the time being, can such analysis be conducted from the primary sources in the State Archives of Serbia and Montenegro, as these will only become accessible to researchers after 2011.

Clearly, supportive evidence from a single case study can, at most, raise the nested homelands thesis to the level of plausibility. To refine the description of the causal mechanisms, determine the explanatory power and outline the scope of the thesis, it is necessary to systematically compare the Kosovo case with nested homelands in the USSR, both those that might have had disintegrative effects (such as Karabakh or Abkhazia) and those with little or no evidence of such effects (such as North Ossetia). While the full development of our theoretical argument demands such comparative historical analysis, such a project goes beyond the scope of this essay.

Fear of the State Security prevented ethnic Albanian intellectuals and politicians from articulating their grievances before the fall of Ranković. Similarly, in 1971 a group of Belgrade University professors who tried to articulate Serbian national opposition to the Communist Party designs on Kosovo were promptly purged.