Source of Interpretration of Census Figures
Ethnic group
Bulgarian Serbian Greek Turkish
Bulgarians 52.3% 2.0% 19.3% 30.8%
Serbians 0.0 71.4 0.0 3.4
Greeks 10.1 7.0 37.9 10.6
Albanians 5.7 5.8 0.0 0.0
Turks 22.1 8.1 36.8 51.8
Others 9.7 5.9 6.1 3.4
Total 100 100 100 100
Bulgarian, Serbian, and Turkish authors all found that their own
ethnic group was in a majority in Macedonia, while the Greek author
claimed a plurality of Greeks.
One is tempted to impute the huge variations (from 71 percent of Serbs
to none!) to willful fabrication, the accusation of choice found in
all nationalist literature.
"Fabrication," however, while certainly a factor, largely misses the
point, since it assumes, in the tradition of statistical realism, a
correct and objective method of counting identities, whose process is
then spoiled by political elements.
Cultural identities in Macedonia were complex, with much of the
population multilingual, religion (Orthodoxy, Catholicism, Islam)
cross-cutting languages, three Patriarchates (Serbian, Bulgarian,
Greek) contending
for the loyalty of Orthodox believers, and much of the population
having a weak "national" consciousness, in the modern sense of the
word.
All four competing powers rejected the existence of a "Macedonian"
ethnic identity, which explains its absence from the Table, even
though archival sources attest that a growing number of people were
beginning to define themselves as such at the time (Brown 1996).
The argument of the Bulgarian and Serbian nationalists was that the
Slavic language spoken in Macedonian was a dialectical version of
their own language, and therefore Slav-speakers were Bulgarian or
Serbian, respectively.
The Greek position used language as well, but less the vernacular than
the language learned at school. Conveniently (if not coincidentally),
Greek schools were prevalent in the area, since Greek was the language
of prestige and commerce throughout the Balkans.
Greek nationalists also used church affiliation as a criterion,
counting all those pledging allegiance to the Greek Orthodox
Patriarchate as theirs, and not recognizing the rival Bulgarian
Orthodox Church (the Exarchate). The Albanians, meanwhile, were either
a minor or non-existent category in these population figures, being
counted as part of the group of each author according to their assumed
religious affiliation (Albanian-speaking populations being of Muslim,
Orthodox and Catholic backgrounds).
The "Turks" who constituted a majority in the Turkish source included
a number of Albanian-speakers.
The figures cited by Friedman were computed from censuses conducted in
Ottoman Macedonia, either by the state or churches, where a direct
question on nationality (ethnicity) was not asked.
The original data, on religion and language, was collected under
techniques hardly statistically valid, since the practice was to
extrapolate from the numbers of males in households and the counting
itself was capricious (Van Gennep 1992: 120).
Moreover, the different authors interpreting the raw data had
different conceptions of the geographical boundaries of Macedonia,
which probably accounts for some of the important discrepancies in the
estimates of "Turks." However scientifically inadequate these
numerical exercises were from a technical standpoint, the various and
conflicting criteria used to define the categories in the Macedonia
case are not the exception,
but the norm of census politics, if perhaps a bit extreme.
.....................
"From Census and Identity: The Politics of Race,
Ethnicity, and Language in National Censuses
Chapter One
Censuses, Identity Formation, and the Struggle for Political Power
David I. Kertzer and Dominique Arel