University of Chicago, USA

Observing the Observers: Language, Ethnicity, and Power in the 1994 Macedonian Census and Beyond

In Sarajevo, before the Yugoslav war, there was a museum at one end of the bridge where Archduke Ferdinand was assassinated. Among the displays at this museum was a political cartoon from the period shortly before the outbreak of the Great War. The cartoon shows a disorderly circle of powder kegs, some with long, dangling fuses, others on their sides with gun-powder spilling out. In the center of the circle formed by these powder kegs are a few thin, ill-shaven, dark-mustachioed men in national costumes of the Balkan nations looking around bewildered. Standing outside the circle, eagerly extending lit matches to them are plump, pale, well-groomed men in the West European formal dress of the Great Powers. Thus was the concept of "Balkan powder keg" understood in former Yugoslavia. [ii] There is a certain irony in the image of the Balkans in the center and the Great Powers at the periphery, since in fact precisely the opposite is and has been the case in virtually every sphere of relations between Southeastern Europe and the rest of that continent. And Macedonia became and remains a potential center of conflict because it is the periphery of all its neighbors, who are themselves on the periphery of Europe. [iii]During the nineteenth and early twentieth centuries, national movements in Albania, Bulgaria, Greece, and Serbia all crystallized in such a way that Macedonia was (and is) at the edge of their overlapping claims. One of the ways that conflict has been expressed is through rival census claims.

In comparison to the current position of Albania and Kosovo, the Republic of Macedonia is both central and liminal. Unlike Kosovo with its shadow government and parallel education system, Macedonia meets the normal requirements for an independent country, but unlike Albania with its unequivocal international status and membership in the United Nations under its own name, Macedonia does not enjoy the normal recognition of an independent European state insofar as only some countries have recognized it under it own constitutional name, while others use the temporary United Nations term "Former Yugoslav Republic of Macedonia". Kosovo can be viewed as a region where an ethnolinguistic Serbian minority dominates the Albanian majority; Albania is a country ruled by its ethnolinguistic majority. [iv] But in Macedonia, which has foreign troops stationed inside its borders (the U.S. and Nordic battalions of UNPREDEP), the very legitimacy of the identity of the majority ethnolinguistic group, i. e. the Macedonians, is still subjected to equivocation, both purposeful and naove. [v] Moreover, some actors would dispute whether the Macedonians constitute a majority or even a plurality in the Republic of Macedonia. While in many respects the situation of Albanian majorities in both Albania and Kosovo can be viewed as economically or politically worse than the situation of the Albanian minority in Macedonia, it is Macedonia that is arguably the most unstable of the three, the country on which both Albanian and Kosovar attention is focused. One expression of the instability in Macedonia is the persistence of conflicting population figures.

The counting of populations has been potentially fraught with political tension for millennia. The Book of Numbers (I:2-3) describes a census for the purpose of preparing for war, and the census mentioned in the Gospel of Luke (II:1-5), was quite probably connected with Roman efforts at consolidating its hegemony in what was then still the kingdom (but later the province) of Judea (cf. Alford 1874[1980]:456-57). From 21 June until mid-July 1994, under intense internaland external political pressure, an extraordinary census took place in the Republic of Macedonia -- the ordinary census having been conducted in 1991, when the Republic was still "Socialist" and a part of what is now Former Yugoslavia. The 1994 census was not funded by the government of the country, as is ordinarily the case with modern censuses in sovereign states, but by supranational organizations -- the Council of Europe (CE), which at that time still refused to admit the Republic of Macedonia, and the European Union (EU), whose policies toward Macedonia have often been dominated by Greece. The extraordinary census of 1994 thus provides an opportunity to view more broadly both the complexity of the Macedonian scene the role of European mediation. [vi] The 1994 Macedonian census raises fundamental issues of which the more recent conflicts such as those over education and language use at the republic level are continuations, and it is thus worthy of a more detailed account as an historical moment around which national and international tensions crystallized. Whatever the developments in Macedonia's future, the 1994 census is one of the key links in the chain of events leading to it. In this paper, I shall examine the 1994 Macedonian census both as an eventin itself, and as a part of the larger context of quests for identity and hegemony in the Balkans. In so doing, I hope to shed light not only on both specific and general questions connected with the concepts of ethnic, linguistic and religious identity, but also on the relationship of the supranational to the national, of the central to the marginal, and of "Europe" to the land mass west of the Urals and north of the Mediterranean. I will suggest that the Western Great Powers that to a great extent determine (and fund) the policies of the those actors designated as the International Community, by imposing their own constructs while continuing marginalize Macedonia are not contributing to its stabilization.

I was working that summer as a senior policy and political analyst covering Macedonia for the Analysis and Assessment Unit organized by Dr. Susan Woodward for the Office of the Special Representative of the Secretary General, Mr. Yasushi Akashi, attachedto the United Nations Protection Forces (UNPROFOR) stationed in Former Yugoslavia. In connection with these duties, I arranged to be authorized as an outside observer of the extraordinary 1994 census in my capacity as a member an international organization in accordance with Article 33 of the Census Law. Although I was not officially connected with any of the census' funding organizations, the majority of their representatives were quite willing to allow me to accompany them on their duties and attend their meetings. As a result, I was able to observeboth the process of the census and the European observers who were officially observing it.

Questions of ethnic identity, citizenship, language rights, and the interrelationships of the concepts of language, religion, and "nationality" were hotly contested in Macedonia. The census was therefore a clearly political event rather than the statistical exercise it was officially claimed to be. And this was not the first time that Macedonian census figures have been the subject of conflict concerning these factors. At the beginning of this century, as at the end, economic and political structures in the Balkans were unstable and/or in transition, wars were being fought, interethnic tensions were high, and Macedonia was the object of conflicting claims supported in part by conflicting census figures. The figures in Table 1 display examples of the figures that were used to bolster these claims to Macedonia at the end of the nineteenth and beginning of the twentieth centuries, as Ottoman power waned and the small states of Southeastern Europe sought to consolidate and expand their respective hegemonies.


ethnic group Bulgarian % Serbian % Greek % Turkish %
Bulgarians 1,181,336 52.31 57,600 2.01 332,162 19.26 896,497 30.8
Serbians 700 0.03 2,048,320 71.35 0 0 100,000 3.4
Greeks 228,702 10.13 201,140 7.01 652,795 37.85 307,000 10.6
Albanians 128,711 5.70 165,620 5.77 0 0 0 0
Turks 499,204 22.11 231,400 8.06 634,017 36.76 1,508,507 51.8
Other 219,571 9.72 166,540 13.86 105,844 6.13 99,000 3.4
Total 2,258,224 100 2,870,620 100 1,724,818 100 2,911,004 100

Conflicting Census Figures for Macedonia: 1889-1905
Sources: d'Estournelles de Constant (1914:28-30) and Saral (1975:75)

Although Dako (1919:75) in his book significantly entitled Albania: The Master Key to the Near East[vii]cites similar figures and refers to the obvious discrepancies as 'amusing', these discrepancies are not entirely arbitrary. Rather, at least to some extent, different authors have selected criteria that would support their point of view.

In the case of Greek and Turkish authors, the choice was based on religion and/or schooling. Any member of the Greek Orthodox Church, or, after 1870, any Patriarchist (as opposed to Exarchists, called "schismatics" by Nicolaodes [1899:26]) as well as anyone who went to a Greek school (and since schooling was controlled by religion, Macedonian Christians were left with little choice until the mid-nineteenth century) was counted by the Greeks as a Greek, hence expressions such as "slavophone Greek" and "albanophone Greek". [viii] The complete absence of Albanians from the Greek figures is explained by their being counted as Turks, Greeks, or Miscellaneous on the basis of religion (Muslim, Orthodox, Catholic; cf. Saral 1975 cited in Table One).

Since the Serbian Orthodox Church and Serbian schools remained relatively weak except in parts of the north and west of Macedonia, Serbian authors selected specific isoglosses, i. e. dialect boundaries based on individual linguistic features, to justify ethnic and therefore territorial claims, as illustrated in Table Two and Figure One. [ix]


  [1]shoulders [2]woman/the woman

Differing isoglosses used to support conflicting territorial claims:
[1] the reflex of Common Slavic *tj, [2] the presence of a definite article

The forms cited in the table are those used in the modern standard languages.

The dialectal situation is considerably more complicated but is irrelevant to the basic point being illustrated.


Map showing approximate locations of isoglosses [1] and [2]
These representations are highly schematic. 
The precise distribution of features is complex but irrelevant to the basic point. See Ivic" (1958:25-49).

The isogloss illustrated by column one -- the reflex of Common Slavic *tj, where Serbian and Macedonian have a single palatal stop or affricate rather than a combination of palatal fricative plus affricate or stop -- was used by Belic" (1919:250) to justify Serbian claims to virtually all of northern and central Macedonia. Belić (1919:250), citing Meillet, writes that the Macedonian dialects are neither Serbian nor Bulgarian and that politics will determine the linguistic fate of Macedonia. However, he then goes on to claim that the north and central Macedonian dialects are basically Serbian while the south is basically Bulgarian on the basis of this single isogloss. Belic" (1919:253-56, 264) ridicules Bulgarian scholars who were claiming all of Macedonian as well as Southern Serbian as Bulgarian dialects by suggesting that his opinion coincided with that of impartial European scholarship, i. e. his interpretation of Meillet. He declared that this was because Serbia had contact with the West while Bulgaria "slept deeply under the Turkish yoke, " and that the resulting difference in intellectual development could not easily be overcome. The genuinely impartial French scholar Vaillant (1938:119), however, observes that Belić's argument is based on one phonetic trait and that most Slavists agree that Macedonian is actually a part of a Macedo-Bulgarian group that has been subjected to the prolonged influence of Serbian. He lists numerous phonological traits which link Macedonian with Bulgarian rather then Serbian, e.g. the fate of the jers (short ĭ and ŭ) and juses (nasal ę and ,o), /vŭ/, and vocalic /l/, and notes that vestiges of /št/ in the /ќ/ area show that the latter reflex is the result of substitution, e. g., in Galičnik gak"i 'britches' but gašnik (cf. Bulgarian gaštnik) 'a belt for holding up gakji'. Vaillant (1938:304-08) concludes that Macedonian is not a dialect of Bulgarian but deserves a separate place in Macedo-Bulgarian group. It is important to note that Vaillant wrote this six years before the political recognition of Macedonian as an independent language

On the other hand, choosing a feature such as the presence of the postposed definite article -- as in the second column -- helped justify Bulgarian territorial claims to the Timok-Morava valley in southern Serbia as well as to Macedonia. (Cf. Glenny 1995:24 for an uncritical repetition of this type of linguistic claim with respect to Macedonian. ) Bulgarian figures assumed that virtually any Slav in Macedonia was Bulgarian, and increased the numbers by assuming higher fertility and incidence of extended families for Slavs than for other groups (Kaštnčev 1900:136-37). Thus, for example, if a given village had fifty Albanian houses and forty Slavic houses, by counting 5 members per Albanian household and 7 members per Slavic household based on the foregoing assumption, we end up with a Slavic majority despite the smaller number of houses: 280 Slavs as opposed to 250 Albanians. This is an early example of how important statistical and demographic assumptions underlying counting procedures can be for the outcome of censuses; such assumptions continue to be characteristic of censuses to this day.

Notably absent from these statistics are any figures representing the views of ethnic Macedonians themselves. [x] Except for the map by Čupovski (1913, cited in Petruševski 1992:83), which in any case is not a statistical document, we have very little evidence of Macedonian views in the published literature except occasional moments such as Pulevski's statementof Macedonian national consciousness (1875:48-49), Misirkov's formulation of Macedonian language and statehood (1903:71), the correspondence concerning the Kostur (Greek Kastoria) school of 1892 (Andonovski 1985), and Upward's (1908:204) account of his trip to Voden (Greek Edhessa). [xi] As Rossos (1994, 1995) makes abundantly clear, the suppression of Macedonian ethnic identity in all its manifestations was not only in the interests of all the small powers that lay claim to the territory, but ultimately also in the interests of the great powers that supported the various small powers and that ultimately had a stake in maintaining the partitions of Macedonia as a viable solution for peace. Brown (1996) discusses heretofore untapped archival sources outside the Balkans that attest to both the existence of a separate sense of Macedonian nationality at the beginning of this century and attempts to dismiss it. In certain respects that situation is being replicated today, and population figures are again being used to bolster conflicting claims ranging from minority rights to irredentism. In particular, the technique of privileging religion over language as the basis of identity, which was used by both Turks and Greeks (and later Bulgarians and Serbs) to hegemonize and assimilate various populations in Macedonia, is again being brought into play, as will be seen below.

From April 1-15 1991, under conditions of impending political disintegration, the Socialist Federal Republic of Yugoslavia conducted its last census. Before the end of that year, while the census data were still being processed, war had broken out in Former Yugoslavia and the Republic of Macedonia had subsequently declared independence. The census itself was carried out in an atmosphere of distrust and animosity. Led by the two largest Albanian-identified political parties in Macedonia, the Party for Democratic Prosperity (PDP, in Albanian PPD) and the smaller but more vocal and radical Peoples Democratic Party (NDP, in Albanian PDP) the majority of Albanians in Macedonia (and elsewhere) boycotted the 1991 census, claiming that they would be purposefully undercounted.

The Party for Democratic Prosperity was founded on 15 April 1990 in Tetovo, now a predominantly Albanian town with close historical and communication links to Kosovo, which is just on the other side of Mount Šar. [xii] Branches of the PDP continued to be founded in various towns throughout western Macedonia during that year. Although proclaiming as its major goal minority rights -- and indeed in the 1990 local elections seven Turks and seven Muslims (presumably Slavic-speaking) were elected to local councils (Antonovska et al. 1994a:57) -- the parties are basically Albanian rights parties that advocate special treatment for Albanians as being not a minority but a "constitutive nation" in Macedonia. [xiii] The boycott was first called for by the Peoples Democratic Party and was observed in the communes (opštini) of Debar, Gostivar, Kičevo, Kumanovo, Ohrid, Skopje, Struga, Tetovo, and Titov Veles (Antonovska et al. 1991). The Bureau of Statistics estimated the data for Albanians in the boycotted communes by means of statistical projections utilizing the data from the 1981 census, natural growth of the population during the inter-census period, migration, and other statistical data (Antonovska et al. 1991). The preliminary results were published in November 1991. Table Three gives comparative statistics for all the post-War censuses conducted on the territory of the Republic of Macedonia. [xiv]The categories "Egyptian" and "Bosniac" represent new sociopolitical realities. [xv]


Census total by year, number, and percentage (rounded upward where necessary)

Figures and percentages for declared nationality (narodnost) in all post-World War Two Censuses conducted in the Republic of Macedonia
Sources: Antonovska et al. 1991, 1994a, 1994b, Latific" et al. 1970, Pekevski et al. 1973, Savezni Zavod za Statistiku 1954, 1981.

Before the preliminary figures for the 1991 census were published, Albanian political actors began an international media campaign declaring not merely that they had been miscounted, but that infact Albanians constituted about 40 per cent of the population of Macedonia, i. e. seven to eight hundred thousand (Nova Makedonija 91/04/20). Albanian political actors were supported in their claim by Greece (MILS [Macedonian Information Liaison Service ] 92/12/30), which denies the existence of a Macedonian language and nationality altogether, particularly on its own territory (Human rights Watch/Helsinki 1994:11). Representatives of other groups also cited larger statistics: Serbs claimed up to 300, 000, Turks up to 200, 000, Roms 200, 000, Greeks 250, 000, Gjupci 30, 000, Bulgarians and Vlahs similar figures (cf. MILS 93/01/13, MILS 93/02/22). Added together, these claims surpassed the total number of inhabitants of Macedonia, even without counting Macedonians. The point was clearly not one of statistical accuracy but rather claims to political power and hegemony.

The success of the Albanian public relations can be seen in the fact that within a year of the publication of the preliminary results of the 1991 Yugoslav census, Dr. Geert-Hinrich Ahrens, a German diplomat with the rank of Ambassador and head of the Working Group for Human Rights and Minorities within the International Conference on Former Yugoslavia (ICFY), called for an extraordinary censusin Macedonia to be supervised by the "international community" (MILS 92/11/25). Ambassador Ahrens made two proposals: One involved only those areas with large Albanian populations, the other concerned the entire country. Such a pair of proposals had many implications and repercussions. The first proposal, by focusing on the Albanians to the exclusion of all the other minorities of Macedonia, which, according to the 1991 census made up 14. 37 % of the population, gave implicit legitimacy to Albanian claims for special treatment in addition to legitimizing Albanian politicians' right to claim discrimination at such a level and to demand a recount, as it were. At the same time, the proposals helped reify as a Macedo-Albanian conflict tensions that had been building since the riots in Kosovo in 1981 but that were not an inherent feature of Macedonian life at all periods.

Ahrens' announcement of November 1992 was followed by nineteen months of uninterrupted dispute. First there was an intense controversy over whether or not to hold the census. [xvi] This agreed upon, there followed prolonged debate over the wording of the census law, which was eventually passed with the support of the Albanian members of parliament. One of the chief issues was language use inthe census, and Article 35 of the census law provided for bilingual forms in Albanian, Turkish, Romani, Vlah, and Serbian in addition to Macedonian. [xvii] Finally, just as the census was actually beginning, there were serious behind-the-scenes negotiations with the Albanian members of parliament, who threatened to call for a boycott, despite the presence of the ICOM (International Census Observation Mission) observers and the expense incurred by the European organizations.

The overseers of the census appointed by the European organizations were officially called the Group of Experts. [xviii] Their fields of expertise, however, did not include knowledge pertaining to Macedonia. Rather they were, for the most part, statisticians and bureaucrats without previous Balkan experience.[xix] Many members of the ICOM team, including some of the highest ranking, told me that they were quite surprised when they discovered that they were embroiled in highly charged political issues, as opposed to a mechanical statistical exercise, and they expressed confusion and dismay over the complex ethnic situation they encountered. In view of the origins of the 1994 Macedonian census described above as well as explicit statements by Albanian political actors, the event was clearly linked toa political issue, namely the claim of Albanian politicians for special (non-minority) status for Albanians within Macedonia based on their large numbers (Xhaferi 1994). The Group of Experts, however, attempted to avoid the impression that it was involving itself in the internal political affairs of a sovereign state by publicly declaring that the census was merely a statistical exercise. It can be argued that by labeling the leadership of ICOM the Group of Experts, while avoiding the direct involvement of any parties that were familiar with Macedonia (but see note 18), the CE was attempting to lay claim to adjudicating authority in Macedonian internal affairs and at the same time project an image of objectivity. [xx]

The lack of knowledge of Macedonia on the part of the CE and ICOM was given symbolic representation in the orientation packet for members of the ICOM team. The only item relating to the country itself rather than ICOM's mission in it was a chart listing Cyrillic printed and cursive letters with the names of the letters in Cyrillic and Latin orthography and labeled simply L'alphabet. The very lack of a qualifying adjective in a sense erases Macedonian from the observer's view, and in fact the chart was not a guide to Macedonian Cyrillic, but actually a table of Russian Cyrillic with the last six letters blanked out. Although the last six letters of Russian Cyrillic do not occur in Macedonian, there are seven other lettersthat are used in Macedonian Cyrillic but not in Russian and were thus missing from the chart. [xxi]

To compound the effect, the names of the Russian Cyrillic letters utilize a vowel whose letter comes at the end of the alphabet, so the names of the letters used a symbol that was not given in the list of letters. This chart not only embodied the lack of concern with which the CE and ICOM approached the Macedonian context in which it presumed to operate, but gave false information to the purveyors of expert knowledge. In focusing on the Albanian question, ICOM lost sight of the Macedonian.

Similarly, the privileging of Albanian claims over all others was symbolically represented on the ICOM observers' control forms for censused households. [xxii] Although the Macedonian control forms had sections for indicating the six ethnic affiliations defined by the languages of the census forms (Macedonian, Albanian, Turkish, Romani, Vlah, Serbian), as did the ICOM control form on enumerators, the ICOM household control form specified only Macedonian and Albanian, the remainder being subsumed under 'Others'. The difference in these forms gave written representation to the different conceptions of ICOM and the Macedonian government concerning the purpose of the census.

On the first day of the census, 21 June 1994, I attended a press conference given by Ambassador Ahrens of ICFY, Werner Haug, Chairman of the Group of Experts Council of Europe/European Union, and Robin Guthrie, Director of Social and Economic Affairs, Council of Europe. In addition to insisting to those assembled that the census was a statistical exercise with no political dimension, the Expert Team focused on Albanian objections to question 6 on form p-1, Citizenship, for which the four possible answers were Macedonian, Alien, Person without citizenship, and Pending status (vo tek). The chief problem was that despite assurances to the contrary, the Ministry of Internal Affairs (MVR) had not succeeded in distributing all citizenship documents by the time the census began. In Eastern Macedonia this was not particularly important. Thus, for example, in Radoviš the local government actually used census enumerators to help distribute citizenship documents. Although the problems of document distribution occurred throughout Macedonia, Albanians maintained that a disproportionately high number of qualified Albanians were without citizenship documents. This problem was complicated by the number of Albanians that had fled oppression in Kosovo but whose status in Macedonia was unregulated. Although a compromise solution was eventually reached, a special MVR form tabulating citizenship was added on 2 July in three group areas, and the ICOM mission was upset at this irregularity. [xxiii] After the press conference, I attended a separate meeting between members of the Group of Experts and PDP members headed by Abdurahman Aliti, who later became president of the PDP. The topic was the threatened last-minute boycott mentioned above. Guthrie spoke in very strong terms to Aliti about the need for his party to cooperate with the census. Aliti unhappily noted that in reality the census did indeed have a political dimension, and that if he or his party openly called for support of the census he/they would be wiped of the political map. (Although Aliti did not state who would do this, the NDP or radicals in the PDP would have been the only logical candidates. ) He said that the best his party could do was promise not to actually call for a boycott, but neither would they call for support. Aliti made it clear that he understood the situation and wanted to see the census work, but he stated that he also saw no point in allowing radicals to destroy his political career. I then accompanied the Experts to their offices at the Bureau of Statistics, where the citizenship question was againthe main order of business. Since no one on the ICOM team knew either Macedonian or Albanian they were at a disadvantage when a question of the wording of the rules concerning the Citizenship question arose and the only rule book that was available wasin Macedonian. Members of ICOM also told me that they began their mission with no idea of its political implications or the tremendous ethnic and cultural complexity of the region. They thought they were going to be overseeing the technical aspects of astatistical exercise.

At the end of that first day, the Group of Experts' discussion about the complications they had encountered also revealed their view of Macedonia as something other than European. One member of the Group joked that they should conduct the census like the one 2000 years ago, when everyone went to her native village, a reference to the Gospel of Luke mentioned at the beginning of this paper. The unintended irony of her comment was that this was precisely what the Macedonians would have wanted and what the Albanians would have feared, since an indeterminate number of Albanians had come to Macedonia from Kosovo and elsewhere since World War Two, especially since the Kosovo uprising of 1981. The citizenship law set the term of residence at fifteen years, i. e. had the effect of excluding the most recent wave of Albanian immigration. During the debate over this law, Macedonian nationalist politicians advocated a term of thirty years, Albanian politicians pressed for five years. The longer term would have excluded the majority of Albanians who had come to Macedonia from Kosovo. [xxiv] Another member of the team, speaking in French, described how the census was conducted in Turkey, where there was a curfew (in French: couvre-feu), requiring everyone had to stay indoors and await the census takers under penalty of a heavy fine. A British member of the team misunderstood the French and thought the Turks burned villages during their census. In both the joke and the misunderstanding, the Balkans in general and Macedonia in particular emerge as a primitive "other", backward or barbaric.

It was at the beginning of another meeting between the Group of Experts and Albanian political leaders the following morning that I asked Ambassador Ahrens if it might not be the case that by internationalizing Albanian claims in Macedonia via the CE/ICFY-sponsored census ethnic tensions were in fact exacerbated. [xxv] Dr. Ahrens responded that he thought the international intervention was beneficial and cited as evidence the fact that as soon as the Council of Europe agreed to fund the census Albanian claims dropped immediately from 40% to 30%, and indeed during the negotiations that I attended, at which Albanian politicians were expressing particular misgiving over the issue of citizenship -- Xheladin Murati, who was head of the PDP when the census began but subsequently resigned, claimed that the question about citizenship was designed to demonstrate that the majority of Albanians in Macedonia were dojdenci 'immigrants'-- the figure they cited as a being the minimum below which they would claim falsification was 25%. I should note that even before the first results were released, the percentage claimed had jumped and after the first results were published, and despite ICOM approval, the figure 40% was again being cited, cf. also Albanian prime minister Alexander Mexi, who reportedly cited the figure 800, 000 (MILS 95/04/13). [xxvi]

At times ICOM approached the Macedonian government with a seriously distrustful, almost adversarial attitude. Since the 1994 census was being conducted as a result of the Albanian boycott of the 1991 census, there was a tendency at ICOM to view Albanian claims as based in fact rather than raising an unresolved question. The Macedonian government was thus sometimes viewed by ICOM guilty unless proven innocent. Censuses conducted by sovereign states are not normally overseen by other organizations, while censuses in colonies are supervised by their colonial rulers, as was the case in 1911 British census of India (cf. Anderson 1913:1-13). The fact that the 1994 census in the Republic of Macedonia was conducted under pressure from and with funding from external organizations (CE/ICFY) put that country in an ambiguous position. On the one hand, the external funding and oversight by individuals who were not citizens of Macedonia put the country in a position similar to that of a non-sovereign entity. On the other hand, Macedonia was treated as a sovereign state engaged in discriminatory behavior. As an example of how ICOM's lack of preparation combined with its tendency to view the Macedonian government with distrust led to incorrect judgments, the following incident can be cited. In July I was approached by ICOM members who informed me that the government was discriminating against Muslims by not listing them as Bosniacs (Bošnjaci) or by not giving their language as Serbo-Croatian. These ICOM members had been in contact with Bosniac political activists who had tried to convince them that all Slavic Muslims in Macedonia are Serbo-Croatian-speaking and/or Bosniacs. When I responded that there was a significant number of Macedonian-speaking Muslims -- popularly known as Torbeš, although they prefer to be called Muslimani -- who do not speak Serbo-Croatian and who do not identify as Bosniac, the ICOM reaction was a combination of surprise and skepticism. In the end they came to understand that the situation was indeed as I had explained it to them, but the very fact that such a misunderstanding could arise demonstrates not only the distrust towards the Macedonians with which the European Experts approached the census but also their difficulty in distinguishing information from misinformation disseminated by some ethno-political actors.

Macedonian Muslims often live in underdeveloped, neglected, and isolated areas, e. g. the municipalities of Debar and Kičevo, where there is no ethnic absolute majority. They have therefore been vulnerable to manipulation by Albanian and Turkish politicians who have convinced some of them that they are Slavicized Albanians or Turks rather than Islamicized Slavs, [xxvii] and that thus they could rely more on Turkish or Albanian political parties to support their economic interests, since in economies of shortage such interests tend to fragment along ethnic lines (cf. Verdery 1993). The emphasis of Macedonian nationalist politicians on the connection between the Macedonian Orthodox Church and Macedonian nationality has further alienated some Macedonian Muslims. [xxviii] Census attempts in Macedonian-speaking Muslim villages (Plasnica and Preglovo, Kičevo municipality; Župa, Debar municipality) ran into cases where a monolingual Macedonian Muslim family would demand a bilingual Albanian or Turkish form with an interpreter but then have to have the Albanian or Turkish translated into Macedonian. These incidents were part of a larger pattern of conscious language shift based on religion, such as the incident in the monolingual Macedonian Muslim village of Bačište (Kičevo municipality), where parents demanded an Albanian school for their children (Nova Makedonija91/05/13; v. also Friedman 1993).

A general problem with the 1994 Macedonian census as with other European censuses was the definition of the categories "mother tongue" and "nationality" (the ICOM control forms used ethnic affiliation andnational affiliationinterchangeably). [xxix] The concepts of ethnicity, nationality, language, and religion have a complex history of interrelationships in Macedonia, one whose complexity continues into the present day. Thus, for example, some Muslim speakers of Macedonian declare their nationality as Albanian or Turkish on the basis of identifying their religion with Turkish or Albanian ethnicity. Similarly, some Christian speakers of Albanian declare their nationality as Macedonian on the equation of Macedonian Orthodox Christianity with Macedonian ethnicity. As might be expected, Albanian ethno-politicians insist that Macedonian-identified Albanian-speakers are Albanians while Macedonians insist that Albanian-identified Macedonian speakers are Macedonians. There was also the citizenship-based category Yugoslav, which until 1991 was steadily growing in popularity, not only among Slavs, but also among non-Slavs. Now that Macedonia is no longer part of Yugoslavia, however, this category has ceased to be valid for most people, since it refers to another country. [xxx]

At least some of ICOM observers were unaware of the difference between Serbo-Croatian and Macedonian when they arrived to observe the census, as reflected in their questions to me. When they finally grasped that the difference was a linguistic one, they concluded that language was therefore the basis of nationality. While language and ethnic or national affiliation coincide to a certain extent in Macedonia, such is clearly not always the case, as can be seen not only from such categories as Muslimbut also from Table Four, which gives statistics for the correspondence between declared nationality and declared mother tongue for the 1953 and 1981 censuses (figures for 1994 have not yet been processed).


Declared Nationality Macedonian Albanian Turkish Serbo-Croat Romani Vlah
Macedonians 853,971 1,986 281 934 277 2,565
Albanians 2,152 153,502 6,569 181 70 1
Turks 32,392 27,087 143,615 534 70 10
Roms 1,040 860 2,066 25 16,456 1
Vlahs 137 4 2 14 0 8,130
Serbs 3,945 0 8 31,070 41 9
Yugoslav 2,152 25 50 563 2 4
Other 322 341 569 5,258 173 31
Total 896,651 183,805 153,160 38,579 17,089 10,751
Macedonians 1,276,878 190 160 547 316  
Albanians 1,218 374,181 3 440 1,697  
Turks 16,608 8,592 60,768 366 94  
Roms 4,160 1,697 808 24 36,399  
Vlahs 1,111 1 0 3 2 5,257
Serbs 8,521 10 3 35,867 14  
Muslim 15,075 4,968 2,038 16,325 308 30
Yugoslav 7,645 1,943 274 2,746 530  
Other 13,282 4,247 2,853 17,031 1,280  
Total 1,334,498 391,829 64,907 63,349 37,780 5,931

Difference between declared nationality and declared mother tongue for the six main languages of the Republic of Macedonia: 1953 and 1981
Sources: Savezni zavod za statistiku 1953a, Savezni zavod za statistiku 1988

By attempting to impose a West European construct equating language with nationality (and nationality with statehood), ICOM helped force on people the kind of choices that have led to the current conflict (cf. Gal 1993:344-45). Moreover, the composition of the census form, which required respondents to declare a single mother tongue, effectively erased the multilingualism that has characterized the Balkans for centuries -- if not millennia -- and that is still a significant feature of Macedonian life is some areas (see Fraenkel 1993).

Lability of identity has long been a feature of life in Macedonia. The oldest generation from Western Macedonia remembers when Christians and Muslims would live under the same roof as part of the same extended family. Before the Mџrszteg agreement of 2/3 October 1903, only Muslims could serve as gendarmes, and such officials had significant power at the local level (cf. Skendi 1968:203, 207, 253). In Christian families, therefore, it was not uncommon for one brother to convert to Islam in order to be in a position to protect the entire family. Everyone ate at a common table, and if, for example, pork were available and a zelnik (pie) was made, the women of the house would put pork in only half the pita and both the Christian and Muslim sides of the family would eat from the same pan. On the other hand, marriages have always been freely contracted along religious lines but across linguistic ones. The children of such "mixed" marriages would grow up bi- or multi-lingual. In recent times, when faced with the necessity of choosing a nationality, choices can follow gender lines, e. g. if a Turkish man marries an Albanian woman, the sons may be Turks and the daughters Albanian, while in other families the choice may be for one son to be Albanian and one to be Turkish. The European concept of nationality, equating ethnicity with language with state, does not correspond to the complex realities of Macedonia (nor of many, perhaps most, other countries), and by focusing on 'nationality' to the exclusion of other characteristics we get contradictory situations such as those of parentsinsisting that their children be schooled in a language that they do not know despite the fact that the primary justification for multilingual education at the elementary level is the idea that children learn best when taught in their mother tongue.[xxxi]

Thepoliticization of the language issue and its confusion with nationality in the 1994 census was highlighted in several incidents that occurred in Albanian-speaking villages in southwestern Macedonia, where citizens objected to the fact that some of the Albanian-speaking enumerators were not ethnic Albanians but rather Roms (Gypsies), Gдupci ("Egyptians"), or Vlahs (Flaka e v'llaz'rimit 94/06/28). Since most Gдupci in southwestern Macedonia have Albanian as their first language and many Roms and Vlahs arefluent in it -- especially in southwestern Macedonia -- the issue was clearly not a question of the right to register in one's mother tongue but rather the demand for an ethnic Albanian, i. e. an instance of ethnic prejudice.

The events leading up to the boycott of the 1991 census, the imposition of the 1994 census, and subsequent developments show a pattern of manipulations and fragmentations of ethnic and linguistic identities utilizing legitimate grievances to benefit certain types of political elites. At the time of the census, my assessment was that it would prove a statistical success but a political failure. Insofar as it has not resulted in any significant changes in the figures -- both official and purported -- according to which ethnically based political relations are determined, this prediction has held true. The ICOM final report, while not uncritical, affirmed that the census was carried out according to "European" or "international" standards. It has been refuted by the Albanian political actors who brought it about, but at the same time they have generally continued to try to work within the existing governmental framework. In January 1995, the constitutional court ruled that article 35 of the census law, which governed language use, was unconstitutional, i. e. contrary to article seven of the constitution, which declares Macedonian the official language and guarantees (or restricts) official minority language use at (or to) the local level (MILS 95/01/26). Thus the census law solved nothing in this respect, and when I returned to Macedonia in December 1995 as part of the fact finding mission for the South Balkan Project of the Center for Preventive Action of the Council on Foreign Relations the question of language use at the federal level was still the focus of significant political tension.

If one of the purposes of the externally sponsored census was to either legitimize or silence Albanian claims and thereby promote in one way or another greater stability in Macedonian society, the presence of European mediators in the on-going dispute is not necessarily serving to promote stabilization. In an editorial published by the outspoken albeit still government-dependent weekly Puls more than half a year after the census ended and cited in MILS (95/02/03), Ambassador Ahrens is cited in the following terms:

"Arens [sic] developed a thesis of a parallel existence instead of a common existence between ethnic groups in Macedonia, particularly between Macedonians and ethnic Albanians. According to Arens, 'there never was a true coexistence', but nationalities in Macedonia 'have always led parallel lives. ' He said hehad the feeling that he probably knows more about Albanian history and culture than the average Macedonian. Also, he had the impression that nationalities had aversions to one and other. He backed this claim by the fact that there are no mixed marriages, and there are ethnic tensions in both public and private communication, especially between Macedonians and Albanians. "

This is a significant departure from Dr. Ahrens admonishment to Albanian politicians at the beginning of the census, when they were still threatening a boycott. At that time, he told them they were in the same boat with the Macedonians, and that if they -- the Albanians -- rocked the boat they would both drown. My own experience with journalists has sensitized me to the fact that what appears in the press is not always that which was actually said, but regardless of its actual source, the statement itself is an exemplary instance of a present construction being projected onto the past. It imposes a view of Macedonian reality that at the same time serves the interest of the local political elite that gives a diplomat his international legitimacy and promotes a version of the history of Macedonia that is not only at variance with concrete evidence -- e. g., the Assistant Minister of Education is the son of an Albanian father and Macedonian mother, the Prime minister's brother-in-law is a Turk, a Macedonian friend of minewho used to work in the government has an Albanian wife -- but helps to reify modern ethnic conflicts.

In a slightly broader context, Todorova (1994:453-454) also projects the present onto the past, albeit for quite different reasons, when she attempts to demonstrate defines the term Balkanism only as "politically and ethnically fragmented" or, citing Bercovici (1932), "Austro-Hungarian political policy relating to the Balkans". [xxxii] There is, however, a widely accepted meaning of the term Balkanism that is precisely the opposite of fragmented. In linguistics, a Balkanism is a feature shared among the unrelated or only distantly related languages of the Balkans. The grammatical structures of the Balkan languages attest to centuries of multilingualism and interethnic contact at the most intimate levels. Thus, for example, features such as the replacement of infinitives with subjunctive clauses shared by Balkan Slavic, Balkan Romance, Albanian, Greek, and even some Balkan Turkish dialects result from people speaking each others' languages. During the 1994 census, Debar proved to be the most intractable commune (for reasons relating more to competition between the periphery and the center than between ethnicities), and in the end it was the only commune in which the census was not completed. And yet, the Albanian and Macedonian dialects of Debar provide a striking example of phonological similarity that results from centuries of bilingualism.

The OSCE mission that was stationed in Skopje in December 1995 provided other examples of the need for outside observers to be better informed. I encountered among members of the Mission the misconception that Macedonian is a "really a Bulgarian dialect" (cf. note 4), an attitude that displays a remarkable insensitivity to the milieu in which they were supposed to operate as mediators. I was particularly curious about an incident that had occurred in the Skopje village of Ognjanci. At the time of the census, I made the following observation concerning that village in mynotes: "A mixed village (Macedonians, Turks, Albanians, Roms). It was reported that inter-ethnic relations are excellent and everyone cooperated happily with the census" (1 July 1994). And yet a little more than a year later the following news item appeared: "Yesterday in the Skopje village Ognjanci, a group of citizens of Macedonian and Serbian nationality, tried to prevent the entrance in the school to 40 children of Albanian nationality. There was no incident because the police without using force dispersed the gathered people. The Ministry for Education has decided to include the teaching on the Albanian language for the Albanian children in the same school. Those protesting think that the school was built by them, and will not allow their children to learn in combined classes with children from different grades. They offer a solution - the children of Albanian nationality to continue their education in the former barrack in the village, until the Ministry reconsiders its own decision and finally decides who the school building will be given to. Minister of Education, Emilija Simoska, for the Macedonian TV stated the Ministry does not intend to succumb to any kind of formal pressure and unless there is some disturbances among local people, assistance from Ministry of Interior will be asked. " [sic] (MILS 95/09/21). The OSCE had involved itself in the affair, which had been resolved peacefully. But when I inquired of one of the Mission's members how it was that inter-ethnic relations had deteriorated so significantly in so short a time, my informant responded: "Maybe they were Serbs. We never found out. All we care about is human rights, and then we move on. "

This same member of the OSCE Mission told me of their experience in Debarska Župa, where, recently, Macedonian-speaking Muslim parents have been demanding Turkish-language schools for their children (cf. the incident in Bačište cited above). The OSCEobserver informed me that they had met the Turkish teacher who called on children "randomly" to demonstrate how well they spoke Turkish. [xxxiii] The observer also told me that the grandparents spoke Turkish, implying that the parents were seeking to return the children to their roots (but see note 26 and the paragraph to which it refers). Since a number of the villages in that district are Macedonian Muslim -- the demand for language shift being motivated by factors cited earlier -- I asked the observer if they had spoken with any of the families. The observer responded that they did not visit any homes and did not care what the home language actually was. Parents have the human right to choose the language of the children's school even if the decision is handicaps the children by requiring them to begin school in a language they do not know. The circumstances that lead to such a situation and the resolution of the deeper causes of the problems it represents were explicitly of no interest. As with the ICOM observers, so, too, the OSCE mission does not appear to be prepared to address the complexitiesof the specific context in which it finds itself.

The 1994 census highlighted, among other things, the ambiguity of the term Europe. Geographically, it refers to a continent bounded by the Mediterranean and Black Seas, the southern slopes of the Caucasus and the western slopes of the Urals. Politically or culturally, however, the term Europe often still has the meaning of 'Western Europe' or 'the Europe of the Great Powers'. Thus, for example, the most powerful political unit on the continent calls itself the European Union, although only Western European nations plus Greece are included in it. It is no coincidence that Greece has embarked on a vigorous internal propaganda campaign stressing its membership in this Europe. The exclusion of the southeastern peninsula of geographical Europe from what can be called political Europe is well known in the Balkans.[xxxiv] The sense of alienation generated thereby was eloquently expressed by the Bulgarian author and journalist Aleko Konstantinov (1895) at the end of a vignette in his famous work Baj Ganjo, which satirizes the adventures of a Bulgarian rose-oil merchant in the Western Europe of his day and subsequently in then newly liberated Bulgaria. In the penultimate sentence of Baj Ganjo žurnalist 'Baj Ganjo as a journalist', Konstantinov writes: Evropejci sme nij, ama vse ne sme dotam!Й 'We're Europeans -- but still not quite!' I heard a comparable use of the term Europe during the 1994 census when an ethnic Albanian politician brought me with him into a restricted building explaining to the guard (in Macedonian) : "Toj e od Evropa" 'He is from Europe'. My companion knew that I was an American and an employee of UNPROFOR, but he identified me as od Evropa because my role at that moment was that of a privileged Western outsider just like a member of ICOM. Insofar as international European organizations succeeded in pressuring Macedonia into conducting a census which those organizations funded and observed, it can be argued that political Europe was exerting authority in geographical Europe's southeastern periphery and particularly in Macedonia as a periphery of peripheries in particular. [xxxv]

This Europe was utilized by both Albanian and Macedonian political actors to further their particular goals. The Albanian politicians mobilized quite legitimate social and political grievances based on very real discriminations against ethnic Albanian citizens of Macedonia, ranging from censorship and restriction of language and property rights to firings and jail sentences -- especially since 1981 -- to further a their own careers and demands for autonomy, federalization, and ultimately irredentism (cf. Xhaferi 1994). Macedonian statisticians and politicians, faced with the choice between an externally imposed census or further destabilization due to a loss of legitimacy in an international community that was at that time permitting Macedonia's economic strangulation while continuing to prevent the full realization of its sovereignty, chose the census. But they then imposed their own condition, namely that the funding be sufficient to cover not merely the nationality question, which was the only one Europe sought to resolve and the only one which Albanian ethno-politicians could use to legitimize their claims on the international scene, but also all those features of the Macedonian economy (e. g. agricultural property, land use, etc. ) that form part of a complete census but that had been omitted from the 1991 census due to insufficient funding in the context of economic crisis.

The 1994 census was in sum a statistical success but a political failure. Although it legitimated the basic statistics of the 1991 census, it did nothing to resolve the issues of political hegemony and access to resources that continue to plague Macedonia. However, it did help to reify a conflict whose roots in Macedonian history are not as deep as some political actors would pretend. In seeking to impose a vision of nationality that does not correspond to Macedonia's complex cultural context, political Europe reproduced its vision of Balkan "otherness" and marginality in Macedonia more than it contributed to its stabilization.

The continued presence of various international organizations and actors in Macedonia and the on-going tensions in the region raises the question of how the current Balkan Crisis can best be resolved. Hayden (1995) in his review of Woodward (1995) makesthe point that scholars specializing in Southeastern Europe -- one can add many of them trained with the help of Federal education grants designated specifically for the creation of a cadre area specialists -- were not consulted when US government policies concerning the former Yugoslavia were discussed and adopted. The activities of international organizations illustrated inthe foregoing exposition reveal a similar exclusion of regional expertise. If the future is to be different from the past, a closer cooperation between scholars and political actors is one way to promote the kind of understanding that could lead to stability.



back to top


Abduramanoski, Islam. 1994. Veštačka delba na Romite. Nova Makedonija 30 October, p. 7.

Alford, Henry. 1980. Alford's Greek Testament: An Exigetical and Critical Commentary, Vol I. Grand Rapids, MI: Baker [reprint of the seventh ed., 1874].

Anderson, J. D. 1913. The Peoples of India. Cambridge: Cambridge University.

Andonovski, Hristo. 1985. Makedonističko dviženje vo Kostursko. Makedonija 22 (387), 26.

Antonovska, Svetlana, et al. 1991. Broj i struktura na naselenieto vo Republika Makedonija po opštini i nacionalna pripadnost: Sostojba 31. 03. 1991 godina. Skopje: Republički zavod za statistika.

Antonovska Svetlana. 1994a. Statistički godišnik na Republika Makedonija 1993. Skopje: Republički zavod za statistika.

Antonovska Svetlana. 1994b. Popis '94: Podatoci za segasšnosta i idninata. Prvi rezultati. Soopštenie 1, Soopsštenie 2.Skopje: Zavod za statistika

Belić, Aleksandar. 1919. La Macedoine. Paris: Bloud & Gay.

Bercovici, Conrad. 1932. The Incredible Balkans. New York: G. P. Putnam.

Brown, Keith. 1995. Of Meanings and Memories: The National Imagination in Macedonia. Ph. D. dissertation. University of Chicago.

Brown, Keith. 1996. Friction in the Archives: Nations and Negotiations on Ellis Island, 1904. Paper presented at the Tenth International Conference of Europeanists. Chicago, March 1996.

Christides, Christopher J. 1949. The Macedonian Camouflage in the Light of Facts and Figures. Athens: Hellenic Publishing Co.

Clissold, Stephen. 1968. A Short History of Yugoslavia: From Early Times to 1966. Cambridge: Cambridge University Press.

Dako, Christo A. 1919. Albania: The Master Key to the Near East. Boston: Grimes.

Dimitrovski, Todor et al. 1978. Za makedonskiot jazik. Skopje: Institut za makedonski jazik.

Duijzings, Ger. 1992. De Egyptenaren in Kosovo en Macedoni'. Amsterdams Sociologisch Tijdschrift 18, 2. 24-38.

Unknown. Forthcoming. "The Making of the Egyptians in Kosovo and Macedonia". The Politics of Ethnic Consciousness. Ed. by Hans Vermeulen & Cora Govers. Basingstoke: Macmillan.

Durham, M. Edith. 1925. The Serajevo Crime. London: George Allen & Unwin.

d'Estournelles de Constant, Baron, et al. 1914. Report of the International Commission to Inquire into the Causes and Conduct of the Balkan Wars. Division of Intercourse and Education, Publication No. 4. Washington, DC: Carnegie Endowment for International Peace. (reprinted 1993 with an introduction by George Kennan).

Fraenkel, Eran. 1993. Urban Muslim Identity in Macedonia: The Interplay of Ottomanism and Multilingual Nationalism.Language Contact, Language Conflict, ed. by Eran Fraenkel and Christina Kramer, pp. 27-41. Peter Lang: New York.

Friedman, Victor A. 1985. Problems in the Codification of a Standard Romani Literary Language. Papers from the Fourth and Fifth Annual Meetings: Gypsy Lore Society, North American Chapter. 56‑75. New York: Gypsy Lore Society.

Unknown. 1993. Language Policy and Language Behavior in Macedonia: Background and Current Events. Language Contact, Language Conflict, ed. by Eran Fraenkel and Christina Kramer, 73-99. Peter Lang: New York.

Friedman, Victor A. and Ian Hancock. 1995. "On the Use of the Terms 'Rom' and 'Romani'. " Newsletter of the Gypsy Lore Society 18, 2. 6-7.

Gal, Susan. 1991. Bart-k's funeral: representations of Europe in Hungarian political rhetoric. American Ethnologist 18, 3. 440-458

Unknown. 1993. Diversity and contestation in linguistic ideologies: German speakers in Hungary. Language in Society 22. 337-59.

Glenny, Misha. 1995. The Birth of a Nation (review of Poulton 1995). The New York Review 16 November. 24-28.

Gourevitch, Philip. 1996. Misfortune Tellers. The New Yorker 8 April. 96-100.

Hadži-Ristik, Miodrag. 1994. Prašanjeto na etnogenezata na Egiptjanite vo Makedonija. in Nova Makeodnija 13-17 September pp. 16, 16, 15, 22, 9.

Hancock, Ian. 1995. A Grammar of Vlax Romani. Columbus: Slavica.

Haugen, Einar. 1968. The Scandinavian Languages as Cultural Artifacts. Language Problems of Developing Nations. Ed. by Joshua Fishman, Charles Ferguson and J. Das Gupta. pp. 267-84. New York: John wiley & Sons.

Hayden, Robert. 1995. Review of Woodward 1995. Slavic Review 54. 1114-15.

Human rights Watch/Helsinki. 1994. TheMacedonians of Greece. New York: Human rights Watch.

Ivić, Pavle. 1958. Die serbokroatischen Dialekte: Ihre Struktur und Entwicklung. The Hague: Mouton.

K'nčov, Vasil. 1900. Makedonija: Etnografija i statistika. Sofia: B'lgarsko knižovno družestvo.

Kaplan, Robert D. 1991. History's Cauldron. The Atlantic Monthly June. 92-104.

Konstantinov, Aleko. 1895. Baj Ganjo. Sofia: Daskalov.

Latifić, Ibrahim, et al. 1970. Vitalna, etnička i migraciona obeležja. Popis stanovništva 1961. Belgrade: Savezni zavod za statistiku.

Limanovski, Niyazi. 1993. Islamizacijata i etničkite promeni vo Makedonija. Skopje: Makedonska kniga.

Lunt, Horace. 1984. Some Socio-Linguistic Aspects of Macedonian and Bulgarian. Language and Literary Theory: In Honor of Ladislav Matejka, Ed. by B. Stolz, I. Titunik, and L. Doležel. (Papers in Slavic Philology 5), pp. 83-132. Ann Arbor: University of Michigan.

Lunt, Horace. 1986. On Macedonian Language and Nationalism. Slavic Review 45, 4. 729-34.

Meillet, A. 1918. Les langues dans l'Europe nouvelle. Paris: Payot.

Meyer, Gustav. 1891. Etymologisches Wљrterbuch der albanesischen Sprache. Strassburg: Trџbner.

MILS=Macedonian Information Liaison Service (News Service)

Misirkov, Krste P. 1903. Za makedonickite raboti. Sofija: Liberalni klub. (Reprinted 1974, Skopje: Institut za makedonski jazik, Posebni izdanija 8. )

Nicolaodes, Cleanthes. 1899. La Macedoine. Berlin: Stuhr.

Pekevski, Boro et al. 1973. Naselenie po narodnost vo SR Makedonija. Popis na naselenieto i stanovite, 1971 god. Skopje: Republički zavod za statistika.

Petruševski, Ilija (ed. ). 1992. Macedonia on Old Maps. Skopje: Detska radost. Pipinelis, P. N. 1963. Europe and the Albanian Question. Chicago: Argonaut.

Popovski, Tošo. 1981. Makedonskoto nacionalno malcinstvo vo Bugarija, Grcija i Albanija. Skopje: Makedonska kniga.

Poulton, Hugh. 1995. Who Are the Macedonians?Bloomington: Indiana University Press.

Pulevski, Gorgi. 1875. Rečnik od tri jezika. Belgrade: Drzžavna štamparija. (Slavic section republished in Blaze Ristovski (ed. ), 1974, Georgija M. Pulevski: Odbrani stranici, Skopje: Makedonska kniga. )

Rossos, Andrew. 1994. The British foreign Office and Macedonian National Identity 1981-41. Slavic Review 53. 369-94

Unknown. 1995. Macedonianism and Macedonian Nationalism on the Left. National Character and National Ideology in Interwar Eastern Europe. Rd. by Ivo Banac and Katherine Verdery, pp. 219-254. New Haven: Yale Center for International and Area Studies.

Sabota. Saturday supplement to the Macedonian-language newspaper Nova Makedonija (Skopje).

Saral, Osman Yavuz. 1975. Kaybettigщimiz Rumeli. Istanbul: Bogщazici.

Savezni Zavod za statistiku. 1954. Stanovnistvo po narodnosti (Knj. 9) : Konacni rezultati popisa stanovnistva od 15 marta 1948 godine. Belgrade: Savezni zavod za statistiku.

Savezni Zavod za statistiku. 1981. Statisticki bilten broj 1295 (1981). Popis stanovnistva, domacinstva i stanova u 1981 godini: Nacionalni sastav stanovnistva po opstinama. Belgrade: Savezni zavod za statistiku.

Savezni Zavod za statistiku. 1988. Statisticki Godisnjak Jugoslavije (Knj. 35)


Skendi, Stavro. 1968. The Albanian National Awakening: 1878-1912. Princeton: Princeton University.

Tanasković, Darko. 1992. The Planning of Turkish as a Minority Language in Yugoslavia. Language Planning in Yugoslavia.Ed. by Ranko Bugarski and Celia Hawkesworth, pp. 140-61. Columbus: Slavica.

Todorova, Maria. 1994. The Balkans: From Discovery to Invention. Slavic Review 53. 453-82.

Upward, Allen. 1908. The East End of Europe. London: John Murray.

Vaillant, A. 1938. Le probleme du slave macedonien. Bulletin de la Societe Linquistique de Paris 39 fasc. 2 no. 116. 194-210.

Verdery, Katherine. 1993. Ethnic relations, economies of shortage, and the transition in Eastern Europe. Socialism, Ideals, Ideologies and Local Practice. Ed by C. M. Hahn, pp. 172-186. London: Routledge.

Woodward, Susan L. 1995. Balkan Tragedy: Chaos and Dissolution After the Cold War. Washington, DC: Brookings Institution.

Xhaferi, Arb'r. 1994. Regjistrimi i jasht'zakonsh'm i popullsis' 1994. Tribuna-Sh2, 23/24. 20-21.



back to top


[i] In: Toward Comprehensive Peace in Southeastern Europe: Conflict Prevention in the South Balkans, ed. by Barnett Rubin. New York: Council on Foreign Relations/Twentieth Century Fund. 1996. 81‑105 & 119‑126. (revised, expanded, and updated version of 125, which was supposed to have appeared in 1995).
[ii]For an example of a Great Power view of the causes of World War One that bears striking resemblance to some modern attempts to lay the entire blame for the collapse of Yugoslavia at a single doorstep see Durham (1925). See also Todorova (1994:460) on recent attempts to resuscitate the myth of Balkan responsibility for World War One.
[iii]Throughout this paper, the term Macedonia will be used in both its geographical and political senses depending on the relevant time period. For the time period up to the Treaty of Bucharest in 1913, Macedonia refers to the geographic region defined by Mount Olympus, the Pindus range, Mount Šar, Mount Rila, the Western Rhodopes, and the river Mesta (Greek Nestos). After 1913, unmodified Macedonia will refer to the part of geographic Macedonia that was given to Serbia by the Treaty of Bucharest and which, with a few border adjustments resulting from subsequent treaties and administrative acts, ultimately became the Republic of Macedonia in 1991. Other parts of geographic Macedonia for the period subsequent to 1913 can be referred to with additional modifiers, e. g. Greek (Aegean) Macedonia, Bulgarian (Pirin) Macedonia, Golo Brdo and Lower Prespa (Albanian Macedonia).
[iv]Although the veracity of the 1989 Albanian census figures on minorities has been questioned, e. g. the census officially registered about 5, 000 Macedonians, but sources in Macedonia insisted the number is twenty to thirty times greater (Nova Makedonija 90/02/02), it is generally acknowledged that Albanians constitute the majority in Albania.
[v]For an objective and scholarly discussion of the linguistic issues by an American expert in Slavic linguistics, see Lunt (1984). Kaplan (1991:103), Poulton (1995:116), and Glenny (1995:24) all fail to grasp the significance of this type of equivocation. Thus, for example, Poulton (1995:116) privileges Bulgarian attempts to claim Macedonian as a dialect of Bulgarian by uncritically citing as a "comprehensive refutation" of Macedonian "as a language distinct from Bulgarian" a Bulgarian work first published as an article in the first number of the journal Baщlgarski ezik in 1978 that Lunt (1984:87-88) characterizes as "Incompetent in terms of linguistic theory, and resting on a poorly organized series of propositions and claims, many of them dubious, exaggerated or false". Lunt (1984:88) notes that: "Aside from a dignified answer by Macedonian linguists (Dimitrovski et al. [1978]), this embarrassing aberration from common sense and sound scholarship aroused little response in print. " Unfortunately, Poulton (1995) fails to cite that response. Similarly, Glenny (1995:24) writes: "Primarily, Bulgarian has a definite article and no case declension (unlike all other Slav languages until the Macedonians codified their Bulgarian dialect into a new language). " Aside from the fact that the sharing of two grammatical features is hardly justification for classifying one language as a dialect of another, in addition to which we should note that these two particular features are shared with the southern Serbian dialects, the fact remains that describing Macedonian as a Bulgarian dialect is like describing Norwegian as a Danish dialect (see Haugen 1968). See also Lunt (1986).
[vi]Although U.S. officials reportedly supported the extraordinary census in their private discussions with Macedonian leaders, there was no public U.S. support or participation. The census was essentially a "European" event taking place on Macedonian territory.
* The Bulgarian figures are from 1900, the Serbian from 1889, and the Greek from 1904. These are cited together in d'Estournelles de Constant (1914:28-30) and represent those of the three states that were indpendent, had territorial claims to Macedonia, and fought one another over those claims in the Second Balkan War in 1913. I have added a Turkish account of the 1905 Ottoman census figures for comparison (Saral 1975:152). The Greek figures omit the sandžak of Skopje (†skџp, vilayet of Kosovo [spuriously hellenized as Kossyphopedion in Nicolaodes 1899:25]), while the Bulgarian figures include the kaza of Tetovo (Kalkan-delen, sandžak of Prizren, vilayet of Kosovo ) and the kazas of Debar (Dibre-i b‰l‰) and Reka (R№kkalar/Zir Nanice) in the sandžak of Debar (vilayet of Bitola/Monastir). All the figures include the remaining kazas in the sandžaks of the vilayets of Salonika and Bitola/Monastir belonging to geographic Macedonia. For ease of comparison, I have added percentages. The selection from d'Estournelles de Constant (1914:28-30) was chosen because it is both typical of the discrepancies and because the republication of this report in 1993 has given it greater currency in the present situation. The category 'Other' includes "Wallachians" (i. e. Romance-speakers now known as Vlahs and including Arumanians and Megleno-Romanians), Jews, Gypsies (modern Roms or Roma), and "Miscellaneous" (Circassians, Armenians, etc. ). For a Greek view of the period that refers to other sources, see Christides (1949:32-33). See also Clissold (1968:136) for additional viewpoints. For figures relating to the post-war period, see Popovski (1981:187, 192-93, 247).
[vii]The mutability of the Western construction of "Near East" as a place of the "oriental" (in Said'ssense) is clearly seen in the fact that at the beginning of this century it could refer to Turkey in Europe, i. e. the Balkans (pace Todorova 1994:455).
[viii]It is significant that while claiming factors other than language as the determiners of nationality, every Greek government has prohibited education in the languages of Greek Orthodox Christians other than Greek. This is because language can function asthe determiner of nationality or as a pathway to altered ethnic self-identification, although it does not always do so. It is likewise significant that the Greek-speaking minority of Albania is never referred as "Hellenophone Albanians".
[ix]Similarities in folklore and folk traditions were also adduced to support Serbian claims.
[x]Also absent here are Albanian and Romanian figures. Although Albanians laid claim to the western half of Macedonia as far as a line running from Kumanovo through Veles, the population figures that I have found in Albanian sources of the period do not attempt any ethnic breakdown but rather imply that the regions in question are simply Albanian with the occasional "colony" from some other ethnic group (e. g. Dako 1919:5). It would appear that the limits of Greater Albanian territorial claims were obtained by drawing a line connected all the outlying Albanian-speaking villages of Montenegro, Serbia, Macedonia, and Epirus and then claiming all the territory inside those boundaries. In some respects, the Albanian situation at the turn of the century resembled the Macedonian: Albanianclaims were marginalized due to their lack of any Great Power support. Moreover, Turkey, which in this case functioned like other small powers such as Greece, Bulgaria and Serbia, attempted to render the Albanians invisible, by claiming them as Turks on the basis of religion (cf. Saral 1975 cited in Table One). In this they would be supported by Greece, which claimed all of Southern Albania on the basis of that part of the population that was Orthodox Christian. This left the Catholics of Northern Albania, which Serbia and Montenegro were ready to absorb. Unlike the Macedonians, however, who, as Slavs speaking dialects midway between Serbian and Bulgarian, could be plausibly assimilated to one or the other despite the fact that they were arguably neither (cf. Vaillant 1938 cited in the explanation of Table Two), the dialects of Albanian could not be linguistically claimed by any neighbors and were therefore instead discredited as being merely a Mischsprache, a mixture of all the languages of the Balkans with almost no indigenous elements (cf. Meyer 1891:ix). To this was added the claim that Albanians had no clearly defined national consciousness (Pipinelis 1963:32) and that the North and South (Geg and Tosk) Albanian dialects were so different as to be incapable of uniting into a single language (cf. Meillet 1918:255). Thus, aside from the use of religion as a means of denying legitimacy to both Macedonian and Albanian ethnic identity, both groups were also linguistically marginalized. In the case of Macedonians, they were told that they spoke either Serbian or Bulgarian, while in the case of Albanians, they were told that their language was not really a language at all. (The Greeks had tried this type of strategy on the Bulgarians at an earlier period, cf. Nicolaodes 1899:60, 120). The principle difference was that Albanians had sufficient national consciousness and political organization to lay claim successfully to a part of the disintegrating Ottoman Empire in which they constituted a majority while the Macedonians were left completely out of these processes. Although the Romanian government showed some interest in supporting Vlah claims, especially after the establishment of a Vlah Church (and therefore millet) in 1905, the Romanians were too far away, and the Vlahs too few in number and not particularly strong in national consciousness. Those in the towns were mainly Greek-identified merchants while the transhumant shepherds of the countryside were hardly in a position to organize (Cf. Brown 1995).
[xi]Upward (1908:204) writes of an interview he conducted in a village two hours ride from Voden: "I asked what language they spoke and my Greek interpreter carelessly rendered the answer Bulgare. The man himself had said makedonski! I drew attention to this word, and the witness explained that he did not consider the rural dialect used in Macedonia the same as Bulgarian, and refused to call it by that name. It was Macedonian, a word to which he gave the Slave [sic] form of Makedonski. " On the next page he continues: "The Exarchist claimed that his party had sixty or seventy houses in the village; the Patriarchist had awarded him fifteen or twenty. "
[xii]See Kaщnčev (1900:210-214) for figures from the end of the last century.
[xiii]A PDP campaign poster from the 1994 Macedonian elections gives symbolic representation to the PDP attempt to appeal to its Albanian constituency and the European actors that have played such a significant role in supporting its claims. The poster's slogan is in Albanian, English, and French. The poster thus erases both Macedonian and the other minority languages of Macedonia, while attempting to portray the party as oriented towards "Europe" (cf. the discussion of the termEurope at the end of this article).
[xiv]The large variations in percentages are due not only to mechanical shifts (immigration and emigration) but also to socio-politically motivated changes in choices in declaring identity (cf. Gal 1993, Tanasković 1992:143). The history and specifics of these phenomena are beyond the scope of this paper and deserve separate monographic treatment. My purpose in calling attention to them here is to emphasize the fact that shifts in demography and shifts in identity are independent phenomena.
[xv]"Egyptians", in Macedonian GдupciEg"upci or Egipk"ani (also Gдuptin) are descended of Roms but do not speak Romani and do not identify themselves as Roms. The ethnonyms themselves are cognate with English Gypsy, which is derived from the claim or belief that the Romani people came fromEgypt. The majority of Egipk"ani live in Ohrid and Struga and speak Albanian, those of Bitola speak Macedonian. Some Egipk"ani attempted to register as a separate ethnic group in the 1981 census, but they were listed as "unknown" (Sabota 920306). The Egipk"ani claim to be descended from Egyptians, but there is no concrete evidence to support this claim. A more likely explanation is that they became sedentarized at a very early date and assimilated linguistically but not ethnically to non-Romani speakers and maintained ethnic separateness from Romani-speakers whether nomadic or sedentarized at a later date (see Friedman 1985, Duijzings 1992, Duijzings Forthcoming, Hancock 1995 cf. also Hadži-Ristik" 1994 and Abduramanoski 1994). The categoryBosniac refers to Serbo-Croatian-speaking Muslims.
[xvi]Niyazi Limanovski, head of the organization of Macedonian Muslims, came out against census. Bekir Zhuta, a PDP minister in parliament and Albanian from Struga, stated that there was no need for help from outside. Both were accused by radical Albanians of trying to divide and conquer, although Zhuta did serve as his party's spokesman in declaring the census invalid after the preliminary results were published. (MILS 94/04/13, MILS 94/11/18).
[xvii]Article 7 of the Macedonian constitution reads: "The Macedonian language, written using its Cyrillic alphabet, is the official language in the Republic of Macedonia. In the units of local self-government where the majority of the inhabitants belong to a nationality, in addition to the Macedonian language and Cyrillic alphabet, their language and alphabet are also in official use, in a manner determined by law. In the units of local self-government where there is a considerable number of inhabitants belonging to a nationality, their language and alphabet are also in official use, in addition to the Macedonian language and Cyrillic alphabet, under conditions and in a manner determined by law. " The nationalities named are Albanians, Turks, Vlahs and Roms in that order. Serbs object to the fact that they are not specifically named but are subsumed under the expression "other nationalities". When the census law was being debated, Macedonian nationalists argued that the forms should only be in Macedonian, since the census was being conducted at the national and not the local level. In the end, however, bilingual forms were used for all five principal minority languages (but see the discussion later in this paper). The census also had the effect of legitimizing certain directions in the codification of Literary Romani -- a process that has been on-going since the mid-seventies in Macedonia (see Friedman 1985) -- and of Literary Arumanian, which has gained significant momentum since 1989. It is interesting to note that the Turkish forms were printed without the diacritics of Turkish orthography. While this would not hinder comprehension, it underlines a certain lack of sensitivity that was apparently not crucial. To the best of my knowledge, no objections were raised by any parties concerning this point.
[xviii]See Todorova (1994:481-82).
[xix]The two exceptions were Greek historians from the Institute of Balkan Studies at Thessaloniki. According to information supplied to me by a member of the Group of Experts, there was originally one Greek appointed to the observation team, but the weekend before the team left for Macedonia, while the chief of the Statistical Bureau of the Council of Europe was away, the vice-chief, who is Greek, added another Greek to the team without the required approval of his superior. The Macedonian government granted both Greeks visas without demure. These facts are significant because at the time Macedonia was under a Greek economic blockade, and holders of Macedonian passports could not enter Greece. Moreover, as indicated above, the official Greek position denies the existence of a Macedonian minority on its own territory (Human rights Watch/Helsinki 1994:11). ICOM thus placed the Macedonian government in the position of granting privileges to citizens of a country that refused to do the same for Macedonians.
[xx] Significant lack of objectivity, however, was ascribed by Macedonian officials to a high-ranking member of the ICOM team who bore significant responsibility for organizing the Mission's work. She was perceived as being particularly difficult -- even nasty. It is a measure of the tensions inherent in the situation that Macedonian officials speculated that this individual's peculiar behaviorwas due to her being of Albanian descent, a rumor which I was unable to verify. The behavior of this individual that I observed myself led me to understand why Macedonian officials regarded her as a source or at least an active supporter of the adversarial attitude described later in this paper. In addition to witnessing her tense, perhaps even hostile, behavior myself, I was told by other members of the ICOM team that they perceived her reactions to some situations as inappropriate. This behavior, however, could have been a reflection of the tensions underlying the census itself or idiosyncrasies of the individual's personality rather than an emotional attitude resulting from her ethnic background.
[xxi]There are also two other Russian letters ( and ') that do not occur in Macedonian.
[xxii]In addition to the household forms filled out by the enumerators on the basis of answers given by those being censused, both enumerators and ICOM observers filled out supplementary forms for the purposes of statistical verification and enumerating household members. The supplementary forms were called control forms.
[xxiii]Cf. Gal (1993:345) on abuses of the 1941 Hungarian census during the post-war period, when the 1941 census data, which was supposed to be confidential, was used to deport not only Germans in Hungary who had openly worked for the Nazis, but also anyone who had claimed German nationality and later anyone who had claimed German as their mother tongue.
[xxiv]There is a certain irony in the fact that some member countries of the EU, e. g. Germany, have citizenship laws that are considerably more exclusionary than the Macedonian one.
[xxv]That same day an article appeared in Nova Makedonija (95/06/22) presenting this question as an assertion, i. e. that international attention had exacerbated ethnic tension. It should be noted that although the European Union and the Economic Commission for Europe were involved in the census, the chief actors were ICFY, which proposed it, and the CE, which funded it.
[xxvi] The polemic over figures also continued during the course of the census itself. An article in the Skopje-based Albanian-language daily newspaper Flaka e v'llaz'rimit (94/07/11) entitled "Over 8, 000 Albanians" reported that one of the municipal instructors in Bitola claimed this figure for that municipality. "Any other figure connected with the number of Albanians given by the Bureau of Statistics will be declared a falsification... I have so informed all communal, republic and international bodies" (my translation). This figurecontinued to be cited in the Albanian-language media despite the fact that the data had not yet been processed. The official preliminary figure for Albanians in municipality of Bitola was 3, 970 (Antonovska et al. 1994b).
[xxvii]Historical evidence supports the latter, not the former, contention, cf. Limanovski (1993).
[xxviii]The precipitous banning of the Muslim veil (zar and feredže ) in the early 1950's also created significant alienation of some Macedonian-speaking Muslims from the state and increased their sense that Macedonian identity was a Christian identity. The current dispute is thus not a new one (see Friedman 1993:88-89). However, there are also Macedonian Muslim organizations actively encouraging Macedonian Muslim identification with Macedonian language and ethnicity.
[xxix]Gal (1993:344-45) discusses how the 1873 Statistical Congress insistence on the requirement of a language question in censuses forced people in Austria-Hungary to think of nationality in terms of language instead of citizenship. Then in the 1941 Hungarian census, a separate nationality question was added, which was even more overtly politically manipulated.
[xxx]An example of how "national" feelings in Macedonia are more complex than portrayed by ethno-political (or international) organizations is the Serbian husband of a Macedonian who, unbeknownst to his wife, had always declared himself a Yugoslav in national censuses. He went through a great personal crisis in connection with the 1994 census, since he did not wish to identify with current Serbian policies and did not consider Yugoslav to be a valid category any longer. He chose to declare himself as a Macedonian.
[xxxi]The problem of identity in relation to both the census and Macedonian foreign relations received the following expression in the satirical journal Osten (95/07/08) : "I'm going to declare myself as a Martian in the census. " "Can you speak Martian?" "No need to. Here you can declare yourself a Turk or Albanian without knowing Turkish or Albanian. It's enough to be a Muslim. And as a Martian, no one can negate me. The Martians were never Greeks or involved in a Serbo-Bulgarian quarrel. " "I dig it, buddy. You'll go far in life" (my translation).
[xxxii]Todorova (p. c. 95/03/15) has corrected this in her subsequent work.
[xxxiii]Since no one in the Mission speaks Turkish or Macedonian, it is not clear how they could judge the children's ability.
[xxxiv]A Greek colleague of mine who was born in 1950 told me that when he was growing up, people routinely referred to Western Europe as "Europe", e. g. '"I'm going to Europe for my vacation", i. e. as if Greece were not part of that entity.
[xxxv]The image of political Europe as a political and/or cultural entity that represents a desired or rejected "other" has been and is widespread in the economically peripheral nations of geographic Europe. See Gal (1991) for a Hungarian example of the same types of attitudes with additional references.