international journal of the sociology of language volume 2012, issue 217 (sep 2012)

2012 · Number 217

Issue edItorsJaffer SheyholislamiSchool of Linguistics and Language Studies Carleton University 215 PA – 1125 Colonel By Dr. Ottawa, ON., K1S5B6 Canada E-mail: [email protected]

Amir HassanpourNear & Middle Eastern Civilisations University of Toronto Room 318, 4 Bancroft Avenue Toronto, ON M5A 4L8 CanadaE-mail: [email protected]

Tove Skutnabb-KangasTronninge Mose 3 4420 Regstrup Denmark E-mail: [email protected]

GeNeral edItorJoshua A. FishmanNew York University E-mail: [email protected] Snail mail address: 3616 Henry Hudson Parkway Apt. 7B-N, Bronx, NY 10463 USA Fax: +1 718-796-8155

assocIate GeNeral edItorOfelia GarcíaThe Graduate Center of the City University of New York 365 Fifth Avenue New York, NY 10016-4309 USA E-mail: [email protected]

‘sINGles’ edItorFlorian CoulmasDeutsches Institut f. Japanstudien Jochi Kioizaka Bld. 2F 7-1 Kioicho Chiyoda-ku, Tokyo 102-0094 Japan E-mail: [email protected]

slslc edItorEmily McEwan-Fujita2 Graham Street Dartmouth NS B3A 3H7 Canada E-mail: [email protected]

INterNatIoNal JourNal of the socIoloGy of laNGuaGethe KurdIsh lINGuIstIc laNdscape: VItalIty, lINGuIcIde aNd resIstaNce

UnauthenticatedDownload Date | 2/25/16 5:48 PM

IJSL 2012 | Number 217

Contents

Amir Hassanpour, Jaffer Sheyholislami and Tove Skutnabb-KangasIntroduction. Kurdish: Linguicide, resistance and hope 1

Jaffer SheyholislamiKurdish in Iran: A case of restricted and controlled tolerance 19

Amir HassanpourThe indivisibility of the nation and its linguistic divisions 49

Desmond FernandesModernity and the linguistic genocide of Kurds in Turkey 75

Welat ZeydanlioğluTurkey’s Kurdish language policy 99

Uğur Ümit ÜngörUntying the tongue-tied: Ethnocide and language politics 127

Ergin ÖpenginSociolinguistic situation of Kurdish in Turkey: Sociopolitical factors and language use patterns 151

Tove Skutnabb-Kangas, Jaffer Sheyholislami and Amir HassanpourConcluding remarks 181

Book review

Amir HassanpourPolitics and language ideology in Kurdish lexicography The Azadi English-Kurdish dictionary/ , by Rashid Karadaghi 189

Farhange Dânešgâhe Kordestân. . Kurdistan University Persian-Kurdish dictionary, by Mâĵed Mardux Ruhâni 189


x Contents

Small languages and small language communities 72

Jim HlavacContinuing and shifting multilingualism in an émigré situation: Language use and at titudes amongst Iraqi Chaldeans and Assyrians in Melbourne 195


DOI 10.1515/ijsl-2012-0047 IJSL 2012; 217: 1 – 18

Amir Hassanpour, Jaffer Sheyholislami and Tove Skutnabb-Kangas

IntroductionKurdish: Linguicide, resistance and hope

Amir Hassanpour: University of Toronto. E-mail: [email protected] Sheyholislami: Carleton University. E-mail: [email protected] Skutnabb-Kangas: Åbo Akademi University. E-mail: [email protected]

Finally Kurdish has made it to the pages of the International Journal of the Sociol-ogy of Language.1 This is an event in the history of Kurdish language studies. There is no article about the language in thirty-six years of publishing since 1974.2 And IJSL is not alone in its omission of Kurdish. In fact, this is the first time in the West that a whole issue of a linguistics journal is devoted to its study.

If we move from journals to books, the picture does not change. Kurdish is visibly missing in the growing literature on the sociology of language and socio-linguistics even though in recent years research on the politics of the language (e.g., Hassanpour 2000; Olson 2009), its linguicide in Turkey (e.g., Hassanpour 1993; Skutnabb-Kangas and Fernandez 2008), Iran and Syria (Hassanpour, Skutnabb-Kangas and Chyet 1996), its struggles over standardization and official-ization (e.g., Hassanpour 1992), its use in the media (e.g., Sheyholislami 2010), its general description (e.g., Kreyenbroek 1992) or its gendered lexical heritage (e.g., Hassanpour 2005) has been published in books, dissertations and disparate jour-nal articles. In a “sociology of language” approach to the study of Kurdish, we may address questions about the precarious life of this language which many phi-lologists and linguists have ignored, quite often deliberately (Hassanpour 2000). The story of the preparation of this special issue of IJSL brings to light significant

1 In writing this Introduction, we have drawn on our ongoing research on the language including Hassanpour (1992, 2011), Sheyholislami (2010, 2011, 2012, forthcoming) and Skutnabb-Kangas (2000, 2008).2 There are special issues on, for instance, Berber (IJSL 123), Slovene (IJSL 124), Macedonian (IJSL 131), Estonian (IJSL 139), Serbian (IJSL 151), Quechua (IJSL 165) and on the sociolinguistic landscapes of Turkey (IJSL 165), Iran (IJSL 148) and Arab countries.

Brought to you by | Rice UniversityAuthenticated

Download Date | 5/20/15 12:01 PM

2 A. Hassanpour et al.

social facts about the position of the language both in the evolving world linguis-tic order and within the Kurdish speech community. We are confident that there is no conspiracy to ignore Kurdish in either IJSL or hundreds of journals devoted to the study of language. It is, in fact, the absence of conspiracies that is of socio-logical interest. We intend to reflect, here, on the social life of the language, its troubled history, and the intricacies of undertaking research on it.

In terms of the number of speakers, Kurdish ranks fortieth among the world’s 6,600 to 7,000 languages.3 The numerical strength of the language has, however, been undermined by the division of its speech area and speakers among five neighboring countries of Turkey, Iran, Iraq, Syria and Armenia, and the adoption, by these states, of policies ranging from deliberate killing of the language or lin-guicide (Turkey since 1925, Iran, especially during 1925–1941, Syria since the mid-1960s) to tolerance (Syria in the mid-1930s and World War II to 1958) and official-ization (locally in the USSR, 1921–1992 and Iraq, 1918–1991, and on the national level in Iraq since 2005). Under this heterogeneous, uneven geopolitical division of power, Kurdish is now one of the two official languages of Iraq while it is de-nied many rights including mother tongue medium education in all neighboring countries.

The Kurdish speech area is in Western Asia, now comprising part of what is known as the “Middle East”. Kurdistan, ‘land of the Kurds’, comprises part of the region, southern Iraq and Mesopotamia, where writing in its alphabetical form was invented some six millennia ago. The archeological record including inscrip-tions in the mountains of Kurdistan are full of written texts in diverse languages which are, with a few exceptions, extinct. It is not well known how Kurdish sur-vived in a region which is both a mosaic of ancient written languages and their burying ground. Having co-existed with the rich literary tradition of Arabic, Per-sian, Syriac, Armenian, Ottoman Turkish and spoken languages such as Neo-Aramaic, Kurdish had a rather late beginning in writing. The earliest known evi-dence dates back to the sixteenth century when two dialects, Kurmanji and Hewrami, began a literary tradition, predominantly in poetic form. Later in the early nineteenth century, another dialect, better known as Sorani (Central Kurd-ish) since the 1960s, developed its written tradition, followed by occasional writ-ing in other dialects.

The three literary traditions were largely poetic with only a few prose works, which were mostly non-narrative. This literary spark in the mountains of Kurdis-tan, much like that in Azeri, Pashtu or Baluchi languages, was overshadowed

3 This ranking is based on an estimation of the number of speakers at 20 million in the early 1980s (Leclerc 1986: 55, 138). According to one calculation (Krauss 1992: 7), the median number of speakers of a language is about 5,000 to 6,000.



Introduction 3

by the brilliant and rich literary traditions of Arabic and Persian, the dominant classical languages from northwest Africa to Central Asia and West India. While literary Kurdish was born and survived under the hegemony of these literary languages, it suffered immensely when they became the official languages of the modern states of Iran, Iraq, Syria and Turkey. If in pre-modern times, these languages inspired the Kurds in launching their literary tradition, the modern states imposed them on Kurdish speakers in linguicidal projects aimed at creat-ing Turkish, Persian (Iranian) and Arab (nation-)states (see the articles in this issue).

Living under the shadow of Arabic and Persian, both associated with Islam, literary Kurdish was the vehicle of two autochthonous religions of Kurdistan known, in the West, as Yezidism (Mesĥefa Reş [Black Book] and Kitêba Cilwe [Book of Revelation] in Kurmanji) and Ahl-e Haqq (various texts in Hewrami).4

Kurdish has been written in a variety of alphabets, including Armenian, Ara-bic, Cyrillic, Roman and Syriac. The first printed translations of the Bible into Kurmanji were published in the Armenian alphabet.5 Kurmanji texts have also been written in the Syriac alphabet (see, for instance, Fuad [1970: 121–123]). This diversity reflects the complex linguistic and literary life of West Asia,6 as well as the post World War I division of Kurdistan among states ranging from monarchi-cal Iraq to socialist USSR, where the choice of alphabets was primarily a political event decided by the government.

The Kurdish speech area (c.f. Figure 1) has experienced many divisions, the more permanent one being the border between the Ottoman and Iranian states in 1639. By the mid-nineteenth century, a small enclave of Kurdish population came under Russian rule. While this border has survived until now (forming Iran-Turkey and Iran-Iraq borders), the Ottoman Empire was dismantled by the end of World War I, and Britain and France created a number of states out of the prov-inces they occupied. The Ottoman part of Kurdistan was re-divided among Iraq (under British Occupation, 1918–1920, and Mandate, 1920–1932), Syria (under French Occupation, 1918–1920, and Mandate, 1920–1946) while the rest remained under Ottoman rule until 1923 when the Turkish nationalist leader Kemal Atatürk abolished the Ottoman regime and replaced it by the Republic of Turkey. The

4 There is a controversy on whether Yezidism is rooted in a written or oral tradition; it is argued, more recently, that the written texts date back to more recent times, e.g., nineteenth century. See, for instance, Kreyenbroek (1995: 1–25).5 This was a translation of the Gospel of Mathew published in 1856 in Constantinople.6 There has been little research interest in comparative studies of the literary and oral traditions of the peoples of West Asia – Arabs, Armenians, Assyrians, Jews, Kurds, Persians, Turks and others.




Fig.

1: M

ajor

Kur

dish

dia

lect

gro

ups

(Has

sanp

our 1

992:

22)



Introduction 5

Kurds of Caucasus became part of the USSR, after Soviet power was extended to Caucasia in 1921.

These divisions have re-shaped the political weight of the dialects and their course of development. Kurmanji (also called Northern Kurmanji or Northern Kurdish) is now spoken by the majority of the Kurds in all countries where Kurd-ish is spoken. Sorani (also called Southern Kurmanji or Central Kurdish) is spo-ken in Iran and Iraq. The third dialect group, variously identified as Southern Kurdish and Kermashani, is spoken primarily in Iran but some of the dialects are also spoken in Iraq. The fourth group consists of Hewrami7 or, in European phi-lology, Gorani (Iran and Iraq) and Zaza or Dimilki (Turkey). Each of these dialects consists of a group of subdialects.

There is no consensus on the nature or significance of these dialectal varia-tions. While the Kurds themselves have not doubted the Kurdishness of the four dialect groups, some Western philologists in the late nineteenth century argued that the Hewrami/Zaza group constitutes a non-Kurdish language (MacKenzie 1961, 1966). More recently, some linguists have made the same claim regarding Kurmanji and Sorani assigning each the status of autonomous languages. These claims are based primarily on genealogical classifications made on the basis of a few phonetic features to the full disregard of the sociological fact of speakers identifying themselves as Kurds and their language as Kurdish (Hassanpour 1998). Equally significant from a sociological perspective is the emergence of a group of Zaza speakers since the 1ate 1980s who claim a non-Kurdish ethnic and linguistic identity. In 2006, a group of Hewrami speakers submitted a petition, endorsed by about 500 signatures, to the parliament of the Kurdistan Regional Government (Iraqi Kurdistan) and demanded to be recognized as a “language mi-nority”; they emphasized, however, that they considered their language and eth-nicity to be “Kurdish” (Sheyholislami 2008). The rift between native speakers and (some) linguists, as well as the changing identities of some Zaza speakers, high-light the limitations of a purely linguistic approach to language which is both a social and linguistic formation.

Writing in Kurdish and its subsequent literary development began under con-ditions of a flourishing feudal order in the sixteenth century.8 The great majority of Kurds lived in rural societies, both tribal and feudal, with a small but signifi-cant urban population. And most of the rural population, the peasantry, was tied to the land in a mode of production similar to serfdom in European societies. This

7 While Kurds and Hewrami speakers call this dialect and its variants “Hewrami”, Western philologists classify Hewrami as a dialect of Gorani (see, e.g., MacKenzie 1966: 4).8 For instance, two Ottoman sultans, one in 1485 and the other in 1515 banned the use of printing in the Arabic alphabet (Oman 1991: 795).




socio-economic system was conducive to fragmentation, diversity of dialects, and plurality of cultures and literary traditions.

The feudal order of Kurdistan had its elaborate system of principalities, i.e., mini-states which ruled over much of Kurdistan; some were independent small dynasties while others were nominally dependent on either the Ottoman or Ira-nian monarchs. The emergence of literary Kurdish dialects is historically associ-ated with the rise of Kurdish political power in the fifteenth and sixteenth centu-ries. This literature was born in the mosque schools and in the courts of Kurdish princes and feudal lords.

Throughout much of its written history, Kurdish literature was scribal, hand-written on paper. The Ottoman and Iranian monarchs were not interested in the use and diffusion of printing when this technology began spreading from Europe to the East in the sixteenth century. Paper and ink and even limited wood-block printing, devised in China and Korea, had made incursions into West Asia long before Gutenberg, but the first printed books in Arabic letters came from the West.

The transition from Kurdish scribal to print culture began in 1898 with the publication of Kurdistan, the first Kurdish newspaper in Cairo. The paper, like the ones that followed it from 1909 to 1923, was predominantly in the Kurmanji dialect.

World War I (1914–1918) and the dismantling of the Ottoman Empire changed the linguistic landscape of Kurdistan. The Ottoman state’s genocide of the Arme-nian and Assyrian peoples in 1915–1923 virtually eliminated the Armenian and Neo-Aramaic languages from northern parts of Kurdistan where Armenians, Assyrians and Kurds had lived for centuries. After the war, the migration of the sizeable Jewish community of north-western Iraqi Kurdistan removed their Neo-Aramaic language and culture from the Kurdish speech area. Although the physi-cal destruction of the Kurds in Turkey was much more limited than that of Arme-nians and Assyrians, they too were subjected to a harsh policy of linguicide and ethnocide. Under these conditions, Kurdish turned into a site of inter-state polit-ical conflict (Hassanpour 1993).

The 1918 re-division of the Kurdish speech area also changed the destinies of the dialects. Kurmanji’s superior position came to an end with its division among five newly forming states and its violent suppression in the Republic of Turkey, which had the largest number of speakers and a flourishing modern-style intelli-gentsia. While the repression of Kurdish can be traced back to the late Ottoman period, republican Turkey pursued a policy of linguicide (see Skutnabb-Kangas and Dunbar 2010), the deliberate killing of the language, after the suppression of a Kurdish revolt in 1925 (see articles by Fernandes; Zeydanlıoğlu; Üngör, this issue). In neighboring Syria, the French Mandatory power tolerated speaking in the language and, for a few years before and during World War II, allowed Kurd-



Introduction 7

ish publishing but rejected demands for mother-tongue medium education. In Iran, too, official policy under Reza Shah was deliberate killing of the language (see Sheyholislami, this issue).

The policy of linguicide in Turkey and Iran did not simply affect the speakers of Kurdish in these countries. Western powers, in alliance with Turkey and Iran and against the USSR, kept silent about the policy of linguicide and ethnocide and, in fact, endorsed it. While Britain and France, whose Mandatory rule over Iraq and Syria was recognized in 1920 by the League of Nations, had pledged to protect the rights of their “Kurdish minority”, they supported the policy of re-stricting their language rights by all means possible. This was in spite of the fact that the League of Nations had, as early as 1925, committed itself to make Kurdish an official local language in Iraq (Question of the frontier between Turkey and Iraq: 89).

Kurdish leaders in Iraq protested the League for failing to pressure Britain-Iraq into complying with their pledges regarding language rights. The British dip-lomatic correspondence of the period, 1920s to the 1950s, provides a detailed documentation of the concerns of the UK about Kurdish nationalism and how granting any “concessions” to this nationalism may play into the hands of “com-munists”. Moreover, France and Britain were determined to prevent the “spill-over” of this nationalism into Turkey and Iran. Many Western academics and mainstream media pursued, in unison with their states, a similar line. After the war, the state, mainstream media, and many in academia, including linguists who studied Kurdish, endorsed the criminalization of the language in Turkey, Iran and Syria in spite of the fact that the suppression of any language violated, among others, the Charter of the United Nations (paragraphs 6.11, 55), UN Univer-sal Declaration of Human Rights (paragraphs 2, 26), and the International Cove-nant on Economic and Political Rights (article 27).9 Linguists and linguistics had not yet taken linguicide seriously. Even when the conceptual and theoretical frameworks for problematizing language killing and language death began to be worked out in the 1960s and 1970s, many linguists specializing in Kurdish re-mained silent (Hassanpour 2000). This was in spite of the fact that the criminal-ization of the language came also with a ban on field work by linguists and other researchers, both native and non-native, especially in Turkey, Iran and Syria. While Western politicians endorsed the project of linguicide, linguists who stud-ied the language decided to remain a-political and build a high wall between lan-guage and politics.

9 For more information on Turkey’s violation of international law see Skutnabb-Kangas and Bucak (1994: 347–370). For information and analysis of the limitations of Turkey’s recent language reforms, see Dunbar and McKay (2002).




It is, thus, not difficult to see why Kurdish was rarely taught in Europe and North America. Unlike Arabic, Persian or Turkish, it was not the language of any state. Moreover, it had been criminalized except in Iraq and the USSR. Lack of academic interest meant lack of publishing interest. Academic libraries cannot, under the circumstances, build a collection adequate for research and teaching of the language. All of this amounts to a dearth of faculty members specializing in the language, lack of teaching material, limited supervision of student research, and little funding of field work. Quite often degree programs in Middle Eastern languages are supported by government funding and, sometimes, support from Middle Eastern governments.

In 1958, the pro-Western monarchy in Iraq was overthrown by nationalist of-ficers, an event which the US and its allies considered a shift to communism. The US, interested in the potential of Kurdish nationalism to undermine the new Iraqi republican regime, declared Kurdish a “strategic language”, which allowed fund-ing the development of textbooks and its teaching for the purpose of familiarizing students and government personnel with the language.10 Even then, both govern-ments and academics continued to support Turkey’s linguicidal policy and en-dorse its claim that there was no Kurdish language and no “Kurdish problem” because the Kurds had been assimilated once and for all.11

The formation of the Iraqi state under the British Mandate together with the suppression of the language in Turkey changed the linguistic landscape of Kurd-istan. Britain allowed the limited use of Kurdish in the media, elementary educa-tion, and local administration. Sorani was the dialect of this official local lan-guage (Hassanpour 1992, this issue). Thus, while state policy in Turkey, Iran and Syria threatened the viability of Kurmanji, in Iraq, the status of Kurmanji was undermined by the ascendency of Sorani. However, if the largest section of Kur-manji speech area had turned into the killing fields of the language in Turkey, the

10 The Office of Education (Department of Health, Education and Welfare) proposed in 1960 to Ernest N. McCarus, Associate Professor of the Department of Near Eastern Languages and Literatures, University of Michigan, to prepare a basic course and a series of readers “for the instruction of students in that language” (Cameron 1967: iii). McCarus (1960: 325) wrote that “Kurdish today has strategic importance because of current political conditions in the Middle East, but it has long been of interest to Westerners for a variety of reasons.”11 See, for instance, article by Morgan Philips Price, Member of British Parliament, in the Manchester Guardian (September 1950): “There is no doubt that the Turkish method of national assimilation coupled with political freedom is bringing results, but it is a drastic remedy that only a strong Government can attempt . . . I doubt now if the Russians will succeed in making any mischief among the “mountain Turks” (the Turkish name for their Kurds) of Anatolia. They may be more successful however, in Persia and Iraq” (quoted in Bulletin du Centre d’Études Kurdes, No. 13, September 1950, p. 11).



Introduction 9

dialect was promoted, under Soviet rule, in the small enclaves of Kurdish popula-tion in Caucasia from 1921 to 1937. Here, Kurmanji was the only dialect spoken. Unlike Turkey, where an urban modern intelligentsia had been in the making after the 1860s, the Kurdish society in Caucasia was tribal, rural and predomi-nantly orate. Numerically, it was the smallest Kurdish population of the five coun-tries; moreover, Kurds were scattered throughout Caucasia and Central Asia, with a larger concentration in Armenia. Progress was swift, however. If illiteracy rates remained high in Turkey and the rest of Kurdistan, literacy was achieved among the Kurdish community of the USSR by the early 1940s when the first generation of the intelligentsia had already emerged.

The first conference on planning the development of Kurdish was convened in Yerevan, capital of Armenia, in 1934. It was decided that Kurmanji would be the language of writing, education and publishing. The norm chosen was the lan-guage spoken by the Kurdish working class of Armenia which would be devel-oped on the basis of the “literary school of Eĥmedê Xanî”, the prominent poet of the seventeenth century.12

Kurdish publishing was more advanced in the USSR than in Iraq, Iran and Syria. Between 1920 and 1985, the number of books per 1000 persons was 2.17 in Iraq, 0.14 in Iran, 0.09 in Syria, 6.41 in the USSR, and 1.13 for all four countries (Hassanpour 1992: 218). Of all Kurdish periodicals published between 1898–1985, 72.4% were in Iraq, though most of them were ephemeral. However, if we use the more accurate measure of the number of journals per 100,000 persons in 1985, the USSR with one regularly published paper, R’ya T’eze [New Road], emerges as the most active site of Kurdish journalism.13 In broadcasting, too, Soviet Kurds were ahead of the rest if measured by the number of broadcasting hours per units of population (Hassanpour 1992: 296).

Kurdish publishing and education were brought to an end between 1937 and 1945 when many Kurds from Armenia and Azerbaijan were deported to Central Asia in 1937 followed by the deportation of Georgia’s Kurds to Central Asia in 1944. Although this World War II-era policy was reversed and publishing and ra-dio broadcasting resumed in 1954, the promotion of the language never returned to the peak of pre-war years.

12 For a report on this conference, see Vil’chevskii (1936) and a summary of the report in Nikitine (1956: 289–293).13 The USSR with one newspaper and a Kurdish population of 115,858 ranked first (0.86%), Iraq with eight journals and a Kurdish population of 3,105,000 (1980) ranked second (0.25%), and Iran ranked third (0.05%) with two magazines and a population of 3,500,000 (1980). For sources and more detail see Hassanpour (1992: 246).




The access of many languages of the world to the new media of motion pic-ture and broadcasting has been quite limited even decades after their emergence. Radio broadcasting and cinema lend themselves to state control, and in the Mid-dle East, broadcasting emerged as a state monopoly. In fact, in Iran and Turkey, broadcasting and film were in the official languages only, and used to Turkify and Persianize the Kurds and other non-Turkish and non-Persian peoples. Still, broad-casting overcame the barrier of borders, and acted as a major factor in creating a national listening public. At the same time, under conditions of the Cold War, Western powers interfered in the destinies of Kurdish language broadcasting (Hassanpour 1992, 1996).

If printing was a late-comer to Kurdistan, Kurdish had an early start in broad-casting which began in the mid-1920s in Soviet Caucasia. Some fifteen years later, Britain and France allowed Kurdish broadcasting under conditions of World War II. This was a response to the extensive Nazi radio propaganda which had begun before the war. Radio Baghdad launched its programming in 1939 with a brief daily Kurdish broadcast. Later, the British-sponsored Sharq al-‘Adna [Near East] station operating in Palestine launched a Kurdish program. Both were in Sorani but a French-sponsored program began broadcasting in Kurmanji in Beirut. These war-time stations came to an abrupt end by the end of the war, although Radio Baghdad continued and expanded its predominantly Sorani program to three hours just before the fall of the monarchy in 1958 (Hassanpour 1992: 281–303; 1996).

While Britain and France closed down their Kurdish broadcasting stations by the end of the war, in the USSR, Kurdish publishing resumed in 1945 and broad-casting began on Radio Yerevan in 1954. The half-hour Kurdish program of Radio Yerevan was entirely in Kurmanji and for a while in the early 1950s, Radio Baku allowed the leaders of the Iran’s Kurdistan Democratic Party, who lived in exile in Azerbaijan, to broadcast their anti-Shah and anti-US speeches and party posi-tions in Kurdish. During decades of harsh suppression of the language in Turkey when possession of recorded Kurdish music and listening to foreign broadcasting were criminalized, many Kurds tuned in to Radio Yerevan. In the dark years of violent repression in Turkey, listeners saw a ray of hope every evening when the station began with Yerevan xeber dede [Yerevan speaks]. In sharp contrast, the United States and Britain treated this program as well as broadcasts from Radio Baku as Soviet propaganda aimed at “stirring Kurdish nationalism”. The confi-dential diplomatic correspondence of the period as well as some Western media reports indicate that the US and UK were entertaining the idea of broadcasting in Kurdish in order to neutralize the impact of Soviet broadcasting. However, this project did not materialize because both governments came to the conclusion that broadcasting in the language, even if the content was American and British



Introduction 11

propaganda, was not politically desirable. It is clear from this correspondence that Turkey regarded British or American broadcasting detrimental to its policy of forced assimilation of the Kurds. The British government agreed and was actually interested in population movements and ethnic cleansing of the Kurds. A secret dispatch from the British Legation in Damascus writes:

I realise how extremely touchy the Turks are about their Kurdish minority and, indeed, one can appreciate their point of view. Such broadcasts would tend to discredit their policy of complete assimilation. The process of assimilation is going on equally, though less rapidly, in Syria. There is a possibility that it might be speeded up if the Syrians ever succeeded in increasing the population of Jezireh [a mostly Kurdish region] by a settlement of Palestin-ians or by other means.14

Instead of Kurdish broadcasting by USA or UK, the Iranian Army launched two local stations in Sanandaj (1951) and Mahabad (1953), both in Sorani. However, this moment of the Cold War, fought over the language of airwaves, further inten-sified when Radio Cairo began a half-hour program in Sorani in 1957. This, too, was seen as a conspiracy by the Soviet Union and United Arab Republic paving the way for a “communist takeover” of the region. While Iran, Turkey and Iraq were actively engaged in diplomatic and propaganda campaigns against the sta-tion, the Iraqi monarchy was overthrown by nationalist officers in July 1958. Radio Baghdad’s Kurdish program was extended to four hours and joined Radio Cairo in exposing Britain, Iran, Israel, Jordan, Turkey, US and UK as enemies of the Kurds and other peoples of the Middle East.

The leaders of Iraq’s Kurdistan Democratic Party, under attack by the govern-ment, launched an armed struggle for autonomy in 1961 which continued until the first US-led war on Iraq in 1991. One of the major demands of the autonomists was language rights including mother tongue medium education on secondary and tertiary levels. By 1991, when a major part of Iraqi Kurdistan came under the control of Kurdistan Regional Government, Kurdish was used in some secondary schools though the Ba’thist regime’s Arabization policy continued. Textbooks and teaching were in Sorani Kurdish.

While Iraqi Kurds launched their longest armed struggle in 1961, the Kurds of Iran did the same in 1979 (see Sheyholislami, this issue), followed by the Kurds of Turkey in 1984. One of the central demands in these autonomist movements was the officialization of the language, and its use as the medium of instruction. These demands have been ignored in Iran, Turkey and Syria. While this issue was going

14 Secret dispatch No. 10622/8/50 from W. H. Montagu-Pollock, British Legation, Damascus, to G. W. Furlonge, Eastern Department, Foreign Office, 16 November 1950 [FO195/2650], reproduced in Destani (2006: 251).




to the press in 2011, none of these three countries allowed the language to be used as a medium of instruction in either public or private educational institutions.

During these armed conflicts, Turkey destroyed no less than 1300 villages and hamlets, and Iraq eliminated 4,009 villages in the course of a genocide known as Al-Anfal. The rural population moved into cities in Turkey and was transferred into concentration camps on major highways. These “forced urban-ization” projects have diminished the rural bases of the language, and changed the dialect mosaic, a situation that remains to be studied. Far from being a matter of civil war, the autonomist struggles turned into regional and international conflicts, and the status of Kurdish continued to be a question of international politics.

Several developments in 1991 changed the linguistic terrain again, this time in favor of Kurmanji. By the beginning of the last decade of the century, Turkey had succeeded in threatening the vitality of Kurmanji as a spoken language. Eight decades of physical and symbolic state violence against the language, including criminalization (speaking was treated as violation of the “indivisibility” of the Turkish nation), dialectisation (claiming that Kurdish is not a language), ruralisa-tion and de-intellectualisation (claiming that it is a rural, uncultured dialect; banning its use in print and broadcast media, even in music), has lead to the loss of language among many speakers, especially in urban areas. Even when it is spoken at home, the ban on mother tongue medium education makes it difficult for many Kurds to achieve fluency in reading and writing their language, and as a result they speak, read and write in Turkish. While this policy has not changed, in 1991, in Turkey, the Kurds were allowed to speak, though not write in their language. Speaking itself was restricted and banned in government offices, par-liament, or election campaigns. However, gradually, writing and limited broad-casting were also tolerated. By the end of the century, recorded music and print literature were produced in large quantities. The private teaching of the language is now allowed but under conditions that make it very difficult to help the lan-guage maintain its vitality. For instance, Kurdish cannot be taught to children younger than twelve years old (see Skutnabb-Kangas 2008). The goal of this poli-cy is to ensure that Kurdish children first get fluency and literacy in Turkish.

Turkey’s change of direction was, in part due to pressure from the European Union, which required respect for cultural rights as a condition for Turkey’s accession to the union (Olson 2009). This led to a limited reform of the legal frameworks of linguicide but these policies continue by both legal and political means. For instance, Kurmanji is not allowed to be used as medium of instruc-tion in education, and this is crucial for the vitality of any language, especially threatened ones; the legal provisions for private teaching of the language are too restrictive to allow any change in the status quo. In spite of these reforms, for



Introduction 13

example, the use of letters w, q, x and é in writing Kurmanji is banned because, not used in Turkish alphabet, they invoke “separatism”. However, the demand for mother tongue medium education is extensive. The teaching of Kurdish language and literature, as subjects, may be allowed at selected institutions of higher education.

These demands are based on two arguments (these are considered in great detail in Skutnabb-Kangas and Dunbar [2010]; there are many additional exam-ples in Skutnabb-Kangas [2000]). The first one is that the present submersion education through the medium of Turkish violates the human right to education. It can, from an educational, sociological and psychological viewpoint, be seen as genocide, according to two of the five definitions of genocide in the United Na-tions International Convention on the Prevention and Punishment of the Crime of Genocide,15 namely Article II(e): “forcibly transferring children of the group to another group”; and Article II(b): “causing serious bodily or mental harm to mem-bers of the group” (emphasis added). Subtractive dominant-language medium education for minority children prevents access to education, because of the lin-guistic, pedagogical and psychological barriers it creates. It often curtails the de-velopment of the children’s capabilities, and thus perpetuates poverty (according to theories about poverty by economics Nobel laureate Amartya Sen); and it is organized against solid research evidence about how best to reach high levels of bilingualism or multilingualism and how to enable these children to achieve aca-demically in school. Dominant-language medium education for minority children can have harmful consequences socially, psychologically, economically, and po-litically. It can cause very serious mental harm: social dislocation, psychological, cognitive, linguistic and educational harm, and, partially through this, also eco-nomic, social and political marginalization. It can often also cause serious physi-cal harm, e.g. in residential schools, and as a long-term result of marginalization – e.g. alcoholism, suicides, incest, violence, illnesses, and short life-span. This education can also be considered to give rise to international criminal responsibil-ity, exploring the application of the legal concept of “crimes against humanity” (see Skutnabb-Kangas and Dunbar 2010: Chapter 7).

The other argument is based on the fact that solid theoretical and empirical research shows that properly conducted mother-tongue-based multilingual edu-cation can lead to high levels of multilingualism (e.g. Kurdish/Turkish/English), good school achievement (with accompanying later job prospects), a positive multilingual identity, and positive attitudes towards self and others (see, e.g.

15 E793, 1948; 78 U.N.T.S. 277, entered into force January 12 1951; for the full text, see http://www1.umn.edu/humanrts/instree/x1cppcg.htm.




Skutnabb-Kangas et al. 2009; Heugh and Skutnabb-Kangas 2010). Thus it can also be part of both poverty and conflict prevention. In addition, it is a linguistic human right.

While the legal reforms of Turkey are, in part, due to the country’s applica-tion to accede to the European Union, the disintegration of the Soviet Union in 1991 curtailed the language rights of the Kurds in the newly independent repub-lics of Armenia, Azerbaijan and Georgia; it brought the end of broadcasting and the demise of publishing in Kurmanji,16 while the 1988–1994 war between Arme-nia and Azerbaijan over the disputed territory of Nagorno-Karabakh led to the further dispersion of the Kurdish population in the region.

In Iraq, once the rule of the central government was replaced by Kurdish self-rule in 1991, the Kurds were for the first time in charge of their language. Signs in the streets were now all in Kurdish – Kurmanji in the west and Sorani in the East. Although textbooks were not available at all levels, the language of instruction from kindergarten to college shifted to Kurdish. The Kurmanji speaking area of Badinan chose to use their dialect instead of Sorani. The second US war against Iraq in 2003 overthrew the Ba’thist regime and further consolidated the position of the Kurdish government and the language. The new, post-Ba’th constitution declares Kurdish one of the two official languages of Iraq. However, Arab nation-alist politicians in Baghdad pay lip service to this constitutional arrangement. In Kurdistan itself, Kurmanji speakers have insisted on the use of their dialect in education, media and administration. At present, although Sorani has the upper hand in terms of the number of publications and broadcasting channels, Kur-manji is making headway in Badinan, in the northwest of Iraqi Kurdistan, while the revival of the dialect in Turkey and the demand for language rights in Syria contribute to the enhancement of both its status and corpus. One may argue that Kurdish is, at present, a bi-standard language with two dialects Kurmanji and Sorani, while its other dialects are also struggling for access to writing, publish-ing and official recognition.

Another important development of the late twentieth century is a new wave of the dispersion of Kurdish speakers throughout the world. Refugees from all parts of Kurdistan have created new diasporic communities from Australia and New Zealand to Sweden and Canada, and have turned Kurdish into a transna-tional language. Equally significant is the proliferation of satellite television channels, social media, and internet use in both the diaspora and Kurdistan (see Sheyholislami 2010, 2011, 2012). The first Kurdish satellite television channel,

16 Occasionally, books are published, e.g., the translation into Kurmanji of the Georgian national epic The Knight in the panther’s skin (Şota Rûstavêlî, Wergirê p’ostê piling, published in T’ibîlîsî, Georgia, 2007).



Introduction 15

Med-TV, was launched in 1994 not by the Kurdistan Regional Government (KRG) but, rather, by a group of Kurds from and in Britain (Ayata 2011: 173–196; Hassan-pour 2003). It was a multilingual, multi-dialectal, channel though primarily in Kurmanji. Turkey has relentlessly campaigned for silencing the channel (Hassan-pour 2003). Failing to do so, Ankara launched its own channel, predominantly in Kurmanji, in 2009. Soon, the two major political parties sharing power in KRG began their satellite broadcasting. There were in 2010 no less than fifteen chan-nels (see Sheyholislami 2011).

The story of the Kurdish language, very briefly outlined above and in more detail in the rest of this issue, provides a picture of the present world linguistic order in which discrimination against languages and their speakers is the norm rather than exception. Many smaller languages are eliminated because the edu-cational system, the media, the market and the state threaten, often in tandem, their viability in the hierarchal linguistic order. In the Kurdish case, the state more than the market, disrupts the dynamics of vitality, especially mother tongue medium education and its use in mass media and administration. Linguistics and linguists cannot remain neutral in this conflictual relationship.

The experience of preparing this special issue endorses what we have said so far. Twice in the process, which began in March 2006, we decided to abandon the idea. A major problem is the dearth of research on the sociolinguistics and sociol-ogy of Kurdish; this is in spite of the fact that political constrains over the lan-guage overshadow whatever structural attractions it may offer. This situation ex-plains the absence of a sizeable pool of potential contributors; our first call for papers resulted in very few submissions. While a new generation of students, in-cluding native speakers, interested in the sociology of the language is emerging in the Middle East and in diasporas, it is not easy to overcome the language divide in compiling edited volumes: submissions written in Middle Eastern languages raise questions of translation and style and demand a longer and more laborious commitment.

The suppression of Kurdish makes it difficult to conduct field work which is indispensable for sociolinguistic research. It has also made it virtually impossible to access census data on the number of Kurdish speakers in Turkey, Iran, Iraq and Syria. Turkey provided census figures for the number of Kurdish speakers until 1965, although the figures may have been meddled with and respondents to the census question about language had been in a difficult position to reveal their linguistic identity. Interested in creating an “indivisible nation”, these states have not allowed the production of linguistic maps or statistical data which reveal their ethnic and linguistic diversity.

Under these conditions, the contributions in this issue do not cover the frag-mented Kurdish sociolinguistic landscape in a balanced manner. Kurdish in




Turkey receives more attention (see Fernandes; Zeydanlığlu; Üngör, this issue); Iraq and Iran each is covered in one paper (see Hassanpour; Sheyholislami, this issue), and Syria, Armenia and the diasporas are mentioned only in passing. This is primarily because research on the Kurdish language of each country is shaped by its political weight, numerical strength, economics and politics of knowledge production, and the state of academic freedom. Researchers’ knowledge of vari-ous languages, and here not least English, also plays a role. Although Kurdish has been most repressed in Turkey, this country’s interest in accession to the Euro-pean Union has led to limited legal reforms which may allow more research espe-cially in the realm of language rights.

In spite of these limitations, we hope that this issue of IJSL will attract more research interest to the social and political constituents of the linguistic mosaic of Kurdistan and the Middle East. This issue may also serve as a corrective or coun-terpoint to papers in two special issues of IJSL 165 on Turkey and IJSL 148 on Iran, each of which carries a paper on language planning. The former does not mention Kurdish at all and one gets the impression that Turkey is a monolingual Turkish society. The latter mentions Kurdish in one book review.

A dominant theme in the papers in this issue is the repression of the language especially in Turkey. Obviously, the question of language rights looms large in these studies. We hope that the advent of Kurdish into this prestigious journal, itself a product of the justice-seeking scholarship of Joshua Fishman, will en-hance the study of the language and promote consciousness about the precarious nature of not only small languages but also larger ones like Kurdish.

We are aware that a new generation of researchers, both native speakers and others, trained in various “sciences of language” is emerging in the countries where the majority of the Kurds live as well as in the expanding diasporas. We are confident that in not too distant future they will produce a wealth of knowledge about the sociology and sociolinguistics of the Kurdish language.

ReferencesAyata, Bilgin. 2011. The politics of displacement: a transnational analysis of the forced

migration of Kurds in Turkey and Europe. Baltimore, Maryland: The John Hopkins University dissertation.

Cameron, George G. 1967. Preface. In Jamal Jalal Abdulla & Ernest N. McCarus, Kurdish basic course: dialect of Sulaimania, Iraq. Ann Arbor: The University of Michigan Press.

Destani, Bjtullah (ed.). 2006. Minorities in the Middle East: Kurdish Communities 1918–1974. Vol. 3: 1941–1967. Slough: Archives Editions.

Dunbar, Robert & F. McKay. 2002. Denial of a language: Kurdish language rights in Turkey. London: Kurdish Human Rights Project.



Introduction 17

Fuad, Kamal. 1970. Kurdische handschriften. Wiesbaden: Franz Steiner Verlag.Hassanpour, Amir. 1992. Nationalism and language in Kurdistan, 1918–1985. San Francisco:

Mellen Research University Press.Hassanpour, Amir. 1993. The internationalization of language conflict: the case of Kurdish. In

Eran Fraenkel & Christina Kramer (eds.), Language contact – language conflict, 107–155. New York: Peter Lang.

Hassanpour, Amir. 1996. The creation of Kurdish media culture. In Peter Kreyenbroek & Christine Allison (eds.), Kurdish culture and identity, 48–84. London: Zed Books.

Hassanpour, Amir. 1998. The identity of Hewrami speakers: reflections on the theory and ideology of comparative philology. In A. Soltani (ed.), Anthology of Gorani Kurdish poetry, 35–49. London: Soane Trust for Kurdistan.

Hassanpour, Amir. 2000. The politics of a-political linguistic: linguists and linguicide. In Robert Phillipson (ed.), Rights to language: equity, power, and education, 33–39. Mahwah, New Jersey: Lawrence Erlbaum Associates.

Hassanpour, Amir. 2003. Diaspora, homeland and communication technologies. In Karim H. Karim (ed.), The media of diaspora, 76–88. London: Routledge.

Hassanpour, Amir. 2005. The (re)production of patriarchy in the Kurdish language. In Shahrzad Mojab (ed.), Women of a non-state nation: the Kurds, 227–263. Costa Mesa, CA: Mazda Publishers.

Hassanpour, Amir. 2011. Preface to the Reader by Amir Hassanpour. In Khanna Omarkhali, Kurdish reader: modern literature and oral texts in Kurmanji, [1]–20. Wiesbaden: Harrassowitz Verlag.

Hassanpour, Amir, Tove Skutnabb-Kangas & Michael Chyet. 1996. The non-education of Kurds: a Kurdish perspective. International Review of Education 42(4). 367–379.

Heugh, Kathleen & Tove Skutnabb-Kangas (eds.). 2010. Multilingual education works: from the periphery to the centre. New Delhi: Orient BlackSwan.

Krauss, Michael. 1992. The world’s languages in crisis. Language 68(1). 5–10.Kreyenbroek, Phillip G. 1992. On the Kurdish language. In P. G. Kreyenbroek & S. Sperl (eds.),

The Kurds: a contemporary overview, 68–83. London: Routledge.Kreyebroek, Phillip G. 1995. Yezidism – its background, observances and textual traditions.

Lewinston, N.Y.: The Edwin Mellen Press.Leclerc, Jacques. 1986. Langue et société. Laval, Canada: Mondia Éditeurs.MacKenzie, David N. 1961. Kurdish dialect studies. Oxford: Oxford University Press.MacKenzie, David N. 1966. The dialect of Awroman (Hawrāmān-ī Luhōn). København: Ejnar

Munksgaard.McCarus, Ernest. 1960. Kurdish language studies. The Middle East Journal 14(3). 335–335.Nikitine, Basil. 1956. Les Kurdes: Étude sociologique et historique. Paris: Librarie

Klincksieck.Oman, G. 1991. Matba‘a. In the Arab world. Encyclopaedia of Islam, new edn., vol. 6, 794–799.

Leiden: E. J. Brill.Olson, Robert. 2009. Blood, beliefs and ballots: the management of Kurdish nationalism in

Turkey, 2007–2009. Costa Mesa, California: Mazda.Question of the frontier between Turkey and Iraq. 1925. Report submitted to the Council by the

Commission instituted by the Council Resolution of September 30, 1925, Lausanne, League of Nations.

Sheyholislami, Jaffer. 2008. Language and nation-building in Kurdistan-Iraq. http://www.kurdishacademy.org/?q=node/712 (accessed 15 November 2010).




Sheyholislami, Jaffer. 2010. Identity, language, and new media: the Kurdish case. Language Policy 9(4). 289–312.

Sheyholislami, Jaffer. 2011. Kurdish identity, discourse, and new media. New York, NY: Palgrave Macmillan.

Sheyholislami, Jaffer. 2012. Linguistic minorities on the Internet. In Kirk St. Amant & Kelsey Sigrid (eds.), Computer-mediated communication across cultures: International interactions in online environments, 235–250. Hershey, PA: IGI Global.

Sheyholislami, Jaffer. Forthcoming. Language policy and nation-building in Iraqi Kurdistan: language as instrument, identity, and rights.

Skutnabb-Kangas, Tove. 2000. Linguistic genocide in education – or worldwide diversity and human rights? Mahwah, NJ & London, UK: Lawrence Erlbaum Associates.

Skutnabb-Kangas, Tove. 2008. Linguistic human rights in education, and Turkey. In The authors’ group (eds.), The linguistic rights of minorities. Diversity. International PEN multilingual electronic collection of poetry, fiction and non-fiction. http://www.diversity.org.mk/collection.asp?id=182 (accessed 12 December 2010).

Skutnabb-Kangas, Tove & Sertaç Bucak. 1994. Killing a mother tongue – how the Kurds are deprived of linguistic human rights. In Tove Skutnabb-Kangas & Robert Phillipson (eds.), Linguistic human rights: overcoming linguistic discrimination, 347–370. Berlin & New York: Mouton de Gruyter.

Skutnabb-Kangas Tove & Robert Dunbar. 2010. Indigenous children’s education as linguistic genocide and a crime against humanity? A global view. Gáldu Čála. Journal of Indigenous Peoples’ Rights 1. http://www.e-pages.dk/grusweb/55/ (accessed 1 April 2012).

Skutnabb-Kangas, Tove & Desmond Fernandes. 2008. Kurds in Turkey and in (Iraqi) Kurdistan: a comparison of Kurdish educational language policy in two situations of occupation. Genocide Studies and Prevention 3(1). 43–73.

Skutnabb-Kangas, Tove, Robert Phillipson, Ajit Mohanty & Minati Panda (eds.). 2009. Social justice through multilingual education. Bristol: Multilingual Matters.

Vil’chevskii, Oleg. 1936. Pervaia vsesoiuzanaia Kurdovedcheskaia konferentsiia i problema literaturnogo iazyka Kurdov SSSR [The first all-union Kurdological conference and the problem of the literary language of the Kurds of USSR]. IAzyk i myshlenie/Le langage et la mentalité VI–VII. 333–337.



http://www.diversity.org.mk/collection.asp?id=182

http://www.e-pages.dk/grusweb/55/

DOI 10.1515/ijsl-2012-0048 IJSL 2012; 217: 19 – 47

Jaffer SheyholislamiKurdish in Iran: A case of restricted and controlled tolerance

Abstract: It has been claimed that the 1979 revolution in Iran transformed the country in many respects. This article aims to examine the extent to which the Islamic Republic of Iran (IRI) has deviated, if at all, from the linguicidal policies of the Pahlavi dynasty towards non-Persian languages in Iran. The article finds, in both the monarchical and IRI regimes, a policy of (a) treating multilingualism as a threat to the country’s territorial integrity and national unity, (b) restricting the use of non-Persian languages, and (3) promoting the supremacy of Persian as a venue for unifying the ethnically and linguistically heterogeneous body politic. While the continuity in the language policy of the two regimes is prominent, dif-ferences will be noted especially in the changing geolinguistic context of the re-gion where Kurdish has achieved the status of an official language in Iraq (since 2005) and has enjoyed some level of tolerance in the linguicidal Turkish state (since 1991). New communication technologies as well as cross-border social and linguistic networking among the Kurds throughout Kurdistan and the world have changed the language environment but not the official policy of “one-nation = one-language”. Persianization of non-Persian peoples continues to be the build-ing block of the Islamic regime’s language policy.

Keywords: Kurdish; Persian; Iran; nationalism; language policy and planning

Jaffer Sheyholislami: Carleton University. E-mail: [email protected]

1 IntroductionHerçen mewaçan Farsî şekeren [Although they say Persian is sweet]Kurdî ce lay min bes şîrinteren [Kurdish is much sweeter if you ask me]Me’lûmen ce dewr dinyay bedendêş [No wonder that in this unjust world]Diłşaden herkes we zuwan wêş [Everyone is happier with their own language](Xanay Qubadî, Kurdish poet, 1700–1759)1

1 Different versions of these verses have been recorded in different sources. This version is from the Kurdish literary periodical Sirwe (Qubadî 1985).

Brought to you by | Purdue University LibrariesAuthenticated


20 J. Sheyholislami

The main slogan of the Iranian revolution of 1979 popularized and advanced by the leading clergymen at the time was “hokumete Eslami” [Islamic government] which meant a “total Islamization of state politics, society, and the individual” (Paul 1999: 190). The motive for the revolution was said to be spiritual and not materialistic, establishing a political order that was neither Western nor Eastern (Soviet style), and rebuilding the Iranian society and individual through “cultural revolution”. For many, this partly meant a drastic disassociation with the monar-chy’s “modernization” project and consequently with the hegemony of Persian (Iranian) nationalism. The Constitution of the Islamic Republic of Iran (hence-forth IRI), Article 19, confirms the new regime’s commitment not to make lan-guage a basis for discrimination: “All people of Iran, whatever the ethnic group or tribe to which they belong, enjoy equal rights; and [colour], race, language, and the like, do not bestow any privilege”.2 The assertion that in Islam, and thus in the IRI, “the question of border, colour, language and race doesn’t exist” ap-peared to be an indispensable constituent of the IRI narrative (Paul 1999: 209).

However, studies have shown that the discourse of “the immemorial Iranian nation” has continued to be reproduced in education (e.g., textbooks) (Ram 2000) and religious and cultural practices (e.g., Friday sermons), including the ways in which Persian fares in the discourse of the “Iranian nation” (Paul 1999). In general, as Paul puts it, “the commitment to the Islamic Revolution can now be understood as generating the very commitment that it was originally meant to replace: that to the nation and home-country” (Paul 1999: 217). Despite these ob-servations, many Persian nationalists, especially those who have remained loyal to the legacy of the monarchy after the Shah, continue to believe that the IRI has abandoned the pre-Islamic “great” and “ancient” Iranian civilization, culture and language (i.e. Persian). Some have even claimed that “the establishment of the Islamic Republic marks the rule of the Arabs over Iran again” or “second Mus-lim occupation” of “the Persian homeland” (Naderpour 1993: paras 40, 43; see, also, Saad 1996: 59). This perception of the IRI is akin to that of the late nine-teenth and earlier twentieth centuries’ architects of the Persian and later Iranian national identity, who believed that “Iranians were raided by Arabs” in the 7th century and as a result lost not only their ancient civilization but their “reason, knowledge, and ethics” (Tavakoli-Targhi 1990: 83).

This article investigates to what extent the IRI has deviated from the Pahlavi’s suppressive policies towards non-Persian languages such as Kurdish. There is a dearth of research on the state of non-Persian languages in Iran. Except for a few

2 IRI Constitution: http://www.iranonline.com/iran/iran-info/government/constitution.html (accesssed 20 August 2012).



Kurdish in Iran 21

studies by researchers committed to the promotion of minority languages in Iran (e.g., Anonby 2004/2005; Bani-Shoraka 2002; Hassanpour 1992; Jahani 2002), the vast majority of studies on Iran’s language policy and planning or sociolinguis-tics focus on Persian which is the mother tongue of about 50% of Iranians (e.g., Hayati and Mashhadi 2010; Kia 1998; Perry 1985; Sadeghi 2001). The article finds that despite some differences the IRI’s language policy is essentially a continua-tion of that of the monarchy. This policy is characterized by (a) treating multi-lingualism as a threat to the country’s territorial integrity and national unity, (b) restricting the use of non-Persian languages, and (3) promoting the supremacy of Persian as a venue for unifying the ethnically and linguistically heterogeneous body politic. While the continuity in the language policy of the two regimes is prominent, differences will be noted especially in the changing geolinguistic con-text of the region where Kurdish has achieved the status of an official language in Iraq (since 2005) and has enjoyed some more tolerance in the linguicidal Turkish state (since 1991). Access to new communication technologies and cross-border social and linguistic networking among the Kurds throughout Kurdistan have changed the language environment but not the official policy guided by the “one-nation = one-language” myth, or what Skutnabb-Kangas calls “the Western nation-state ideology” (2000: 319). Persianization of non-Persian peoples con-tinues to be the building block of the IRI’s language policy.

2 Theoretical issuesMany of the (nation-)states that emerged in the twentieth century modeled vari-ous aspects of their nation-building projects on those of the Europeans, espe-cially France because it has been held as one of the first “successful” modern (nation-)state building projects where homogenization has gone “smoothly” (Kymlicka and Straehle 1999: 73). This observation can be exemplified in the case of Iran. For example, Farhangestâne Zabân va Adabe Fârsi [the Academy of Per-sian Language and Literature] was “tacitly modeled on the Académie française”, in 1935 (Perry 1985: 308). The country’s twentieth century higher education, which produced policy makers and linguists among other professionals, was also designed along French lines. Thus, it is not surprising if Iran’s language policy has also been modeled on that of France. However, reflecting on the French Revo-lution, Billig (1995: 27) states: “The Rights of Man and the Citizen did not spread to the rights of Bretons and Occitans to use their own tongue in the schoolrooms of France: the northern langue d’oil was enforced, with the backing of legal stat-utes, over la langue d’oc”. According to Haugen (2003: 416), “. . . French revolu-tionaries passed a resolution condemning the dialects as a remnant of feudal



22 J. Sheyholislami

society”. Other “modern” states, such as the UK, have influenced more recently established states’ language policy as well. Billig (1995: 27) continues: “In the nineteenth century, Welsh and Lowland Scottish were officially banned in British schools. . . . The Argentinean government, in a curious by-way of national history, discouraged the use of Welsh in Patagonia” (see also Kilborn 1993). In the twen-tieth century, Catalan and Basque were banned during Franco’s reign in Spain, and Kurdish was proscribed in Turkey for nearly a century (see articles in this issue, Fernandes; Üngör; Zeydanlıoğlu). Iran also banned minority languages, including Kurdish, at least during the Reza Shah period, 1925–1941 (Hassanpour 1992).

The aim here is not to discern a few states with explicit oppressive language policies only and then blame them for the injustices that Iran or other states have inflicted upon linguistic minorities. According to Fishman, “even the much vaunted ‘no language policy’ of many democracies is, in reality, an anti-minority-language policy, because it delegitimizes such languages by studiously ignoring them, and thereby, not allowing them to be placed on the agenda of supportable general values” (Fishman 2001: 454). The aim rather is to highlight the fact that it is mostly since the birth of the modern (nation-)state that countries have perpetu-ated symbolic violence against languages and physical violence against their speakers in the name of the unity of the nation and the territorial integrity of the state. These states have supposedly followed “the nationalist principle” which,

. . . requires that the political unit and the “ethnic” one be congruent. In other words, given that ethnicity is basically defined in terms of shared cultures, it demands that everyone, or very nearly everyone, within the political unit be of the same culture, and that all those of the same culture be within the same political unit. Simply put one culture, one state. (Gellner 1997: 45)

For Gellner national literacy is indispensable to a modern, industrialized, (nation-)state, and it could only be achieved if the population has been “social-ized into a high culture” which in turn “can only be achieved by a fairly mono-lithic education system” (1983: 140, emphasis added). Gellner is correct to suggest that the modern and industrialized communities need to communicate sufficient-ly in order to maximize production and development. However, history and re-search have shown that such need does not have to be met in one language only and at the expense of other cultures and languages. It does not have to be through a “monolithic” education system.

Language is much more than a means of communication. Bourdieu, in the context of early France, observes: “It would be naïve to attribute the policy of linguistic unification solely to the technical needs of communication between the



Kurdish in Iran 23

different parts of the territory, particularly between Paris and the provinces. . . . The conflict between the French of the revolutionary intelligentsia and the dia-lects or patois was a struggle for symbolic power in which what was at stake was the formation and re-formation of mental structures” (Bourdieu 1991: 48). In other words, in a multilingual country, imposing one language on the whole pop-ulation at the expense of other languages is legitimizing a particular political authority, a certain style of social control, and an ideological way of socializ-ing the public that results in their indoctrination into a more or less similar way of seeing, understanding and interpreting the world, of seeing themselves and others, and thus of having the same identity, that of the dominant group. One of the most effective ways of assimilation and “integration” has been formal edu-cation (Skutnabb-Kangas 2000). As Fishman suggests: “More than most other authoritative specialists, the authorities of the educational system are deeply im-plicated in planned language shift. . . . Education [is] a very useful and highly ir-reversible language shift mechanism” (Fishman 2006: 320).

If states have used formal education to annihilate linguistic diversity, advo-cates of linguistic diversity, linguistic human rights, and bilingual education3 argue that in the context of minority languages formal education should be utilised to foster language maintenance, development and transition from one generation to the next. From this perspective, in a multilingual society, additive bilingualism, “where [two] languages continue to develop during the school years” (Cummins 2009: 22), can lead to not only the development of minority languages but also the students’ overall well-being: academic, cognitive and so-cial development (Skutnabb-Kangas et al. 2009).

This is not to suggest that education is the only domain within which lan-guages can survive and develop, nor that education alone can protect a language. The use of language in different domains such as the home, the community and the media are all important to various degrees. However, for many researchers none of these are as important as education in the modern world. This impor-tance of education in the maintenance of a language is eloquently put in this quote from the Renaissance, Welsh scholar Glyn Lewis:

Polyglottism is a very early characteristic of human societies, and monolingualism a cul-tural limitation. It is doubtful whether any community or any language has existed in isola-tion from other communities or languages. . . . If there is one thing we learn from a historical study of languages in contact it is that the languages which appear to contribute most and

3 “[T]he term bilingual education refers to the use of two (or more) languages of instruction at some point in a student’s school career. The languages are used to teach subject matter content rather than just the language itself” (Cummins 2009: 19).



24 J. Sheyholislami

survive longest . . . are usually supported and reinforced by powerful institutions, of which the schools . . . are among the most influential. (Cited in Hornberger 2009: 197; emphasis added)

Acknowledging the importance of schools, Halliday writes: “[s]ocieties, and gov-ernments, can avoid taking explicit political decisions about national languages and language policy; they can simply let things take their course. But somebody has to take conscious decisions about what is to happen in schools, at least as regards media and subjects of instruction” (Halliday 2007: 220). These decisions are often made deliberately by states and their parliaments and governments. For example, some of these policies are enshrined in countries’ constitutions. However, language management can also occur “through the on-going practices of politicians, bureaucrats, educators and other state authorities” (Ricento 2007: 212).

Elevating the status of minority languages in a policy statement or constitu-tion is a positive step in that it may serve as an acknowledgment of the existence of a language and its speakers, or perhaps it could be a remedy for healing some of the “injuries” (Fishman 1997: 334) that many speakers of minority languages endure as the result of marginalization and stigmatization created by a policy of monolingualism in a multilingual society. Such decisions and policies may also clarify how minority languages are treated in a multilingual society, for example, whether they are promoted, accommodated, tolerated or suppressed (Kloss [1977], cited in Ricento [2007: 221]). However, decisions and policies alone may not change the actual use of languages or their long term maintenance (De Va-rennes 1996). Language policy needs to be accompanied by proactive language planning that creates equal opportunities for all the languages of a state. The two go hand in hand (Hornberger 2009). Thus in the context of Kurdish and other minority languages in Iran it is important not only to look at the constitution and the officials’ positions and attitudes towards minority languages and Persian, but also to examine the extent to which languages are used in a variety of domains – especially education.

3 Kurdish in IranThe Ethnologue, 16th edition, lists seventy-six languages for Iran.4 The number of speakers for these languages and speech varieties vary greatly, ranging from a

4 http://www.ethnologue.com/show_country.asp?name=ir (accessed 10 October 2010).



Kurdish in Iran 25

few thousand to over 30 million. In addition to the official language, Persian, there are several sizable linguistic groups in Iran. Three of these linguistic groups have been more vocal with respect to demanding language rights: Azerbaijani, Baluchi and Kurdish. The latter has had the longest history, perhaps, of struggling for language rights in the country.

The Iranian census does not provide information about the ethnicity or lan-guage of the population, but it is safe to suggest that Kurdish is the third larg-est language in Iran (about 6–7 million speakers), outnumbered by Persian (about 30 million) and Azerbaijani (about 13–14 million) (The Ethnologue, 16th edition; McDowall 2004; Bani-Shoraka 2002). Although Kurdish speakers can be found in the capital city of Tehran and other major urban centers like Tabriz, the vast majority of them live in what Kurds themselves often refer to as Eastern Kurdistan, Kurdistan East or Iranian Kurdistan which encompass the provinces of Ilam, Ker-manshah (Kirmashan), Kordestan, and West Azerbaijan. There are also over 500,000 Kurmanji variety speakers living in the province of Khorasan in Iran’s northeast. All four dialect groups of Kurdish and their sub-dialects, except for Zaza (see the Introduction to this issue), are found in Iran. Their speakers make up the second largest Kurdish population after Turkey. However, whereas the vast majority of Kurds in Turkey, Syria and Armenia speak Kurmanji Kurdish (Northern Kurdish), the majority of Kurds in Iran (similar to Iraq) speak Sorani Kurdish (Central Kurdish). The second largest dialect group of Kurdish in Iran is identified as Kermashani and Southern Kurdish (spoken in and around the cities of Kermanshah and Ilam) followed by Kurmanji (spoken around the cities of Orumiyeh, Khoy, Makoo, and in Khorasan). The Hewrami, or Gorani (Gurani), variety (spoken in and around the towns of Nowsoud, Paveh, and Juwanro) has the smallest number of speakers (close to 100,000) (see Zolfaqari 2010). See Figure 1.

4 Iran’s language policy

The first constitution of Iran (1906) declared Persian, the language of about 50% of the population, the official language of all Iranians. It also prescribed that all members of parliament “had to possess the ability to speak Persian, read and write Persian, and be Iranian subjects of ‘Iranian extraction’ ” (Kia 1998: 32). Ar-ticle 19 of the supplementary fundamental laws of 1907 added that compulsory instruction in Persian had to be regulated by the [M]inistry of [S]ciences and [A]rts” (Kia 1998: 32). However, this policy was not implemented until the govern-ment in Tehran became much more centralized.



26 J. Sheyholislami

Fig.

1: K

urdi

sh S

peec

h Ar

eas

in Ir

an (H

assa

npou

r 199

2: 9

)



Kurdish in Iran 27

4.1 Reza Shah, 1925–1941

The first Pahlavi ruler was Reza Khan (1925–1941), who came to power in the wake of a British-backed coup d’état in 1921 (Curtis 2010) and in 1925 declared himself Reza Shah. He, similar to Kemal Atatürk in Turkey (see Zeydanlioğlu this issue), was committed to turning the loosely integrated state into a highly centralized regime. A part of this centralization process was to integrate and assimilate the country’s non-Persian peoples through, “among other coercive measures, the ex-clusive use of the Persian language in education, administration, and the mass media” (Hassanpour 1992: 126; cf. Jahani 2002; Bani-Shoraka 2002). Prescribing to the ideology of one-nation = one-language, the state banned non-Persian lan-guages. Government institutions including schools were decisively directed to use Persian only in their written communications and teaching materials. In their memoirs, prominent Kurdish poets and writers tell compelling stories about both physical and symbolic violence they endured, not just for using the language in writing but also for possessing Kurdish books or periodicals (Noori 2006; Salih 2007). Assimilation policies even reached beyond the borders of Iran. For exam-ple, the Iranian government complained that the officialization of Kurdish, even on the local level, in neighboring Iraq in 1931 was a British plot to advance Kurd-ish independence and consequently the disintegration of Iran (Hassanpour 1992: 128).

Despite the oppression, shortly after the abdication of Reza Shah some Kurd-ish materials were clandestinely printed by a newly founded Kurdish organiza-tion in the early 1940s (Hassanpour 1996: 55). It was a strong indication that the Kurds were not willing to abandon their language under pressure or assimi-late into the dominant culture. A Kurdish poet, Noori (2006: 7–8), who was a young man during the Reza Shah’s reign, writes in his memoirs that it was the state’s proscription of Kurdish that made him even more determined to write in Kurdish.5

5 Noori (2006: 7) recalls: “I will never forget that horrible day, when my father suddenly came home, and said: Mihemed! Where are the Kurdish books? I retrieved them like very precious items that were hidden among my other books and things, and I gave them to my father. The oven was on. My father quickly threw them into the flames. I had no idea why he did that. My little soul was burning along with them. Suddenly, a group of gendarmerie burst into the house and started searching everywhere. They asked my father: it has been reported that you have gone to Iraq and brought with you Kurdish books. Where are they? Where have you hidden them? . . .”



28 J. Sheyholislami

4.2 Mohammad Reza Shah, 1941–1979

Although Mohammad Reza Shah’s regime started to relax the coercive assimila-tion of minorities including Kurds, Persianization remained the core policy of the central government and its regional and local institutions from the military to education. In the central government’s discourse the Kurds started to be referred to as “true born (asîl ) Aryans” (Hassanpour 1992: 128) and “Irȃniȃne asil ” [origi-nal Iranians] who had supposedly been protecting the Iranian borders for centu-ries. Despite this, similar to other “Iranian” minority languages such as Baluchi (Jahani 2002), Kurdish was called a “dialect” of Persian. Demands for language rights, particularly education through the medium of mother tongue, both in Kurdistan and other regions (e.g., Azerbaijan), were either ignored or rejected outright.

The post-World War II conditions characterized by a weakened Iranian cen-tral government created the opportunity for the Kurds and the neighboring Azer-baijanis to declare their own autonomous governments. During the short-lived Kurdish Republic of 1946 Sorani Kurdish was declared as the official language, steps were taken to modernize the language (e.g., word coinage), textbooks were brought from Iraqi Kurdistan, teachers were trained, and Kurdish was the main medium of instruction (Hassanpour 1992). In addition to the newspaper Kurdis-tan, which was published every other day, several bi-weekly, monthly and occa-sional magazines were published. Although the republic lasted for only eleven months, its impact on enhancing Kurdish language was remarkable (Hossain 2002).

Even after the fall of the Republic in December 1946, the Shah, unlike his father, continued his relaxed language policy albeit in very controlled and limited measures. This, however, was not due to any substantial change in the state’s policy but rather a product of the socio-political changes of the time. A few peri-odicals (e.g., Koohestan 1944–1947) and books were published intermittently ei-ther because the central government was too weak to prevent them, or they were published (e.g., the periodical Kurdistan in the mid 1960s, in Tehran) with the direct support of SAVAK (the Shah’s secret police) for political ends. Similar moti-vations, for example to offset the Soviet broadcasting in Kurdish (Hassanpour 1996) or the influence of Kurdish cultural development and nationalism in Iraq, also prompted limited broadcasting in Kurdish and Azerbaijani in the 1950s. None of these were meant to enhance the status of Kurdish. It was still considered a “dialect” of Persian, was not taught in schools, and more importantly was not the medium of instruction in the formal school system. Signs of language shift from Kurdish to Persian in major urban centers such as Ilam, Kermanshah and Sanandaj were visible. The vast majority of people spoke Kurdish in the privacy of



Kurdish in Iran 29

their homes and on the streets but they could not read or write their language, and those who did mostly kept it to themselves; it was not a valued skill. As May (2006) tells us this should not come as a surprise:

If majority languages are consistently constructed as languages of “wider communication” while minority languages are viewed as (merely) carriers of “tradition” or “historical iden-tity”, as was the case in early LP, it is not hard to see what might become of the latter. Minor-ity languages will inevitably come to be viewed as delimited, perhaps even actively un-helpful languages – not only by others, but also often by the speakers of minority language themselves. (May 2006: 257)

In fact, during the Shah time in Iran, the vast majority of the Kurdish youth were not even aware that their language was written or that it could function in a “modern” world, for example in education, public service, movie production, and so forth. Hassanpour (1992) is compelled to categorize the situation of Kurd-ish in Iran during both monarchy regimes as “linguicide”. At best, the status of minority languages during the second monarchy could be characterized as restricted and controlled tolerance. This is also applicable to the IRI, as will be illustrated.

4.3 Islamic Republic of Iran (IRI) since 1979

Kurds participated actively in the 1979 revolution which toppled the Shah in Feb-ruary 1979. They gathered around their political organizations, and most of them, from the leftist to the most conservative ones, demanded self-rule (xudmuxtarî or khodmokhtari) within the borders of the Iranian state rather than independence or self-determination. However, from the beginning tensions between the Kurds and the central government were high. According to Ahmadzadeh and Stansfield (2010: 18), “the new theocratic government did not have an understanding of the particular ethnic issues which characterized their multi-ethnic country, and the conflict quickly escalated”. When on November 17, 1979 one of the bloodiest con-flict periods between the two sides ended, the Kurds optimistically entered into negotiations with the central government. The Kurdish delegation prepared a declaration consisting of 26 articles as the basis for possible negotiations. The demand for education through the medium of the mother tongue and the use of Kurdish in government offices in the Kurdish region was second on the list (Pash-koy Kurdistan 2010). Intermittent negotiations which continued for a few months, however, produced no agreements.

For almost two years, the Iranian central government could not regain full control over Kurdish towns and cities. During this period Kurdish literary



30 J. Sheyholislami

soci eties were established and plans were underway to teach Kurdish at the ele-mentary level and to adults, to use the language as the medium of instruction, and to establish the “University of Kurdistan” (Hassanpour 1992: 332). Teaching Kurdish started with minimal resources in Kurdish populated cities such as Ma-habad, Bookan, and Sardasht. Political organizations were teaching Kurdish to their cadres and members. Ad hoc classes and courses were operating throughout Kurdistan. In the summer of 1979, for example, about 1000 people sat in the am-phitheatre of the Youth House (Khaney Jawanan) in Mahabad for several days to learn reading and writing in Kurdish. In the summer of 1980, the Kurdish Demo-cratic Party of Iran (KDPI), the largest political organization at the time, estab-lished Kořî Fêrkirdin û Perwerdey Seranserî Kurdistan [Teaching and Training Council of Kurdistan] to develop Kurdish books for both elementary schools and adults (Pashkoy Kurdistan 2010: 2).

In November 1981 the military took control of the last Kurdish city but it could not control all villages for another three years. The two largest Kurdish par-ties, KDPI and Komala, published Kurdish textbooks for elementary grades and also books for adults. Until 1985 these organizations trained what they called “Mamostay Shorish” [Revolution Teacher] who taught Kurdish, alongside Per-sian, to children and adults in villages. A Kurdish writer and translator, Naser Hisami, who played an active role in the production of Kurdish reading books and training Kurdish teachers believes that “the campaign [to teach Kurdish and in Kurdish] was one of the most valuable contributions of the Kurdish move-ment in those years because it proved that in a free Kurdistan education must be in Kurdish” (personal communication, September 2010). Also reflecting on this campaign, Abdullah Hassanzadeh, previous Secretary General of KDPI and a renowned Kurdish translator, states that the promotion of education in the mother tongue showed to the central government that the Kurds could not wait for the state’s permission and were willing to do whatever it took to maintain their language, not just in spoken but also in written form (Pashkoy Kurdistan 2010: 2).

By 1985 the Iranian state was more or less in total control of Kurdish cities and villages. Tensions remained high between the Kurdish population, who had lost thousands of their loved ones in the war, and the central government who felt that their “Islamic” revolution had been “betrayed” by the Kurds. Once again, Persian became the sole language of instruction throughout Iranian Kurd-istan. Many Kurdish children considered Persian “the language of soldiers and Pasdars [i.e. Revolutionary Guards] and [were] afraid of it” (Pashkoy Kurdistan 2010: 4). Under these circumstances, Kurds had to find ways to use their language in publications, media, and even in schools, without being labeled counter-revolutionaries or separatists. Kurds and other minority groups have been hoping



Kurdish in Iran 31

that one day they will be allowed to have their language and literature taught in schools because the IRI constitution grants this right. Article 15 of Section 2 of the constitution of IRI reads:

The official language and script of Iran, the lingua franca of its people, is Persian. Official documents, correspondence, and texts, as well as text-books, must be in this language and script. However, the use of regional and tribal languages in the press and mass media, as well as for teaching of their literature in schools, is allowed in addition to Persian.

This article does not explicitly specify the medium of instruction in schools; however, one can infer that when “text-books” are required to be in Persian then education must be through the medium of Persian. This has been the case in practice, and it is similar to the monarchy’s era: believing in an education system that Cummins calls “subtractive education through the medium of a dominant language” which “can have harmful consequences socially, psycho-logically, economically and politically” (Cummins 2009: 20). The second line is very explicit that “regional and tribal languages” are “allowed” to be used “in the press and mass media” and also “for teaching of their literature in schools”. It is important to also note that the article makes no reference to the teaching of non-Persian languages as a subject. Nor does it contain any pro-visions about the use of non-Persian languages in administration and public services. A state’s administrative work and public services depend on written texts and the article is very specific that such texts “must be” in Persian. Again, similar to the monarchical period, public services are not provided in non- Persian languages.

Despite these serious limitations, the article has been viewed by some peo-ple, even from Kurdish oppositional groups, as a positive step in that it officially acknowledges linguistic diversity in Iran, something that the monarchy failed to do in the Constitution (Pashkoy Kurdistan 2010). However, one can argue that such an acknowledgment cannot be anything more than a “liberal mask”. It has been argued that “very often, multilingual societies which apparently tolerate or even promote heterogeneity in fact undervalue or ignore the linguistic diversity of their populace. A liberal orientation to equality of opportunity for all may mask an ideological drive towards homogeneity, a drive which potentially marginalizes or excludes those who either refuse, or are unwilling, to conform” (Blackledge 2005: 34–35). Article 15 has been viewed positively by some, perhaps because of the fact that some researchers have taken it at face value without investigating the extent to which it has been implemented even in limited ways (e.g., Hayati and Mashhadi 2010: 32). In the following section the use of Kurdish in media and education will be examined.



32 J. Sheyholislami

4.3.1 The use of Kurdish in the media

Broadcasting continues to be under a state monopoly. Kurdish radio and televi-sion programs (from one to a few hours a day) are aired from broadcasting centers in Tehran, Sanandaj, Mahabad, Orumiye, Kermanshah and Ilam radio and televi-sion stations. Most Kurdish intelligentsia, novelists, poets and journalists are un-happy with the Kurdish these stations use, and believe that there are some delib-erate attempts to harm the language. The content of the state-owned media in Kurdish is closely monitored by authorities, often through translation or by other means, because most chairs and directors of these radio and television stations are from other parts of Iran and do not speak Kurdish.6

In a similar vein, all publications have to be approved by the Ministry of Cul-ture and Islamic Guidance. Between 1985 and 1996, in addition to occasional pub-lications of books on religion and culture, a few bilingual (Kurdish-Persian) peri-odicals (e.g., Armanj and Ashâbe Enqelâb) were published by state-sponsored institutions, but very few people outside Tehran knew about them. The most im-portant Kurdish periodical in this period was Sirwe [The Morning Breeze]: Literary and Cultural Journal, which was published quarterly in Orumiyeh by the Salahad-din Aiyubi Publications, the Centre for the Dissemination of Kurdish Culture and Literature. The magazine was widely distributed among Kurdish audiences main-ly because it was known to be headed by a renowned and extremely popular Kurdish poet, Hemin, best known for his patriotic and love poems since the 1946 Kurdish Republic. Many literary figures, students, and artists from Kurdish com-munities in Iran, Iraq and diasporas would accept Hemin’s invitation to write in the magazine. After the death of Hemin in 1986, the popularity of Sirwe continued and grew even stronger when Ahmad Ghazi, another well-known Kurdish educa-tor, translator, writer and poet became the general editor. About fifteen years later, observers still praise the periodical for being a source of learning to read and write in Kurdish for a generation of writers and translators that currently contribute to the development of the language (Peyami Kurdistan 2005; Hesenpur and Bayezidi 2010).

On the other hand, many Kurdish researchers and literary figures, especially those who were in exile or active in Kurdish oppositional groups, believed that because Sirwe was supported financially by the provincial government it could not possibly have a genuine objective and consequently could not be beneficial

6 I worked at the Iranian National Radio and Television, Mahabad and Tehran branches, from 1981–1986 in various capacities such as writer, translator, radio plays actor, and announcer. Also, in recent years, I have continued to investigate Kurdish media as part of my research (e.g., Sheyholislami 2011).



Kurdish in Iran 33

for Kurdish language and culture (Pishko 1986). To some critics, Sirwe’s pub-lisher, Salahaddin Publications, was similar to the Saidian Publications that was allowed to publish a limited number of books on religion and culture during the monarchical regime. In 2004, Sirwe was moved from Orumiye to Tehran and later to Sanandaj where its final issue, number 281, was published in June 2010, after the government reportedly withdrew its financial support.

There can be little doubt that despite Hemin’s and his colleagues’ good inten-tions, the central government had its own propaganda objective in supporting Sirwe. In fact the magazine’s license was issued to a member of the Revolutionary Guards. The government was desperate to reach Kurdish audiences after several years of civil war in the region. The authorities probably hoped that having a con-trolled, well-circulated Kurdish magazine would quench the thirst of Kurdish readers and writers for a literary venue, so that they would not resort to clandes-tine publications and media run by Kurdish oppositional groups. However, once again, the Kurds used the opportunity and used Sirwe to develop their language and increase dramatically the number of youth who could read and write in their mother tongue.

By the beginning of the 1990s when hundreds of thousands of Iraqi Kurds took refuge in Iranian Kurdistan, and when the Kurdish movement in Turkey was reaching its utmost strength, authorities in Iran started to show more relaxation with respect to Kurdish cultural and linguistic activities. Between the early 1990s and the mid-2000s more than twenty periodicals (mostly bilingual, Kurdish-Persian) were published, including Awêne (Tehran), Aştî, Sîrwan, Rojhelat, Merdum, Aso (Sanandaj), Zrêbar (Mariwan), Didgah, Riskan, Peyamî Kurdistan, Mehabad (Mahabad), and Pûşpeŕ (Sardasht). During the government of reformist president Khatami, 1997–2005, there was another wave of monthly, quarterly or occasional periodicals known as govarî xuwêndkaran [university student maga-zines], such as Rave, Rojev, Ruwange, Hejîn, and Mukrîyan. Due to either financial difficulties or political restrictions most of these periodicals ceased publication during the Ahmadinejad presidency. Currently there are a few periodicals, for ex-ample in the city of Mahabad, published by university students. However, be-sides Mehabad (whose editor is occasionally called to the Ettelaat [the regional office of the secret police] over unapproved content) no other Kurdish publication in that city has permission to publish. Publication of Kurdish books on religion, history, folklore and literature are occasionally allowed.

Published materials can serve the language if they are accessible, for exam-ple, through two channels: public libraries and privately owned bookstores. In-formation obtained from the libraries of six Kurdish populated cities confirm that about half the libraries (i.e. in the cities of Bookan, Mahabad and Saghez) carry some old Kurdish books that were either donated by individuals or acquired from



34 J. Sheyholislami

government-sponsored publishers. However, except for a library in Bookan, no other library has catalogued the books or has them on their shelves for public use. None of the libraries subscribe to any Kurdish periodical. Although privately owned bookstores carry some Kurdish titles, reports from Iranian Kurdistan indi-cate that these bookshops are under constant surveillance by the authorities. For example, according to the website www.24katjimer.com in November 2010, Mohammad Hakimi, the owner of the bookshop Hejar in the city of Saghez, was arrested for “collecting and selling Kurdish books”. According to the report, his books were confiscated, and in addition to a two-year jail term he has been fined a considerable amount of money.7

The maintenance and use of the Kurdish language has also been the focus of literary societies and clubs. Depending on political circumstances in the region, since 1979 Kurds have had some opportunities to establish local literary societies and hold poetry and story reading sessions, usually on the occasion of special ceremonies or cultural events. From the early 1990s more literary societies started to emerge. However, these societies have remained as small circles of Kurdish fic-tion or poetry lovers and they have very little means of engaging the public, most of who do not read or write in Kurdish. The First Scientific Congress of Kurdish Language Teaching in Iran (Yekemîn Kongrey Zanistî Fêrkarî Zimanî Kurdî le Êran) held in 2002 aimed at setting up a system to teach Kurdish (e.g., designing a com-mon curriculum for all Kurdish courses at all levels, identifying or producing text-books for teaching Kurdish, and so forth). However, according to Rahmani (2005), this attempt, similar to others that were made to teach Kurdish in a systematic way, turned out to be “only a mirage” because it “gave people lots of hope” but no positive outcome.

Iranian Kurds have welcomed all kinds of opportunities to use their language in a wide variety of domains including cinema. However, this has proven to be even more difficult. For example, in response to a question in a documentary, the internationally acclaimed Kurdish-Iranian filmmaker Bahman Ghobadi states:

No one has given me permission to make Kurdish language films. However, slowly and with putting on smiles I managed to make the first Kurdish language movie in the history of the world cinema. . . . I kept my smiles to make the second, and the third, but . . . when I was just about to start shooting my fourth film in Kurdish they called me [and said]: “your movie must not be in Kurdish. If it is, however, the Kurdish portion must not be more than 20%”. I told them: I’m a Kurd, an Iranian Kurd, and it should be the Kurds’ right to have their own language. Now that I’ve made half of the film Nîwe Mang [Half Moon], they have confiscated it, and they do not allow me to make my fifth movie. I don’t exactly know the reason. But, I think if I were not a Kurd this certainly would not have happened . . . I don’t know what my

7 http://www.24katjimer.com/Detail.aspx?id=855&LinkID=3 (accessed 25 November 2010).



Kurdish in Iran 35

crime is. I’m very Iranian; I’m an Iranian Kurd. I make movies to make the conditions for the Kurds in Iran better. I want the Kurds have the same rights as the Persians. . . . [My translation]8

4.3.2 Education

Contrary to some claims (e.g., Hayati and Mashhadi 2010), although the constitu-tion permits the teaching of minority literatures in schools there is no evidence that this has been implemented. In fact there is no evidence that any of the minor-ity languages such as Azerbaijani, Kurdish and Baluchi have been officially taught in any elementary or secondary public or private school that operates under the auspices of the Iranian Ministry of Education and Training. In an email interview, a teacher with thirty years’ experience living in Iran, puts it this way: “So far, many open gatherings and meetings in Mahabad, Sanandaj, Tehran and other cities have taken place [to discuss the possibility of teaching Kurdish as a subject in schools] but they have not resulted in any positive outcome” (A. M., June 2010).

However, teaching Kurdish in universities (and also in private schools) has been in the news in recent years (e.g., Pashkoy Kurdistan 2010). On 15 March 2004, the weekly Kurdish-Persian periodical Aştî reported on an important de-velopment reflected in its main headline in Persian: “Ta’sise yak rešte-ye jadide tahsili dar dânešgâhhâ-ye Iran: Zabân va adabiyâte Kordi” [The establishment of a new major in Iranian universities: Kurdish language and literature]. Amin Shabani, one of Sanandaj’s representatives in the IRI parliament, on 5 May 2010 suggested that the teaching of Kurdish as a subject should start at the middle and high school levels in addition to offering university courses on Kurdish. However, none of these suggestions have been implemented.9

In 1997 the Azad University of Sanandaj offered, for the first time, two credits as part of the general/elective credits (18 out of 136 for BA) for one term. The sec-ond attempt to run courses on Kurdish language and literature took place in 1998–1999 in two universities in Tehran: Shahid Baheshti and ‘elm va San’at. Kurdish students at the University of Shahid Baheshti established Binkey Ziman û Edebî Kurdî [the Centre for the Kurdish Language and Literature]. With the help of two graduate students who had taught Kurdish previously, the centre managed to run two classes once a week, in both universities, one on the Kurdish language (50 students) and the other on literature (35 students). The two courses were

8 http://www.youtube.com/watch?v=9mTQrvPNvXc (accessed October 14 2010).9 http://kurdistantoday.ir/NewsDetail.aspx?itemid=1189 (accessed 15 July 2010).



http://www.youtube.com/watch?v=9mTQrvPNvXc

36 J. Sheyholislami

“extra-curricular” activities and no academic value was attached to them. They did not continue in the following year because the two graduate students who were teaching the course completed their studies and returned to their hometowns.

In 2003 some reformist Kurdish members of parliament convinced the then Minister of Education and Training, Mostafa Mo’ein, to allow the University of Kordestan, located in Sanandaj, to offer a BA in Kurdish Language and Literature. Two faculty members were assigned to design a curriculum, which was later ap-proved by the Higher Education Development Council (Šurâye Gostareše Amuzeše Âli) of the Ministry at its 467th meeting on 27 October 2003. The BA program was entered on the list of the programs for which students could compete in the country-wide university entrance exam (Konkur) in 2004. It meant that classes for the BA in Kurdish Language and Literature would officially begin in the fall of 2004. This never took place, however. Two of the instructors, who were assigned to welcome the students of the new BA program on the first day of classes, were “extremely disappointed” when they could not find any students that day; they were not yet aware that the Ministry had cancelled the program (S. S. personal communication, June 2010). A similar program for Azerbaijani that was devel-oped at the University of Tabriz experienced the same fate. The evidence pre-sented here again illustrates that there have not been any significant changes in the state’s policy towards the use or teaching of non-Persian languages in higher education. During the monarchical period in the early 1970s two courses on Kurd-ish language were offered at the University of Tehran (Hassanpour 1992: 331).10

Independent of the school system, since the mid-1990s there have been some informal and ad hoc private Kurdish teaching classes in some parts of Iranian Kurdistan (Pashkoy Kurdistan 2010: 2; Ranjbar 2006). The most well-known soci-ety offering these classes is the Cultural and Teaching Society of Soma: Kurdish Learning School (Binkey Ferhengî w Fêrkarî Soma: Fêrgey Zimanî Kurdî), better known as Soma. Soma has provided Kurdish learning classes at its branches in several cities of Iranian Kurdistan. The centre has published its own textbooks, and teaches Kurdish orthography (both Arabic and Latin based), grammar and literature at five levels to students aged nine and older. After completing all levels and courses students will be tested, and upon passing they receive a certificate indicating that they have completed Soma’s program. These graduates, if they so wish, may become teachers where they studied or move on to other towns and villages to teach on a voluntary basis. It has been reported that Soma has taught

10 Hassanpour convincingly argues that “[t]he course offerings can be explained by the political context of confrontation between Iran and Iraq in the 1970–75 period” (Hassanpour 1992: 331).



Kurdish in Iran 37

thousands of people to read and write in Kurdish (R. M. Personal communication, May 2010; Pashkoy Kurdistan 2010). Although all teachers11 work on a voluntary basis, to cover rent and material expenses students pay a small tuition fee to reg-ister in each course.

How this center started, what it has accomplished, and how it has lasted until now, November 2010, is interesting because it exemplifies Iran’s limited, re-stricted and controlled tolerance towards non-Persian minority languages. In 1994, a written request for permission to open a center for teaching Kurdish was submitted to the officials in the city of Mahabad. After six years of numerous follow-ups and “interrogations” (R. M. personal communication, May 2010), a one-year permission was obtained in 2000. After a few weeks, authorities in-formed the center that they had changed their mind and that they wished to re-voke the permission. However, Soma’s director told them that a few hundred people had already registered for the Kurdish courses and it was too late to cancel them. The permission was never renewed but the center has continued its activi-ties except for one period of several months when the authorities put a hold on its activities. A well-informed teacher working with Soma describes the center’s rela-tionship with the authorities:

[The state] has not provided any facilities to the school. They have even created various obstacles [to stop us from keeping our centre]. We are under constant surveillance, but be-cause we have not crossed any red lines and have not given them any excuses, they have let us continue with our activities. If this centre was [in any other business] by now we would have been offered thousands of dollars in loan and grant money from banks and others, but [because we teach Kurdish] now we owe more than $20,000.00; we have difficulty paying the rent. . . .

According to Jamal Pourkarim, one of Soma’s previous teachers and directors, who was not living in Iran when he spoke with a Kurdish periodical about Soma (Pashkoy Kurdistan 2010), ordinary people have been helping the centre as much as possible, for example by providing free space to hold classes.

The Soma case illustrates, more than anything else, how willing the people are to maintain their mother tongue, especially in written form. Soma’s students learn to read and write their mother tongue despite many difficulties, with full knowledge that this skill will not secure them a better job or the ability to enter a college or university. Perhaps this is what Fishman (1997: 336) calls “ethnocul-tural devotion” for the “beloved language”. When asked about the motivation of

11 Soma’s 200 teachers may hold a college diploma or a BA but they have all been educated in Persian. They have learned to read and write Kurdish and have become familiar with Kurdish literature either on their own or through Soma and literary clubs ad hoc Kurdish classes.



38 J. Sheyholislami

these students taking Kurdish classes, a Soma teacher provided the following answer:

The main motivation for the students is intrinsic. I believe that 99 percent of the students attend Soma’s classes because of their desire to return to the self, their roots and national sentiments. The public is more informed these days, their connections have expanded greatly, they are exposed to mass communications; Kurdish girls and boys thus search for their ethnic and national identity.

Overall, Soma appears to be a success story in terms of creating a positive at-titude toward the mother tongue among youth and the population at large. Most young people before the revolution could not read and write Kurdish. Despite this, a considerable number of well-informed Kurdish educators, writers, poets and journalists in the region are not happy with the way Soma teaches Kurdish. The writing system that it uses, for example, is said not to be in line with any of the writing systems that the Iranian Kurds have used for the past few decades in their periodicals and books. The critics are concerned that Soma’s unorthodox practices could confuse Kurdish readers and writers, damage the language’s slow standardization process, and in general harm the language rather than promote its development and maintenance.

Although Kurdish, similar to other non-Persian languages of Iran, is not the medium of instruction in formal education, nor is it taught as a subject, it is used in teacher and student interactions, including the teaching of a variety of subjects. The degree and nature of the practice, however, is situated and varies. Whereas one teacher says that he uses Kurdish in class to explain the subjects he is teach-ing (S. M. July 2010) another says that he could use Kurdish but he does not be-lieve that it would help the students since it may confuse them if they are taught the same lesson in two languages, Persian and Kurdish. Both teachers are from the same town, Mahabad. Another teacher from Sanandaj, the capital of the Kord-estan province, writes that the vast majority of teachers not only do not use Kurdish in their teaching but continue using Persian even when they talk to their students outside the classroom. He writes: “[a]lthough [Kurdish is] occasionally [used in teaching] by those who really care [about their students and language] the action is perceived illegal by the state officials. Every year or so official memo-randa are sent to schools reiterating that they should avoid using any language other than the official language” (M. S. July 15 2010). For example, the Central Of-fice of Education of Kordestan-Saghez, on 22 November 2010, sent a circular12 to

12 http://saghez.org/index.php?besh=dreje&id=1978 (accessed 30 November 2010).



Kurdish in Iran 39

schools in the region and urged teachers not to use Kurdish in their classes but only Persian, “the national language”.13

Privately owned institutions have also been instructed not to use Kurdish in teaching and training courses. For example, in 2007, the Kordestan branch of Rahgosha Institute14 warned driving schools in the province not to use Kurdish in their training sessions:

We have been informed that a few instructors use the local language of Kurdish in their teaching. Since Iran encompasses various ethnic groups with local [speech] varieties, ac-cording to the constitution the official language of Iran is Persian, and teaching at all school levels must be conducted in the official language of the country. It is necessary to issue an order [destoor] to (inform) all instructors that under no circumstance do they have the right to teach using the local language. [My translation and emphases]

These warnings issued within the context of the IRI to discourage and prohibit the use of Kurdish in schools are very similar to those of the monarchy era (see Hassanpour 1992: 126).

5 Discussion and conclusionKurdish and other non-Persian languages were banned in the Reza Shah period (1925–1941). A policy of restricted and controlled tolerance started to be adopted when Mohammad Reza Shah came to power in 1941. This meant that non-Persian minorities could speak their language at home and within their communities. There were also limited and state-controlled publications and broadcasting in ad-dition to ad hoc university courses on Kurdish. None of these policy relaxations, however, in any way jeopardized the hegemony of Persian. In addition to being the dominant language of media, broadcasting, cinema, publications, public ser-vices and administration, more importantly, Persian remained the medium of in-struction at all school levels. Nor was Kurdish taught as a subject.

The new constitution of the IRI in 1979 reiterated the core language policy of the Pahlavi era: that Persian was the only official language of the country and the only language of all education and official documents (i.e., public services, ad-ministration, private corporations and businesses). But, it also added that “the use of local and ethnic languages in the press and mass media is allowed. The

13 While appreciating the place of “the Kurdish language” in the Iranian “national culture” the circular says that its objective is to “respect the rights of colleagues and guest students”.14 The official website of the Rahgosha Institute can be accessed here: http://btnrahgosha.ir/page.php?141 (accessed 7 October 2010).



40 J. Sheyholislami

teaching of ethnic literature in the school, together with Persian language in-struction, is also permitted.” However, this paper has shown that the article has not been implemented. Furthermore, examples presented in this study illustrate that the state has not only failed to promote non-Persian languages but it con-tinues to interfere with the work of those speakers who want to learn, teach and maintain their language at their own expense, for example in the case of private Kurdish courses. This example illustrates that legal measures in support of a mi-nority language are not enough. One wonders if this is one of those cases where recognition of minority languages in a legal document could actually work against their overall objectives, where “change of status can be used as a political instrument to neutralize those pressing for recognition of their language by re-ducing the rallying power of their cause” (Romaine 2007: 219). Billig is suspicious of measures taken by the majority to relax “legal suppression of language” be-cause according to him it is often done “either in the interests of recapturing a harmless heritage, or to ward off demands from separatist or irredentist groups” (Billig 1995: 27).

At first glance there seems to be a considerable difference between the lan-guage policy of the monarchy and that of the IRI. In addition to Article 15 there have been two positive developments in the IRI period. There have been more Kurdish publications and literary activities, and also private courses for the past several years. It is safe to suggest that most Kurds welcome these developments because they may help youth to positively identify with their language. However, upon closer examination, the differences between the IRI and the monarchy’s language policies become much less significant. The state continues to deny speakers of non-Persian languages what Skutnabb-Kangas and Phillipson call “the core linguistic human rights” which are necessary in order to prevent lan-guage loss, because these rights “can enable the diversity of the linguistic ecology to evolve in processes of modernisation rather than being sacrificed” (Skutnabb-Kangas and Phillipson 2008: 3–4).

Table 1 presents these core linguistic human rights (Skutnabb-Kangas and Phillipson 2008: 3–4) in relation to language policies in three time periods in Iran (Reza Shah, Mohammad Reza Shah, and IRI). It shows that except for a partial difference in the first item (“positive identification with a (minority) language by its users, and recognition of this by others”), the rest of the core human linguistic rights have had the same fate in Iran since 1925: they have been denied. In the first row under IRI there are two sets of both minus and plus signs. The top set means that minority language speakers have started to positively identify with their own languages, but this is not due to any major change in the state’s lan-guage policy. Rather it is an outcome of the linguistic minorities’ commitment and efforts to develop their languages, their persistent demands for language



Kurdish in Iran 41

rights, and recent regional and transnational geolinguistic changes (see below). The second set of minus and plus signs acknowledges that article 15 of the IRI Constitution is a “recognition” of the existence of non-Persian languages in Iran. It even allows these languages to be used in mass media and the teaching of their literature. However, because the speakers of non-Persian languages are denied most of the core linguistics rights it is clear that “others” (i.e. the state) are not ready or willing to recognize that minorities such as Kurds positively identify with their own languages. Furthermore, the majority of Persian elites and academics, similar to the Shah era, continue to call non-Persian languages such as Kurdish and Baluchis “regional dialects”, “local dialects”, “ethnic languages” and “vari-ants of Persian” (e.g., Hayati and Mashhadi 2010: 27).

Based on the evidence presented here it becomes clear that the IRI has not diverged from the language policies of the monarchy in any significant way: (a) multilingual education is completely dismissed and viewed as a threat to the country’s territorial integrity; (b) the use of non-Persian languages is restricted and controlled; (c) supremacy of Persian as a venue for unifying the ethnically and linguistically heterogeneous body politic is perpetuated not only by the state apparatus but also by the majority of Iranian academics who have written on language policy and planning, the Iranian film industry, publishers and other forms of media.

In the early 1980s there was some doubt about the loyalty of the IRI to the “great” Persian language and the myth of one-nation = one-language. However, in 1988, the current Iranian Supreme Leader, Khamene’i, in an important speech reassured Iranians that at least with respect to language the new regime had no intention of diverging from the nationalism of the Pahlavi monarchy. Speaking to the delegates of the conference “The Greatness of the Persian Language, and the

Core linguistic human rights I II III*

– “positive identification with a (minority) language by its users” − − − +– “and recognition of this by others”

“learning a (minority) language in formal education”− − − +

– “not merely as a subject ” − − −– “but as a medium of instruction” − − −– “additive bilingual education, since learning the language of the

state or the wider community is also essential”− − −

– “public services, including access to the legal system, in minority languages or, minimally, in a language one understands”

− − −

*I = Reza Shah (1925–1941); II = Mohammad Reza Shah (1941–1979); III = IRI (1979–)

Table 1: Core linguistic human rights in Iran in relation to Kurdish



42 J. Sheyholislami

Necessity of Protecting it”, the then President of State Khamene’i says that Per-sian “is the language of [true]”, and that as the “national language” it constitutes “the most important and original determinant of cultural identity” (Paul 1999: 211). In the same speech, he casts doubt as to whether Arabic has the same “expressional capacity” as Persian, and says: “I don’t know if (the poet) Hafez could really be translated into Arabic” (Paul 1999: 211). As Paul argues “[n]ot only does Khamene’i, in calling language the most important determinant of national identity, ignore religion and revolutionary commitment, but his assertion that Persian might be of ‘superior (expressive) capacity’ to Arabic could even be con-sidered blasphemous” (Paul 1999: 211). To further ensure Persian’s “superior (ex-pressive) capacity” and to protect it from the intrusion of “foreign” elements, Farhangestân-e Zabân va Adab-e Fârsi [The Academy of Persian Language and Literature] that was established during the Reza Shah reign continues its mis-sion.15 This is in addition to another institution, Šura-ye Gostareš-e Zabân va Adabiyât-e Fârsi [The Assembly for the Expansion of Persian Language and Lit-erature], which was established in 1987 for the purpose of teaching Persian not only in Iran and other Persian-speaking countries but throughout the world.16

Among many other measures taken to protect the supremacy of Persian and to maintain the marginalization of minority languages is the “Ban on the Use of Foreign Names, Titles and Expressions” law that was adopted by the Iranian Par-liament in 1996.17 According to this law, all businesses and Government-backed organizations are barred from using foreign names and terminologies in their signs, advertisements, labels, titles and other forms of language use (e.g., official speeches and interviews).18 The objective of the law is said to be “the protection

15 The official website of The Academy of Persian Language and Literature http://www.persianacademy.ir (accessed 11 October 2010).16 The official website of The Assembly for the Expansion of Persian Language and Literature: http://www.persian-language.org/aboutus-page.html (accessed 13 November 2010).17 The actual law in Persian can be accessed here: http://rc.majlis.ir/fa/law/show/119218. See also the report “If Iranian Bill becomes law, no phoning for sandwiches” in The New York Times, 6 December 1996: http://www.nytimes.com/1996/12/06/world/if-iranian-bill-becomes-law-no-phoning-for-sandwiches.html (accessed 13 November 2010).18 According to Alireza Karimi, the General Director of the Advertising and Information Branch of the Ministry of Islamic Culture and Guidance, the law only started to be implemented in 2009: http://www.ershad-is.ir/archive/1-latest-news/1370-2010-06-17-08-17-12.html (accessed 13 November 2010). According to Karimi, in the period of one year, about 24,000 official notices and warnings were issued to businesses and institutions that used “foreign names” in their signs, advertisements, product names, and so forth. There are other reports by executive branches that indicate the law was implemented in 2005 in some parts of the country, for example in the small towns and cities around Tehran such as Varamin, Sawjublagh and Shimiranat: http://www.ad.gov.ir/print.php?sn=news&pt=full&id=6497&catId=2&action



http://www.persianacademy.ir

http://www.nytimes.com/1996/12/06/world/if-iranian-bill-becomes-law-no-phoning-for-sandwiches.html

http://www.nytimes.com/1996/12/06/world/if-iranian-bill-becomes-law-no-phoning-for-sandwiches.html

http://www.ad.gov.ir/print.php?sn=news&pt=full&id=6497&catId=2&action=print

Kurdish in Iran 43

of the Persian language’s strength (qovvat) and authenticity (asâlat)” (Badashian 2010). All “legislative, executive and criminal enforcement branches, organiza-tions, state corporations and institutions . . .” (e.g., Radio and Television) (sub-section 2) are directed to establish “Councils for the Protection and Expansion of the Persian Language” (subsection 10). These councils have the responsibility to identify foreign words used in their institutions and decide on selecting ap-propriate Persian equivalents, in consultation with the Academy of the Persian Language and Literature. According to Subsection 4, words from “Arabic or other languages” that have become parts of the Persian language do not constitute “for-eign words”. It is implicitly understood that the “foreign words” in this law refer to words from Western languages that have penetrated the Iranian linguistic environment in recent years (e.g., computer, mobile phone, e-mail, etc.). How-ever, writing in an Azerbaijani newspaper, Heydarbaba, published in Toronto, an Iranian-Canadian “lawyer and legal expert” argues that this law designates non-Persian languages such as Azerbaijani and Kurdish as foreign as well. Subsection 6 states: “those who speak, in addition to Persian . . . [one of the] local and ethnic varieties that are common to some regions in Iran are allowed to use names from their own language[s] or variet[ies] to name their products, institutions and prem-ises in their own regions” (emphases added). Badashian (2010) asks: “Does this mean that a product which has been produced in Azerbaijan or Kurdistan and which bears an [Azeri] or Kurdish name cannot be consumed in other parts of the country? Or, does this mean that a regional corporation with an indigenous name cannot have branches under the same name in other parts of the country?” To add to this criticism it is also worth mentioning that in this law and the discourse surrounding it, while Persian continues to be called “the national language”, “the official language”, “the second most important language of Islam”, and even the “mother tongue” of all Iranians,19 minority languages are referred to as “local and ethnic varieties” and are freely interchanged with “ethnic languages” or sim-ply “dialects” ( guyešha). Given this, Badashian (2010: 8) doubts if the objective of the law is only “the protection of the Persian language’s strength and authen-ticity”. For her, the law has other unstated objectives: “a gradual assimilation of non-Persian Iranian languages so that homogeneity and erosion of cultural diver-sity can be achieved . . . .” The supremacy of Persian continues to be nurtured by

=print (accessed 13 November 2010). Those in violation of the law received official notices to comply with the law or have their premises sealed by the security forces.19 For example by Alireza Karimi, the General Director of the Advertising and Information Branch of the Ministry of Islamic Culture and Guidance: http://www.ershad-is.ir/archive/1-latest-news/1370-2010-06-17-08-17-12.html (accessed 13 November 2010).



http://www.ad.gov.ir/print.php?sn=news&pt=full&id=6497&catId=2&action=print

http://www.ershad-is.ir/archive/1-latest-news/1370-2010-06-17-08-17-12.html

http://www.ershad-is.ir/archive/1-latest-news/1370-2010-06-17-08-17-12.html

44 J. Sheyholislami

the Iranian state at the expense of minority languages that are spoken by close to half of the country’s population.

The suppressive language policy that Iran has pursued since 1925 has left the speakers of non-Persian languages two choices: assimilation or resistance. At least since the Kurdish Republic of 1946 the Kurds have demonstrated that they are willing to resist assimilation, and they have continued to put language rights at the top of their list of demands. This, however, has not been without regrettable consequences. Escalated conflicts between the central government and the Ira-nian Kurds in the first few years of the early 1980s alone resulted in the loss of tens of thousands of lives, not to mention other devastations, destructions and long-term psychological damages that a war-torn community or country experi-ences (Ahmadzadeh and Stansfield 2010). However, in recent years demands for language rights have been expressed in more peaceful ways. More importantly, the Kurds are no longer the only national minority that demand language rights in Iran; other non-Persian linguistic groups such as Azerbaijanis and Baluchis, whose languages enjoy official status in Azerbaijan and Pakistan respectively, have joined them (Bani-Shoraka 2002; Jahani 2002; Shaffer 2000).

After being banned for decades, Kurdish in Turkey has started, since 1991, to be tolerated in limited and controlled ways. Since 2005 Kurdish has become one of the official languages of Iraq and it is the medium of education, economic ac-tivity, administration and public services in the Kurdistan Regional Government. Furthermore, more and more Kurds of different countries and regions have be-come connected through new communication technologies, diasporic communi-ties, and the relatively easy transnational movements of people and objects. Currently there are at least five 24/7 satellite television stations (i.e. Komala TV, Rojhelat TV, Tishk TV, Aso Sat, Koord Channel) belonging to Kurdish oppositional groups broadcasting to Iranian Kurdish audiences. In addition, Kurds are con-nected to the World Wide Web (Sheyholislami 2010, 2011, 2012).

All signs indicate that language policy in Iran needs a major reform in favor of non-Persian languages. This needs to take place at all three levels of language planning (i.e. status, corpus, and acquisition): granting minority languages such as Kurdish, Azerbaijani and Baluchi official status (either at the national or re-gional level); providing these languages with necessary means and resources to engage in corpus planning and modernization; and promoting and supporting the acquisition of these languages in the formal school system where they must not only be taught as subjects but more importantly be used as media of instruc-tion. The latter in particular will set the new language policy fundamentally apart from that of the monarchy era. Research shows that sound language policy and planning should and can aggregate welfare for all the citizens of a (nation-)state like Iran, if necessary steps are taken sooner rather than later. With language



Kurdish in Iran 45

shift, death and murder later could mean too late. It is, however, doubtful – if not impossible – that any major changes will arise in Iran’s language policy without a fundamental change in the country’s political and administrative structure. Changing the latter into a federated system within which linguistic minorities may achieve some degree of political autonomy and freedom to manage their own education system and language could be a promising start.

ReferencesAhmadzadeh, Hashem & Gareth Stansfield. 2010. The political, cultural, and military re-

awakening of the Kurdish nationalist movement in Iran. Middle East Journal 64(1). 11–27.Anonby, Erik John. 2004/2005. Kurdish or Luri? Laki’s disputed identity in the Luristan province

of Iran. Kurdische Studien 4/5. 7–22.Badashian, Setare. 2010. Naqdi bar qânune madeye vâhede “Mamnu’iyate bekârgiriye asâmi,

anâvin va estelâhâte bigâne” mosavebe 1375 bâ ta’kid bar hoquqe Irâniâne ĝaire Fars [A critique of the ban on the use of foreign names, titles and expressions law, adopted in 1996, with an emphasis on the rights of the non-Persian Iranians]. Haidarbaba 2(20). 8, 12.

Bani-Shoraka, Helena. 2002. The Iranian language policy of the twentieth century: the case of Azerbaijani in Tehran. In Annika Rabo & Bo Utas (eds.), The role of the state in west Asia, 141–150. Istanbul: Swedish Research Institute in Istanbul.

Billig, Michael. 1995. Banal nationalism. London: Sage.Blackledge, Adrian. 2005. Discourse and power in a multilingual world. Amsterdam: John

Benjamins.Bourdieu, Pierre. 1991. Language and symbolic power. J. Thompson (ed.), G. Raymond &

M. Adamson (trans.). Cambridge: Harvard University Press.Cummins, Jim. 2009. Fundamental psycholinguistic and sociological principles underlying

educational success for linguistic minority students. In Tove Skutnabb-Kangas, Robert Phillipson, Ajit K. Mohanty & Minati Panda (eds.), Social justice through multilingual education, 19–35. Toronto: Multilingual Matters.

Curtis, Mark. 2010. Secret affairs: Britain’s collusion with radical Islam. London: Profile Books.

De Varennes, Fernand. 1996. Language, minorities and human rights. The Hague: Martinus Nijhoff.

Fishman, Joshua, A. 1997. Language and ethnicity: the view from within. In Florian Coulmas (ed.), The handbook of sociolinguistics, 227–343. Oxford: Blackwell.

Fishman, Joshua A. 2001. From theory to practice (and vice versa): review, reconsideration and reiteration. In J. A. Fishman (ed.), Can threatened languages be saved? Reversing language shift revisited: a 21st century perspective, 451–483. Celvedon, UK: Multilingual Matters.

Fishman, Joshua A. 2006. Language policy and language shift. In Thomas Ricento (ed.), An Introduction to language policy: theory and method, 311–328. Malden, MA: Blackwell.

Gellner, Ernest. 1983. Nations and nationalism. Ithaca, NY: Cornell University Press.Gellner, Ernest. 1997. Nationalism. New York: New York University Press.



46 J. Sheyholislami

Halliday, Michael A. K. 2007. Language and education (Collected works of M. A. K. Halliday 9). J. Webster (ed.). London: Continuum.

Hassanpour, Amir. 1992. Nationalism and language in Kurdistan, 1918–1985. San Francisco: Mellen Research University Press.

Hassanpour, Amir. 1996. The creation of Kurdish media culture. In Peter Kreyenbroek & Christine Allison (eds.), Kurdish culture and identity, 48–84. London: Zed Books.

Haugen, Einar. 2003. Dialect, language, nation. In Christina Bratt Paulston & G. Richard Tucker (eds.), Sociolinguistics: the essential readings, 411–422. Malden, MA: Blackwell Publishing.

Hayati, A. Majid & Amir Mashhadi. 2010. Language planning and language-in-education policy in Iran. Language Problems & Language Planning 34(1). 24–42.

Hesenpur, Enwer & Hesen Bayezidi. 2010. Jinan û Govarî Sirwe: Le witûwêjêk da le gel Miryemî Qazî [Women and Sirwe magazine: in an interview with Mariam Ghazi]. http://24katjimer.com/Detail.aspx?id=525&LinkID=3&AspxAutoDetectCookieSupport=1 (accessed 18 October 2010).

Hornberger, Nancy. 2009. Multilingual education policy and practice: ten certainties (grounded in indigenous experience). Language Teaching 42(2). 197–211.

Hossain, Himdad. 2002. Rojnamenûsî Kurdî serdemî Komarî Dêmokratî Kurdistan 1942–1947 [Kurdish journalism during the Democratic Republic of Kurdistan 1942–1947]. Sulaimani: Dezgay Chap û Pexşî Serdem [Sardam Publications].

Jahani, Carina. 2002. State control and its impact on language in Balochistan. In Annika Rabo & Bo Utas (eds.), The role of the state in West Asia, 151–163. Istanbul: Swedish Research Institute in Istanbul.

Kia, Mehrdad. 1998. Persian nationalism and the campaign for language purification. Middle Eastern Studies 34(2). 9–36.

Kilborn, Richard. 1993. “Speak my language”: current attitude to television subtitling and dubbing. Media, Culture and Society 15. 641–660.

Kymlicka, Will & Christine Straehle. 1999. Cosmopolitanism, nation-states, and minority nationalism: a critical review of recent literature. European Journal of Philosophy 7(1). 65–88.

May, Stephen. 2006. Language policy and minority rights. In Thomas Ricento (ed.), An introduction to language policy: theory and method, 255–272. Malden, MA: Blackwell Publishing.

McDowall, David. 2004. A modern history of the Kurds, 3rd edn. London: I. B. Tauris.Naderpour, Nader. 1993. In hope of a third movement. [Translated by Farhad Mafie]. http://

www.oldrowzane.wpinewsroom.com/000_maqalat/2403/40319Nader_Naderpour.htm (accessed 20 August, 2012).

Noori, Mohammad. 2006. Şewêk û Xewêk: Bîreweryekanî Miĥemmed Nûrî [One night and one dream: Mohammad Noori’s memoirs]. Salahaddin Ashti (ed.). Sulaimani, Iraqi Kurdistan: Jin Publications.

Pashkoy Kurdistan. 2010 February 19. Organî Hîzbî Dêmokratî Kurdistan. Be boney rojî cihanî zimanî zigmakî. Amadekirdinî Qadir Wirya [Kurdistan supplement. Organ of the Democratic Party of Kurdistan. On the occasion of the International Mother Language Day. Prepared by Qadir Wirya].

Paul, Ludwig. 1999. “Iranian nation” and Iranian-Islamic revolutionary ideology. Die Welt des Islams, New Series 39(2). 183–217.

Perry, John R. 1985. Language reform in Turkey and Iran. International Journal of Middle East Studies 17(3). 295–311.



http://www.oldrowzane.wpinewsroom.com/000_maqalat/2403/40319Nader_Naderpour.htm

http://www.oldrowzane.wpinewsroom.com/000_maqalat/2403/40319Nader_Naderpour.htm

Kurdish in Iran 47

Peyami Kurdistan. 27 August 2005. Rojnamewanî Kurdî w akamêkî trajîk: Witûwêj le gel Suware Futûĥî [Kurdish journalism and a tragic ending: interview with Suware Fotoohi]. Peyamî Kurdistan 12. 12–15.

Pishko, F. 1986. Mergi Hemin [Hemin’s death]. Pêşeng. 12–18.Qubadî, Xana. 1985. Herçen mewaçan Farsî şekeren. Sirwe 1(1). 2.Rahmani, Obaid. 2005. Her serabê bû: Dû sał be ser damezranî Koŕî Zanistî Zimanî Kurdî da

[It was only a mirage: two years after the establishment of the Scientific Society for the Kurdish Language]. http://hedih.blogspot.com/2005/07/blog-post.html (accessed 15 July 2010).

Ram, Haggay. 2000. The immemorial Iranian nation? School textbooks and historical memory in post-revolutionary Iran. Nations and Nationalism 6(1). 67–90.

Ranjbar, H. 2006. Çalakiyekanî Encumenî Edebî Leylax [Activities of the Laylakh Literary Society]. Mahabad: Social, Cultural & Literary Magazine 66. 41.

Ricento, Thomas. 2007. Models and approaches in language policy and planning. In Marlis Hellinger & Anne Pauwels (eds.), Handbook of language and communication: diversity and change, 211–240. Berlin & New York: Mouton de Gruyter.

Romaine, Suzanne. 2007. The impact of language policy on endangered languages. In Matthias Koenig & Paul De Guchteneire (eds.), Democracy and human rights in multicultural societies, 217–236. Burlington, VT.: UNESCO.

Saad, Joya Blondel. 1996. The image of Arabs in modern Persian literature. London: University Press of America.

Sadeghi, Ali Ashraf. 2001. Language planning in Iran: a historical overview. International Journal of the Sociology of Language 148. 19–30.

Salih, Sadigh (ed.). 2007. Bîreweryekanî Miĥemmed Şapesendî [Mohammad Shapasandi’s memoirs]. Sulaimani, Iraqi Kurdistan: Jin Publications.

Shaffer, Brenda. 2000. The formation of Azerbaijani collective identity in Iran. Nationalities Papers 28(3). 449–477.



Sheyholislami, Jaffer. 2012. Linguistic minorities on the Internet. In Kirk St. Amant & Kelsey Sigrid (eds.), Computer-mediated communication across cultures: international interactions in online environment, 235–250. Hershey, PA: IGI Global.

Skutnabb-Kangas, Tove. 2000. Linguistic genocide in education – or worldwide diversity and human rights? Mahwah, NJ: Lawrence Erlbaum Associates.

Skutnabb-Kangas, Tove & Robert Phillipson. 2008. A human rights perspective on language ecology. In Angela Creese, Peter Martin & Nancy Hornberger (eds.), Encyclopedia of Language and Education, Vol. 9, Ecology of Language, 2nd edn., 3–14. New York: Springer.

Skutnabb-Kangas, Tove, Robert Phillipson, Ajit K. Mohanty & Minati Panda (eds.). 2009. Social justice through multilingual education. Toronto: Multilingual Matters.

Tavakoli-Targhi, Mohamad. 1990. Refashioning Iran: language and culture during the Constitutional Revolution. Iranian Studies 23(1–4). 77–101.

Zolfaqari, Ardashir. 2010. The Avromani-speaking area in Iran. Iran and the Caucasus 14. 323–330.



http://hedih.blogspot.com/2005/07/blog-post.html

DOI 10.1515/ijsl-2012-0049 IJSL 2012; 217: 49 – 73

Amir HassanpourThe indivisibility of the nation and its linguistic divisions

Abstract: Kurdish has four “geographical” dialects divided arbitrarily and forci-bly among five neighboring countries of Turkey, Iran, Iraq, Syria and Armenia. It has three literary dialects, two standardizing varieties, numerous norms and three alphabets. Further complicating this linguistic landscape since 1918 is the crisscrossing of dialect areas by international borders and subjecting them to state policies ranging from linguicide (Turkey, Iran, Syria) to officialization on the local (Iraq before 2005; USSR) and national levels (Iraq since 2005). Under these conditions, dialect divisions were overshadowed by the linguicidal situation which threatened the survival of the language. The formation of the Kurdistan Regional Government in 1991 and the officialization of Kurdish as one of the two state languages of Iraq in 2005 have removed the external (state) threat, and raised, once more, the question of the dialect base of the standard language. While Iraqi rulers had in the past used dialect pluralism as justification for deny-ing Kurdish official status, now the Kurds themselves have to cope with the lin-guistic fragmentation of their nation. This article examines the conflict over the adoption of one or two of the major dialects, Sorani and Kurmanji, as the official standard language in Iraq.

Keywords: dialect and language; nationalism and dialects; standardization; sociolinguistic theory

Amir Hassanpour: University of Toronto. E-mail: [email protected]

In his well-known article entitled “Dialect, language, nation” Einar Haugen (1966: 928) noted that “. . . the French revolutionaries passed a resolution condemning the dialects as a remnant of feudal society. The dialects, at least if they threaten to become languages, are potentially disruptive forces in a unified nation; they appeal to local loyalties, which could conceivably come into conflict with national loyalty.” While the Kurdish nationalists of the twentieth century were not as radical as their French predecessors in opposing feudalism, they have not wavered in rejecting dialect divisions and equating national unity with the unity of language.



50 A. Hassanpour

1 Statement of the problemKurdish is made up of at least four “geographic dialect” groups, which are some-times labeled according to their location in Northern, Central and Southern Kurd-istan. Before the rise of Kurdish nationalism in the twentieth century, the geo-graphic fragmentation of the language was not seen as a problem. The dialects were not politically or socially hierarchized into, for instance, literary and spo-ken, standard and vernacular, official and non-official, urban and rural, or supe-rior and inferior. Linguistic and literary heterogeneity was, in fact, the norm in pre-nationalist Kurdistan before the twentieth century; dialect divisions were not seen as an obstacle to Kurdish unity in so far as political, linguistic and cultural fragmentation was the order of the day in the predominantly rural feudal society. Two dialects (Kurmanji and Hawrami) had begun literary life independently in the sixteenth century followed by a third one, Sorani, in the nineteenth century. Even the transition from scribal to print culture in 1898 in Kurmanji did not ini-tially lead to hierarchization. However, with the formation of the Iraqi state under the British Mandate in the 1920s, dialect divisions did turn into a site of conflict between the evolving Arab monarchy and the Kurds, and later among the Kurds themselves. The Iraqi regime used the existence of two major dialects as grounds for denying Kurdish the status of a local official language. With the end of Iraqi state rule over much of the Kurdish region in 1991 and the formation of the quasi-independent Kurdistan Regional Government, Kurdish became the language of this administration, and before long Kurds found themselves divided over the choice of one of the dialects as the official, standard language. Some speakers of Sorani, which has been the more extensively used medium since 1918 in Iraq, claim that their dialect has already acquired the status of a de facto official lan-guage and reject the possibility of officializing Kurdish as a bi-standard language as it is the case in Norwegian. They equate national unity with linguistic unity: the universal law is the indivisibility of the nation cemented by one language and one (official) dialect. Under these conditions, the language/dialect distinction looms large in the life of Kurdish speakers. This article is a study of the conflict over dialectal fragmentation of the language.

2 Questions of theoryThe ties that bind language to nation(alism) are well known (see, e.g., Fishman 1973; Jones 2001; Millar 2005) to the extent that Marxist theory treats language as one of the four essential constituents of the nation. However, if nationalism thrives on linguistic homogeneity, sociolinguistics finds its raison d’être in



Kurdish in Iraq 51

heterogeneity. Distinguishing itself from linguistic theory, which is interested in “completely homogeneous” speech communities,1 sociolinguistic theory looks for variation and heterogeneity in the structure of language and its use. In socio-linguistics and dialect studies, “systematic heterogeneity” is seen as “an inherent universal of language structure” (Francis 1983: vii). Still, sociolinguistics has pur-sued a rather limited concept of the social. For instance, it makes a distinction between social and geographic (or regional, areal) dialects, privileges the former, and ignores the latter as if they are not socially constituted (Butters 1997). Pre-occupied with correlations between social variables such as class, gender, or pro-fession and linguistic variables especially in phonology, morphology, grammar, semantics and discourses (see, e.g., Kerswill 2004), “variationist sociolinguis-tics” fails to see geographical heterogeneity as a socially significant variation in language.

Much of the research on Kurdish has been historical and descriptive not in-formed by sociolinguistic theory and evidence. Western studies of the language began in the eighteenth century by philologists, members of the diplomatic corps stationed in the region, and missionaries whose approach was diachronic, genea-logical and comparativist. In (dis)counting a dialect as Kurdish, these early phi-lologists relied primarily on what MacKenzie (1961: 72) later called patterns of “historical sound change”. Using this yardstick himself, he found out that not only Hawrami and Zaza were non-Kurdish, but also the two largest dialect groups, Kurmanji and Sorani, did “not constitute a single, unified language” (MacKenzie 1986: 479). Here, the underlying assumption about the distinguishing feature(s) of language is singularity and unity in phonology (sound change) and, at best, morphology though rarely grammatical structure. This assumption betrays a pol-itical and ideological conflict between this brand of linguistics and the intuitions of the native speakers of the dialects who identify their speech as Kurdish. Some students of the language have acknowledged the conflict. For instance, C. J. Edmonds (1957: 10), the British political officer stationed in Iraqi Kurdistan, who was fluent in Kurdish and later taught the language at the University of London, noted that “[t]he European authorities generally maintain that Gorani [Hwrami] is not Kurdish and that the people who speak it are not Kurds; but the people themselves feel themselves as Kurds in every way.” Another linguist, Todd, in a study of Dimili [Zaza], says: “Dimili speakers today consider themselves to be

1 According to Chomsky (1965: 3), “Linguistic theory is concerned primarily with an ideal speaker-listener, in a completely homogeneous speech-community, who knows its language perfectly and is unaffected by such grammatically irrelevant conditions as memory limitations, distractions, shifts of attention and interest, and errors (random or characteristic) in applying his knowledge of the language in actual performance.”



52 A. Hassanpour

Kurds and resent scholarly conclusions which indicate that their language is not Kurdish. Speakers of Dimili are Kurds psychologically, socially, culturally, eco-nomically, and politically” (Todd 1985: vi).

It may not be imperative to reconcile the conflict between autochthonous lan-guage identities and linguistic theorizations of dialect/language. However, in the context of Kurdistan, any theorization of “dialect” assumes the form of a political undertaking that touches on the question of state violence against Kurdish and its speakers. Here, nation-states ruling over the Kurds have justified their projects of linguicide by claiming that Kurdish is not a language because it is a conglomera-tion of dialects – a debased Turkish dialect (in Turkey), a “pure” dialect of Persian (in Iran), or a quasi-language with no single standard (in Iraq). Dialectization is thus the strategy that both the linguicidal states and some Soranists use to main-tain national unity although for quite different purposes.

3 The Kurdish linguistic landscape in IraqBritain created the Iraqi state soon after occupying three southeastern provinces of the Ottoman Empire in 1917–1918 and naming them Iraq. Arabic and Kurdish were the largest languages spoken within the new borders followed by Turkoman and varieties of Neo-Aramaic spoken by Assyrians and Jews. The Kurds were allowed some degree of self-rule including the use of their language in adminis-tration, education and print media. This policy, based on geopolitical consider-ations, was in line with the British strategy of fanning the flames of Arab and Kurdish nationalisms against Ottoman Turkey during World War I (Edmonds 1925: 83–84). Two years later, in 1920, the newly formed League of Nations confirmed Britain’s Mandatory rule over Iraq subject to annual reviews of the ad ministration of the country by the League’s Permanent Mandates Commission.2 Respect for the rights of minorities was one requirement for satisfactory administration. How-ever, during the twelve years of its Mandatory rule (1920–1932), Britain created Iraq as a monarchical regime in which power, political as well as linguistic, was centralized and Arabic was given the status of the only official national language. Britain and Iraq confirmed their respect for the language rights of the Kurds in numerous pledges, from 1918 to 1932, made to the League and the Kurds.3 How-

2 For a study of Britain’s creation of the Iraqi state see Sluglett (2007) based on a survey of rich archival material.3 For a survey of these pledges see the confidential dispatch from Sir F. Humphrys, High Commissioner for Iraq, Baghdad, to Lord Passfield, Secretary of State for the colonies, 23 April 1930, and its attachments reproduced in Destani (2006: Vol. 1, 178–196).



Kurdish in Iraq 53

ever, both Britain and the Iraqi government persistently restricted the use of Kurdish, the second largest language spoken by about 20% of the population, and made knowledge of Arabic a requirement for citizenship in the new nation (Hassanpour 1992: 102–119). This policy led to unceasing conflicts between the Iraqi state and the Kurds as well as Assyrians and Turkomans who resisted forci-ble integration into the new state. If the imposition of Arabic was, for Britain, a project of constructing an “Iraqi nation”, it meant, for the Kurds, a project of as-similating them into an Arab state. The agenda of the Assyrians and Kurds was building their own nations, and failing to do so, they demanded un restricted lan-guage rights.

In recognizing the language rights of the Kurds, the British Mandate authori-ties pursued a policy of demoting rather than promoting Kurdish. This policy, adopted especially after the resolution of a border conflict with Turkey in 1926, included: (1) refusing to make Kurdish, alongside Arabic, one of two official lan-guages of Iraq, and, instead, giving it a local official status with numerous restric-tions on its use; (2) limiting native-tongue teaching to only a few grades in pri-mary schools; (3) restricting its use to only a few schools, and only in one part of the Kurdish provinces; (4) limiting its administrative use, mostly to munici-palities, and only in a few cities; (5) intervening in Kurdish language cultivation activities by blocking the standardization efforts of Kurdish language reformers;4 and (6) using the dialectal divisions of Kurdish as a means to deny the Kurds comprehensive language rights and to restrict the geographical scope of its use. This article is a survey of the conflict over the two major dialects, Kurmanji and Sorani.

Kurdish leaders and intellectuals were fully aware of the consequences of the linguistic dimension of Britain’s nation-building and state-building project in Iraq, and resisted it step by step though without success. Britain granted Iraq independence in 1932 and the only “concession” that the independent Iraqi state made to non-Arabic speakers was a “Local Languages Law”, which Bagh-dad reluctantly approved. The law, even in its drafting stages, was criticized by the Kurds for its malevolent intentions. The attitude of the Kurds is graphi-cally inscribed in a poem by Salam Ahmad ‘Azbani (1892–1959) entitled “To the League of Nations” (Bo KomeÎî Eqwam, September 1931), in which he accuses

4 For information on the way Iraqi government authorities deliberately demoted the status of the Kurdish language, see a report by C. J. Edmonds, political officer in Kurdistan: “The Kurdish question”, attached to Secret No. S. A. 321 (Ministry of the Interior, Baghdad, the 12th May, 1929, to The Secretary to His Excellency the High Commissioner for Iraq, Baghdad) in Kurdish Policy File No. 13/14 VI‘ Secret, pp. 189–199 (quotation from p. 194), Baghdad High Commission File, National Archives of India, New Delhi.



54 A. Hassanpour

the organization of “double-dealing”, “deceit” and “playing into Britain’s hands” (Hassanpour 1992: 113).

The policy mapped during the Mandate period continued throughout the mo-narchical era (1932–1958) and the various republican regimes until the formation of the Kurdistan Regional Government (KRG) in 1991. During this period of seven decades, the main conflict in the realm of language was between Baghdad and the Kurds who demanded full linguistic rights including recognition of Kurdish as an official language. See Figure 1.

3.1 Dialect politics under the British Mandate, 1920–1932

The dialect used in local administration, education and print media in 1918 was Sorani. It is, on the all-Kurdistan level, the second largest dialect group after Kur-manji, which is spoken by approximately 60 percent of all the Kurds. Kurmanji is spoken in all countries where Kurds live while Sorani is spoken only in Iraq and Iran. Two smaller dialect groups are Zaza/Hawrami and a group of southernmost dialects. The re-division of Kurdistan in 1918 among five neighbouring countries, Turkey, Iran, Iraq, Syria and the USSR, changed the relationship between dialects including their numerical weight within a single country (see below).

Literary development as well as transition from scribal to print culture came first in Kurmanji (see Introduction, this issue); this includes the first newspaper, Kurdistan (1898) and the first printed book (Alifbayê Kurmancî [Kurdish alpha-bet], 1909) published by the Kurds themselves.5 Also, the first recorded songs on the newly invented cylinders were in Kurmanji (Simon and Wegner 2000). The earliest attempt at reforming the alphabet came in 1913 in the Kurmanji magazine Rojî Kurd, “Kurdish Sun”. Thus, before the creation of the Iraqi state in 1918, Kur-manji had the upper hand in literary development especially prose, journalism, translation, and in print culture. This was to a large extent because the speakers of the dialect were more numerous, more urban, and more literate, a situation that led to the emergence of a new intelligentsia after the Ottoman state’s admin-istrative reforms of the mid-19th century had brought the first state-run, modern type schools including military ones to Kurdish cities.

The 1639 division of Kurdistan between the Ottoman state and Iran cut the dialects vertically, north–south, leaving all of them, except Zaza, on each side. The re-division of the Ottoman side of Kurdistan in 1918 into Turkey, Iraq and

5 By “Kurdish book”, I mean here a printed work, regardless of number of pages, produced by Kurdish authors and targeted at Kurdish readership. This definition excludes publications for academic purposes, missionary literature or government publications.



Kurdish in Iraq 55

Fig. 1: The line separating Kurmanji and Sorani dialects



56 A. Hassanpour

Syria further fragmented the Kurmanji dialect, which was now found in all the countries where Kurdish was spoken including Soviet Caucasus and Central Asia. Sorani remained divided as it was before World War I between two countries, Iraq (formerly Ottoman Turkey) and Iran. Even more significant was the changing sta-tus of Kurmanji: its dominant position came to an end not only because of its new fragmentation but, more seriously, due to its violent suppression in the Republic of Turkey, where it had the largest number of speakers and a flourishing modern-style intelligentsia. While the repression of Kurdish can be traced back to the late Ottoman period, republican Turkey pursued a harsh policy of linguicide, the de-liberate killing of language. The names “Kurds”, “Kurdistan” and “Kurdish lan-guage” were banned, and the use of the language was criminalized even in the privacy of home (for more detail and documentation, see Zeydanlıoğlu; Üngör; and Fernandes, this issue). In neighbouring Syria, the French Mandatory power tolerated speaking in the language and, for a few years before and during World War II, allowed Kurmanji publishing and broadcasting but rejected demands for native language education. In Iran, too, official policy under Reza Shah (1925–1941) was the deliberate killing of the language. Only in the USSR, the small and scattered enclaves of Kurmanji speakers in Armenia, Azerbaijan, Georgia and Turkmanistan were allowed to use their language in schools and the media.

In Iraq, the choice of Sorani as the dialect of administration and education was not based on linguistic considerations or criteria such as “purity”, “authen-ticity”, “beauty”, literary record, or the print output of the dialect. Had literary record and print culture been the criteria, Kurmanji should have been the official language. The choice of Sorani was, rather, based on political considerations. In Iraq, speakers of Sorani were more urbanized and more active in the emerg-ing Kurdish nationalist politics (c.f. Table 1). By contrast, the Kurmanji speech area was more tribal-rural, and numerically smaller than the rest of the Kurdish speech community. Thus, while state policy in Turkey, Iran and Syria threatened the viability of Kurmanji, its status in Iraq was undermined by the ascendency of Sorani.

The political weight of Sorani, within Iraq, was due to the significance of the city of Sulemani (Sulaymaniya) as the main center of the emerging Kurdish na-tionalist movement. The city had been, since the last decade of the 19th century, the seat of the only military school of the Mosul province of the Ottoman Empire. At the time of the British occupation (1918–1920) and mandate (1920–1932), a Kurdish religious and feudal leader, Shaikh Mahmud, had established its autono-mous rule centered at Sulemani.

After the occupation of Baghdad in March 1917, the British authorities pub-lished newspapers in Arabic, Persian, Kurdish and English. The Kurdish paper, Têgeyiştinî Rastî [Understanding truth], was in the speech of Sulemani, a sub-



Kurdish in Iraq 57

dialect of Sorani. British politicians favored the formation of a centralized state, but unable to confront Turkish army incursions into northern Iraq, they tolerated Shaikh Mahmud hoping that his autonomous government would be a barrier to Turkey’s threat. The dialect of Sulemani was used in the publications of this gov-ernment which held power between autumn 1918 to June 1919 and October 1922 to July 1924.

The only available official British reference to the choice of Sorani as the of-ficial language is in a document dating from 1931 (Great Britain 1920–1931: 230): “School books have been translated into the Kurdish of Sulaymaniya liwa – the only literary Kurdish at present existing in Iraq. . . .”

The use of the dialect in education, publishing and administration continued without conflict with Kurmanji until 1931, when Britain declared its intention to grant Iraq independence. The Kurds, Assyrians and Turkomans, although resent-ing Britain’s disregard of their rights, opposed the independence project because, in their view, the Arab regime in Baghdad would be more repressive in the ab-sence of surveillance by Britain and the League of Nations. Britain responded to these worries by pressuring the Iraqi government to enact legislation granting the Kurds, Assyrians and Turkomans language rights. Such legislation was one of the preconditions for gaining “independence” and membership in the League of Na-tions. The first draft of the law presented to the League did not raise the question

Sorani-speaking urban centers

A B C D1947 1965 1977 1987

Arbil 27,036 90,320 175,200 445,937Sulemani 33,510 86,822 175,413 364,096Koy Sanjaq 7,703 10,338 14,368 39,484Halabja 6,395 11,206 22,411 37,049Shaqlawa 3,355 6,814 9,707 17,565Rawandiz 3,320 5,807 6,242 8,916Raniya 1,301 4,090 8,594 49,986

Kurmanji-speaking Urban Centers

Dahok 5,621 16,998 36,512 114,322Zakho 14,249 14,790 28,536 60,655Aqra 5,579 8,659 n/a n/aAmadiya 2,511 2,578 4,845 4,872

* Sources: Sousa (1953: 35–36); Najm al-Din (1970) and Iraq Republic (1966); Hafid (2000: 19–26, 113–117, 200–203)

Table 1: Sorani and Kurmanji speaking towns of Iraq*



58 A. Hassanpour

of dialects. However, unwilling to officialize Kurdish, the Iraqi government de-layed ratification of the draft law by “discovering” that Kurdish had two different dialects (Sluglett 2007: 143–144).

Though the Mandate authorities were well aware of the intentions of the Iraqi government, they continued pressuring Baghdad to pass the law in order to con-vince the League of Iraq’s qualifications for independence and membership into the international organization. However, the approved text carried a new Article 8 which stipulated: “. . . in the Qadhas [administrative units which together make a liwa, which in turn make a mutassarifiya, ‘governorate’] of Mosul liwa referred to in this Law the people may choose the type of Kurdish language they desire within one year from the coming into force of this Law” (the text of the law ap-pears in Great Britain [1931: 73–75]). Sir Francis Humphrys, British High Commis-sioner of Iraq, defended the delays in legislation by claiming unjustifiably in the League of Nations that “even in Sulaimaniya liwa itself there were several dialects spoken.”6 He claimed, moreover, that the addition of the new article was done after “consultation with representative Kurds” (League of Nations 1931: 119). This was not true, either. For instance, one well-known “representative Kurd”, Amin Zaki, Deputy from Sulaymaniya and a known historian and literary figure, was not consulted. He was, in fact, surprised to learn, in a meeting with Humphrys (May 20, 1931), about Article 8. Zaki criticized the action and later submitted, in writing, his criticism of the law and its divisive political intention to the High Commissioner. His views are important as they were, and still are, indicative of much Sorani opinion on the dialect question in Iraq. According to Zaki (1935: 66–99):

1. The general principle which other nations have followed in dialect selection is the [adoption of the] most eloquent ( fesîh) and orderly (rêk û pêk) dialect of the language. In Turkey, as we see, it is the dialect of Istanbul which is prevailing in all the offices of the government in [outlying areas of] Anatolia, although it is quite different from the dialect of this region. Another example is the acceptance of the eloquent [classical] Arabic language in the offices of the government of Iraq while we all know it is quite different from the popular dialect. . . . If this general rule is observed one must not heed the slight difference between the dialect of the people to the east and west of the Great Zê [Zab River] and not break up the unity of the Kurdish language. In selecting the eloquent dialect according to scientific principles, the Eastern Kirmanji dialect [Sorani] must be accepted since it is very close to the eloquent

6 This claim is in sharp contrast with the opinion of Edmonds (1937: 487), the British Mandate’s expert on the Kurds and their language, who believed that the dialects of “Soran, Baban, Ardelan and Mukriyan” which cover, in addition to those referred to by Humphrys, the dialects across the Iranian border, “form a single linguistic group”. All these dialects are mutually intelligible.



Kurdish in Iraq 59

Mukri dialect, and must be made the official language of all the offices and institutions ex-isting in Iraqi Kurdistan. 2. If the western Kirmanji (Badinan) language is adopted for the people in the five qadhas of Mosul, it will cause the following harms: a) the unity of the Kurdish language will be destroyed. This will be a cause of dissidence and hatred among the Kurds and it will never indicate the good will of the government toward them, b) it will result in the retrogression of education in the five qadhas because there are no textbooks in the Badinan [Kurmanji] dialect and there are not any competent individuals to be assigned to teach in the schools and serve in the offices; very probably, they cannot have access to textbooks for quite a long time and, thus, there will be no benefit in their teaching; it is by no means unlikely that, due to this situation, they might avoid their language and [thus] be forced into accepting another language, c) naturally, the government must 1. either spend twice as much for pro-ducing textbooks in two dialects . . . and double spending for preparing printed papers, translations of regulations and laws, etc. for both sides – which [this government] is far from undertaking, or 2. not listen to the demands of the two sides; this will result in the prevention of all the three liwas and five qadhas from making progress and will be a cause for grievance. . . . I can say, on the basis of my previous experience, that this ill state of affairs will not take long to surface, because this government which has never felt pity for the Kurdish people and has never allotted them a just share of educational funds . . . now will not spend double amounts for the education of the Kurds in the three liwas and five qadhas. That is why I request you to look at this task sympathetically and intervene so that only one dialect will be accepted for all Iraqi Kurdistan.

The law was, however, ratified and, according to the Annual Report of 1931 (Great Britain 1931: 19), “a committee was set up in each of the Kurdish qadhas of Mosul liwa to advise the Government as to the form of Kurdish which the inhabit-ants desired should be adopted for use under Article 8 of the Local Languages Law”. A year later, “delegates from these qadhas met in March in Mosul liwa headquarters and by seven votes to two they decided in favour of the use of the indigenous Bahdinan [Kurmanji] dialect” (Great Britain 1932: 5). According to a “secret and very urgent memo”:

On 1/3/1932 a meeting took place in the office of this Mutasarrifiyet [governorate] by dele-gates elected from all the Kurdish Qadhas except Zibar where nobody had . . . desired to attend the meeting in question [due to weather conditions]. . . . After discussion and con-versation for the adoption of the form of the Local Language no agreement has been reached at in opinion, since seven of the members proposed that the Bahdini Language should be regarded as official language against two others who voted for the Soorani Language. . . .7

7 No. 205 of 3.3.1932 from Mutasarrif [governor of] Mosul to the Ministry of Interior, File No. 13/14, Vol. XII, Kurdistan Policy, p. 2, National Archives of India, New Delhi.



60 A. Hassanpour

Two delegates from Shaikhan, three from Aqra and two from Dahok voted for Bahdini [Kurmanji] and two delegates from Zakho voted for Sorani (Great Britain 1932: 3).

3.2 Monarchical Iraq, 1932–1958: the demise of Kurmanji

Soon after independence, the suspicion, shared by both Kurds and some Mandate authorities, that Baghdad had raised the dialect question in order to avoid lan-guage legislation and to Arabize the Kurdish speech areas of Mosul proved to be true. Just before the legislation and after it, there was no native language educa-tion in Kurdish qadhas of Mosul. According to a “list of schools in the Kurdish areas of Iraq”, prepared by the Iraqi government and submitted to the League of Nations by Britain in 1931, there were eleven schools in the Kurmanji (Badini) speaking towns in which instruction and textbooks were in Arabic. In these schools, according to this source, “[i]nstruction [is] in Arabic – Kurdish is used as the medium of explanation especially in the elementary classes.”8 After indepen-dence, the government had no interest in changing this status quo.

Although Sorani speaking areas were also subject to Arabization, the dialect made considerable progress in standardization by the end of the period – a re-formed alphabet, a purified vocabulary (Abdulla 1980; Hasanpoor 1999), a grow-ing prose literature, and uninterrupted use in broadcasting and journalism (for a detailed survey, see Hassanpour [1992]). Its position vis-a-vis Kurmanji was even further strengthened when it was adopted as the language of the short-lived Kurd-ish Republic of 1946 in Iran. By contrast, the progress made by Kurmanji in Syria was interrupted in 1946 when this country achieved independence (Hassanpour 1992: 136–139); in the USSR, Kurmanji was promoted until 1937 but suppressed from 1937 to the end of World War II, when its use resumed in media and, to a lesser degree, education. While broadcasting in the dialect by Radio Yerevan (Armenia) and the newspaper R’ya T’eze [New Road] enhanced the status of Kur-manji, the anti-Soviet policies of Iraq and neighboring countries did not allow interaction between the speakers of the dialect in these countries. Radio Yerevan remained the main venue of contact between the Kurmanji of the Soviet Union and the neighboring states.

8 The table is in a document entitled “Appendices to the British Government’s observations on the petition, dated April 19, 1931, from Taufiq Wahbi Beg, League of Nations, Permanent Mandates Commission. Iraq. June 22nd, 1931, Geneva”, appended to a secret despatch from E. R. Ludlow-Hewitt, Acting High Commissioner for Iraq, to the Secretary of State for the Colonies, 5 June 1931 reproduced in Destani (2006: Vol. 2, 256–259).



Kurdish in Iraq 61

3.3 Republican Iraq, 1958–1991: the unsettled hegemony of Sorani

A major change in the relation between the two dialects took place soon after the overthrow of the monarchy on July 14, 1958. Radio Baghdad began broadcasting in Kurmanji and a number of privately published Kurmanji journals appeared. These developments led to a heated discussion of the dialect base of the stan-dardizing language.

According to the only available official explanation for broadcasting in Kur-manji, the Kurdish station had failed to serve “all the Kurdish compatriots before the Revolution [1958 coup d’état] since it did not broadcast in the Bahdinan [Kur-manji] dialect which is spoken by Kurdish compatriots in Mosul liwa and parts of Arbil liwa.” Broadcasting time was, consequently, increased to four hours equally divided between the two dialects (Iraq Republic 1959: 257).

The question had already been discussed in the Kurdish press by October 1958, when a writer and political activist, Mihamadi Mala Karim, asked, “In what dialect should we write our language?” The writer referred, by analogy, to the role of the dialect of the Quraish tribe in Arabia which had provided the base for the widely spread literary Arabic. He argued that Sorani could play a similar role in view of the progress made in Iraq (article in Hîwa, quoted in Nêrevan [1959: 143]).

The first congress of Kurdish teachers held in September 1959 discussed the question, and resolved that Sorani was the base of the official literary lan-guage in Iraq. A minority opted, however, for the use of both dialects (Rojî Nö, Vol. 2, No. 3, June 1961: 33–34). The resolution was confirmed again at the second teachers congress held in 1960 (Hetaw, Vol. 7, No. 185, 15 September, 1960: 28).

Apparently responding to the resolutions in an article entitled “The question of the unification of the written Kurdish language, Kurmanji or Sorani?,” the head of the Kurdish Students Society in Europe, Vanley (1959: 5) noted that his Sorani-speaking compatriots would be surprised to read about this question since, he argued, they had never considered the issue in its national, i.e., all-Kurdistan pro-portions. Vanley, himself a speaker of Kurmanji, suggested that the British Man-datory power and the monarchy had conspired with Turkey to prevent the use of Kurmanji because of its repercussions among the Kurds of Turkey, who were speakers of the same dialect (Vanley 1959: 9; private correspondence dated 4 May, 1984). He argued that Kurmanji was even more important than Sorani and that its use in Iraq must be promoted alongside that of Sorani.

The reasons for the promotion of Kurmanji in Iraq were thus outlined by Van-ley: (a) it is spoken by the great majority of the Kurdish people, 62%, while Sorani



62 A. Hassanpour

is at best the speech of 38%; (b) it has very “clear (and numerous) grammatical rules” which were scientifically studied and fixed in many books of grammar printed in the Soviet Union, Syria and Paris; (c) it is already written in well-adapted Latin characters, while Sorani is written in the unsuitable Arabic alpha-bet; and (d) Kurmanji is most probably nearest, he asserts, to the “Ancient Aryan” languages and especially to “Zend-Avesta”.

Sorani counter-arguments were, according to Vanley, based on (a) its high musicality, (b) the simplicity of its grammatical rules, and (c) the progress it had made after thirty years of written literary experience in Iraq. He concluded that even “practically disregarding dialects advantages, it is IMPOSSIBLE to impose, under the present political conditions, one of them on the whole Kurdish people” (Vanley 1959: 7). Even on the regional level in Iraq where Kurmanji speakers are in a minority, Sorani should not be imposed on them (Vanley 1959: 8–9) for these reasons: (1) to avoid the injustice of depriving them of the right to have schools and newspapers in their dialect; (2) to make mass education possible; (3) the Kur-manji of Iraq is not an isolated dialect but part of the “biggest dialect on a thor-ough national level”; (4) to compensate, as a patriotic duty, for the suppression of the language in Turkey where Kurmanji alone is understood; and (5) to make pos-sible a progressive solution of the unification of the written language on a true national basis.

Vanley’s article, written in English, was published in Arabic translation in pamphlet form a few months later in 1960. It received an immediate response from Hazhar (1960) who noted, among other things, that (a) Sorani speakers did not intend to impose their dialect on the whole nation, (b) the officialization of Sorani was due to the more intensive nationalist struggle of Sorani speakers in Iraq and Iran and a result of relatively more advanced development of capitalist relations of production in these areas, and (c) numerical strength could not be a determining factor in the development of a language. Hazhar concluded that Sorani should not be imposed on Kurmanji speakers but, rather, the promotion of Kurmanji should be supported so that increased literary and cultural contact might facilitate the choice of one of the dialects (Hazhar 1960: 27) after national independence (i.e., a unified Kurdistan) is achieved.

Soon relations between the republican Iraqi government and the Kurds dete-riorated. Responding to a wave of repression by the government, some Kurdish leaders launched armed resistance in September 1961 and soon demanded au-tonomy within the borders of Iraq. This long, intermittent, autonomist war (1961–1991) provided an opportunity to see how the Kurds themselves deal with the dia-lect question without government interference. The most influential individual leader, Mulla Mustafa Barzani, a native speaker of Kurmanji, spoke in Sorani ap-parently because the Sorani element in the movement was more prominent. Kur-



Kurdish in Iraq 63

manji was, however, used in clandestine broadcasting and occasionally in the press alongside Sorani in the areas under Kurdish control. Correspondence and other written communication, training, and instruction in the schools were in Sorani. Kurmanji continued to be used marginally in the clandestine press of the political parties involved in the autonomous war.

Two different views were expressed on why Kurmanji was introduced into the mono-dialectal circumstances of Iraq in 1958. Mihamadi Mala Karim, a writer, literary critic and historian, blamed it, later in 1973, on the Kurds themselves. He noted that some patriotic intellectuals who had a hand in running the Baghdad radio station had simply made an honest mistake for which, later, the Kurdish nation had to pay a high price (quoted in Nebez [1976: 13]).

A similar view is held by Rasul (1971: 52–53) though he traces the “mistake” back to 1956 when a communist party clandestine newspaper, Azadî Kurdistan [Freedom of Kurdistan], began publishing articles in the two dialects “because the publishers and writers had not understood the truth of the problem” (Rasul 1971: 52). Rasul held himself “to some extent” and the poet Dilan “to a larger ex-tent” responsible. He emphasized that the leadership of the then clandestine Kurdish Democratic Party had, by contrast, a “scientific view” of the dialect ques-tion and considered Sorani the basis of the literary language.

Nebez (1976: 13) holds a different view on how broadcasting in Kurmanji be-gan. He claims that it was “a special political plan which the clique of [Prime Minister] Qasim executed with the support of some opportunist Kurds . . .” Ac-cording to Nebez (private correspondence, 24 April 1985), Kurmanji was being used in the Kurdish press before 1958 as a spontaneous rather than planned ef-fort. Some Kurmanji speaking journalists wrote in their dialect while a Sorani speaker like Giw Mukriyani occasionally printed Kurmanji material because of his nationalist urge for the equal treatment of all Kurds. Many others believed, ac-cording to Nebez, that writing in all the dialects would spontaneously and gradu-ally lead to the emergence of a unified language. The use of Kurmanji in Azadî Kurdistan was also, according to Nebez, part of this trend (private correspon-dence, 24 April 1985).

According to Nebez (private correspondence, 24 April 1985), the Kurdish Democratic Party supported instruction in Sorani in all the Kurmanji speaking areas. Hamza Abdulla, Secretary General of the Party, and himself a Kurmanji speaker, believed that writing in Kurmanji would divide Kurdish into two lan-guages. The Communist Party of Iraq opposed, according to Nebez, the intro-duction of Sorani into the Kurmanji areas because, they believed, dialect differ-ences were too many. The real aim of the communist party was, according to Nebez, competing with the [nationalist] Kurdish Democratic Party. Hence, the strengthening of the Kurdish nationalist movement was not in the interest of the



64 A. Hassanpour

Communist Party which had translated and distributed Vanley’s writing in de-fense of Kurmanji (private correspondence, 24 April 1985). According to a third, non-influential group led by Nebez, Sorani need not be imposed on Kurmanji speakers even though writing in all dialects would not lead to spontaneous unifi-cation. A prolonged scientific endeavor led by a language academy enjoying pop-ular support would be necessary for eventual unification (private correspon-dence, 24 April 1985).

According to Rasul (1971: 52–53), the divide-and-rule policy was readily adopted by the Iranian government in the state-sponsored paper Kurdistan where “instead of two parts [dialects] they built several”. This policy of “writing in five dialects” had already been repudiated a decade earlier by the Kurdish Students Society in Europe (Rojî Nö, Vol. 2, No. 3, June 1961: 29–34).

After 1958, Kurmanji was used more widely in the press, book publishing and broadcasting. Thus, while only two books were published in the dialect before 1958, a total of 25 titles appeared between 1958 and 1977 (Hassanpour 1992: 193). There has been at least one feature in Kurmanji in many periodicals published in Iraq since the 1970s. Sorani retained, however, its dominant position in all the functions of the Kurdish language (Hassanpour 1992: 167–348). The Kurdish sec-tion of the Iraqi Scientific Academy, which was a state organ under the Ba’th re-gime (1968–2003), did not show interest in the issue during its active years, 1973–1978, although the question of the unification of the dialects was discussed in one of its meetings (al-Basir 1984).

While the Iraqi Government was able to demote Sorani and deny Kurmanji official status, the Kurmanji speaking community had become increasingly de-tribalized by the 1960s and an intelligentsia eager to read and write in their native dialect was in the making during the 1950s and 1960s (Hassanpour 1992: 440–445). The available evidence from both contemporary Kurdish and offi-cial Mandate sources does not support Vanley’s claim, cited above, that the adoption of Sorani was a compromise between Turkey and Britain intended to protect the Kurmanji-speaking Kurds of Turkey from the repercussions of the linguistic freedom enjoyed by the Kurds of Iraq. In fact, Turkey and Britain/Iraq were in a state of war until the settlement of their dispute over the oil-rich Mosul province in 1926, and this (1918–1926) is a period of British support for the use of Sorani (Hassanpour 1992: 103–106). Actually, available evidence sug-gests that both Iran and Turkey tried to prevent the Kurds of Iraq from gaining language rights in any dialect. Moreover, Britain and France both aimed at re-stricting the progress of the Kurdish language (especially native language educa-tion) as a means of containing Kurdish nationalism, which they regarded as a threat to the territorial integrity of all the Kurdish-inhabited countries; all the states in the region as well as the US and Britain considered Kurdish nationalism



Kurdish in Iraq 65

a site of communist activism and a means for “Soviet incursion” into the Middle East.9

3.4 The conflict over dialects in the Kurdistan Regional Government

The US-led war on Iraq in 1991 led to the removal of Baghdad’s rule over major parts of Iraqi Kurdistan. The Kurdish political parties which had engaged in armed struggle for autonomy since 1961 took over the administrative structures left behind by the Ba’thist regime, and by 1992 conducted elections to form the Kurdistan Regional Government. The word “Regional” implied that it was still part of Iraq and not a secessionist entity.

Once Saddam’s power evaporated, a visible change was the replacement of Arabic by Kurdish in administration and education and the substitution of out-door Arabic signs and writings by Kurdish ones. However, this Kurdicization re-kindled the older dialectal cleavage between Sorani and Kurmanji. Badinan, the Western areas of Iraqi Kurdistan, chose to write outdoor signs (in shops, restau-rants, movie theatres, and other public spaces) in their own dialect, Kurmanji, while the eastern parts continued to use their Sorani dialect. This was a signifi-cant event: for the first time, Kurmanji speakers declared, in an informal or spon-taneous plebiscite, that they wanted to use their own dialect in all functions. Also, this event allowed, for the first time, the two sides to draw the borderline between their dialects. Travelling from the Sorani into Kurmanji areas or vice-versa, it is easy to see the border in every line of outdoor writing.

However, this fragmentation of the Kurdish speakers into two dialect areas was overshadowed, in the 1990s, by the hardships of the war-torn Iraqi Kurdistan and the civil war between the two major parties, Kurdistan Democratic Party and the Patriotic Union of Kurdistan, which had founded the KRG, and divided it between themselves.10 The second US war on Iraq in 2003, which toppled the Ba’thist regime, brought KRG once more into the framework of the Iraqi state but confirmed the status quo in which the Kurdish region retained its autonomous structure. Moreover, the new Iraqi constitution of 2005 recognized Kurdish as one of the two official languages of the country. These developments brought the dia-lect division to the fore again. While Iraqi Kurds, devastated by the genocidal campaigns known as Anfal (meaning ‘spoils of war’ in the Quran) in 1988, were

9 Evidence for this analysis is readily available in the British diplomatic correspondence about the Kurds, part of which is reproduced in Destani (2006).10 For a detailed survey of the political structure of the KRG, see Stansfield (2003).



66 A. Hassanpour

resettling in their villages and towns and the KRG was consolidating its rule, in-ternal contradictions began surfacing. For instance, the Kurdicizaton of educa-tion and administration in Badinan raised, once again, the question of dialects. The speakers of Kurmanji demanded, increasingly, that education and adminis-trative services be offered in their own dialect. This demand met resistance from some Sorani speakers. Very soon speakers of Hawrami, too, demanded language rights.

In early 2008, a group of 53 identifying themselves as “Kurdish writers, liter-ary figures, academics and journalists” published a petition addressed to the KRG president and cabinet, the parliament and political parties in which they de-manded the officialization of Sorani (Ĥesenpûr 2008).11 Their arguments in favor of Sorani were a repetition of those made in the 1960s as can be seen in the fol-lowing synopsis.

The petitioners claimed that the previous Iraqi regime had “split the Kurds through dividing their language” and that the “chauvinist discourse of some Arab and Turkish racists” denies the Kurds the ability to politically unite and standardize their language. However, Kurds have already standardized their language on the basis of Sorani (Middle Kurdish), which has evolved as the cultural and education language of the three largest provinces of Iraqi Kurdistan. No longer a “local” variety, Sorani has turned into a medium for philosophy and political and social sciences, and is the language of thousands of books and hundreds of periodicals, and already functions as a bridge between Kur-manji (Northern Kurdish) and southern dialects (South Kurdish). Moreover, Sorani has been the “language of the first Kurdish government,” that of Sheikh Mahmud, as well as the 1946 Kurdish Republic in Iran; it was also considered the official Kurdish language by the congress of teachers in 1959, a position en-dorsed by the leader of the Kurdish autonomist movement Mustafa Barzani as well as the Kurdistan Democratic Party in the wake of the March 1970 agree-ment between Baghdad and the Kurds. No nation can consolidate itself without a single-dialect official language. The role of non-official dialects is to enrich the official language.

Based on this line of reasoning, the petitioners claimed that Sorani was a de facto official language and it was the responsibility of the top leadership, i.e., the presidency of KRG, the parliament, and the council of ministers, to transform this de facto status into de jour. This can be done, the petitioners suggested, through legislating Sorani as the official language and establishing a national institution for studying and archiving Kurmanji and other dialects.

11 See the text and its background and analysis in Ghazi (2009).



Kurdish in Iraq 67

The petition led to considerable debate in the print and broadcast media and internet outlets; however, the KRG officials tried not to respond to this sensitive issue, although a few made their “personal position” known either in favor or against it. Faced with the counter-argument that Kurdish is, like Albanian, Arme-nian, or Norwegian, a bi-standard language (see, e.g., Hassanpour 1992: 435–437) and that both dialects should be official, the Soranists declared that they were concerned only with Iraqi Kurdistan, the jurisdiction of the only existing Kurdish government. This problematization allows the Soranists to raise the issue of majority-minority relations, and claim that Sorani speakers outnumber Kurmanji speakers, making it impractical to officialise a minority dialect. Still some Soran-ists expose their desire for the supremacy of their dialect. One claimed that speak-ers of Kurmanji in Turkey, too, should adopt Sorani as their national language.12 To sum up, Soranists present two main arguments: (1) Sorani is more advanced than Kurmanji because it has been, since 1920, the uninterrupted vehicle of edu-cation, publishing and administration, and has, thereby, evolved as a standard language; (2) the adoption of two dialects as the official language will divide the Kurdish nation into two pieces.

Soranists, however, have failed to substantiate the claim that their dialect has already scaled the heights of standardization. In fact both dialects have been undergoing standardization, and this process, in Kurdish as in other languages, is unending. Standardization, far from being a one-time event, is a process involv-ing de-standardization and re-standardization.13 Sorani’s main accomplishment has been reforming the alphabet (phonemizing the Arabic-based alphabet) and the vocabulary (purification and new coinages) with the former more successful

12 Shakely (2010), for instance, guesses that, among Turkey’s 12 to 15 million Kurds, there are no more than 12 to 15 thousand who can read in Kurdish and the sum total of writing in Kurmanji and in the Roman alphabet does not exceed 10% of the output in Sorani. He then suggests that the establishment of the KRG in Iraq “should have turned into an opportunity for the Kurds of Turkey to make use of the cultural and linguistic experience of Iraqi Kurdistan. In such a relationship, it is natural that an ignorant and uninformed people would make use of the experience of a knowledgeable and informed people. They could easily teach themselves [Sorani] Kurdish and gradually transmit the [Sorani] Kurdish language and culture and alphabet to the youth of Turkey’s Kurdistan and create a great cultural movement in this part of Kurdistan . . . and the Kurdish alphabet [the Arabic-based alphabet instead of the Roman alphabet used by Kurmanji speakers of Turkey] would thus turn into a symbol of resistance and libratory political struggle against the Turkish occupiers and be a return to the roots of Kurdish nationalism, language and culture.”13 For a study of “non-standardization”, “de-stanadardization” and “restandardization” in Germanic languages, see Linn and McLelland (2002).



68 A. Hassanpour

than the latter.14 The alphabet reform is, however, shared by both dialects. While orthography remains the most un-standardized realm of Sorani, it was codified in the Kurmanji of the Soviet Union.15 In the realm of vocabulary, both dialects have been subjected to extremist purification (except Kurmanji in the former USSR), which has seriously harmed their lexical development (for instance, both have focused on replacing Kurdicized loans with coinages; see book review, this issue); moreover, many new coinages in Sorani are shared by Kurmanji.

The Soranists also fail to grasp the implications of recent social, political and technological transformations for the changing linguistic landscape of Kurdistan. On the political side, Kurmanji speakers themselves have decided, in the absence of central state power, to use their dialect as a form of standard Kurdish, and have already made significant advances in making it the medium of instruction in primary and secondary schools. They see this as a question of language rights, and the KRG is not in a position to limit the use of their dialect except through coercion.

The ongoing revolution in communication technologies shapes the conflict over dialects. Kurmanji speakers are actively present in the social media and en-gage in new genres of writing ranging from texting to blogging. These media have, unlike print media, expanded the interactive use of language in unprece-dented ways. It would be unrealistic to expect the Kurmanji speaking younger generations to text and talk in Sorani. The internet has already opened, for Kur-manji and other dialects, new spaces for writing in new styles and genres. Even if Sorani is officialized, Kurmanji writers, speakers, singers, actors and their new reading, listening, and viewing publics will continue to use their dialect in do-mains that are outside the control of the KRG. While the traditional print media have been seen as architects of nation(alism) (see, e.g., McLuhan 1964: 155), it is often argued that new electronic media have a tendency to fragment audiences and nations.

Kurmanji speakers find it difficult to be persuaded by the Soranists’ claim that their dialect is more prestigious by reason of its massive publishing output.

14 It is difficult not to see Soranist mythifications of the history of the language when they see Kurmanji as an impoverished dialect left behind irreversibly by the stunning speed of Sorani progress. This great leap includes, according to one Soranist, the following: it is the “richest” dialect with “an extensive and rich literature” and uninterrupted literary development; it was the official language of the few Kurdish governments in the twentieth century; it has been the medium of education in Iraqi Kurdistan for ninety years; and it is “to a very large extent advanced in journalism and book publishing” with “great and important masterpieces compiled and many literary and scientific masterpieces translated into it” (Shakely 2010).15 By the early 1980s, lexicographer Bakaev (1983) had discussed theoretical issues underlying the codification of Kurmanji orthography and provided an orthographic dictionary.



Kurdish in Iraq 69

As noted earlier, Kurmanji boasts a much longer history in both print and broad-cast media. Kurmanji had full hegemony in print culture from the beginning to 1918 and the first radio station (in the Caucasus in 1925) and the first satellite TV channel (Med-TV, 1995) were launched in Kurmanji. Sorani, struggling to replace Arabic in science, higher education, administration and the courts, is not visibly ahead of Kurmanji and can hardly be a major attraction. After two decades of Kurdish self-rule, adequate teaching resources such as textbooks for tertiary edu-cation are not available.

Another significant change is the increasing ascendency of Kurmanji in the wake of the limited freedom speakers have gained in Turkey. Although Turkey continues its linguicidal policy, legal reforms since 1991 have allowed increasing publishing, broadcasting and film production in the dialect.

One development in the region is the relaxation of border crossings which has allowed the Kurds more mobility in travelling and settling in Iraqi Kurdistan. This is largely due to the political developments in the region since 1991 and the formation of the Kurdistan Regional Government. Another change is the increas-ing contact of these fragments due to nationalist struggles in Iraq, Iran and Tur-key, which has permitted more non-mediated, face-to-face contact among dialect speakers. Equally significant is the way satellite television and internet as well as the social media have evaded these borders which, according to one observer, used to “bleed” due to forcible division of the population (Kashi 1994). Soranists, however, seek to shield the borders around Iraqi Kurdistan. Under the banner of rejecting “pan-Kurdish” or “Greater Kurdistan” politics, they reduce Kurmanji speakers in Iraq to minority status, and thereby deny them language rights. Their stated goal is to unify the Kurdish nation through the creation of a single- standard, official language; their concern for national unity is, however, betrayed by reducing the “national language” to a local language less than one-fourth of its whole. It is more likely, however, that this project of nation-building through a single-dialect official language may lead to fragmentation as it happened in the Czech/Slovak case or the Serbian and Croatian context. This explains why the Kurdistan Regional Government has chosen not to take sides and, instead, main-tain the status quo as long as possible.

4 Discussion and conclusionsIn the linguistic landscape of Kurdish, geographical dialects have no predeter-mined or fixed status underwritten by their territorial expanse or numerical strength; this is a situation in which the geographical is political and the political is social; in other words, geographical dialects transform into politically and



70 A. Hassanpour

socially constituted entities. The larger entity, Kurmanji, loses its numerical and territorial supremacy and the smaller, Sorani, assumes ascendency and expands territorially. Here, the larger transforms into smaller and is treated by some Sora-nists as a “dialect”, while the smaller evolves as the “language” of the nation. At the same time, this situation, a product of the post-World War I geostrategic divi-sion of the Kurdish speech area, is being undermined, since 1991, by political developments which disturb the linguistic order of the past nine decades. Now, the Kurmanji speakers of Iraqi Kurdistan, smaller in number than Sorani speak-ers, demand official recognition of their dialect in education and administration. Equally significant is the trend of re-linking of the Kurmanji dialects of Iraq and Turkey and even Iran and Armenia.

If the ascendency of Sorani was a product of World War I geostrategic re-mapping of the region, new political developments such as the relaxation of linguicide in Turkey and the political mobilization of the Kurds in this country for the right to native-language education are bringing Kurmanji to the fore again. Today, the hegemony of Sorani in Iraq is challenged not only by its Kurmanji speakers but also by the revitalization of Kurmanji across the borders.

It is already discernible, in the landscape painted above, that the linguistic and the geographic intersect in political and social terrains where divergent ac-tors pursue conflicting interests. In pre-nationalist times, the Hawrami literary dialect, spoken by a small population in a mountainous enclave in Southern Kurdistan, assumed the role of a koine, i.e., a language used in a territory larger than its speech area, and turned into one of the literary languages of some Sorani literati. Today, however, the very fact of smaller number of its speakers thrusts the dialect’s pre-nationalist literary eminence into its geographic retreat and dimin-ishes the prestige it once enjoyed. Its current use in the internet, broadcast and print media is unlikely to help it shake off the fetters of geography. Even here, however, the dialect is not held back by the mere fact of geography or numerical weakness. That it achieved a prestigious status in the feudal regime of Ardalan and Baban principalities with mass illiteracy and no mass media but tends to fade away in media-saturated Kurdistan of the twenty-first century attests to the primacy of the social. In the feudal order, orality was the dominant mode of com-munication, and the tiny group of the literati was able to use any dialect or lan-guage for limited writing needs such as religion, literature and administration. In both cases, under the feudalism of the past and in its present mode of decline, the social underwrites the geographic and the demographic. In the ongoing conflict between the two major dialects, too, this complex intersection of the linguistic, the geographic and the social (in the broadest sense of the term) draws the socio-linguistic map of Kurdistan.



Kurdish in Iraq 71

The focus of this article has been on the conflict over the dialect bases of a language undergoing standardization. The changing status of each dialect was by and large an outcome of changes in the political destinies of the dialect speak-ers and their speech area, which were often beyond their control. No doubt, one change in status, in this case the use of Sorani dialect in administration, media and education in 1918, not only raised its status but also initiated alphabetic, lexical, semantic and grammatical codifications, which in turn enhanced its standing. In discussions of the status of the two dialects cited above (see Sections 3.1 to 3.4), advocates of one or the other dialect used benchmarks such as “elo-quence”, “unity”, “clarity” (grammar), “simplicity”, “richness” (vocabulary) and “musicality”. This study has not examined the (socio)linguistic “correlates”, if any, of such evaluation criteria. While linguistics has little to say about “elo-quence”, “musicality” or “unity”, these points of reference are sociologically and politically significant in so far as they help dialect speakers to align their forces in the battle over hegemony. At the same time, once set in motion, they act as sign-posts in the selection of norm and the codification of form or structure. In fact, in the current debate, Soranists appeal increasingly to the codification of form, especially phonemization of the alphabet or new terminologies and registers, as evidence of the progress of their dialect and its transition to the rank of (standard) language. While sociolinguistics and linguistics offer much in accounting for codified structures, codification is itself motivated by non-linguistic consider-ations. In the Kurdish case, for instance, terminological development and alpha-betic reform are projects of extremist purism, obviously a nationalist reaction to decades of linguicide, ethnocide, and genocide (Hassanpour 2005).

The officialization of Kurdish as one of the two state languages of Iraq in 2005 and the relaxation of linguicide in Turkey since 1991 have further politicized or, rather, polarized the dialectal divisions of the language. This polarization is in conflict with the aspirations of many Kurdish nationalists who equate the indivis-ibility of the nation with the indivisibility of the language. Looking at this contra-diction with an eye on the conflict between theory and practice, it is safe to state that if Kurdish remains fragmented dialectally, the “sciences of language” are no less divided. Advances in the study of language have led not to holistic theories of language but to more hyphenated fields.

ReferencesAbdulla, Jamal Jalal. 1980. Some aspects of language purism among Kurdish speakers.

Heslington, York: University of York dissertation.Bakaev, Ch. Kh. 1983. Osnovy Kurdskoi orfografii. Moscow: Izdatel’stvo “Nauka”.



72 A. Hassanpour

al-Basir, Kamil Hasan. 1984. Zimanî Netewayetî Kurdî [The Kurdish national language]. Baghdad: KZÊ Press.

Butters, Ronald. 1997. Dialectology and sociolinguistic theory. In Alan Thomas (ed.), Issues and methods in dialectology, 1–13. Bangor, Wales: Department of Linguistics, University of Wales Bangor.

Chomsky, Noam. 1965. Aspects of the theory of syntax. Cambridge, MA: MIT Press.Destani, Bjtullah (ed.). 2006. Minorities in the Middle East: Kurdish communities 1918–1974.

Slough: Archive Editions.Edmonds. C. J. 1925. A Kurdish newspaper: Rozh-i-Kurdistan. Journal of the Royal Central Asian

Society 12(1). 83–90.Edmonds, C. J. 1937. A bibliography of Southern Kurdish, 1920–36. Journal of the Royal Central

Asian Society 24(3). 487–497.Edmonds, C. J. 1957. Kurds, Turks and Arabs. London: Oxford University Press.Fishman, Joshua. 1973. Language and nationalism: two integrative essays. Rowley, Mass.:

Newbury House Publishers.Francis, Winthrop Nelson. 1983. Dialectology: an introduction. London: Longman.Ghazi, Hassan. 2009. Standardisation and the question of the Kurdish varieties: the language

debate in Iraqi Kurdistan. Paper presented to the international conference on The Kurds and Kurdistan: Identity, Politics, History, University of Exeter, 2–3 April. http://www.kurdishacademy.org/?q=node/582 (accessed 13 March 2011).

Great Britain. 1920–1931. Special Report by His Majesty’s Government in the United Kingdom of Great Britain and Northern Ireland to the Council of the League of Nations on the progress of ‘Iraq during the period 1920–1931. London: His Majesty’s Stationary Office. [Colonial No. 58, 1931].

Great Britain. 1931. Report by His Majesty’s Government in the United Kingdom of Great Britain and Northern Ireland to the Council of the League of Nations on the administration of ‘Iraq for the year 1931. London: His Majesty’s Stationary Office. [Colonial No. 74, 1932].

Great Britain. 1932. Report by His Majesty’s Government in the United Kingdom of Great Britain and Northern Ireland to the Council of the League of Nations on the administration of ‘Iraq for the period January to October, 1932. London: His Majesty’s Stationary Office, 1933.

Hafid, Sallahaddin. 2000. Abûrîî Kurdistan, 1977–1990 [The economy of Kurdistan, 1977–1990]. Slemani, Iraqi Kurdistan: Senterî Lêkołînewey Straticîî Kurdistan.

Hasanpoor, Jafar. 1999. A study of European, Persian and Arabic loans in standard Sorani. Uppsala, Sweden: Reports on Asian and African Studies 1.


Hassanpour, Amir. 2005. Kurds. In Dinah L. Shelton (ed.), Encyclopedia of genocide and crimes against humanity (Macmillan Reference USA), Vol. 2, 632–637. Detroit: Thomson Gale.

Haugen, Einar. 1966. Dialect, language, nation. American Anthropologist 68(6). 922–935.Hazhar, S. (pseudonym). 1960. Soranî ya Kirmancî: Wełamî “Mas’ala tawĥīd al-lugha al-

Kurdīyya” [Sorani or Kurmanji: A response to “The question of unification of the Kurdish language”]. Sulaymaniya: Zhin Press.

Ĥesenpûr, Emîr. 2008. Kurdî wek zmanêkî cût-standard [Kurdish as a bi-standard language]. Rojname [Silemani, Iraq] 206, April 24. 11.

Iraq Republic. 1959. Thawrat 14 tammûūz fī āmihā al-awwal [The July 14th Revolution in its first year]. Baghdad: Dar al-Akhbar Press.



Kurdish in Iraq 73

Iraq Republic. 1966. Al-ta’dād al-’āmm lil-ssana 1965: Al-natā’ij al-awwalīya [General census of 1965: preliminary results . . .]. Baghdad: Ministry of Interior, 12 January.

Jones, Gavin. 2001. Language nation. American Literary History 13(4). 776–788.Kerswill, Paul. 2004. Social dialectology/Sozialdialektologie. In Klaus Mattheier, Ulrich Ammon

& Peter Trudgill (eds.), Sociolinguistics/Soziolinguistik: An international handbook of the science of language and society, 2nd edn, Vol. 1, 22–33. Berlin & New York: De Gruyter.

Kashi, Ed. 1994. When the borders bleed: the struggle of the Kurds. New York: Pantheon.League of Nations. Permanent Mandates Commission. 1931. Minutes of the Twentieth Session

held at Geneva from June 9th to June 27th. Geneva: Series of League of Nations Publications, VI.A. MANDATES.

Linn, Andrew & Nicola McLelland. 2002. Standardization: studies from the Germanic languages. Amsterdam: John Benjamins.

MacKenzie, David N. 1961. The origin of Kurdish. Transactions of the Philological Society. 68–86.

MacKenzie, David N. 1986. Kurds, Kurdistan v. – language. The Encyclopaedia of Islam 5. 479–480.

McLuhan, Marshall. 1964. Understanding media: the extensions of man. New York: Mentor Books.

Millar, Robert M. 2005. Language, nation and power: an introduction. Basingstoke: Palgrave Macmillan.

Najm al-Din, Ahmad. 1970. Ahwal al-sukkan fi al-’Iraq [The condition of population in Iraq]. Cairo: al-Jaylawi Press.

Nebez, Jamal. 1976. Zimanî yekgirtûy Kurdî [The unified Kurdish language]. Bamberg, Germany: National Union of Kurdish Students in Europe.

Nêrevan. 1959. Notes sur la presse kurde d’Iraq. In Orient, 3e trimestre, 10. 139–148.Rasul, ‘Ezaddin Mistafa. 1971. Serincê le zimanî edebîî yekgirtûy Kurdî [A glance at the unified

literary Kurdish language]. Baghdad: Salm~n al-’A’zamî Press.Shakely, Ferhad. 2010. Zimanî Kurdî le astaney serdemêkî tazeda [The Kurdish language in the

threshold of a new era]. In Hewler. No. 3, 2010-01-06, posted in http://www.hawler.in/more.php?id=3039 (accessed 8 November 2010).

Simon, Artur & Urlich Wegner (eds.). 2000. Music! 100 Recordings. 100 years of the Berlin Phonogramm-Archiv. 1900–2000. Mainz, Germany: Wergo.

Sluglett, Peter. 2007. Britain in Iraq: contriving king and country. London: I. B. Tauris.Sousa, Ahmad. 1953. Atlas of Iraq. Baghdad: Surveys Press.Stansfield, Gareth R. V. 2003. Iraqi Kurdistan: political development and emergent democracy.

London: Routledge-Curzon.Todd, Terry Lynn. 1985. A grammar of Dimili (also known as Zaza). Ann Arbor, Michigan:

University of Michigan dissertation.Vanley (Vanly), Ismet Cheriff. 1959. The question of the unification of the written Kurdish

language: Kurmanji or Sorani? Kurdistan (published by Kurdish Students Society in Europe) November. 5–10.

Zaki [Beg], [Mihamad] Amin. 1935. Dû teqellay bêsûd [Two fruitless efforts]. Baghdad: Mariwani Press.



http://www.hawler.in/

DOI 10.1515/ijsl-2012-0050 IJSL 2012; 217: 75 – 98

Desmond FernandesModernity and the linguistic genocide of Kurds in Turkey

Abstract: Zygmunt Bauman, Alexander Laban Hilton and Paul Havemann, amongst others, have argued that genocide is intimately linked to modernity. Modern discourses on development, modernization and western science as well as key meta-narratives of modernity (advancing the teleological myth of progress and civilization), “gardener’s visions” and the very categorization and standard-ization of national languages (crucial to the biopolitical formation of global pop-ulations under the system of modern nation-states) have all legitimated and ef-fected policies and practices that have been genocidal in their nature and scope. This article examines and details the extent to which all these identified aspects of modernity can be observed in the case of Turkey. The findings indicate that linguistic/cultural and physical genocide of Kurds in Turkey has taken place (over the past eight and a half decades) as a direct consequence of the Kemalist/Ataturkist modernity project. Language policy – which has advocated linguistic imperialism alongside linguistic genocide – has been a critical tool for the cre-ation of the modern Turkish nation-state.

Keywords: Kurds; modernity; genocide; triage

Desmond Fernandes: The Campaign Against Criminalising Communities. E-mail: [email protected]

1 IntroductionIn recent years, a number of scholars, drawing upon a range of case studies and wider structural analyses, have engaged in a number of debates and concluded that “genocide” – inclusive of linguistic genocide – “and modernity are closely interwoven” (Hinton 2007: 420). They have argued that modern discourses on de-velopment, modernization and western science as well as key meta-narratives of modernity (advancing the teleological myth of progress and civilization), the “gardener’s vision” (Bauman, cited in Noakes [2010: 1]) and the very categoriza-tion and standardization of national languages (crucial to the biopolitical forma-tion of global populations under the system of modern nation-states) have all

Brought to you by | University of North DakotaAuthenticated


76 D. Fernandes

legitimated and effected policies and practices that have been genocidal (inclu-sive of linguistically genocidal) in their nature and scope.

These studies are of considerable relevance to linguists who are engaged in linguistic human rights advocacy and who are seeking to analyze how and why linguistically genocidal policies, educational programs and practices have been conceptualized and implemented (and often legitimized) by nation states as part of wider cultural and physically genocidal plans to westernize, develop, modern-ize and civilize societies. This paper presents the key conceptual findings of Visvanathan (1988), Solomon (2010) and Havemann (2005) and integrates them for the first time into a case study analysis of the genocide of Kurds in modern Turkey.

The findings of this study emphasize the manner in which genocidal (inclu-sive of linguistically genocidal) processes in Turkey against the Kurdish Other have not been accidental by-products of the state’s modernity project: they have been central aspects of the drive to transform society. Linguistic genocide, in this sense, is analyzed within the wider context in which Kurds have been geno-cidally targeted. The presentational style that has been adopted – which is used in a number of journals, papers and academic publications by scholars such as Bourke (2000), Ahmed (2003), Shoup (2006), Banerjee (2007), Laing (2008), Zeydanlıoğlu (2008, 2009), Uçarlar (2009) and Skutnabb-Kangas (2010) – extensively draws upon selective and block quotations to highlight key findings.

This study emphasizes the value and relevance of reflecting upon the manner in which the linguistic genocide (alongside other forms of cultural and physical genocide) of Kurds in modern Turkey has taken place at a time when, all too distressingly:– The Turkish state still persists in branding such debate as thought crime

(Fernandes 2010a).– Many visiting as well as resident genocide scholars, linguists, journalists,

academics, MPs, editors, publishers and human rights analysts in Turkey have been reluctant to even address the Kurdish modernity and/or genocide (inclusive of the linguistic genocide) question due to the oppressive situa-tion that exists (Beşikçi 2009: 2; Fernandes 2007, 2010a).

– People who have dared to engage in such thought crime have found them-selves being removed from their university posts in Turkey or (if they are visiting Turkey) denied entry to the country or detained, deported and subjected to harrowing, lengthy and expensive court cases, criminalization, death threats and/or even murder by state inspired forces (Fernandes 2007, 2010a, 2010c; KHRP 2010). As Rafferty (2005: 1) confirms: “The ‘Kurdish question’ – otherwise known as the genocide of . . . Kurds – is one of the most contentious issues in Turkey today.”



Linguistic genocide in Turkey 77

– Several publishers in Turkey, fearful of criminalization, have exercised self-censorship and mutilated a number of international texts (when translating them into Turkish) that address these concerns, thereby depriving readers (including linguists) from accessing key findings that could be of use to their future work (Üngör 2007; Fernandes 2010a).

– Many Turkish universities, fearful of state targeting, have continued to stifle academic debate in these areas and barred a number of students from researching these subject areas (Lofti 2007: 1; Fernandes 2010a).

– “The physical, linguistic, and cultural genocide committed by Turkey against the Kurds” is generally “treated with silence and/or considered controversial. The status of the Turkish government in denying their actions has created pressure on the United States and other Western Nations governments, universities, and media organizations to treat this holocaust as delusions of the Kurdish people” (Swartz 2007: 1).

– Herman and Peterson (2010: 88) have additionally identified “a remarkably deep” ideological “bias but also a consistent, even a rigid one over a long period of time” that has resulted in genocide not really being significantly addressed or debated in much of the US mainstream press, as far as “Turkey’s treatment of its Kurds in the contemporary period” is concerned.

2 Key terms and definitionsKey terms need to be defined at the outset of the study. Modernity can “best be described as a set of interrelated processes” that “characteriz[e] the emergence of ‘modern society’. Politically, modernity involves the rise of secular forms of government . . . Economically, [it] refers to capitalist expansion and its derivatives. . . . Socially, [it] entails the replacement of ‘traditional’ loyalties” with “ ‘modern’ ones” and culturally, it “encompasses the movement . . . to an emphat-ically secular and materialist worldview” which, “in many ways”, was “epito-mized by Enlightenment thought” (Hinton 2002: 7, 8). “Social scientists”, accord-ing to the Encyclopedia of Science, Technology and Ethics (2005, cited in Book Rags [2006: 1]), “describe modernity as a particular form of culture or society de-pendent on and supportive of science and technology, with the process of creat-ing such a society defined as modernization. . . . Modernization is a . . . term for a concept known in the nineteenth century as the ‘civilizing’ process, and during the first half of the twentieth century as ‘Westernization’ ”.

Genocide is defined in this paper along the lines set by the 1948 United Na-tions Genocide Convention, as well as by Lemkin when he originally coined the term:



78 D. Fernandes

By ‘genocide’ we mean the destruction of a nation or of an ethnic group. . . . Generally speaking, genocide does not necessarily mean the immediate destruction of a nation, ex-cept when accomplished by mass killings of all members of a nation. It is intended rather to signify a coordinated plan of different actions aiming at the destruction of essential founda-tions of the life of national groups, with the aim of annihilating the groups themselves. . . . Genocide has two phases: one, destruction of the national pattern of the oppressed group; the other, the imposition of the national pattern of the oppressor. (Lemkin 1944: 80).

In sociolinguistic terms, Lemkin identified the following acts as genocidal when assessing Nazi policies and practices:

In the incorporated areas . . . local institutions of self-government were destroyed and a German pattern of administration imposed. Every reminder of former national character was obliterated. Even commercial signs and inscriptions on buildings, roads, and streets, as well as names of communities and of localities, were changed to a German form. . . . Nationals of Luxemburg . . . were required to assume in lieu thereof the corresponding Ger-man first names; or, if that is impossible, they must select German first names. . . . The de-struction of the national pattern in the social field has been accomplished . . . by German-ization of the judicial language and of the bar. . . . The local population is forbidden to use its own language in schools and in printing. According to the decree of August 6, 1940, the language of instruction in all Luxemburg schools was made exclusively German. (Lemkin 1944: 83, 85)

Linguistic genocide is “ ‘prohibiting the use of the language of the group in daily intercourse or in schools, or the printing and circulation of publications in the language of the group’. This was how linguistic genocide was defined in Article III(1) of the final draft of what became the [Genocide] Convention” (Skutnabb-Kangas 2000: 1). Although this article was voted down for questionable political reasons when the Convention was finally accepted, “those states then members of the UN were in agreement that this was how the phenomenon could be de-fined” (Skutnabb-Kangas 2000: 1). Policies of assimilation – aimed at eradication of indigenous/minority education – which “linguistically, often also culturally” result in transference “to the majority group”, can also be held to be genocidal, according to Articles II(e) and II(b) in the present convention (Skutnabb-Kangas 2000: 1) and Lemkin (Docker 2004: 13).

Ethnocide, as defined by Lemkin (who coined the term alongside genocide) and several other scholars, is often held to be synonymous with the term and phenomenon of genocide (Lemkin 1944; Lukunka 2007). “Constructive genocide” is defined in the following manner:

In some instances, racism becomes so dangerous and extremist that it becomes directed against the very existence of a people – nationally, ethnically, and culturally, and thus par-takes of some of the attributes of genocide without the direct acts of [physical] annihilation.




Racism . . . which denies the very existence of its victims, can safely be termed, in law, ‘con-structive genocide’. When a people . . . are not recognized as existing, when they are denied their homeland, their national existence and identity, and the basic rights and fundamental freedoms accorded to other peoples – what, in such circumstances, remains of them and for them as a people? They become non-people and the individuals non-persons. Is this not in effect . . . constructive genocide? (Al-Qasem 1977: 13)

Phillipson defines linguistic imperialism as:

. . . a theoretical construct, devised to account for linguistic hierarchization, to address issues of why some languages come to be used more and others less, what structures and ideologies come to be used more and others less, what structures and ideologies facilitate such processes, and the role of language professionals. . . . Linguistic imperialism is a sub-type of linguicism. . . . Linguistic imperialism takes place . . . where language interlocks with other dimensions, cultural (particularly in education, science and the media), economic and political. (Phillipson 1997: 238–239)

3 Modernity and genocide: theoretical considerations

“Modernity is fundamentally about conquest, ‘the imperial regulation of land, the discipline of the soul, and the creation of truth’ (Turner 1990: 4), a discourse that enabled the large-scale regulation of human identity” – even in a linguistic sense – “both within Europe and its colonies” (Ashcroft et al. 2000: 145). Such regulation and standardization was often effected in the name of promoting west-ern science, the nation state, modern empires, civilization and progress. Alvares (1988: 32) argues that this application of reductionist science sought structurally to “reduce . . . diversity by eliminating it, and introducing more simplified, mech-anized designs instead. . . . The process of elimination . . . [in] . . . the domain of language” was, consequently, all too often rationalized and undertaken “to stan-dardize the popular language and eliminate anarchy in the domain of the peo-ple’s speech” and the people’s use of ‘other’ languages (Alvares 1988: 32). “Most significantly, all ordinary experience” was to be “recast in the ‘official language’, stamped with official approval, to be considered worthy of human use” (Alvares 1988: 32).

To Devy (2009: 46), “the rhetoric of modernity and the ideology of progress boiled down to what Gandhi called violence”. The resulting “ethical framework surrounding technological civilization”, when not countered or effectively ques-tioned, has lead to “nature and people who are termed backward” – often, for using unofficial and/or unrecognized languages and embracing outmoded values



80 D. Fernandes

and belief systems – being “routinely destroyed. . . . Modern civilization is based on unrestricted violence” (Devy 2009: 46). For Visvanathan:

Theories like racism in anthropology, orientalism in linguistics, IQ in psychology, social Darwinism in political economy and biology, are bracketed off as ‘pseudo-sciences’ or as distortions of normal [modern] science. I suggest an alternative explanation. The maraud-ing genius of science needs these spaces – these ‘pseudo-sciences’ – for the free play of its imagination. This collective unconsciousness of science constitutes an integral part of the scientific experiment. Marking it off saves science as a phenomenon but contributes little to our understanding of it. It does not explain why these theories so often recur in science. One can see the same trend in the modern discourse on development. Development should be regarded as a [modern] scientific project. It represents the contemporary rituals of the labo-ratory state. As a project, it is composed of four theses, ingrained in the logic of western science, of modernity as technocracy. One can call them:1. The Hobbesian project, the conception of a society based on the scientific method;2. The imperatives of progress, which legitimize the use of social engineering on all those

objects defined as backward or retarded;3. The vivisectional mandate, where the other becomes the object of experiment which in

essence is violence and in which pain is inflicted in the name of science;4. The idea of triage, combining the concepts of rational experiment, the concept of obso-

lescence and of vivisection – whereby a society, a subculture or a species is labeled as obsolete and condemned to death because rational judgment [by those overseeing the modern nation-state] has deemed it incurable.

Development as a [modern] technocratic project includes all four themes. (Visvanathan 1988: 258–259)

As he argues:

In fact, if concepts could ever be death warrants, the above glossary could be regarded as genocidal. . . . Lurking quietly within modernity-as-a-scientific-project is the idea of triage. . . . If progress demands the summoning of the Other into modernity, triage is the dispensing with of the Other. . . . Societies and cultures are now being destroyed because they are considered refractory to the scientific gaze. . . . The western encounter with the other ends in its eventual logic as erasure. . . . Science [in this context] has no place for the defeated except as objects of an experiment. . . . Social triage . . . is a deliberate decision or act of a state to define a target group such as a minority within its territory as dispens-able. The decision, however, must also be articulated on rational grounds. For, though tri-age is genocide, it involves the rational imposition of death on those regarded as refractory to the scientific gaze. It is in this sense that the term helps us to understand the particular quality of violence of which scientific rationality is capable. (Visvanathan 1988: 259, 271, 272)

For Visvanathan, “the nation-state cannot permit ethnicities which serve as com-peting sites for power. . . . The Hobbesian project which encapsulates modernity as a creation myth was literally a contract between state and science to manufac-




ture the idea of a mass society of equal and uniform individuals. Modern society was monocultural in more ways than one” (Visvanathan 1988: 276). With the re-alization of such “projects”, as May (1999: 1, 2) has observed: “Not surprisingly, education – as a key institution of the nation-state – has played a central part historically in the subjugation of indigenous languages and cultures and the re-lated assimilation of indigenous peoples into the dominant or ‘common’ language and culture of the nation-state”. “In the process, indigenous languages and cul-tures were specifically proscribed, demeaned and diminished” – indeed, often subjected to linguistic/constructive genocide and linguistic imperialism (Fer-nandes 2010a) – “by the state via its education system. . . . Consequently, indige-nous languages” – when acknowledged as “existing”, that is – “and cultures [often] came to be constructed as antediluvian and unnecessary in the modern world – a vestige of ‘primitive’ cultures best left in the past. In contrast, ‘national’ languages and cultures – or, more specifically, the languages and cultures of dominant ethnic groups – were viewed as the apogee of modernity and progress” (May 1999: 1, 2).

For Solomon (2010: 44): “Historically-speaking, it goes without saying that language policy has been a critical tool for the creation of the modern nation-state and a constant site of state intervention”. Indeed, “in what has virtually been a universal process”, modern nation-states have “established themselves linguistically by the elimination of difference through standardization – along with the concomitant displacement of minority populations and the appropria-tion of minority lands” (Solomon 2010: 44). And “education”, as Alvarado (2010: 1) observes, is such that it “plays a vital role in shaping both language standard-ization and its primacy over alternative language use”.

For Solomon (2010: 45), “looking at the history of modern linguistic trans-formation, postcolonial writers have shown not only how the colonial and post-colonial state mobilized language in the creation of ‘invented traditions’, but also how the establishment of national literary and linguistic traditions . . . in metropolitan social formations originated as a technique of colonial governance” – in which other languages were often subjected to linguistic imperialism and linguistic/constructive genocide. To Havemann (2005: 57, 59), colonization is “a key feature of modernity” in which indigenous peoples:

. . . have been [perceived as] chronic obstacles to modernization to be overcome by whatever means – typically by violence concealed behind liberal legalities. . . . Modernity [in this context] . . . generates waste: both the physical detritus of industrialisation . . . and those human beings who impede the level of growth and degree of order required. For centuries, such people have been disposed of on a genocidal scale. The survival of the modern form of life depends on the proficiency and dexterity of the techniques for waste disposal of both kinds. (Havemann 2005: 57, 59)



82 D. Fernandes

Consequently, in its imperial/paleo-imperial/colonial context, “the law defines both the citizen’s bundle of rights and the excluded’s absence of rights. . . . [The excluded] occupy a zone of exception wherein the sovereign suspends its law’s protection from them and their land or lives may be taken with impunity” (Have-mann 2005: 59, 60). Torture, incarceration (in camps and prisons), deculturation, forced resettlement, forced cultural transfer of children through educational lan-guage programmes, cleansing and/or disappearance within a genocidal context become rationalizing instruments that are used by modern vivisectionist, labora-tory states.

Modernity related “discourse[s] dehumanize potential targets and construct them as “enemies of the state”, “terrorists” and “traitors”, effectively placing them outside of the state’s protection and denying them their citizenship rights” (Zeydanlıoğlu 2009: 5). Technologies tied in to the western, colonizing, moder-nity project – railways, dams, telegraph systems, aircraft, poison gas – have also been used to crucially advance quite specific genocidal agendas (Fernandes 2010a). To Hinton:

European [modernity linked] expansion was largely driven by a desire for new lands, con-verts, wealth, slaves, and markets, [and] some scholars refer to the resulting annihilation of indigenous peoples as ‘development’ or ‘utilitarian’ genocides. . . . This devastation was le-gitimated by contradictory discourses that simultaneously asserted that the colonizers had the ‘burden’ of ‘civilizing’ the ‘savages’ living on their newly conquered territories and that their deaths mattered little since they were not fully human.

Metanarratives of modernity supplied the terms by which indigenous peoples were con-structed as the inverted image of ‘civilized’ peoples. Discourse about these ‘others’ was fre-quently structured by a series of value-laden binary oppositions (see also Bauman 1991; Taussig 1987): modernity/tradition, civilization/savagery, us/them, centre/margin, civilized/wild, humanity/barbarity, progress/degeneration, advanced/backward, developed/underdeveloped . . . . Maybury-Lewis [2002] describes how the inhumane and genocidal treatment of indigenous peoples was often framed in meta-narratives of modernity, particu-larly the notion of ‘progress’. (Hinton 2002: 9, 10)

Certainly, it needs to be recognized that, within these types of ideological frame-works and planning contexts, cultural destruction even in the post-1945 period has become an accepted key process that has often been advocated in par-ticular modernization-linked development programmes. Escobar (1994: 4) has revealed the way in which “one of the most influential documents” of the post-Second World War period on “development”, prepared by a group of experts con-vened by the UN with the objective of designing concrete policies and measures for the economic development of underdeveloped [“Third World”] countries, “suggested” – indeed, factored in – “no less than a total restructuring of ‘un-




developed’ societies”. It clarified that: “There is a sense in which rapid economic progress is impossible without painful adjustments. Ancient philosophies have to be scrapped. Old social institutions have to disintegrate. Bonds of caste, creed and race have to burst, and large numbers of persons who cannot keep up with progress have to have their expectations of a comfortable life frustrated” (Escobar 1994: 4).

As Visvanathan (1988: 277) has argued: “Underlying modernization is a sub-stratum of intolerance. The variegated traditions of . . . the nomadic, the tribal, the pastoral and the peasant” conceptually and practically “have to be bulldozed into a flatland called modernity and there is little time for consultation. . . . To the laboratory state, these people are . . . ethnics practising styles of life . . . which are refractory to science”. Consequently:

According to the logic of development, they must either acculturate or disappear. . . . In the process, [allegedly] peaceful development [of this kind] has created more refugees. . . . What we are in fact confronting here is development as slow genocide. . . . Intrinsic to all such technocratic projects is the idea . . . [that] a mechanical scheme [can be] . . . imposed on a culture without any consideration for the traditions of the community. . . . [And with this] . . . large dams literally become experiments on the people. . . . The technology of most large dams is basically vivisectional. . . . For the scientist-technocrat, the development of all land is inevitable. . . . The movement from cultural destruction through obsolescence to triage-as-erasure is a short step. (Visvanathan 1988: 277, 278, 279)

For Visvanathan (1988: 280), then, the example of “the elimination of the Ache Indians” in Paraguay “has raised in a fundamental way the problem of genocide through development. The process of resettlement, involving slow death through deculturation, does fall within the clauses of the Genocide convention. Item three of the Genocide convention of the UN includes: ‘Deliberately inflicting on the group conditions of life calculated to bring about its physical destruction in whole or in part’ ”. Consequently, “to remove tribal people from their natural habitat would be cultural ethnocide. . . . The fact [is] that the laboratory state now deems certain cultures dispensable” (Visvanathan 1988: 280) and, in this respect, “the notion of calculated dispensability, of erasing people from the commons of the world” (Visvanathan 1988: 280) becomes a rational, modernizing bureaucratic/accounting consideration (Neu and Therrien 2003). Rojas (1996: 1), in reviewing Rostow’s 1960 “modernization theory” – which has so influenced Turkey’s Cold and post-Cold War development-cum-counter-insurgency programme in the pre-dominantly Kurdish East (Fernandes 2010a) – has observed the manner in which it conceptualizes action against opponents of its top-down development vision. “Social disturbances” which may take place in “the form of peaceful agitation, political violence, nationalism, revolution or guerrilla warfare” are perceived to



84 D. Fernandes

be anti-progress and inimical to the interests of the modern, western state. “It follows that crushing human beings involved in these social disturbances” – even in a genocidal sense, if need be (Fernandes 2010a) – “takes the form of ‘humani-tarian actions’ to preserve social order and social peace to maintain the balance [of the] family-civil society-state” (Rojas 1996: 1). Maybury-Lewis, moreover, con-cluded in 2002 (cited in Dean 2009: 1069) that all too often “[nation] states feel they cannot modernize effectively if they tolerate indigenous cultures in their midst”.

4 The modernity project in Turkey and the genocidal targeting of Kurds

It is important to appreciate that Kemal (Atatürk), the founder of modern Turkey, together with the ruling elite, embarked almost immediately upon a modernity-driven project that was inspired by US and European imperialist, western civiliza-tional experiences (Fernandes 2010a). “Turkish modernity, in its Republican movement”, certainly “arrogated to itself enlightenment values of rationality, progress and universality. So, for example, the 1931 statutes of the Republican People’s Party (the only party in the parliament)” at the time, “stated that: ‘The party has accepted the principle that all laws, regulations and procedures used in the administration of the state should be prepared and implemented . . . in ac-cordance with the foundations of and the forms of science and technology in modern times’ ” (Houston 2001: 89). Houston (2001: 89) clarifies that, “in its bid to create civilization”, the state “took upon itself the task of liberating the people from tradition”.

For Bart (2005: 1), “ethnocide quickly became the raison d’être of . . . the mod-ern Turkish state. . . . Not only have successive Turkish governments denied the genocide they perpetrated against Armenians, Aramaeans and Assyrians but they have attempted to destroy the Kurdish nation in every conceivable way bar total [physical] extermination. . . . Turkey lives in a permanent . . . republic always bordering on insanity and genocide”. As I have noted elsewhere: “Kemal em-barked on a genocidal course in order to effect greater Turkish unity, reinforce his own authoritarian power base, and realize hyper-nationalistic, colonialist and modernist ambitions. The same methods have been applied elsewhere against the Other – as ‘a basic mechanism of empire and the national state’ (Simpson 1993: 3, 4)” (Fernandes 1998: 66). The modernity project itself, as numerous ana-lysts have shown, also drew inspiration in places from the earlier Ottoman/CUP and War of Independence period (Fernandes 2010a; Hobsbawm 1989).




Concerning the nature of this project, Atatürk is recorded as emphasizing that “the Turks have only ever gone in one direction – towards the West (cited in Spiegel 2010: 1) . . . For everything in the world – for civilization, for life, for suc-cess, the truest guide is knowledge and science (cited in Kinzer 2001: 36). . . . Gentlemen, uncivilized people are doomed to be trodden under the feet of civi-lized people. . . . We will live as a progressive and civilized nation in the arena of civilization” (cited in Zeydanlıoğlu 2008: 155, 160), where “we will take science and technology from wherever it is and insert it in the head of each member of the nation. There is no restriction and condition on science and technology. . . . If [ignorance] is not eliminated, we will stand on the same spot. If something is standing on the same spot, this means that it is going backwards” (Atatürk, cited in cited in Zeydanlıoğlu [2008: 160]).

For Conversi (2006: 326), Atatürk “could only conceive development as utter, remorseless and complete Westernization”. As far as “the founders of the Turkish Republic” were concerned, “the European . . . experience of the past century was central to their project. . . . Emphasis was given to developing a sense of nation-hood based on the Turkish language” (Kirisci 2004: 276). Indeed, Tachjian (2009: 2) concludes that “it was just as important to ‘Turkify’ it economically, linguisti-cally and demographically as it was to liberate the land. Indeed, the aim of the leaders was to establish a [modern] nation-state that was based exclusively on Turkish identity. Consequently, the presence of other ethno-national groups, the question of their cohesion and investment in the development of their commu-nity became insupportable”. Colonial genocides against Kurds, Armenians, As-syrians, Chaldeans, Syriacs, Greeks and Others have, not surprisingly, taken place (Fernandes 2010a, 2010b, 2010c).

For Üngör:

. . . the key discursive devices which the Kemalist centre employed to represent their rela-tionship with the Kurdish periphery was ‘civilization’. . . . The non-Turkish population of the eastern provinces was looked down upon as primitive and inferior, fit [only] for colonial rule by a Turkish [westernized] master nation which operated in the name of progress and ratio-nality. They were viewed, moreover, as inherently treacherous and anti-Turkish and hence threats to security against which Turkish state and army personnel had to be permanently on guard. (Üngör 2008: 32)

“Forging a uniform nation out of the heterogeneous Ottoman population meant official intolerance to differences. Intolerance meant either forceful assimilation” – i.e. through cultural and linguistic/constructive genocide (see Fernandes 1998, 2007, 2010a) – “or clearance of different groups that resisted assimilation” (Ergil 2009: 1). Clearance, as Musa Anter and others have clarified, often meant physical



86 D. Fernandes

genocide of Kurds (Fernandes 2007). “So assimilation followed by eviction and decimation of the ‘others’ became inbuilt characteristics of the republican re-gime” (Ergil 2009: 1).

“It was the discourse of westernization/modernization”, Mesud Yeğen (1999) contends, “that led the Republic to perceive” – and de-legitimize – “the Kurd-ish resistance as a resistance of pre-modernity and brigands” (Uçarlar 2009: 117). Atatürk, indeed, is recorded as justifying repressive – indeed, genocidal – state action in the following way: “Could a civilized nation tolerate a mass of people who let themselves be led by the nose by a herd of shaykhs, dedes, sayyids, chelebis, babas and amirs?” (McDowall 1996: 196). The physical existence of Kurds, moreover, could even be rejected and denied by Turkish state-discourse though the application of modern pseudo-scientific theories (Fernandes 2010a).

“After the elimination and forced removal of the Kurdish elites from the East”, using such rationalizations, “the Kemalists saw the remaining Kurdish population . . . as ‘raw material’ for the Turkish nation” (Üngör 2008: 33) to be disposed of at will (Nezan 1993; Fernandes 1998, 2010a). The Turkish constitu-tion, moreover, consecrated “Kemal’s voluntarist fiction, according to which Turkey is strictly Turkish” (Chaliand 1994: 30). As the modern state “was founded on an extreme fascist kind of nationalism, like that of the Nazis” in some re-spects (Hayri, as cited in Akturk et al. [2001: 479]) and that of Mussolini in Italy (Tirman 2005; Beşikçi, as quoted by Van Bruinessen [2005: 28]), impulses that made genocide a conscious strategy amongst the Kemalist elite became even more pronounced (Fernandes 2010c). Orientalism and other western pseudo-scientific theories justified and inspired genocidal assaults against the Other. For Zeydanlıoğlu (2008: 159): “The Kemalists took on what I call the ‘White Turkish Man’s Burden’ in order to carry out a civilizing mission”:

The making of the Turkish nation went hand in hand with the “forgetting, postponing and canceling” of the Kurdish ethnic identity ([Mesud] Yeğen 1999: 120) . . . and the suppression of the Kurdish ethnic identity was made possible by a state knowledge-production that re-lied on European Orientalist constructs and racial theories. Etienne Copeaux (1998: 52) has underlined that Turkish historiography and linguistics are ‘children of Western Orientalism; they are its products’. (Zeydanlıoğlu 2008: 161, 162, 163)

In practical terms, “Turkish Orientalism was crystallized” in Kemalist pseudo-scientific theories: “These theories were disseminated widely throughout society, especially in school textbooks, and still continue to influence the discourse of Turkish nationalism today” (Zeydanlıoğlu 2008: 164). The application of torture – in a genocidal context (Fernandes 2010a) – has also been “directly linked to the making and maintaining of Turkey as a homogenous nation-state of Turkish speakers” (Zeydanlıoğlu 2009: 76).




In terms of the manner in which metanarratives of modernity supplied the terms by which indigenous Kurdish peoples were constructed as the inverted im-age of “civilized” peoples, Ayce Akturk has detailed the manner in which “Kurds were called kuyruklu kurt, meaning ‘Kurd with a tail’, or kiro, meaning ‘uncivi-lized, uneducated, rude, worthless, one who knows nothing’. Meanwhile”, the Kemalist inspired, modern mass “education system and media indoctrinated people that Turks were the heroes of the world” (Akturk et al. 2001: 479, 480). With the Republic becoming “the great storyteller of the nation, sponsoring the grand narratives of nationalism, independence and secularism”, Kurdish dis-courses “were cast as villains” (Houston 2001: 89). “During the Tunceli [Dersim] rebellion”, unsurprisingly, “it was said: ‘What the Republican regime has been doing in Tunceli’ ” – i.e. its linguistic/cultural and physically genocidal assault against Kurds (Socialist Party of Kurdistan [PSK] 2008; Fernandes 1998, 2010a, 2010c) – “ ‘is not a military operation, but the march of civilization’ ” (Mesud Yeğen 1999: 560). “The Turkish position was that these ‘primitives’ and ‘bandits’ should give way to modern civilization, just like the American Indians had. This should be effected by their assimilation to the supposedly superior Turkish cul-ture and the physical elimination of those who resisted” (Van Bruinessen 1994b: 167, 168). “After the Dersim rebellion had been suppressed, other Kurdish regions being ‘civilized’ from above knew better than to resist” (Van Bruinessen 1994a: 12, 13).

“Military reports call[ed] all people of Dersim indiscriminately ‘bandits’ ” even as “the Law on Resettlement provide[d] the legal framework for a policy of ethnocide” (Van Bruinessen 1994: 149, 150, 152, 153). “Inönü, right hand and successor of Atatürk”, indeed, “expressed the official position: ‘We are frankly [n]ationalist. . . . We must Turkify” – even in socio-linguistic terms – “the inhabit-ants of our land at any price, and we will annihilate those who oppose the Turks or ‘le turquisme’ (Barkey and Fuller 1998: 10)” (Jongerden 2001: 81).

Concerning the impacts of the Southeast Anatolia Regional Development (GAP) Project, intended to modernize, civilize and develop the predominantly Kurdish East via the construction and operation of 19 power stations, 22 dams and linked developments, the top-down “project has always been underpinned by the long standing assimilation policies of the Turkish state with regard to [indige-nous] Kurdish people – their forced inclusion into mainstream Turkish culture and society” (Ronayne 2005: 36), using genocidal processes and techniques (Fer-nandes 2010a). For Gerger (1997: 18), modern Turkish national, secular values as advanced by Kemalism had ensured that “the nation-building process degener-ated into a permanent indoctrination campaign brainwashing successive genera-tions with the most extreme varieties of nationalistic and racist ideologies”. Salih (2006: 1) confirms that “the ideology of Kemalism has been enshrined in the 1982



88 D. Fernandes

Constitution as sacrosanct which cannot be amended; even proposals to do so may constitute a criminal offence”.

5 The Kurdish genocide, 1924–2010The consequences of this paleo-imperialist march towards modernity and civili-zation have been devastating. Not only the Kurds, but others have been subjected to ongoing genocidal assaults as modern laws, administrative, developmental, educational and accounting/propaganda systems, penal codes (some fashioned on Mussolini’s fascist codes) and rational military and counter-insurgency meth-ods and techniques have been applied (Fernandes 2010a, 2010b, 2010c). “In Besikci’s (1990) view, the law” on the pacification and modern reform of Tunceli undoubtedly “served to legitimate genocide” (Van Bruinessen 1994b: 183).

It is also important to appreciate that Atatürk, as its first President, “saw the unification and modernization of education as the key” (Commission on Social Issues 2008: 1) towards framing culture within a modernist-Turkish nationalist straight-jacket. Under this guiding framework, “the diversity of languages in Anatolia was an obstacle to the construction of a homogeneous cultural identity that would become the basis of a national one. Thus, the imposition of Turkish language” – in schools, law courts, press and media outlets and all public recre-ational and work spaces – “became the most significant instrument of the state for creating a Turkish national identity. The new link between the state, its citi-zens and the national identity was enforced by the obligation of Turkish as the national language, whose alphabet [even] replaced Arabic letters with the Latin script in November 1928” (Uçarlar 2009: 120):

The Latin script was introduced not only to undermine the power of religious leaders . . . but also to break ties with the Ottoman past in order to accelerate the reforms in favor of westernization. . . . Furthermore, the expected increase of literacy was supposed to serve the construction and spread of the concept of [the modern] nation. . . . Moreover, the Turk-ish language was [itself] purified from the Arabic and Persian words that represented the Islamic and ‘backward’ Ottoman past. (Uçarlar 2009: 121, 122)

For Shafak: “In the name of modernization, our language shrunk tremendously (cited in Lea 2006: 1). . . . Very few people in Turkey question today the Turkeyfi-cation of the language that we went through. I find that very dangerous because I think that linguistic cleansing is something comparable to ethnic cleansing” (Shafak 2005: 1). To facilitate these modern transformations, “ethnic and linguis-tic studies were also institutionalized by the state. In 1931, the Society for the Study of Turkish History . . . was established, and morphed” four years later “into




the Turkish Historical Foundation. The aim of the foundation” was to create “a national [modern] education in the service of political aims” (Uçarlar 2009: 121, 122).

To Uçarlar: “Every attempt by the Turkish elite to eliminate the hegemony of the Kurdish elite over the Kurdish people also aimed to destroy the political, eco-nomic and social elements of Kurdishness, as well as the consciousness of Kurd-ishness among the people. . . . It is not so striking that the Kurdish language” was targeted “in the service of [modern] nation and state building” (Uçarlar 2009: 125).

In terms of the “painful adjustments” that were deemed to be necessary to progress to the level of contemporary civilization, certain Kurdish sources have estimated that “over half a million [Kurdish] people” were “deported, of whom nearly half died en route” between 1925 and 1928 alone (Lustgarten 2003: 6). Dur-ing the aftermath of the failed Sheikh Said uprising of 1925, seen by many as a nationalist and religious response by Kurdish factions to the secular and Turkifi-cation linked reforms of the modern state (Leicht 1998; Fernandes 2010c), Randal (1999: 121) has concluded that “hundreds of Kurdish villages were burned, and between 40,000 and 250,000 peasants died in the ensuing ‘pacification’. Over the next dozen years or so, perhaps a million Kurdish men, women and children were uprooted and shipped to Western Anatolia”. “Large parts” of the “Kurdish population were sent to concentration camps in the western provinces” (Frodin 1944: 5).

The Turkish Prime Minister reportedly stated in 1938: “We will carry out a military operation in Dersim. . . . There will be an extermination action. . . . Our army . . . will begin maneuvers in the area, ridding it of its inhabitants. In this way, the problem will be pulled up by its roots” (Dersimi 1999 [1952]: 289). Atatürk, in a speech at the opening of parliament in 1936, similarly clarified that: “We have to remove this abscess [Dersim, renamed Tunceli in Turkish] at its roots. To deal with this problem, we will give wider powers to the government” (White 2000: 79). Such “powers” led to further genocidal massacres, slow death measures, Turkish place name-changing, forced assimilation and forced resettlement (Fer-nandes 1998, 2010c).

In linguistically and culturally genocidal terms, in March 1924 – i.e. one year before the first Kurdish rebellion/uprising – “the public use of Kurdish and the teaching of Kurdish” was “prohibited. Influential Kurdish landowners and tribal chiefs were forcibly resettled in the west of the country” (Zürcher 2004: 178). Mod-ern law courts refused to accept Kurdish (Fernandes 2010a). Article 12 of the 1924 constitution also “closed the Parliament to the Kurds who would resist forgetting, delaying or canceling their identity and language” (Uçarlar 2009: 121, 122). “Turk-ish place names began to replace Kurdish ones” (McDowall 1996: 191).



90 D. Fernandes

According to the 1925 Plan for the Reformation of the East, “the cities and towns where Kurds live were listed, and speaking Kurdish there was banned” (Bayrak [1993: 486, 487], cited in Malmisanij [2006: 6]). Kurdish speakers found themselves being fined according to a tariff for every Kurdish word spoken (Fer-nandes 1998). “Kurdish language, music and national costume were outlawed. . . . Like it or not, everyone within Turkey’s borders were by legislation declared to be Turks. The words Kurds and Kurdistan like Armenia and Pontus were forcibly erased from dictionaries and literature” even as “many slogans were coined: ‘One Turk is worth the whole world! . . . Turkish blood is clean, pure and superior!’ ” (Baksi 1986: 103). Broadcasting and publishing in the Kurdish language was pro-hibited (Fernandes 2010c) even as “the newly established nationalist institution called People’s House (Halkevi) would gear Turkish identity and Kemalist ideol-ogy to the popular audience” (Üngör 2008: 33).

“The compulsory adoption of surnames in 1934” served to “turn numerous Kurdish families into Türks, Öztürks, Tatars, or Özbeks” (Van Bruinessen 1997: 6). Alınak reiterates the view that Turkification of Kurds via the schooling system and forced resettlement were core objectives advanced by İnönü and Marshal Çakmak during the 1930’s (Önderoğlu 2010: 1). For Jongerden:

In the 1930’s and 1940’s, government policy in Dersim . . . resembles the conquest and occupation of enemy territory. . . . The building of an educational structure was given priority. . . . It was even suggested that Kurdish children be sent to boarding schools where they would speak exclusively in Turkish. . . . Right up to the present day, boarding schools are established in the Kurdish areas in order to have more control over the children’s educa-tion and to enforce a switch of their identity. (Jongerden 2003: 77, 78)

In Dersim, after the 1937–1938 genocidal onslaught, “the Turkish army kidnapped many Kurdish children who were under the age of seven and placed them in Turk-ish families in western Turkey” (Koivunen 2002: 99). The Tunceli law ensured that educational establishments engaged in assimilating orphan Kurdish girls strictly enforced the teaching of Turkish, whilst banning the Kurdish language (Fernandes 2010a). The Resettlement Law of the 1930’s, moreover, “abrogated any legal recognition of Kurdish tribes and their leaders, thus permitting the au-tomatic sequestration of their immovable assets. All settlements in which Kurdish was the mother tongue” were to be “dissolved, and the displaced Kurds were to be resettled” – as part of the Turkification drive – “in localities where they would make up no more than 5% of the population. . . . It was further prescribed that ‘those who speak a mother tongue other than Turkish will be forbidden to form villages, quarters or groups of artisans and employees’. The intention was to de-stroy Kurdish identity” and language “in its entirety” (Lustgarten 2003: 7). The law also aimed “at the dismembering of the Kurdish community even down to the




family unit. It has been judged that parents, married sons and married grandsons shall be evicted to different areas” (Turkish Human Rights Association 1996: 19).

Other culturally and physically genocidal plans, policies and practices have been evident since the 1930’s (Fernandes 2007, 2010a). Forty nine Kurdish intel-lectuals, for example, were arrested in 1959 as part of a wider initiative that was aimed at intentionally murdering 1,000 Kurdish intellectuals (Anter 1991). During the 1990’s, Kurdish intellectuals were once again subjected to assassination and disappearance (Fernandes 2010a). During the 1940’s, a report by the Inspector General of the First Inspectorate recommended that more Kurdish leaders from the East be deported even as “Turkish language boarding schools for Kurdish children” were to be constructed, “where all traces of Kurdish culture” and lan-guage “could be expunged” (McDowall 1996: 209, 210). The 1949 Provincial Ad-ministration Law further “authorized” the changing of “names of places and this authority was used quite liberally. Moreover, article 16 of the 1972 Population Law prohibited giving Kurdish names to new-borns” (Yeğen 2008: 3).

After the 1960 coup, “the military regime in 1961 systematically started to change Kurdish place names into Turkish and establish regional boarding schools in order to assimilate the Kurdish population” (Uçarlar 2009: 129). The Forced Settlement Law that was passed at the time “stipulated that this was done in order to ‘carry out certain social reforms’ ” that would “ ‘demolish the order of the Middle Ages that exists in Turkey’ ” (Zeydanlıoğlu 2008: 65). General Gürsel, as head of the military regime, “lauded a book . . . which claimed that the Kurds were in fact of Turkish origin”, and declared whilst standing on an American tank that: “ ‘There are no Kurds in this country. Whoever says he is a Kurd, I will spit in his face’ ” (Zeydanlıoğlu 2008: 64, 65).

“Waves of place name changing”, indeed, “occurred – and were initiated – [even] under so-called liberal governments” (Jongerden 2009: 10). Various gov-ernment initiatives were aimed at stopping Kurds from listening to foreign broad-casts in Kurdish and even accessing Kurdish educational courses internationally (Fernandes 2010c). Uçarlar (2009: 133, 134) confirms that “the military adminis-tration (1980–3) banned strictly the use of Kurdish language. . . . The assimilation of Kurdish children into the Turkish language was fostered through the dissemi-nation of compulsory schooling. The Kurdish names of villages that [had] re-mained intact after the changes of the 1960’s were adjusted into Turkish. Kurdish families were forced to give Turkish names to their children”, and this pressure was still being applied during the first decade of the 21st century (Fernandes 2007, 2010c). Torture has continued to be applied in a genocidal context (Fernandes 2010a), even as Article 89 of the post-coup constitution “prohibited the right of Kurds to political representation. . . . The constitution also legally enshrined the ban on the Kurdish language” (Zeydanlıoğlu 2009: 81).



92 D. Fernandes

Helsinki Watch (1990: 37) has further detailed the manner in which, “in May 1989, the National Security Council launched a campaign denying the existence of a distinct Kurdish nation and a Kurdish language. Pamphlets were issued and distributed to schools in the south-east” to reinforce this message. Skutnabb-Kangas (cited in Fernandes [2006: 34]) concluded in 2002 that Turkey’s policy “still fit[ted] two of the definitions of genocide in the UN International Convention on the Prevention and Punishment of the Crime of Genocide. . . . [What is happen-ing] is genocide, according to the UN definition. . . . In addition, Turkey is of course also committing linguistic genocide according to the specific definition on linguistic genocide”. More recent assessments have drawn the same conclusions (Skutnabb-Kangas 2005, 2010; Skutnabb-Kangas and Fernandes 2008; Skutnabb-Kangas and Dunbar 2010). Even by mid-2010, Cengiz Aktar confirmed that “teach-ing Kurdish at [public] school[s] is not at all on the agenda” of the government and state (Aktar 2010: 1).

With regard to the nature of the state’s genocidal policies in Turkish Kurdis-tan between 1984–1997, some estimates suggest that over three million Kurds were forcibly displaced and subjected to mental harm, tens of thousands of peo-ple were killed, over 4000 settlements were fully or partially destroyed and thou-sands of people disappeared (Fernandes 2010a, 2010b). In development terms, too, the Southeastern Anatolian Project has been used to facilitate “an ethnic and cultural genocide against Kurds” (Tataii 2010; see also Fernandes 2010a). The genocidal actions of the Turkish state during the 2000–2010 period have also been recognized, as such, by a number of genocide scholars, policy analysts, law-yers, human rights campaigners, political organizations and movements (Fer-nandes 2010a, 2010b, 2010c). To Gerger (cited in Cudi [2010: 1]), writing in August 2010: The US seems “to have reached some sort of an understanding” with the governing Justice and Development Party (AKP) “concerning the Kurdish issue”:

The previous state strategy was to nationalist-Kemalists, ‘total liquidation through violence’. . . . Now with the active aid of president Obama, the liberal coalition under the AKP government tried ‘Açılım’ which meant a new phase – liberal phased liquidation. . . . [But] even this created serious cleavages within the ruling classes and now it seems that they have met again at the old strategy of nationalist total liquidation through force and violence. (Cited in Cudi [2010: 1])

6 ConclusionIt is evident that Kurds in Turkey have been subjected to linguistic and other forms of genocide. Social triage, the Hobbesian project and the vivisectional man-date have all been evident. The influence of modernity in shaping and legitimiz-




ing the genocidal process in Turkey has been substantive. For Schulter, it is im-portant to recognize that:

. . . while the European Holocaust of 1939–1945 against Jews and other ‘inferior’ peoples rightly serves as an ideal case of genocide, the persecution of . . . the Kurds under the Turk-ish Republic, is also genocide in the original and proper sense of the term as coined by the jurist Raphael Lemkin . . . Turkish policy in Northern Kurdistan . . . might serve as an ex-ample of the attempted cultural destruction of a ‘national pattern’ by forced . . . ‘Turkifica-tion’. In fact, Lemkin’s original description of genocide, with its focus not only on the sys-tematic slaughter and starvation of the Jews but also on the imposition of the German language in places such as Luxembourg, might have cited Turkish policy in Northern Kurd-istan. (Schulter 2000: 1)

Today, culturally and linguistically genocidal policies and practices are still in place and the spectre of physical genocide looms once again (Fernandes 2010a, 2010b). For Havemann (2005: 61): “It may be comforting to claim that genocide was a facet of early/simple/industrial modernity and that it does not happen any more. The law, state and dominant culture selectively forget, engaging in histori-cal denial: they deny the immediacy of genocide and ethnocide or that what went on in Australia”, for example, “ought to be described as ‘genocide’ ”. However, “modernization has always produced and legitimated atrocities and suffering. . . . In Australia”, he concludes – as I do in this article with regard to Turkey – “the apparently ‘civilizing’ imperatives of modernity . . . amount to genocide. . . . Until we overcome denial by acknowledging the truth, we can never get to ‘the place called reconciliation’ ” (Havemann 2005: 61, 79).

ReferencesAhmed, Nafeez. 2003. Behind the war on terror. East Sussex: Clairview.Aktar, Cengiz. 2010. Education in mother tongue. Hurriyet 13 August.

http://www.hurriyetdailynews.com/n.php?n=education-in-mother-tongue-2010-08-13 (accessed 13 August 2010).

Akturk, Ayce, Hakki Haylir & Ahmed Tigran. 2001. A foot in Australia, three souls in Kurdistan. In Gina Lennox (ed.), Fire, snow and honey: voices from Kurdistan, 477–492. Sydney: Halstead Press.

Al-Qasem, Anis. 1977. Racism and world peace. In Executive Council, International Organization for the Elimination of All Forms of Racial Discrimination (ed.), Zionism and racism: proceedings of an international symposium, 7–16. London: Billing & Sons.

Alvarado, Emmanuel. 2010. Language and politics. New Renaissance 29 September. http://www.ru.org/book-reviews/language-and-politics.html (accessed 30 July 2010).

Alvares, Claude. 1988. Science, colonialism and violence: a Luddite view. In Ashis Nandy (ed.), Science, hegemony and violence: a requiem for modernity, 68–122. Delhi: Oxford University Press.



94 D. Fernandes

Anter, Musa. 1991. Hatiralarim [My memoirs]. Stockholm: Wesanen Welat.Ashcroft, Bill, Gareth Griffiths & Helen Tiffin. 2000. Post-colonial studies: the key concepts.

London & New York: Routledge.Baksi, Mahmut. 1986. Kurdish voice: Shivan Perwer. Location not listed: Helin House.Banerjee, Subhabrata. 2007. Corporate social responsibility: the good, the bad and the ugly.

Cheltenham: Edward Elgar Publishing.Barkey, Henry & Graham Fuller. 1998. Turkey’s Kurdish question. Boston: Rowman & Littlefield.Bart, Daniel. 2005. The Turkish condition. Kurdistan Observer 26 May.

http://www.armeniandiaspora.com/showthread.php?28091-The-Turkish-Condition (accessed 30 July 2010).

Bauman, Zygmunt. 1990 [1989]. Modernity and the holocaust. Cambridge: Polity.Bauman, Zygmunt. 1991 [1989]. Modernity and the holocaust. Ithaca: Cornell University Press.Bayrak, Mehmet. 1993. Kürtler ve Ulusal-Demokratik Mücadeleleri: Gizli Belgeler-Araştırmalar-

Notlar. Ankara: Öz-Ge Yayınları.Beşikçi, Ismail. 1990. Tunceli kanunu 1935 ve Dersim jenocidi [The Tunceli law of 1935 and the

Dersim genocide]. Istanbul: Belge.Beşikçi, Ismail. 2009. Europe submitted to the Ankara criteria. PEN-Kurd 2–9.

http://www.pen-kurd.org/englizi/ali-kilic/Europe-submitted-to-the-Ankara-criteria.pdf (accessed 30 July 2010).

Book Rags. 2006. Modernization. Book Rags. http://www.bookrags.com/research/modernization-este-0001_0003_0/ (accessed 30 July 2010).

Bourke, Joanna. 2000. An intimate history of killing. London: Granta.Chaliand, Gerard. 1994. The Kurdish tragedy. London: Zed.Commission on Social Issues. 2008. John Dewey in Turkey. Social Issues 20 October.

http://deweycsi.blogspot.com/2008/10/republic-of-turkey-was-proclaimed-on.html (accessed 30 July 2010).

Conversi, Daniele. 2006. Genocide, ethnic cleansing and nationalism. In Gerard Delanty & Krishan Kumar (eds.), The Sage handbook of nations and nationalism, 320–334. London, Thousand Oaks & New Delhi: Sage.

Copeaux, Etienne. 1998. Türk tarih tezinden Türk-islam sentezine [The Turkish history thesis of the Turkish-Islam synthesis]. Istanbul: Tarih Vakfi Yurt Yayinlari.

Cudi, Nucan. 2010. Gerger: New round for Turkey power struggle. ANF 5 August. http://en.firatnews.com/index.php?rupel=article&nuceID=714 (accessed 5 August 2010).

Dersimi, Nuri. 1999 [1952]. Kürdistan tarihinde Dersim [Dersim in the history of Kurdistan]. Berlin: Mezopotamien Verlag.

Dean, Bartholomew. 2009. Machetes in our hands, blood on our faces. Anthropological Quarterly 82(4). 1069–1072.

Devi, G. N. 2009. The G N Devi Reader. Hyderabad: Orient Blackswan.Docker, John. 2004. Raphael Lemkin’s history of genocide and colonialism. Paper presented at

the United States Holocaust Memorial Museum, Washington, 26 February.Ergil, Doğu. 2009. Official view of the Turkish state towards the minorities in general. Kurdish

Info 25 November. http://www.kurdish-info.eu/News-sid-Dogu-Ergil-official-view-of-the-turkish-state-towards-the-minorities-in-general-14087.html (accessed 30 July 2009).

Escobar, Arturo. 1994. Encountering development: the making and unmaking of the third world. Princeton: Princeton University Press.

Fernandes, Desmond. 1998. The Kurdish genocide in Turkey, 1924–98. Armenian Forum 1(4). 57–108.



http://www.bookrags.com/research/modernization-este-0001_0003_0/

http://www.bookrags.com/research/modernization-este-0001_0003_0/

http://deweycsi.blogspot.com/2008/10/republic-of-turkey-was-proclaimed-on.html

http://www.kurdish-info.eu/News-sid-Dogu-Ergil-official-view-of-the-turkish-state-towards-the-minorities-in-general-14087.html

http://www.kurdish-info.eu/News-sid-Dogu-Ergil-official-view-of-the-turkish-state-towards-the-minorities-in-general-14087.html


Fernandes, Desmond. 2006. Turkey’s US backed war on terror: a cause for concern? Variant. 33–36.

Fernandes, Desmond. 2007. The Kurdish and Armenian genocides: from censorship and denial to recognition? Stockholm: Apec.

Fernandes, Desmond. 2010a. Modernity, modernization and the genocide of Kurds and Others. Stockholm: Apec.

Fernandes, Desmond. 2010b. The Armenian, Assyrian, Syriac, Greek, Kurdish, Greek Cypriot and other genocides: recognition and denialism. Stockholm: Apec.

Fernandes, Desmond. 2010c. The politics of genocide recognition and denialism: the Armenian, Assyrian, Syriac, Greek, Kurdish, Greek Cypriot and other genocides. Stockholm: Apec.

Frodin, Johan. 1944. Neuere kulturgeografische wandlungen in der ostlichen Turkei. Zeitschrift der gesellschaft fur erdkunde 79(102). 1–20.

Gerger, Haluk. 1997. Crisis in Turkey (MERA Occasional 28). Amsterdam: MERA.Havemann, Paul. 2005. Denial, modernity and exclusion: indigenous placelessness in

Australia. Macquarie Law Journal 5. 57–80.Helsinki Watch. 1990. Destroying ethnic identity: the Kurds of Turkey. Washington: Helsinki

Watch.Herman, Edward & David Peterson. 2010. The politics of genocide. New York: Monthly Review.Hinton, Alexander Laban. 2002. The dark side of modernity: toward an anthropology of

genocide. In Alexander Laban Hinton (ed.), Annihilating difference: an anthropology of genocide, 1–39. Berkeley & London: University of California.

Hinton, Alexander Laban. 2007. Genocide and modernity. In Conerly Casey & Robert Edgerton (eds.), A companion to psychological anthropology: modernity and psychocultural change, 419–435. Oxford: Blackwell.

Hobsbawm, Eric. 1989. The age of empire, 1875–1914. London: Abacus.Houston, Christopher. 2001. Islam, Kurds and the Turkish nation state. Oxford & New York:

Berg.Jongerden, Joost. 2001. Resettlement and reconstruction of identity: the case of the Kurds in

Turkey. The Global Review of Ethnopolitics 1(1). 80–86.Jongerden, Joost. 2003. Violation of human rights and the Alevis in Turkey. In Paul White & Joost

Jongerden (eds.), Turkey’s Alevi enigma: a comprehensive overview, 71–89. Leiden: Brill.Jongerden, Joost. 2009. Crafting space, making people: the spatial design of nation in modern

Turkey. European Journal of Turkish Studies 10. http://ejts.revues.org/Index4014.html (accessed on 30 July 2010).

KHRP. 2010. KHRP condemns Turkey’s ejection of EU Turkey civic commission chair. KHRP Press Release 26 April. http://www.khrp.org/khrp-news/550-khrp-condemns-turkeys-ejection-of-eu-turkey-civic-commission-chair.html (accessed 26 April 2010).

Kinzer, Stephen. 2001. Crescent and star. New York: Farrar, Straus & Giroux.Kirişçi, Kemal. 2004. The Kurdish question and Turkish foreign policy. In Lenora Martin &

Dimitris Keridis (eds.), The future of Turkish foreign policy, 277–314. Massachusetts: MIT.Koivunen, Kristiina. 2002. The invisible war in north Kurdistan. Helsinki: University of Helsinki

PhD thesis.Laing, Jacqueline. 2008. Information technology and biometric databases: eugenics and other

threats to disability rights. Journal of Legal Technology Risk Management 3(1). 9–35.Lea, Richard. 2006. In Istanbul, a writer awaits her day in court. The Guardian 24 July.

http://www.guardian.co.uk/books/2006/jul/24/fiction.voicesofprotest (accessed 30 July 2010).



http://ejts.revues.org/Index4014.html

http://www.khrp.org/khrp-news/550-khrp-condemns-turkeys-ejection-of-eu-turkey-civic-commission-chair.html

http://www.khrp.org/khrp-news/550-khrp-condemns-turkeys-ejection-of-eu-turkey-civic-commission-chair.html

http://www.guardian.co.uk/books/2006/jul/24/fiction.voicesofprotest

96 D. Fernandes

Leicht, Justus. 1998. A balance sheet of Kemalism. WSWS 17 November. http://www.wsws.org/articles/1998/nov1998/turk-n17.shtml (accessed 30 July 2010).

Lemkin, Raphael. 1944. Axis rule in occupied Europe: laws of occupation – analysis of government – proposals for redress. Washington: Carnegie Endowment for International Peace.

Lofti, Manal. 2007. Kurd: the forbidden word in Turkey. Asharq Al-Awsat 16 December. http://www.aawsat.com/english/news.asp?section=3&id=11177 (accessed 30 July 2010).

Lukunka, Barbra. 2007. Ethnocide. Online encyclopedia of mass violence 3. http://www.massviolence.org/Ethnocide (accessed 30 July 2010).

Lustgarten, Anders. 2003. This is the only valley where we live: the impact of the Munzur dams. London: KHRP/Cornerhouse.

Malmisanij, M. 2006. The past and the present of book publishing in Kurdish language in Turkey. Location not listed: Next Page Foundation.

May, Stephen. 1999. Introduction. In Stephen May (ed.), Indigenous community-based education, 1–7. Clevedon: Multilingual Matters.

Maybury-Lewis, David. 2002. Genocide against indigenous peoples. In Alexander Laban Hinton (ed.), Annihilating difference: an anthropology of genocide, 43–53. Berkeley & London: University of California Press.

McDowall, David. 1996. A modern history of the Kurds. London: IB Tauris.Neu, Dean & Richard Therrien. 2003. Accounting for genocide. Blackpoint & London: Fernwood

& Zed.Nezan, Kendal. 1993. Kurdistan in Turkey. In Gerard Chaliand (ed.), A people without a country:

the Kurds and Kurdistan, 38–94. London: Zed.Noakes, Jeremy. 2010. Review of architects of annihilation. Reviews in History 373.

http://www.history.ac.uk/reviews/review/373 (accessed 30 July 2010).Önderoğlu, Erol. 2010. Trial against Kurdish politician stirs up history. BIA News Center 25 May.

http://bianet.org/english/minorities/122226-trial-against-kurdish-politician-stirs-up-history (accessed 30 July 2010).

Phillipson, Robert. 1997. Realities and myths of linguistic imperialism. Journal of Multilingual and Multicultural Development 18(3). http://www.multilingual-matters.net/jmmd/018/0238/jmmd0180238.pdf#search=’phillipson%20linguistic%20imperialism (accessed 30 July 2010).

Rafferty, Jean. 2005. Zarakolu. Norsk PEN. http://www.norskpen.no/pen/Zarakkolu2.html (accessed 30 July 2010).

Randal, Jonathan. 1999. After such knowledge, what forgiveness? My encounters with Kurdistan. Boulder: Westview Press.

Rojas, Róbinson. 1996. Modernisation theory and the laws of social change. Róbinson Rojas Archive. http://www.rrojasdatabank.info/capital8.htm (accessed 30 July 2010).

Ronayne, Maggie. 2005. The cultural and environmental impact of large dams in southeast Turkey. Galway & London: National University of Ireland & KHRP.

Salih, Karim. 2006. Possibilities of prosecuting Turkish leaders for crimes against humanity and war crimes. KurdishMedia.com 3 November. http://www.collectifvan.org/article.php?id=5389 (accessed 30 July 2010).

Schulter, Margo. 2000. Pirozbe newroz: A Jewish greeting to Mehdi Zana. Kurdish Observer 18 March. http://www.kurdishobserver.com/2000/03/18/hab06.html (accessed 30 November 2007).



http://bianet.org/english/minorities/122226-trial-against-kurdish-politician-stirs-up-history

http://bianet.org/english/minorities/122226-trial-against-kurdish-politician-stirs-up-history

http://www.multilingual-matters.net/jmmd/018/0238/jmmd0180238.pdf#search=�phillipson%20linguistic%20imperialism

http://www.multilingual-matters.net/jmmd/018/0238/jmmd0180238.pdf#search=�phillipson%20linguistic%20imperialism

http://www.rrojasdatabank.info/capital8.htm


Shafak, Elif. 2005. Linguistic cleansing. New Perspectives Quarterly 22(3). http://www.digitalnpq.org/archive/2005_summer/05_shafak.html (accessed 30 July 2010).

Shoup, Daniel. 2006. Can archaeology build a dam? Sites and politics in Turkey’s south-east Anatolia project. Journal of Mediterranean Archaeology 19(2). 231–258.

Simpson, Christopher. 1993. The splendid blond beast: money, law and genocide in the twentieth century. New York: Grove.

Skutnabb-Kangas, Tove. 2000. Linguistic genocide in education – or worldwide diversity and human rights? Mahwah, NJ: Lawrence Erlbaum Associates.

Skutnabb-Kangas, Tove. 2005. Endangered linguistic and cultural diversities and endangered biodiversity. Paper presented at the conference on Cultural Diversity and Linguistic Diversity, Diyarbakir, 20–25 March. http://www.pen-kurd.org/Diyarbakir-seminar/tove-endangered-linguistic-and-cultural-diversities.html. Turkish version: Tehlike Altındaki Dilsel ve Kültürel Çeşitlilik ve Tehlike Altındaki Biyoçeşitlilik – Çeşitliliğin Korunmasında Dilsel İnsan Hakları Eğitiminin Rolü. Bîr. Kovera lêgerîn û lêkolînê. Aroçtirmo – İnceleme dergisi 3. 7–37. http://www.kovarabir.com/dr-tove-skutnabb-kangas-tehlike-altindaki-dilsel-ve-kulturel-cesitlilik (accessed 5 August 2010).

Skutnabb-Kangas, Tove. 2010. Crimes against humanity in education, and applied linguistics – corporate globalisation or geopolitical knowledge glocalisation? Plenary presentation at the British Association for Applied Linguistics, Aberdeen, 9–11 September.

Skutnabb-Kangas Tove & Robert Dunbar. 2010. Indigenous children’s education as linguistic genocide and a crime against humanity? A global view. Gáldu Čála. Journal of Indigenous Peoples’ Rights 1. http://www.e-pages.dk/grusweb/55/ (accessed 5 August 2010).


Socialist Party of Kurdistan (PSK). 2008. Statement on the 50th anniversary of the UN Genocide Convention. Paper presented at the UK House of Commons, 9 December.

Spiegel. 2010. Turkish EU minister on the Armenian genocide controversy. Spiegel Online International 16 March. http://www.spiegel.de/international/world/0,1518,683701,00.html (accessed 30 July 2010).

Solomon, Jon. 2010. Reappropriating the neoliberal university for a new Putonghua (common language). Edufactory Web Journal 0. http://www.edu-factory.org/edu15/webjournal/n0/Solomon.pdf (accessed 30 July 2010).

Swartz, Dana. 2007. Human rights are not protected by Turkey’s constitution. KurdishMedia.com 5 November. http://forum.politics.be/showthread.php?t=66984 (accessed 30 July 2010).

Tachjian, Vahé. 2009. The expulsion of non-Turkish ethnic and religious groups from Turkey to Syria during the 1920’s and early 1930’s. Online Encyclopedia of Mass Violence. http://www.massviolence.org/Article?id_article=330 (accessed 30 July 2010).

Tataii, Steve. 2010. Turkey’s GAP project is an ethnic and cultural genocide against Kurds. Ekurd.net 31 March. http://www.ekurd.net/mismas/articles/misc2010/3/state3706.htm (accessed 30 July 2010).

Taussig, Michael. 1987. Shamanism, colonialism and the wild man: a study in terror and healing. Chicago: University of Chicago.

Tirman, John. 1997. Ataturk’s children: several new books cast light on Washington’s genocidal ally in the Middle East. Boston Review, December. http://www.photius.com/thus/ataturk.html (accessed 30 July 2010).



http://www.digitalnpq.org/archive/2005_summer/05_shafak.html

http://www.digitalnpq.org/archive/2005_summer/05_shafak.html

http://www.pen-kurd.org/Diyarbakir-seminar/tove-endangered-linguistic-and-cultural-diversities.html

http://www.pen-kurd.org/Diyarbakir-seminar/tove-endangered-linguistic-and-cultural-diversities.html

http://www.kovarabir.com/dr-tove-skutnabb-kangas-tehlike-altindaki-dilsel-ve-kulturel-cesitlilik

http://www.kovarabir.com/dr-tove-skutnabb-kangas-tehlike-altindaki-dilsel-ve-kulturel-cesitlilik

http://www.e-pages.dk/grusweb/55/

http://www.spiegel.de/international/world/0,1518,683701,00.html

http://www.spiegel.de/international/world/0,1518,683701,00.html

98 D. Fernandes

Tirman, John. 2005. Ataturk’s children. Boston Review December 1997–January 1998. http://www.bostonreview.net/BR22.6/Tirman.html (accessed 30 July 2010).

Turkish Human Rights Association (IHD). 1996. Habitat 2: the alternative report. Ankara, Istanbul & Diyarbakir: IHD.

Turner, Bryan (ed.). 1990. Theories of modernity and postmodernity. London: Sage.Uçarlar, Nesrin. 2009. Between majority power and minority resistance: Kurdish linguistic

rights in Turkey. Lund: Lund University.Üngör, Uğur Ümit. 2007. Excuse me, did you say genocide?: on the poverty of debates on mass

violence in Turkey. The Armenian Weekly 73(16), 24 April. http://www.hairenik.com/armenianweekly/gin042107_08.htm (accessed 30 July 2009).

Üngör, Uğur Ümit. 2008. Seeing like a nation-state: Young Turk social engineering in Eastern Turkey, 1913–50. Journal of Genocide Research 10(1). 15–39.

Van Bruinessen, Martin. 1997. Race, culture, nation and identity politics in Turkey: some comments. Paper presented at the Mica Ertegun Annual Turkish Studies Workshop on Continuity and Change, Department of Near Eastern Studies, Princeton University, 24–26 April.

Van Bruinessen, Martin. 1994a. Genocide in Kurdistan? The suppression of the Dersim rebellion in Turkey (1937–8) and the chemical war against the Iraqi Kurds. In George J. Andreopoulos (ed.), Conceptual and historical dimensions of genocide, 141–170. Pennsylvania: University of Pennsylvania Press.

Van Bruinessen, Martin. 1994b. Genocide of Kurds. In Israel Charny (ed.), The Widening Circle of Genocide. Vol. 3: Genocide – A Critical Bibliographic Review, 165–192. New Brunswick: Institute on the Holocaust and Genocide.

Van Bruinessen, Martin. 2005. Ismail Beşikçi: Turkish sociologist, critic of Kemalism and kurdologist. The Journal of Kurdish Studies 5. 19–34.

Visvanathan, Shiv. 1988. On the annals of the laboratory state. In Ashis Nandy (ed.), Science, hegemony and violence: a requiem for modernity, 257–288. New Delhi: Oxford University Press.

White, Paul. 2000. Primitive rebels or revolutionary modernisers? The Kurdish national movement in Turkey. London: Zed.

Yeğen, Mesud. 1999. The Kurdish question in Turkish state discourse. Journal of Contemporary History 34(4). 555–568.

Yeğen, Mesut. 2008. Banditry to disloyalty: the Kurdish question in Turkey. Ankara: Seta.Zeydanlıoğlu, Welat. 2008. The white Turkish man’s burden: orientalism, Kemalism and the

Kurds in Turkey. In Guido Rings & Anne Ife (eds.), Neo-colonial mentalities in contemporary Europe? Language and discourse in the construction of identities, 155–174. Newcastle upon Tyne: Cambridge Scholars.

Zeydanlıoğlu, Welat. 2009. Torture and Turkification in the Diyarbakir Military Prison. In Welat Zeydanlıoğlu & John Parry (eds.), Rights, citizenship and torture: perspectives on evil, law and the state, 73–92. Oxford: Inter-Disciplinary.

Zürcher, Erik. 2004. Turkey: a modern history. London: I. B. Tauris.



http://www.hairenik.com/armenianweekly/gin042107_08.htm

DOI 10.1515/ijsl-2012-0051 IJSL 2012; 217: 99 – 125

Welat ZeydanlıoğluTurkey’s Kurdish language policy

Abstract: This article examines the Turkish state’s assimilationist policy towards the Kurds and the Kurdish language in Turkey. It studies how the Turkish nation-alist elites, the Kemalists, have throughout the 20th century systematically sup-pressed the Kurdish language as part of their aim to construct a homogenous nation-state of Turkish speakers. It shows that this linguicidal policy was strongly informed by the traumatic collapse of the Ottoman Empire and the consequent Kemalist emphasis on complete ethno-linguistic homogeneity as criteria for be-ing “Turkish”, “Western” and “civilised”. The article discusses the various “Turki-fication” strategies of the authorities, such as banning the Kurdish language, the denial of the existence of the Kurds, changing the names of towns and villages, the forced re-settlement of Kurds and the assimilation of Kurdish children. It crit-ically analyses the recent developments in Turkey’s Kurdish language policy and the reform efforts of the current government as part of the country’s EU candi-dacy. The article reflects however, that whilst looking good on paper, these re-forms have had little impact in reality and Kurdish speakers in Turkey are still systematically denied their basic human and linguistic rights.

Keywords: Turkey; Kurds; Kurdish question; Kemalism; linguicide

Welat Zeydanlıoğlu: The University of Uppsala. E-mail: [email protected]

1 Language and the making of the Turkish nationAs the Ottoman Empire gradually fragmented and what remained of it was trans-formed into the Republic of Turkey following the Treaty of Lausanne in 1923,1 the Turkish nationalist cadres that established this new state inherited a population diverse in terms of ethnicity, language and religion. Although some homogenisa-tion had been achieved following the Armenian genocide (1915–1917) (Akçam 2004; Bloxham 2005a; Dadrian 2004), which largely ended the Armenian pres-ence in Anatolia, and the exchange of more than two million people between Greece and Turkey (1923) (Aktar 2000: 17–66), Turkey continued to remain a

1 For the Treaty of Lausanne see: http://www.mfa.gov.tr/lausanne-peace-treaty.en.mfa.

Brought to you by | Georgetown UniversityAuthenticated

Download Date | 5/22/15 3:35 AM

100 W. Zeydanlıoğlu

heterogonous country. This composition naturally continued to reflect the hetero-geneity of the Ottoman Empire, which, like other empires, had contained a multi-tude of different ethnic, linguistic and religious groups within its once large territories. This was well represented in the official language of the Empire, Otto-man Turkish (Osmanlıca), which was a version of Turkish with extensive borrow-ings from Arabic and Persian as well as other languages. The Turkish state that was built on the remaining lands of the Empire came to be strongly dominated by a nationalist ideology that emphasised the unity, secularity and the indivisibility of the Republic and the Turkish nation, aiming for the complete homogenisation of the society. The Kemalists, named after Mustafa Kemal Atatürk (meaning the “father of the Turks”), the founder of the country, intended to create a western and secular nation-state for the Turks and thereby leave behind their “Oriental” Ottoman past. As such, the inherited ethno-religiously diverse society, out of which a new nation was to be moulded, constituted an anachronism for the Ke-malists. As Ümit Cizre has pointed out, “the national community, from the inau-guration of the Republic, had to be constructed out of an embarrassing diversity of a demographic reality which was a legacy of the Ottoman mosaic. The confus-ing range of ethnic, linguistic, and sectarian attachments produced insecurities and anxieties over the question of the constituents of an identity in an inarticu-late and uncertain social world” (Cizre 2001: 229).

Atatürk himself was convinced that the weakness and ultimate demise of the Ottoman Empire was due to its multicultural nature and the Ottoman accep-tance of this diversity had made it vulnerable to foreign manipulation and the search for independence by minority groups such as Kurds and Armenians (Muller 1996: 175). According to Atatürk, Turks could only reach “the contempo-rary level of civilisation” by becoming a secular and homogenous nation-state of Turkish speakers. The nationalist military officers, who formed the nucleus of the Kemalist elite, were strongly influenced by the discourse of the enlighten-ment as well as rising fascist and nationalist currents in Europe. Traumatised by the humiliating experience of the collapse of the Ottoman Empire, they were certain that strong nations were the only legitimate means of organising a state. A nation could only consist of a secular society that shared the same ideal, lan-guage, territory and culture. Ziya Gökalp, who laid the theoretical foundations of Turkish nationalism, claimed at the time, “today in Europe only those states which are based on a single-language group are believed to have a future” (Gökalp 1981 [1918]: 81). Similar views were in circulation as expressed by earlier nation-alist pioneers, who were aware of the powerful role played by language in the construction of a national identity (Aydıngün and Aydıngün 2004). As early as 1878, the famous Ottoman nationalist poet Namık Kemal argued, “while it is a necessity that we try to annihilate all the languages of our country except



Turkey’s Kurdish language policy 101

Turkish, should we instead issue the Albanians, the Laz and the Kurds an alpha-bet to be used as a weapon of division? Language may perhaps be a firmer barrier than religion in preventing the transformation of a people” (Arai 1994: 18; my translation).

As this example shows, the policy of “Turkification” (türkleştirmek) pre-dates the founding of the Republic of Turkey and has its foundations in the late 19th and early 20th centuries when it was developed by nationalist intellectuals, in particular those active in the Committee of Union and Progress, (İttihat ve Terakki Cemiyeti), who argued for example that Kurds had no separate lan-guage or history, and could therefore not be considered a genuine nation. Socio-logical and anthropological research was commissioned in order to justify the Turkification of Anatolia in general and the Kurds in particular (Dündar 2001; Üngör 2008). This strategy clearly had the vital role of guaranteeing the domi-nance and hegemony of the Turks over other ethnicities, such as Kurds, and to deny them the legitimacy to establish their own separate states as outlined in the Treaty of Sèvres (1920).2 Such strategies continued and were further systematised by the Kemalists, most of whom had been active in the Committee of Union and Progress. Following the establishment of the Republic, the Kurdish language, identity and the geographical area of Kurdistan were gradually denied and the official argument developed that there were no Kurds in Turkey, only those who had forgotten their “Turkishness”. The denial of the existence of Kurds and simul-taneously clamping down on the Kurdish language and culture shaped the core of the Turkish state’s Kurdish policy, which continued unabated throughout the 20th century.

The Kemalist nation-building project was implemented in an authoritarian and top-down manner by the nationalist state elites, who perceived themselves as the representatives of progress and modernity. The largely rural, poor and il-literate masses ravaged by years of warfare and displacement could hardly be trusted to modernise themselves and had to be modernised “in spite of them-selves”. From the 1920s onwards, through the state and its apparatuses, the “en-lightened” ruling elites emphasised notions of nationalism, civilisation, science, modern education, rationality and secularism and devoted themselves to replac-ing the “old” with the “new”. In this sense, one can argue that, although Turkey was never formally colonised and that in fact a war of liberation was fought against Western invading powers (1920–1922), the Kemalist elites were function-ing within a clear Orientalist and Euro-centric discourse and were the key agents

2 For the Treaty of Sèvres see: http://www.pollitecon.com/html/treaties/Treaty_Of_Sevres_1.htm.




in the reproduction and deployment of an ethno-colonial vision within Turkey. As such one can refer to a Turkish version of the “White Man’s Burden” that aimed to “civilise” and culturally and linguistically colonise a purportedly “backward” and “Oriental” society by internally eradicating the retrograde influence of Islam and primordial identities (for a detailed analysis see Zeydanlıoğlu [2007]; [2008]). In Turkey, this “self-Orientalisation” (Dirlik 1997: 111) articulated itself in the attempt to launch a full-scale “civilisational shift”, a modernising revolution against the “old order”, which was to be carried out in all aspects of life through far-reaching radical reforms in areas ranging from politics, education, and law to attire, music, literature, architecture, the arts etc.

Naturally, as in most other nationalisms, language was at the heart of the Kemalist nation-building project. Modern Turkish as the national lingua franca shared by all the citizens was crucial in the dissemination of Kemalism as the of-ficial ideology and in overcoming ethnic, religious and linguistic peculiarities of the Ottoman past deemed to be lethal obstacles to the creation of a new nation. In this way, the pluralism associated with the Ottoman Empire was cast as an anach-ronism, as a remaining component of “Oriental civilisation”. The ground was thus laid for the dramatic and forced Latinisation of the Turkish language as well as its use as the defining attribute of modern Turkish national identity. Accord-ingly, the Kemalist language reforms aimed to guarantee the “purification” of the Turkish mind from “backwardness” and “religiosity” and bring forward the true Turkish identity and impose Turkish as the dominant language in all areas of life. The purification of Turkish also meant its westernisation, and the “scientific” study and transformation of Turkish was to mirror this process. As Geoffrey Lewis has pointed out, the aim of this radical reformation was to “break Turkey’s ties with the Islamic east and to facilitate communication domestically as well as with the Western world” (Lewis 1999: 27). The most important reform in terms of lan-guage was the Language Revolution (Dil Devrimi) of 1928, which introduced the Latin alphabet replacing the hitherto used Arabic letters. According to Yunus Nadi, a prominent member of the Kemalist intelligentsia and the founder of the Kemalist party-state newspaper Cumhuriyet [Republic], the aim of the alphabet reform was “to unite Turkey with Europe in reality and materially” (Ahmad 1993: 82). As Atatürk himself put it, so long as Turkish was written from right to left it could never properly express the ideals of European civilisation. The picturesque involutions and intricacies of Arabic script afforded a psychological background to the Oriental mentality which stood as the real enemy of the Republic” (Çolak 2004: 73).

The Language Revolution had the important role of advancing national cul-ture and the idea of a pure Turkish language. The Law on the Adoption and Ap-plication of the Turkish Alphabet (Türk Harflerinin Kabulü ve Tatbiki Hakkında




Kanun) (1928)3 ordered all state institutions and private societies to use the new Turkish alphabet in all their activities (Yildiz and Fryer 2004: 22). The revolution was to ensure that all citizens could consider themselves part of the new nation through a common language. Being able to speak Turkish was the single most important criterion for being considered Turkish, as Atatürk noted, “it is difficult to believe a person who claims to belong to Turkish culture and society if they don’t speak Turkish” (Okutan 2004: 181). Other laws passed had the same pur-pose. The Law on the Unification of Education (Tevhid-i Tedrisat Kanunu) (1924), with its roots in nineteenth century Ottoman reforms, secularised and centralised the education system introducing mixed gender education. This law banned the medrese, traditional religious institutions that had provided education in non-Turkish languages such as Kurdish. The Surname Law (Soyadı Kanunu) (1934) stipulated that all citizens had to adopt either Turkish surnames or surnames de-rived from Turkish.

Throughout the 1920s and 1930s the Kemalist regime mobilised all its forces to promote Turkish and spread the ideas and products of the language revolution. For example, during the “Citizen, Speak Turkish!” (Vatandaş, Türkçe Konuş!) campaign, launched in 1927 and which peaked in 1937, posters were hung on walls, bulletins were distributed on the streets and public declarations were made advocating that all Turkish citizens should speak Turkish (Sadoğlu 2003: 275–290). An important institution at the heart of the language revolution was the Turkish Language Institute (Turk Dil Kurumu) founded in 1932 through the initia-tion and encouragement of Atatürk, with the task to create a cohesive national language to shape a homogeneous nation. The aim of The Language Institution was the creation of pure Turkish (öz Türkçe) through eliminating Persian and Ara-bic words and influences and in their place inventing new words or introducing “pure” Turkish words assembled from various Turkish dialects. The Language Institution was part of an arsenal of other Kemalist institutions such as the Turk-ish History Institution (Türk Tarih Kurumu), with the primary aim to write and disseminate the new national history of the Turks. These and other institutions enabled the state to maintain control over all knowledge production. With the Turkish History and Language Institutions at its heart, the 1930s saw the launch of several conferences that were to provide the “science” behind the invention of the new Turkish language, nation, history and myths of origin for the new proud citizens of Turkey and at the same time deny any legitimacy to competing discourses of nationalism. These conferences resulted in the “Turkish History Thesis” (Türk Tarih Tezi) and the “Sun-language Theory” (Güneş-Dil Teorisi). The

3 For the Turkish original of the Law see: http://www.mevzuat.adalet.gov.tr/html/463.html.




basic argument of the History Thesis was that the Turks had historically been “culture creators” and disseminators of “civilisation” and that their homelands (Central Asia and Anatolia) were the cradle of human civilisation. The Sun-Language Theory, supported by the “findings” of the History Thesis, claimed that Turkish was the source of all languages. Although such ideas had a longer history dating back to earlier nationalist thought in the late 19th century Ottoman Em-pire, it was in this period that they were systematised with the potential to reach a wider section of the population disseminated through state apparatuses and the new national media. İbrahim Necmi Dilmen, the General Secretary of the Turkish Language Institution, for example, laid out the main purpose of the Sun-Language Theory during The Third Turkish Language Conference in 1936, argu-ing, “it would be impossible not to find the traces of Turkish cultural presence in the cultures of all languages known by history. By relying on these main lines of history, the fact that Turkish is the ultimate source of all languages will be estab-lished” (Üçüncü Türk Dili Kurultayı 1937: 64–65).

Through such arguments the “greatness” of the Turks was being “proven”, thereby debunking European Orientalist discourses on Turks as a “backward” race that had made no significant contribution to civilisation. It was important for Kemalist ideologues to prove to “the West” that Turks were not backward or mem-bers of the “yellow race” but in fact members of the “civilised White race”. As such, “rather than being the result of a profound change of mentality, the use of that terminology was an attempt to neutralise the racist condemnation of the Turks from within by utilising the same jargon” (Aytürk 2004: 19). The Turks were “carriers of civilisation” and it was a natural duty for the Turks to “spread civilisa-tion”, particularly to other “backward peoples” (Aydın 2001: 346). In the case of the Sun-Language Theory, this meant that Turkish as the source of all languages and the language of civilisation had to replace all “uncivilised” languages. For example, the scholar Celal Sahir Bey, during the proceedings of the First Turkish Language Conference in 1932, stated that the Theory was concerned with study-ing the languages of civilised nations and “without doubt, languages spoken by uncivilised tribes cannot be considered” (Birinci Türk Dil Kurultayı Müzakere Zabıtları 1933: 438). As Alastair Bonnett has argued, Turkish nationalism in this period “deployed a form of orientalism in which the East is cast as a separate and primitive realm, to be distinguished from both the West . . . as a model . . . and ‘the Turk’ as an idealised ethno-national identity” (Bonnett 2004: 74–76). The author-ities went to great lengths to disseminate these new ideas across society, in schools and school textbooks and the theories came to form the basis of Turkish historiography, official arguments and policies, also influencing literature, the-atre and journalism of the period. Until the 1970s, historical or sociological re-search that did not confirm these theories were not published (Nezan 1993: 76). In




fact, they continue to influence contemporary state, military, academic and pop-ular discourses in Turkey to varying degrees (Copeaux 1998).

2 The Turkish state and the Kurdish languageThe Kurdish question can be said to have constituted the most important chal-lenge to the Turkish Republic since its foundation. The Kurds, as the largest cul-turally and linguistically distinct non-Turkish people, have borne the brunt of the Kemalist nation-building project, which has resulted in the denial of their very existence and the banning of their language and culture, despite the fact that dur-ing the Ottoman Empire Kurdish tribes and principalities had significant self-rule and were officially recognised as an important component of the Muslim popula-tion of the Empire. However, in the early decades of the Turkish Republic, Kurds ceased to exist as a distinct ethnic group in official Turkish discourse and system-atic attempts were made to forcibly “turkify” them, with the Kurdish language as the primary target. As Donald Bloxham has outlined, “the assault on the Kurds was cast in terms of the broader fight against the influence of religion and reac-tionary traditionalism (and indeed many Kurds objected vociferously to seculari-sation) as republican Turkey established a tradition of refusing to admit its ethnic cleavage, depicting the millions of Kurds as ‘mountain Turks’ ” (Bloxham 2005b: 230).

In official Turkish state discourse, the existence of Kurds was denied, or con-structed as a threat to the very essence of the state and its identity. This discursive hegemony was achieved without the actual pronunciation of the words “Kurds”, “Kurdistan” or the “Kurdish question”. Instead terms such as “Mountain Turks”, “the East”, “banditry”, “reactionary politics”, “tribal resistance” or “regional backwardness” were deployed in order to represent Kurds as culturally and eco-nomically primitive, backward Muslims, tribal bandits, smugglers or as simple peasants exploited by feudal landlords (Yeğen 1999). This hegemonic construc-tion, backed by the physical force of the Turkish army, created the very basis for the suppression of Kurds. In terms of language, since the primary marker of dif-ferentiation between Turks and Kurds was language, the elimination of the Kurd-ish language (as well as other autochthonous non-Turkish languages), became the main aim of the Turkish nation-building project. Kurds were to successfully become “Turks”, not only by taking on Turkish as their new language and Kemal-ism as their new ideology, but also by rejecting and forgetting their mother tongue, identity, culture and heritage. Accordingly, the modern history of Turkey can be read as not only the denial and oppression of the Kurdish ethnic identity, but particularly the long-term policy of annihilating the Kurdish language.




Turkey’s Kurdish language policy has therefore been referred to as “linguicide” or “linguistic genocide”, the deliberate extermination of a language (Hassanpour 1992, 1993, 2000; Skutnabb-Kangas and Bucak 1995; Hassanpour, Skutnabb-Kangas and Chyet 1996; Skutnabb-Kangas 2000, 2005; Skutnabb-Kangas and Fer-nandes 2008; Skutnabb-Kangas and Taylor 2009; Skutnabb-Kangas and Dunbar 2010). As Tove Skutnabb-Kangas and Sertaç Bucak have cogently pointed out, “To kill a language you have to either kill the individuals speaking it or make these individuals change their mother tongue. Turkey tries to change the mother tongue of the Kurds and make Turkish their mother tongue” (Skutnabb-Kangas and Bucak 1995: 362).

In response to the state’s centralising, secularising and homogenising poli-cies there were a significant number of Kurdish uprisings (1925, 1927–1930 and 1937–1938), all of which were brutally crushed, followed by deportation, forced re-settlement and the massacring of a large number of Kurds. A report prepared by the Interior Minister following the Sheikh Said rebellion in 1925, for example, called the “Eastern Reform Plan” (Şark Islahat Planı), proposed that Kurdish prov-inces be ruled “in a colonial way” and that the region be turkified through forced resettlement. As part of the “reform” of the region, the Settlement Law (İskan Kanunu) of 1934 ordered the dispersion of Kurds in order to break up their social cohesion. The Settlement Law Interim Committee stressed that it was the primary duty of the Turkish Republic to exalt the Turks, who all shared the same race, mentality and language. Those “Turks” who had forgotten their “Turkishness” should be forced to appropriate it (Yıldız 2001: 242–253). The following year a law was passed which was aimed at “liberating” and “protecting” the “pitiful” peo-ple of the Dersim province by dismantling the tribal structure of the area, in order to better “civilise” (temdin) and “assimilate” (temsil ) its inhabitants (Yildiz 2001: 258–259). All such coercive social engineering policies were being pursued in breach of the Treaty of Lausanne. Its much quoted Article 39 stipulates:

No restriction shall be imposed on the free use by any Turkish national of any language in private intercourse, in commerce, religion, in the press, or in publication of any kind or at public meetings. Notwithstanding the existence of the official language, adequate facilities shall be given to Turkish nationals of non-Turkish speech for the oral use of their own lan-guage before the courts.

3 Military coups and language policiesTurkey saw important political transformations in the second half of the 20th century, such as the transition to multi-party rule and the election of the Demo-




crat Party (Demokrat Parti, DP) in 1950. The DP successfully attracted the sup-port of various sectors of society that had been silenced during the Kemalist single party rule (1923–1950). However, the Turkification of Kurds as a state policy persisted and Turkey’s linguicidal Kurdish policy remained unchanged and was reinforced with each military coup (1960, 1971 and 1980). Kemalism continued to form the official basis of the state’s ideology and the Kurdish ethnic identity continued to be systematically denied and persecuted. In 1959, Law No. 7267 stip-ulated that “village names that are not Turkish and give rise to confusion are to be changed in the shortest possible time by the Interior Ministry after receiv-ing the opinion of the Provincial Permanent Committee” (Yildiz and Fryer 2004: 23). Article 54 of the new 1961 Constitution, prepared by the military regime which toppled the democratically elected DP government and hanged the Prime Minister Adnan Menderes, repeated the previous constitutional provision that, “Everyone bound to the Turkish state through the bond of citizenship is a Turk”.4 The new constitution secured the permanency of Kemalism as the official ide-ology underlining the irreversibility of Atatürk’s revolution. Article 58 of the Law Concerning Fundamental Provisions on Elections and Voter Registers stated, “it is forbidden to use any other language or script than Turkish in propaganda disseminated in radio or television as well as in other election propaganda” (Yildiz and Fryer 2004: 26). Such laws were systematically deployed to imprison Kurdish speakers and those who dared to question the status quo. Following the coup, 485 prominent Kurdish intellectuals that criticised state policies were arrested and sent to a concentration camp in central Turkey without trial, while others were sent to exile in western parts of the country (Nezan 1993: 65). The Turkish Workers’ Party (Türkiye İşçi Partisi, TIP), for example, which managed to gain 15 seats in the parliament in 1965, came to be the first political party to address the Kurdish question and challenge this taboo subject. However, this led to the banning of the party for “encouraging separatist activities” (Gunter 1990: 17; Nezan 1993: 68–70). This period also saw the names of many Kurdish vil-lages and settlements change to Turkish as well as the launch of Turkish radio stations, hoping to spread the Kemalist ideology and counter popular radio broadcasts in Kurdish from neighbouring countries (Nezan 1993: 68–70). Board-ing schools were built in particular in the Kurdish provinces (Beşikçi 1970: 552–553), in order to cut off Kurdish children from their families and community and as in other educational and military institutions, they were strongly encouraged and more often forced to forget their mother tongue and exposed to propaganda

4 The English version of the 1961 Constitution can be accessed at: http://www.anayasa.gen.tr/1961constitution-text.pdf.




that Kurds were “bad”, “dirty” and “primitive”, while at the same time also main-taining that there were no such things as Kurds. Students were to feel ashamed of their language, culture and background (Skutnabb-Kangas 1981: 308–313). As emphasised so far, these strategies were part of a state policy to turkify the Kurd-ish population of Turkey in order to prevent the growth and spread of Kurdish nationalism.

In fact, a recently unearthed secret official report by Turkish journalists clearly confirms this policy. Prepared by the State Planning Organisation soon after the 1960 coup in order to “solve the problem of Kurdish separatism and re-gional underdevelopment”, titled “The Principles of the State’s Development Plan for the East and Southeast”, the report suggests various strategies in order to assimilate the Kurds (Dündar and Akar 2008). The extensive report proposes to eliminate Kurdish ethnic awareness by encouraging ethnic mixing through vol-untary and forced migration. It underlines the importance of commissioning so-ciological and anthropological studies of the region as well as financial incen-tives in the transformation of the ethnic composition of the Kurdish region. This is a clear example of how sciences have always been considered by the Turkish authorities as a “tool” to serve “state interests” and social engineering policies. Some of the proposals of the report were: the re-settlement of “those who con-sider themselves to be Kurdish” with “surplus population” from the Black Sea region in order to favour Turkification; to sever the links between “those who consider themselves to be Kurdish” in Turkey from Kurds in Iran, Syria and Iraq; to broadcast radio programs prepared by propaganda specialists and local (Kurd-ish) songs with Turkish lyrics; to persuade the local population that in racial terms the Turkish political system is the most beneficial for them; to inform the international intellectual community that Turkey does not have a Kurdish prob-lem; to immediately establish a Turkology Institute to produce scholarly work that will prove that Kurds are of Turkish origin and that the Turkish version of the history of the East be disseminated; publications to be produced and dissemi-nated which argue that Kurds originate from Turanian tribes. Ironically, such crude and oppressive assimilationist strategies and coercive practices often had the opposite effect of raising ethnic awareness among many Kurds, and the 1970s in particular saw a rise and radicalisation of Kurdish and leftist movements that challenged the official ideology.

In this highly politically dynamic period, various new political, religious and extremist movements, organisations and publications emerged. Strong popula-tion growth, urbanisation and the spread of political ideologies played an impor-tant role in politicising urban youth and strengthening peripheral voices. The Kurdish question easily re-emerged in this political climate with leftist journals, newspapers and books drawing attention to the Kurds’ plight and enabling Turk-




ish and Kurdish intellectuals to challenge the official line. However, all such ac-tivities were systematically suppressed, with many writers and publishers im-prisoned for “propagating separatism” (Gunter 1990: 4–19; Nezan 1993: 66–68). Fanned by economic recession and instability, a rise in social unrest, strikes, demonstrations and political assassinations resulted in another coup in 1971, where the government was forced to resign following an ultimatum from the Turkish army. The military urged the government to carry out reforms “in a Ke-malist spirit”. Thousands of Turkish and Kurdish activists were arrested. The Diyarbakır military court sentenced more than a thousand “Kurdish separatists”, while counter-insurgency units were “raking through the Kurdish provinces one by one; several thousand peasants were pursued, arrested and tortured” (Nezan 1993: 78). The chaotic and violent decade of 1970s came to an end with the brutal military intervention of 12 September 1980, a milestone in the social engineering of the Turkish society and the severe suppression of Kurdish ethnic identity and language.

The systematic repression of politically dissident Turks and Kurds and the policy of assimilating the Kurds reached its peak with the 1980 coup, and the Kurdistan region continued to be ruled “in a colonial way”. The military junta declared that its goal was “to exterminate communism and separatism”. In 1982, a highly secret booklet internally distributed by the Turkish Land Forces Com-mand identified the Kurds as the primary “divisive and destructive force” arguing that the word Kurd comes from the noise created by “Mountain Turks” of east-ern Turkey when they walk on the snow (Kara Kuvvetleri Komutanlığı 1982: 43). Officials ordered Kurdish folk songs to be sung only in Turkish to avoid “separat-ism” and public speaking or printing in Kurdish was banned and thousands of newspapers, magazines and books on Kurds were confiscated and burnt. David McDowall points out that by 1986, 2,842 more Kurdish villages had been given Turkish names (McDowall 2000: 424). In fact, a study of this state policy has shown that between 1940 and 2000, the names of more than 12,000 villages, amounting to every third village in Turkey, were “turkified”, with a particular concentration in the Kurdish provinces and the Black Sea regions (Tunçel 2000). In the 1980s, the notorious Diyarbakır Military Prison, with a sign at its en-trance hall ordering “Speak Turkish, speak it a lot” (Türkçe konuş, çok konuş), became a concentration camp where thousands of Kurds were brutally tortured and “turkified”, with many killed or maimed (Zeydanlıoğlu 2009). A fact-finding mission of the main opposition Social Democratic People’s Party (Sosyal De-mokrat Halk Partisi, SHP) in 1986 reported that all of “eastern Turkey had become a sort of concentration camp where every citizen was treated as a suspect, and where oppression, torture and insult by the military was the norm” (Taşpınar 2005: 100–101).




The new 1982 Constitution5 and other legislation stipulated under strict mili-tary guidance, with a heavy emphasis on Turkish ethnicity, legitimised such widespread suppression of Kurdish identity during this period and indeed con-tinue to play a crucial role in the repression of the Kurdish language today. The Preamble of the Constitution states, “No protection shall be accorded to an activ-ity contrary to the national interest of the Turks or their existence, the principle of the indivisibility of Turkey with its state and territory, the historical and moral values of Turkishness [Türklük] or the nationalism, principles, reforms and mod-ernism of Atatürk”. The fundamental principles of Turkey’s national and territo-rial integrity and political unity, Kemalist nationalism and the official language are systematically embedded in the Constitution and the Turkish legal system. Article 2 of the constitution stresses that the Republic of Turkey is bound to the nationalism of Atatürk, while Article 3 provides that the Turkish state, with its territory and nation, is an indivisible entity, and that its language is Turkish. Ar-ticle 4 prohibits not only the amendment of these provisions, but also the pro-posal to amend them. Article 26 stated that “no language prohibited by law shall be used in the expression and dissemination of thought” (amended in 2001). Article 42, which is still in force today, provides that “no language other than Turkish shall be taught as a mother tongue to Turkish citizens at any institutions of training or education”. Crucially, Article 28/2 specified at the time, “no publica-tions or broadcasts may be made in any language prohibited by law” (also amended in 2001). The legally prohibited languages in question were “languages other than those which are the primary official languages of states recognised by the Turkish State” (Law 2932) meaning in particular Kurdish. Law 2932, which was not annulled until 1991, stipulated “the mother tongue of Turkish citizens is Turkish” (Skutnabb-Kangas and Bucak 1995: 355–356). Article 174 currently af-firms the commitment to retain Kemalism by providing that no provision of the constitution can be interpreted so as to render unconstitutional a number of Atatürk’s key Reform Laws, and safeguard the secular character of the republic. Furthermore, Article 81 of the Law 2820 on Political Parties, 1982, which also still remains in force today without any amendments, stipulates that political parties cannot “claim that there exist minorities in Turkey. It is forbidden to protect or develop non-Turkish cultures and languages” (Skutnabb-Kangas and Bucak 1995: 356).6

5 For the Turkish original of the 1982 Constitution and a description of all the amendments to date see: http://www.anayasa.gen.tr/1982ay.htm. For the English version see: http://www.anayasa.gen.tr/1982constitution.htm.6 For the Law on Political Parties see the website of the Turkish Parliament: http://www.anayasa.gen.tr/2820sk.htm.



http://www.anayasa.gen.tr/2820sk.htm


The 1982 Constitution and the various legal provisions accompanying it con-tinue to haunt Turkey’s difficult democratisation effort to this day and justify the suppression of the Kurdish language. Such laws provided and continue to pro-vide the legal basis for Turkey’s policy of denying and oppressing the Kurdish identity and persecuting Kurdish speakers. However, instead of silencing Kurdish nationalism, these laws have played a crucial role in fuelling the largest Kurd-ish insurgency in the modern history of Turkey launched by the Kurdistan Work-ers’ Party (Partiya Karkerên Kurdistan, PKK). Since the launch of its armed rebel-lion in 1984 for an independent Kurdish state, the PKK has successfully tapped into the frustrations and grievances of a large number of Kurds affected by the above described laws and practices. The immensely destructive civil war that has ensued between the PKK guerrillas and the Turkish army, with an estimated death toll of more than 40,000 mostly civilian Kurds, only came to an end when the PKK leader Abdullah Öcalan was captured in Kenya in 1999, although military operations and sporadic clashes continue to this day.

4 Easing restrictions?The military junta “left office” in 1983 after having brutally but temporarily si-lenced all possible sources of dissent, however, the 1982 Constitution and other legislation guaranteed them an “omnipotent presence” and immense powers in influencing various aspects of life in Turkey. This was certainly the case when it came to “national issues” such as the Kurdish question, where the military ulti-mately had the final say and there was very little room for politicians to go against the official line on the Kurds. Turgut Özal led the Motherland Party (Anavatan Partisi, ANAP) to victory in the 1983 general election and he faced staunch oppo-sition from the military and Kemalist political elite when he attempted to find some sort of political rather than military solution to the Kurdish question. De-spite this resistance, Özal managed to introduce a number of limited reforms under his leadership which included repealing Law 2932 that banned the Kurdish language. At least on paper, this repeal meant that the use of Kurdish for non-political purposes was no longer illegal, such as speaking Kurdish, publishing newspapers in Kurdish or playing Kurdish music. However, education and broad-casting in Kurdish were still prohibited by law (Kirişci and Winrow 1997: 137).

It should be noted, however, that despite these reforms it was under Özal’s leadership that the notorious “Village Guards” (Korucular) system was intro-duced in 1985, which recruited approximately 100,000 Kurdish paramilitaries to fight the PKK (Bozarslan 2000: 24). In 1987, a State of Emergency (Olağanüstü Hal ) was declared in the Kurdish region which bestowed the regional governor with




special powers, such as evacuating villages, restricting the press and overseeing civil trials against security force personnel (Gunter 1990: 83). This State of Emer-gency remained in effect until 2002. Indeed on the very same day in 1991 that the language ban on Kurdish had been repealed, the Anti-Terror Law 3713 was passed which defined terrorism so vaguely that not only were the PKK directly targeted by this legislation but also anyone involved in the promotion of Kurdish lan-guage or culture (Terörle Mücadele Kanunu 3713). Article 8 of the Act enabled prosecutors to charge individuals on the basis of engaging in “verbal and written propaganda [that] aims to destroy the national unity and the indivisibility of the Turkish Republic” and has been systematically used against Kurdish politicians, intellectuals and activists. The Terror-Law was also the first clear sign of the start of the Turkish state’s “low intensity war” against the Kurds and the continued militarisation of the Kurdish question (Koivunen 2002).

By the early 1990s, it had become clear that the state had failed to put an end to popular dissent and the growing Kurdish nationalist mobilisation, with the PKK transformed into a mass movement. In a letter in 1993, Özal warned the then Prime Minister Süleyman Demirel that, “the Turkish Republic is facing its gravest threat yet. A social earthquake could cut one part of Turkey off from the rest, and we could all be buried beneath it” (Pope 1993). This was a serious threat to the Kemalist nation building project and a clear sign that policies of coercive assimi-lation had failed to achieve the anticipated result. The limited gestures of the Özal era, most of which stayed only on paper and others which ultimately failed to alter core policies came to an end with his suspicious death in 1993 and the sup-pression of the Kurds as an entrenched state policy continued unabated.

The 1990s were marked by thousands of deaths and immense material and environmental cost. During the civil war the Kurdish provinces were transformed into a militarised zone by the Turkish army. As Gulistan Gurbey has put it, “the military state of emergency and the use of force by the government in Kurdish areas in the southeast of Turkey lasted for decades. Forced expulsion from their homes and farms and resettlement in other localities, banning the expression of Kurdish identity, arbitrary arrest and persecution, and torture have become the everyday experience for Kurds in Turkey” (Gurbey 2005: 137–138). In its “dirty war” against the PKK, the Turkish army deployed the Kurdish paramilitary Vil-lage Guards, intelligence organs, criminal elements and Special Forces (Özel Timler) without any differentiation between civilians and combatants, murdering scores of Kurdish businessmen, intellectuals, activists and politicians. Mirroring earlier resettlement policies, the 1990s witnessed the systematic siege and de-struction of more than 3000 villages and settlements resulting in the displace-ment of 3 to 4 millions Kurdish peasants (McDowall 2000: 440–441). All this had a disastrous effect on the Kurdish society and language as Kurds were displaced




across the region and the rest of Turkey, with the natural environment in which Kurdish had thrived for centuries significantly destroyed (for the relationship be-tween displacement and language loss see Williams [1988]). At the same time, the state’s coercive policies galvanised Kurds and fuelled the Kurdish nationalist movement, with pro-Kurdish parties persistently gaining the majority of votes in the Kurdish region.

In the political arena, Kurdish politicians were repeatedly abused, harassed and prosecuted and several murdered and all pro-Kurdish parties were system-atically closed by prosecutors, only to reappear under a different name. As Watts has underlined, “No political party in Turkish history had ever sustained parlia-mentary representation while promoting formal recognition of a Kurdish people as its central political platform” (Watts 1999: 636). This was best symbolised by the imprisonment of Leyla Zana, the first Kurdish woman elected to parliament, who scandalised the Kemalist establishment at her swearing-in ceremony in 1991 by speaking in Kurdish and wearing a headband with the traditional Kurdish colours. Adding to her oath in Kurdish, she declared, “I have sworn this oath for the sake of brotherhood between the Turkish and the Kurdish people” (van Bru-inessen 2001: 107). Through this act, Zana became the first person to speak Kurd-ish in the parliament and was convicted at Ankara State Security Court in 1994 of membership of an illegal armed organisation, the PKK, under Article 168/1 of the Turkish Penal Code and sentenced to 15 years’ imprisonment. Zana was released in 2004 but has been continually prosecuted and was sentenced to 10 years’ im-prisonment in 2008 for various political speeches (BBC News 2008).

5 The beginning of a new era?Since the late 1990s, with Turkey’s EU bid and the required democratisation re-forms there has been a very slow and gradual easing of Turkey’s prohibitive Kurd-ish language policy, although the process remains riddled with contradictions as politicians and activists continue to be prosecuted and restrictions on the Kurdish language remain. While, on paper, certain reforms ease restrictions on freedom of expression or cultural or linguistic rights, the staunchly Kemalist judicial bureau-cracy and security forces on the ground ensure that old policies of suppression continue. In fact, the European Court of Human Rights (EctHR) has repeatedly condemned Turkey for violating the universal rights of its citizens and recently declared Turkey the worst violator of human rights among the 47 signatory states to the European Convention on Human Rights (Today’s Zaman 2010).

The EU’s Copenhagen criteria, which requires that the candidate state has institutions to preserve democracy, the rule of law, human rights, and the




recognition and protection of minority rights, has led to a number of reforms, however, there have been serious problems with their actual implementation. Turkey’s candidacy for EU membership was initially accepted in 1999 and the EU held its first talks with the republic in 2004. Following these talks, Turkey pledged to carry out constitutional reforms in line with EU standards. The Justice and De-velopment Party (Adalet ve Kalkınma Partisi, AKP), with its roots in political Is-lam, which came to power in 2002, has spearheaded a number of reforms in order to comply with EU accession criteria. These have been introduced through “Har-monisation Laws” which have resulted in various amendments to the 1982 Con-stitution and other legislation. However, in 2007 the Kemalist military threatened to oust the AKP from power and the party narrowly evaded closure in 2008, ac-cused of violating the constitution by attempting to change the country’s secular character. It has also recently emerged that since the AKP came to power, high-ranking generals have hatched several plots to create instability and chaos in the country in order to undermine the government and create an atmosphere favour-able for another coup (Erdem 2010). Therefore, the AKP has had to tread a deli-cate path when pursuing reforms, particularly relating to the Kurdish question, which perhaps explains in part why the implementation of these reforms has often been slow and unsystematic. It is also worth noting that not a single one of these constitutional and legislative amendments ever refer to the Kurdish lan-guage or Kurds specifically.

The amendment of Article 26 of the Constitution was a significant reform as this article had proscribed the use of prohibited languages “in the expression and dissemination of thought”. This led to changes in other existing legislation which had previously banned Kurdish as a “prohibited language”, such as the harmoni-sation law which amended legislation on radio and television broadcasting:

Although Turkish will be the basis of TV and radio broadcasts, broadcasts in different lan-guages and dialects used by Turkish citizens in their daily lives is made possible. Further-more, it has been emphasised that such broadcasts cannot be against the fundamental principles in the Republic enshrined in the Constitution and the indivisible integrity of the state with its territory and nation. (Kurdish Human Rights Project 2005: 17)

However, as these reforms were taking place, the widespread suppression of Kurds continued. For example, in 2001, whilst constitutional amendments im-proving human rights were being voted on in parliament, organisers of a Kurdish football tournament faced prison sentences of up to five years after footballers played in kits sporting the Kurdish national colours (Huggler 2001). Although re-forms made it possible for newspapers and books to be produced in Kurdish, their political content led to court cases against many authors and publishers. For ex-




ample, Azadiya Welat [Free Country], the only Kurdish daily newspaper in Tur-key, “has had to re-establish itself 3 times in the 12 years it has been in existence, and had 25 cases against them in that 12-year period” (Kurdish Human Rights Project 2005: 22). Similarly, changes in legislation granted parents the right to give Kurdish names to their children as long as they were not “subversive” or containing letters not in the Turkish alphabet (Q, W and X; letters used in the Kurdish alphabet). Kurdish politicians continued to be prosecuted for speaking Kurdish; a member of the pro-Kurdish Democratic People’s Party (Demokratik Halk Partisi, DEHAP) was sent to jail for 6 months in 2005 for simply saying “I” in Kurdish at an official function and in the same year there were “60 cases pending against the DEHAP president in Ankara for saying ‘Rojbaş’ (Hello) in Kurdish, versus ‘Merhaba’ which is used in Turkish” (Kurdish Human Rights Project 2005: 26) (this party has since also been banned). Abdullah Demirbaş, the mayor of the municipality of Sur in Diyarbakır, was dismissed in 2007 and the municipal coun-cil dissolved for providing multilingual services in the municipality. The “sepa-ratist” services in question included bilingual publications for children in Kurd-ish and Turkish, tourist guides in six languages, and a promotional video of Diyarbakır in Turkish, Kurdish and English (Toumani 2008). More recently in 2009, the Kurdish politician Orhan Miroğlu was found guilty for speaking Kurdish during the campaign for elections in 2007. The court held that, among others, Miroğlu had breached Law 298 on Basic Provisions on Elections and Voter Regis-ters which prohibits the use of languages other than Turkish in electoral propa-ganda (Önderoğlu 2009). Thus, reforms have failed to change the reality on the ground as prosecutors have continued to launch cases against Kurdish politicians and activist for speaking Kurdish or for campaigning for Kurdish rights.

Reforms with regards to the Kurdish language education have faced similar restrictions. The 2003 Law on Teaching in Different Languages and Dialects Tra-ditionally Used by Turkish Citizens in their Daily Lives (for the English translation of the law see BİA News [2003]) allowed private courses teaching Kurdish to be opened, but as with other reforms, serious caveats were introduced. For example, courses could only last for 10 weeks and no more than 18 hours per week and were for adult students only; teachers had to be native speakers of Turkish and have a diploma (it was not clarified how these diplomas were to be obtained); students were to pay high fees and the buildings in question were to meet unusu-ally strict regulations and that the private courses were not to receive financial support from the state (Yildiz and Muller 2008: 87–88). This legislation was a good example of the Turkish authorities’ contradictory position vis-à-vis the Kurdish language, as law-makers were going to great lengths to prevent an actual change in Turkey’s language policy, through the very law that was to make this possible. Not surprisingly, Turkey’s brief experience with the few private Kurdish




language courses that were eventually set up after immense bureaucratic hurdles failed as a result, with certain mainstream Turkish newspapers and news agencies presenting this as a “proof” that Kurds were not interested in learning their own language (on the representation of Kurds in the Turkish media see Zeydanlıoğlu and Demir [2010]; Sezgin and Wall [2005]). Kurdish parents, stu-dents and activists in response launched a large-scale campaign for public Kurd-ish education and called for changes to the constitution, successfully collecting hundreds of thousands of signatures that were then submitted to the parliament. The authorities responded to this campaign by arresting hundreds of students for “supporting an illegal organisation” under Article 169 of the Penal Code (Mater 2002). Thus, the above described half-hearted “concessions” have often come too late and generally fall short of Kurdish expectations, especially when one con-siders the fact that Kurdish is now an official language of federal Iraq. As Hamit Bozarslan has underlined, “they deepen the feeling of humiliation; no private teaching institution, for example, has yet obtained the right to teach in Kurdish. In Batman, for instance, the courses could not start for one technical reason; the outside door was 85 cm. large, instead of the 90 demanded by the authorities” (Bozarslan 2005: 133).

In 2004, Turkey’s public service broadcaster the Turkish Radio and Televi-sion Corporation (Türkiye Radyo ve Televizyon Kurumu, TRT) was allowed to broadcast in Kurdish, albeit only for two hours per week, always with Turkish subtitles and with no programmes to target children. In parallel, various other small private TV and radio stations continued to be prosecuted for broadcasting Kurdish music or Kurdish programmes. For example, the local private TV station, Gün TV, which broadcasts from Diyarbakır, has continuously faced legal harass-ment with more than 60 cases filed against it since its foundation in 2001 (Kurd-ish Human Rights Project 2005: 23). On 20 March 2009, the station was fined and taken off air for 12 days for not providing Turkish subtitles to a political de-bate program (CNN Türk 2009). Similarly, Roj TV, the Denmark based Kurdish satellite station, continues to be banned in Turkey, which considers the channel to be a PKK mouthpiece. Several Kurdish politicians have been prosecuted for having appeared on Roj TV or for protesting against its closure in Germany (AFP 2007).

In January 2009, however, state-run channel TRT-6 commenced its broadcast exclusively in Kurdish, with Prime Minister Erdoğan congratulating the launch of the channel with a few words in Kurdish. The content of the “family channel” TRT-6 consists of programs broadcast on Kurdish culture, literature, cuisine, mu-sic and history. There are also general interest programs on health, travel, nature, religion and cartoons for children as well as talk shows, news and debates. There is a conscious effort by the channel not to be “political” with words such as Kurd-




istan and Kurdish names of cities and settlements strictly avoided. Soon after its launch, however, the channel ran into problems, as sceptics had predicted. The popular chat show Rojînname, presented by the Kurdish pop-star Rojîn, came to a sudden end in April after Rojîn resigned. Rojîn issued a statement saying that she was under immense pressure by the directors, heavily censored and treated as a potential criminal, with the show deprived of a meaningful content (Today’s Zaman 2009a). TRT rejected her claims, although since then the channel’s pro-duction company has been dismissed (Fırat New Agency [ANF] 2010). Overall, the majority of commentators considered TRT 6 to be a historic change in Turkey’s attitude to the Kurdish language and the general response was positive. However, some Kurdish politicians remained sceptical arguing that it was AKP’s electoral investment in the region, while the PKK called for the channel to be boycotted (Today’s Zaman 2009b).

One cannot deny the symbolic significance of TRT 6 in terms of rendering Kurds and their language “visible” in Turkey after decades of denial and repres-sion and to this date it can be said that the channel remains one of the few actual achievements of the government. However, such positive developments are con-tinuously undermined by other measures, which cast a dark shadow over positive steps and create distrust in the reform process. For example, during a live broad-casting of the pro-Kurdish Democratic Society Party (Demokratik Toplum Partisi, DTP) meeting in the parliament in March 2009, Ahmet Türk, the chairman of the party, announced that he would deliver the rest of his speech in Kurdish. The state-owned TRT1 channel immediately cut the broadcast, the announcer stating that this would breach the Constitution and the Law on Political Parties. The irony was not lost on Ahmet Türk, who pointed out, “When [Kurdish party] mem-bers salute someone in their own language, they are prosecuted or investigated. When a mayor speaks to his people in their own language, he is prosecuted. But when the prime minister speaks Kurdish, nobody says anything. We don’t think this is right. This is a two-faced approach” (Sobecki 2009).

As with other examples provided, this also clearly shows the problems of the uneven reform process in Turkey, which has been a process of giving with one hand while taking with the other. As Kerim Yildiz and Mark Muller have concluded:

Turkey, then, whilst having made concessions in the field of cultural and linguistic rights which at first sight appear groundbreaking, can on closer inspection be seen to be doing little more than paying lip service to the pro-EU reform process. At root, she re-mains committed to promulgating official Turkish nationalism, and tied up in paranoia over increased cultural and linguistic rights spelling the break-up of the Turkish Republic. She has a great way to go before cultural pluralism is realized. (Yildiz and Muller 2008: 89)




In August 2009, the AKP government launched a “Kurdish initiative”. How-ever, following criticism from nationalist circles, it became the “national unity and brotherhood project”, with the government declaring that it was deter-mined to solve the Kurdish question. Despite strong resistance from the nation-alist opposition accusing the government of “treason” and collaborating with terrorists, the AKP politicians declared their dedication to continue with the “ini-tiative”, raising expectations that the AKP government would finally take serious steps towards solving the Kurdish question. In an emotional speech to party members in the parliament, Erdoğan declared that Turkey would have been in a different place if it had managed to find a solution to the Kurdish question stating, “now we are saying that Turkey has to confront the problem. Turkey has to per-manently solve this problem. . . . We believe in this. We have taken steps, and we will continue to, no matter what price we might have to pay” (Today’s Zaman 2009c).

As part of the “initiative” the government anticipated that in the long term it would bring about the eventual disarmament and re-integration of PKK guerrillas in the mountains of Iraqi Kurdistan back into Turkish society and solve the Kurd-ish question through granting individual and cultural rights. It was emphasised by officials that such steps were to be an essential part of the process of Turkey’s overall democratisation with a new constitution and in tandem with EU reforms. Official sources added that the package was to remove restrictions in education and broadcasting in Kurdish. For example, it was pledged that Kurdish was to become an elective course in schools and private TV and radio stations were to be permitted to broadcast in Kurdish (Today’s Zaman 2009d). As part of the initia-tive, restrictions banning prisoners from speaking Kurdish among themselves or with visitors were soon lifted, at least in theory, as it has since been difficult to assess how effectively this has been implemented. Restrictions on religious ser-mons in Kurdish were also lifted in the period that followed. AKP politicians also signalled that the Law on Political Parties would be amended to allow Kurdish to be used in political speeches, campaigns, slogans, posters etc (CNN Türk 2010). In September 2009, Turkey’s Higher Education Board (Yüksek Öğretim Kurulu, YÖK) accepted the application of Artuklu University in Mardin to establish an “Institute of Living Languages” to provide postgraduate education primarily in Kurdish but also in other regional languages. However, despite such steps, the government remained apprehensive. For example, the “Living Languages Insti-tutes” were originally to be called “Kurdish Institutes” but shortly afterwards came to be referred to, somehow ironically, as “Institutes of Living Languages” instead. As a hopeful 2009 drew to a close, the government’s Kurdish initiative was dealt a lethal blow by the country’s Constitutional Court, which banned the pro-Kurdish DTP in December for having links with the PKK (BBC 2009a) and




dozens of senior politicians were banned from politics for five years. The closure of DTP was followed by the mass-arrest of Kurdish activists and politicians, in-cluding several mayors of Kurdish towns, accused of having links with the PKK (BBC 2009b).

The DTP has since been replaced by its new incarnation the Peace and Demo-cracy Party (Barış ve Demokrasi Partisi, BDP). The military and political situation has deteriorated and the “Kurdish initiative” has come to a standstill, with the PKK accusing the government of using the “initiative” to silence the Kurdish movement as the government has failed to respond to its cease-fires. Meanwhile, Turkey’s only daily Kurdish language newspaper, Azadiya Welat, was again the target of Turkish prosecutors. Vedat Kurşun, the former managing editor of the newspaper, who was imprisoned in 2009 for previous “offences”, received a fur-ther prison sentence of 166 years and six months by the 5th High Criminal Court of Diyarbakır in May 2010. Kurşun was convicted under Article 7 of the infamous Anti-Terror Law for “spreading propaganda for an illegal organization” (Önderoğlu 2010). An Istanbul court had recently suspended the Kurdish language news-paper for two months under another article of the Anti-Terror Law. A further two editors of the newspaper, as well as editors of various other Kurdish and dissident leftist publications are currently facing separate charges and lengthy prison sen-tences (Reporters without Borders 2010a, 2010b). In a statement on the case, the press freedom NGO Reporters without Borders summarised Turkey’s contradic-tory policies as follows:

We condemn the persecution of this newspaper and its editors. These disproportionate punishments expose the contradictions of the government’s policies, especially last year’s initiative that was supposed to give more rights to Turkey’s 25 million Kurds (a quarter of the country’s population) and draw them closer to international standards. The Turkish authorities seem unable to shed their repressive attitudes even when the country’s only Kurdish-language daily newspaper is concerned. The sentence speaks volumes. (Reporters without Borders 2010a)

Such developments are not only casting a dark shadow over the initiative but also raising suspicion that the government is planning to “solve” the Kurdish question without the legally elected representatives of millions of Kurds. This has been most apparent in the mass show trial of 152 Kurdish politicians for their al-leged affiliation with the PKK. Throughout the trial the defendants have insisted on speaking Kurdish and their right to interpreters has been rejected by the court on the grounds that the defendants also speak Turkish. When the defendants have responded in Kurdish, their microphones have been switched off and the proceedings officially recorded as having taken place “in an unknown language”. The court also rejected a report prepared by an expert on minority rights which




argued that legal defence in Kurdish was lawful under Article 39 of the Treaty of Lausanne (Üstündağ 2010).

6 ConclusionDevelopments so far indicate that there have been no fundamental changes in terms of Turkey’s Kurdish policy. In its totality, it can be argued that Turkey’s linguicidal policy continues and that the future of the Kurdish language in Turkey remains under threat because of this very policy. There remain serious political obstacles inhibiting Kurdish speakers from passing on their language to future generations. The military and political elites ultimately continue to consider the persistent survival of the Kurdish language as a “problem” and an “obstacle”; a painful reminder that Turkey has not successfully managed to create a homoge-neous Kemalist nation-state of Turkish speakers. Accordingly, public education in Kurdish in Turkey remains prohibited. Article 42 of the Constitution still main-tains that “no language other than Turkish shall be taught as a mother tongue to Turkish citizens”. While a 24 hour state-channel broadcasts in their language, many Kurds continue to face daily a web of legal restrictions and an insecure and hostile social context informed by a hysterical Turkish nationalist discourse. A militarist Turkish nationalism that forces Kurds to “become Turks” and yet con-siders those who resist this policy as “enemy others” permeates all levels of soci-ety. The Kurdish language in Turkey continues to be one of the most oppressed languages in the world. Campaigners, activists, writers, lawyers and politicians struggling for the survival of the Kurdish language and against Turkey’s repres-sive policies are persistently harassed and persecuted. Various Kurdish linguistic and cultural organizations continue to be prosecuted and suspended for their activities. The limited pro-EU reforms have failed to change the reality on the ground as the Constitution and various other laws, such as the notorious and extremely vaguely defined Anti-Terror Law, allow prosecutors ample room to si-lence dissident voices. In short, Kurds continue to be deprived of their most basic rights in Turkey and persecuted simply for being Kurds.

The positive steps of 2009 by the AKP government raised great expecta-tions, but the government unfortunately backtracked on its own project when faced with difficulties, failing to take the necessary brave steps. AKP’s so far failed initiative again bitterly shows the difficulty of finding a just and demo-cratic solution to the Kurdish question amid entrenched militarist policies and authoritarian habits, ultra-nationalist opposition and bureaucratic unwilling-ness for genuine change. As long as the question of Kurds’ human, political, cul-tural and linguistic rights are not attended to and a just democratic solution not




found, then the resultant endemic conflict will continue to haunt Turkey for the foreseeable future. Today, it is obvious that Turkey is going through an impor-tant but highly turbulent and difficult process of transformation with the out-come very difficult to predict. The need for a democratic Turkey within which Kurds and their language can exist freely, however, seems more urgent today than ever.

ReferencesAgence France-Presse (AFP). 2007. Prosecutor seeks 15 years for Kurdish mayors over Denmark

letter. 3 April. http://www.institutkurde.org/en/info/latest/prosecutor-seeks-15-years-for-kurdish-mayors-over-denmark-letter-624.html (accessed 1 November 2010).

Ahmad, Feroz. 1993. The making of modern Turkey. London: Routledge.Akçam, Taner. 2004. From empire to republic: Turkish nationalism and the Armenian genocide.

London: Zed Books.Arai, Masami. 1994. Jön Türk dönemi Türk milliyetçiliği [Turkish nationalism in the Young Turk

era]. İstanbul: İletişim Yayınları.Aktar, Ayhan. 2000. Varlık vergisi ve ‘türkleştirme’ politikaları [The capital tax and policies of

‘Turkification’]. İstanbul: İletişim Yayınları.Aydın, Suavi. 2001. Cumhuriyet’in ideolojik şekillenmesinde antropolojinin rolü: Irkçı

paradigmanın yükselişi ve düşüşü [The role of anthropology in the ideological formation of the republic: the rise and fall of the racist paradigm]. In Ahmet İnsel (ed.), Modern Türkiye’de siyasi düşünce: Kemalizm [Political thought in modern Turkey: Kemalism], 344–369. İstanbul: İletişim Yayınları.

Aydıngün, Ayşegül & İsmail Aydıngün. 2004. The role of language in the formation of Turkish national identity and Turkishness. Nationalism and Ethnic Politics 10(3). 415–432.

Aytürk, İlker. 2004. Turkish linguists against the West: the origins of linguistic nationalism in Atatürk’s Turkey. Middle Eastern Studies 40(6). 1–25.

BBC News. 2008. Turkey jails Kurdish politician. 4 December. http://news.bbc.co.uk/1/hi/world/europe/7765734.stm (accessed 1 November, 2010).

BBC News. 2009a. Turkish top court bans pro-Kurdish party. 11 December, http://news.bbc.co.uk/1/hi/world/europe/8408903.stm (accessed 1 November 2010).

BBC News. 2009b. Turkey holds dozens in anti-PKK raids. 24 December. http://news.bbc.co.uk/1/hi/world/europe/8430044.stm (accessed 1 November 2010).

Beşikçi, İsmail. 1970. Doğu Anadolu’nun düzeni: Sosyo-ekonomik ve etnik temeller [The order of Eastern Anatolia: socio-economic and ethnic foundations]. İstanbul: E Yayınları.

BİA News. 2003. Regulation on teaching different languages. 10 December. http://bianet.org/english/women/27271-regulation-on-teaching-different-languages (accessed 1 November 2010).

Birinci Türk dil kurultayı müzakere zabıtları 1932 [Theses proceedings of the first Turkish language congress 1932]. 1933. Ankara: The Turkish Language Institution. http://www.tdkkitaplik.org.tr/kurultay01/K010901.pdf (accessed 1 November 2010).

Bloxham, Donald. 2005a. The great game of genocide: imperialism, nationalism, and the destruction of the Ottoman Armenians. Oxford: Oxford University Press.



http://www.institutkurde.org/en/info/latest/prosecutor-seeks-15-years-for-kurdish-mayors-over-denmark-letter-624.html

http://www.institutkurde.org/en/info/latest/prosecutor-seeks-15-years-for-kurdish-mayors-over-denmark-letter-624.html


Bloxham, Donald. 2005b. Changing perceptions of state violence: Turkey’s “westward” development through Anglo-Saxon eyes. In Richard Littlejohns & Sara Soncini (eds.), Myths of Europe, 223–234. Amsterdam: Rodopi.

Bonnett, Alastair. 2004. The idea of the West: culture, politics and history. Hampshire: Palgrave Macmillan.

Bozarslan, Hamit. 2000. Why the armed struggle? Understanding the violence in Kurdistan of Turkey. In Ferhad Ibrahim & Gulistan Gurbey (eds.), The Kurdish conflict in Turkey: obstacles and chances for peace and democracy, 17–30. New York: St Martin’s Press.

Bozarslan, Hamit. 2005. The Kurdish issue in Turkey following the 2003 Iraqi war. In Mohammed M. A. Ahmed & Michael M. Gunter (eds.), The Kurdish question and the 2003 Iraqi war, 123–135. Costa Mesa: Mazda Publishers.

Bruinessen, Martin van. 2001. From Adela Khanum to Leyla Zana: women as political leaders in Kurdish history. In Shahrzad Mojab (ed.), Women of a non-state nation: the Kurds, 95–112. Costa Mesa: Mazda Publishers.

Cizre, Ümit. 2001. Turkey’s Kurdish problem: borders, identity, and hegemony. In Brendan O’Leary, Ian S. Lustick & Thomas M. Callaghy (eds.), Right-sizing the state: the politics of moving borders, 222–252. Oxford: Oxford University Press.

CNN Türk. 2009. Gün TV’ye 12 gün yayın durdurma cezası [Gün TV ordered to cease broadcasting for 12 days]. 20 March. http://www.cnnturk.com/2009/turkiye/03/20/gun.tvye.12.gun.yayin.durdurma.cezasi/518725.0/index.html (accessed 1 November, 2010).

CNN Türk. 2010. Kürtçe propaganda yasağı kalkabilir [The ban on propaganda in Kurdish could be lifted]. 31 January. http://www.cnnturk.com/2010/turkiye/01/31/kurtce.propaganda.yasagi.kalkabilir/561729.0/index.html (accessed 1 November 2010).

Copeaux, Etienne. 1998. Türk Tarih Tezinden Türk-İslam Sentezine [From the Turkish history thesis to the Turkish-Islamic synthesis]. İstanbul: Tarih Vakfı Yurt Yayınları.

Çolak, Yılmaz. 2004. Language policy and official ideology in early republican Turkey. Middle Eastern Studies 40(6). 67–91.

Dadrian, Vahakn. 2004. The history of the Armenian genocide: ethnic conflict from the Balkans to Anatolia to the Caucasus. Providence, RI: Berghahn Books.

Dirlik, Arif. 1997. The postcolonial Aura: Third World criticism in the age of global capitalism. Oxford: Westview Press.

Dündar, Can & Rıdvan Akar. 2008. Ecevit ve gizli arşivi [Ecevit and his secret archive]. İstanbul: İmge Yayınları.

Dündar, Fuat. 2001. İttihat ve Terakki’nin Müslümanları iskan politikası (1913–1918) [The Committee of Union and Progress’s policy of settling Muslims]. İstanbul: İletişim Yayınları.

Erdem, Suna. 2010. Suitcase of documents reveals plan for yet another very Turkish coup. The Times. 28 January. http://www.timesonline.co.uk/tol/news/world/europe/article7005391.ece (accessed 1 November 2010).

Fırat New Agency (ANF). 2010. TRT 6’da deprem [Earthquake at TRT 6]. 12 January. http://www.firatnews.nu/index.php?rupel=nuce&nuceID=19993 (accessed 1 November 2010).

Gökalp, Ziya. 1981 [1918]. Turkish nationalism and Western civilisation. London. Greenwood Press.

Gunter, Michel. 1990. The Kurds in Turkey: a political dilemma. San Francisco: Westview Press.Gurbey, Gulistan. 2005. Implications of Turkey’s constitutional reforms for the Kurds. In

Mohammed M. A. Ahmed & Michael M. Gunter (eds.), The Kurdish question and the 2003 Iraqi war, 136–162. Costa Mesa: Mazda Publishers.



http://www.cnnturk.com/2009/turkiye/03/20/gun.tvye.12.gun.yayin.durdurma.cezasi/518725.0/index.html

http://www.cnnturk.com/2009/turkiye/03/20/gun.tvye.12.gun.yayin.durdurma.cezasi/518725.0/index.html

http://www.cnnturk.com/2010/turkiye/01/31/kurtce.propaganda.yasagi.kalkabilir/561729.0/index.html

http://www.cnnturk.com/2010/turkiye/01/31/kurtce.propaganda.yasagi.kalkabilir/561729.0/index.html

http://www.timesonline.co.uk/tol/news/world/europe/article7005391.ece

http://www.timesonline.co.uk/tol/news/world/europe/article7005391.ece

http://www.firatnews.nu/index.php?rupel=nuce&nuceID=19993

http://www.firatnews.nu/index.php?rupel=nuce&nuceID=19993



Hassanpour, Amir. 1993. The internationalization of language conflict: the case of Kurdish. In Eran Fraenkel & Christina Kramer (eds.), Language contact-language conflict, 107–155. New York: Peter Lang.

Hassanpour, Amir. 2000. The politics of a-political linguistics: linguists and linguicide. In Robert Phillipson (ed.), Rights to language: equity, power, and education, 33–39. Mahwah, NJ: Lawrence Erlbaum Associates.

Hassanpour, Amir, Tove Skutnabb-Kangas & Chyet, Michael. 1996. The non-education of the Kurds. International Review of Education 42(4). 367–379.

Huggler, Justin. 2001. Turkey lifts ban on Kurdish broadcasts in effort to impress EU. The Independent. 28 September. http://www.highbeam.com/doc/1P2-5182813.html (accessed 1 November 2010).

Kara Kuvvetleri Komutanlığı. 1982. Türkiye’de bölücü ve yıkıcı akımlar [Divisive and destructive currents in Turkey]. Ankara: Kara Kuvvetleri Komutanlığı Basımevi.

Kirişçi, Kemal & M. Gareth Winrow. 1997. The Kurdish question and Turkey: an example of a trans-state ethnic conflict. London: Frank Cass.

Koivunen, Kristiina. 2002. The invisible war in north Kurdistan. Helsinki: University of Helsinki dissertation. http://ethesis.helsinki.fi/julkaisut/val/sospo/vk/koivunen/theinvis.pdf (accessed 1 November 2010).

Kurdish Human Rights Project. 2005. Recognition of linguistic rights? The impact of pro-EU reforms in Turkey. London: Kurdish Human Rights Project.

Lewis, Geoffrey. 1999. The Turkish language reform: a catastrophic success. New York: Oxford University Press.

Mater, Nadire. 2002. Appeal for Kurdish language courses refuted. Bianet.org. 15 February. http://bianet.org/english/english/7769-appeal-for-kurdish-language-courses-refuted (accessed 1 November 2010).

McDowall, David. 2000. A modern history of the Kurds. London: I. B. Tauris.Muller, Mark. 1996. Nationalism and the rule of law in Turkey: the elimination of Kurdish

representation during the 1990s. In Robert Olson (ed.), The Kurdish nationalist movement in the 1990s: its impact on Turkey and the Middle East, 173–199. Lexington, KY: The University Press of Kentucky.

Nezan, Kendal. 1993. Kurdistan in Turkey. In Gerard Chailand (ed.), People without a country: the Kurds and Kurdistan, 61–68. New York: Olive Branch Press.

Okutan, M. Çağatay. 2004. Tek Parti döneminde azınlık politikaları [Minority policies during the Single Party rule]. İstanbul: İstanbul Bilgi Üniversitesi Yayınları.

Önderoğlu, Erol. 2009. ECHR will tackle ban on propaganda in Kurkish. Bianet.org. 11 February. http://bianet.org/english/freedom-of-expression/112506-echr-will-tackle-ban-on-propaganda-in-kurdish (accessed 1 November 2010).

Önderoğlu, Erol. 2010. 166.5 years’ sentence for Kurdish journalist Vedat Kurşun. Bianet.org. 14 May. http://bianet.org/english/minorities/121970-166-5-years-sentence-for-kurdish-journalist-vedat-kursun (accessed 1 November 2010).

Pope, Hugh. 1993. Voice from grave airs a Kurdish solution: Ozal letter published advocating forced migration to defeat PKK. The Independent. 13 November. http://www.independent.co.uk/news/world/voice-from-grave-airs-a-kurdish-solution-ozal-letter-published-advocating-forced-migration-to-defeat-pkk-1503951.html (accessed 1 November 2010).



http://bianet.org/english/freedom-of-expression/112506-echr-will-tackle-ban-on-propaganda-in-kurdish

http://bianet.org/english/freedom-of-expression/112506-echr-will-tackle-ban-on-propaganda-in-kurdish

http://bianet.org/english/minorities/121970-166-5-years-sentence-for-kurdish-journalist-vedat-kursun

http://bianet.org/english/minorities/121970-166-5-years-sentence-for-kurdish-journalist-vedat-kursun

http://www.independent.co.uk/news/world/voice-from-grave-airs-a-kurdish-solution-ozal-letter-published-advocating-forced-migration-to-defeat-pkk-1503951.html




Reporters without Borders. 2010a. Former editor of Turkey’s sole Kurdish daily facing up to 525 years in prison. 2 April. http://en.rsf.org/turkey-former-editor-of-turkey-s-sole-02-04-2010,36936.html (accessed 1 November 2010).

Reporters without Borders. 2010b. Another trial against Kurdish minority publication. 2 August. http://en.rsf.org/turkey-another-trial-against-kurdish-02-08-2010,38078.html (accessed 1 November 2010).

Sadoğlu, Hüseyin. 2003. Türkiye’de ulusçuluk ve dil politikaları [Nationalism and language policies in Turkey]. İstanbul: İstanbul Bilgi Üniversitesi Yayınları.

Sezgin, Dilara & Melissa A. Wall. 2005. Constructing the Kurds in the Turkish press: a case study of Hurriyet newspaper. Media, Culture & Society 27(5). 787–798.

Skutnabb-Kangas, Tove. 1981. Bilingualism or not: the education of minorities. London: Multilingual Matters.

Skutnabb-Kangas, Tove. 2000. Linguistic genocide in education or worldwide diversity and human rights? London: Lawrence Erlbaum Associates Publishers.

Skutnabb-Kangas, Tove. 2005. Linguistic genocide. In Dinah Shelton (ed.), Encyclopedia of genocide and crimes against humanity, 653–654. New York: Macmillan Reference USA.

Skutnabb-Kangas, Tove & Sertaç Bucak. 1995. Killing of a mother tongue: how the Kurds are deprived of linguistic human rights. In Tove Skutnabb-Kangas & Robert Phillipson (eds.), Linguistic human rights: overcoming linguistic discrimination, 347–353. Berlin & New York: Mouton de Gruyter.

Skutnabb-Kangas, Tove & Robert Dunbar. 2010. Indigenous children’s education as linguistic genocide and a crime against humanity? A global view. Guovdageaidnu/Kautokeino: Gáldu, Resource Centre for the Rights of Indigenous Peoples.


Skutnabb-Kangas, Tove & Shelley Taylor. 2009. The educational language rights of Kurdish children in Turkey, Denmark and Kurdistan (Iraq). In William Ayers, Therese Quinn & David Stovall (eds.), Handbook for social justice in education, 171–190. New York: Routledge.

Sobecki, Nichole. 2009. Turkey’s forbidden tongue. Globalpost.com. 5 March. http://www.globalpost.com/dispatch/turkey/090305/turkeys-forbidden-tongue?page=0,0 (accessed 1 November 2010).

Taşpınar, Ömer. 2005. Kurdish nationalism and political Islam in Turkey: Kemalist identity in transition. New York: Routledge.

Terörle Mücadele Kanunu 3713 [The Anti-Terror Law 3713]. 1991. Published in the Official Gazette on 12 April. http://www.mevzuat.adalet.gov.tr/html/809.html (accessed 1 November 2010).

Today’s Zaman. 2009a. Kurdish pop star Rojin parts ways with state broadcaster TRT 6. 15 April. http://www.todayszaman.com/tz-web/detaylar.do?load=detay&link=172488 (accessed 1 November 2010).

Today’s Zaman. 2009b. Outlawed PKK threatens TRT’s Kurdish employees. 10 February. http://www.todayszaman.com/tz-web/detaylar.do?load=detay&link=166446 (accessed 1 November 2010).

Today’s Zaman. 2009c. Government to continue Kurdish Initiative, no matter the cost. 15 August. http://www.todayszaman.com/tz-web/news-184055-govt-to-continue-kurdish-aqua-aquainitiative-no-matter-the-cost.html (accessed 1 November 2010).



http://en.rsf.org/turkey-former-editor-of-turkey-s-sole-02-04-2010,36936.html

http://en.rsf.org/turkey-former-editor-of-turkey-s-sole-02-04-2010,36936.html

http://www.globalpost.com/dispatch/turkey/090305/turkeys-forbidden-tongue?page=0,0

http://www.globalpost.com/dispatch/turkey/090305/turkeys-forbidden-tongue?page=0,0

http://www.todayszaman.com/tz-web/news-184055-govt-to-continue-kurdish-aqua-aquainitiative-no-matter-the-cost.html

http://www.todayszaman.com/tz-web/news-184055-govt-to-continue-kurdish-aqua-aquainitiative-no-matter-the-cost.html


Today’s Zaman. 2009d. Outline of Kurdish Initiative emerges at security summit. 18 September. http://www.todayszaman.com/tz-web/news-187472-101-outline-of-kurdish-initiative-emerges-at-security-summit.html (accessed 1 November 2010).

Today’s Zaman. 2010. Turkey is worst human rights violator, EctHR says. 29 January. http://www.todayszaman.com/tz-web/detaylar.do?load=detay&link=199996 (accessed 1 November 2010).

Toumani, Meline. 2008. Minority rules. The New York Times. February 17. http://www.nytimes.com/2008/02/17/magazine/17turkey-t.html?_r=1&pagewanted=all (accessed 1 November 2010).

Tunçel, Harun. 2000. Türkiye’de ismi değiştirilen köyler [Renamed villages in Turkey]. Fırat Üniversitesi Sosyal Bilimler Dergisi 10(2). 23–34.

Üçüncü Türk Dili Kurultayı 1936 tezler, müzakere zabıtları [Theses and proceedings of the Third Turkish Language Congress 1936]. 1937. Ankara: The Turkish Language Institution. http://www.tdkkitaplik.org.tr/kurultay03/K0302001.pdf (accessed 1 November 2010).

Üngör, Uğur Ümit. 2008. Seeing like a nation-state: Young Turk social engineering in eastern Turkey, 1913–1950. Journal of Genocide Research 10(1). 15–39.

Üstündağ, Erhan. 2010. Kurdish speaking defendants silenced at court. Bianet.org. 5 November. http://www.bianet.org/english/minorities/125899-kurdish-speaking-defendants-silenced-at-court (accessed 10 November 2010).

Watts, Nicole F. 1999. Allies and enemies: pro-Kurdish parties in Turkish politics, 1990–1994. International Journal of Middle East Studies 31(4). 631–656.

Williams, H. Colin (ed.). 1988. Language in geographic context. Clevedon: Multilingual Matters.Yeğen, Mesut. 1999. Devlet söyleminde Kürt sorunu [The Kurdish question in state discourse].

İstanbul: İletişim Yayınları.Yıldız, Ahmet. 2001. Ne mutlu Türküm diyebilene: Türk ulusal kimliğinin etno-seküler sınırları,

1919–1938 [How happy is he who is able to say he is Turkish: the ethno-secular borders of the Turkish national identity, 1919–1938]. İstanbul: İletişim Yayınları.

Yildiz, Kerim & Georgina Fryer. 2004. The Kurds: culture and language rights. London: The Kurdish Human Rights Project.

Yildiz, Kerim & Mark Muller. 2008. The European Union and Turkish accession: human rights and the Kurds. London: Pluto Press.

Zeydanlıoğlu, Welat. 2007. Kemalism’s others: the reproduction of orientalism in Turkey. Cambridge: Anglia Ruskin University dissertation.

Zeydanlıoğlu, Welat. 2008. The white Turkish man’s burden: Orientalism, Kemalism and the Kurds in Turkey. In Guido Rings & Anne Ife (eds.), Neo-colonial mentalities in contemporary Europe? Language and discourse in the construction of identities, 155–174. Newcastle upon Tyne: Cambridge Scholars Publishing.

Zeydanlıoğlu, Welat. 2009. Torture and Turkification in the Diyarbakır Military Prison. In Welat Zeydanlıoğlu & John T. Parry (eds.), Rights, citizenship and torture: perspectives on evil, law and the state, 73–92. Oxford: Inter-Disciplinary Press.

Zeydanlıoğlu, Welat & Ipek Demir. 2010. On the representation of “others” at Europe’s borders: the case of Iraqi Kurds. Journal of Contemporary European Studies 18(1). 7–23.



http://www.todayszaman.com/tz-web/news-187472-101-outline-of-kurdish-initiative-emerges-at-security-summit.html

http://www.todayszaman.com/tz-web/news-187472-101-outline-of-kurdish-initiative-emerges-at-security-summit.html

http://www.bianet.org/english/minorities/125899-kurdish-speaking-defendants-silenced-at-court

http://www.bianet.org/english/minorities/125899-kurdish-speaking-defendants-silenced-at-court

DOI 10.1515/ijsl-2012-0052 IJSL 2012; 217: 127 – 150

Uğur Ümit ÜngörUntying the tongue-tied: Ethnocide and language politics

Abstract: This article addresses Young Turk language policy towards Kurdish in the interwar period. It argues that most Young Turk nationalists treated Turkey’s Kurdish minority as assimilable raw ethnic material, as a result of which Kurds became the object of large-scale cultural and linguistic policies aimed at “Turkifi-cation”. This article charts how these language policies infringed upon Kurdish life. It will (a) briefly introduce the Young Turk cultural revolution of 1913–1950, (b) discuss how the Young Turk dictatorship perceived the Turkish language as a vehicle for cultural assimilation, and (c) provide a detailed account of one ex-ample of a boarding school for Kurdish children. It concludes that there is evi-dence that a policy of cultural genocide against Kurds was implemented but rela-tivizes its impact by discussing the Kurds’ ambivalent reception of those policies.

Keywords: cultural genocide; language policy; Kurds; Turkification

Uğur Ümit Üngör: Utrecht University. E-mail: [email protected]

1 IntroductionThis chapter will address the history of Young Turk language policy towards Kurdish in the interwar period. It will argue that most Young Turk nationalists treated Turkey’s Muslim minorities as assimilable raw ethnic material. They ad-hered to the conviction that individual human beings are born with no innate or built-in mental content, in a word, “blank”: not only was their entire resource of knowledge built up gradually from their socialization by the outside world, this socialization could be engineered from above. In other words, their ideologues considered the population fully malleable. In interwar Turkey, assimilative na-tionalism triumphed in social and political discourse, and was shared by the col-lective dictatorship of the party-state. The regime abandoned its belief in socio-logical categories above biological ones when it came to the non-Muslims, such as Jews, Greeks, Armenians, and perhaps also Syriacs/Assyrians. Although there were attempts to “Turkify” these groups culturally, they were generally essential-ized in their identifications and considered largely nonassimilable. Moreover, as



128 U. U. Üngör

they were privileged to maintain their own educational infrastructure, the regime had limited jurisdictional means to extend its reach into their schools and spread Turkish nationalism. The non-Turkish Muslim population, consisting of Arabs, Circassians, Lazes and others, but especially Kurds, became the object of large-scale cultural and linguistic policies aimed at “Turkification”.

This article will chart how these language policies infringed upon Kurdish life. It will (a) briefly introduce the Young Turk cultural revolution of 1913–1950 (Zürcher 1992), (b) discuss how the Young Turk dictatorship perceived the Turkish language as a vehicle for cultural assimilation, and (c) provide a detailed account of one example of a boarding school for Kurdish children. It will conclude that there is evidence that a policy of cultural genocide against Kurds was imple-mented but relativizes its impact by discussing the Kurds’ ambivalent reception of those policies.

2 Theoretical backdropThe theoretical backdrop for this chapter can be traced to an apparent contradic-tion or unresolved paradox in the study of linguistic change and language poli-tics. Whereas one school of thought argues that the global decline of indigenous and minority languages attests to improvements in processes of global societal integration of speakers of those groups, another group of scholars sees this de-cline mainly in terms of deliberate linguistic destruction or system-produced cul-tural decline. Exemplary of the former group is Abram de Swaan’s comparative study Words of the world: the global language system (De Swaan 2001). In this book, he sketches a global hierarchy of language use, identifying super-central (e.g. English), central (e.g. Turkish), and peripheral languages (e.g. Kurdish). De Swaan argues that in order to improve their economic position in the global economy by increasing their communicative efficacy, speakers of peripheral languages will attempt to acquire central and super-central languages. This un-planned but directional (a.k.a. “blind”) process produces voluntary language abandonment and decline of peripheral languages. Importantly, De Swaan warns against language sentimentalism and argues that languages are not extinguished or destroyed but abandoned by their speakers in their emancipatory efforts of upward social mobility. In contrast, scholars of the second school have rejected this approach in favor of a model of hegemony and decline, invoking “minority protection” and ecological metaphors of “diversity” to sustain an ideal of maxi-mum linguistic heterogeneity. Tove Skutnabb-Kangas (2000), for example, in her comprehensive study, interprets states’ language policies towards indigenous and minority languages as leading to the disappearance of linguistic and cultural



Untying the tongue-tied 129

diversity. Subtractive education, she argues, assaults the native languages and thereby damages the cultural lives of the peoples at its receiving end. In her view, the decline of indigenous and minority languages in processes of global linguis-tic change are tantamount to “linguistic genocide”. She identifies the English language as the main culprit of global linguistic de-heterogenization (Skutnabb-Kangas 2000). These partly opposed paradigms do not appear to be reconcilable in the foreseeable future.

In a similar vein, Young Turk cultural and educational policies too are a hotly debated and contested terrain with fierce disagreement on their meaning, inten-tions, and consequences. The debate ranges from issues such as racism and cul-tural othering to power relations. One group of scholars argues that Young Turk policies were neither openly racist in content nor particularly successful in prac-tice. According to one of them, the “regime lacked the resources or cadres to es-tablish an effective educational system in this region and inculcate a Turkish identity, myths and language into the local population” (Lieven 2002: 357). An-other expert argues that modernization of the educational system was “one of the most important and commendable” of Young Turk “reforms”, one that achieved “impressive results” (Winter 1984: 192–193). In this beneficial and inclusivist pro-cess of “modernization”, another scholar argues, “an important factor was edu-cation, in which the Young Turks achieved their greatest successes” (Lewis 2001: 229). Turkish cultural and educational policies towards the ethnic minorities in the eastern provinces (such as Kurds) in this tradition are typically explained as “acculturation”, the exchange of cultural features that results when groups come into continuous firsthand contact, leading to cultural change among both groups, but with both groups remaining distinct (Heper 2007).

But a second group of scholars criticizes this approach for its alleged Turco-centric bias, regime apologia, and the whitewashing of the symbolic violence inherent in systems of cultural domination. They argue the opposite, namely that Young Turk cultural policies were hegemonic, exclusivist, oppressive, and racist towards non-Turks (Hassanpour 1991; Çapar 2006). One sociologist uses labels such as “colonization” and “cultural genocide” to describe Young Turk assimi-lationist policies in Eastern Turkey (Beşikçi 2004). These approaches use post-colonial theory and draw upon Frantz Fanon’s accounts of French education in Algeria to explain the feelings of dependency and inadequacy that non-Turkish people experience in a state permeated by Turkish nationalism. They emphasize the loss of native cultural originality and identity, and the imposition of an alien culture. This purportedly engenders an inferiority complex in the minds of mi-norities, who will try to appropriate and imitate the cultural code of “Turkish-nationalist cultural colonialism” (Zeydanlıoğlu 2007). In this perspective, the Young Turks colonized the eastern provinces and collectively disempowered the



130 U. U. Üngör

population, subjugated their bodies, peripheralized their economies, and most of all: assaulted and destroyed their culture. How can these sharp oppositions be reconciled?

3 The Young Turk cultural revolutionThe genesis of Young Turk educational philosophy was rooted in the regime’s nationalist ideology according to which reshaping Ottoman society into a homog-enous Turkish nation state was necessary and desirable. Throughout the Young Turk era, Turkish ethnic nationalism informed and guided government policies, and the application of nationalist principles and doctrine was at the forefront of Young Turk educational practices. As such, education was the centerpiece of the Young Turk party’s political platform for the “internal colonization” (dahili kolo-nizasyon) of the eastern provinces. Throughout the Young Turk era, education served two purposes: Turkification and the spread of the regime’s propaganda (Tunçay 1995: 239–240). This was of such importance that Mustafa Kemal Atatürk (1881–1938) personally spearheaded the national campaign for education. In his opening speech to the 1921 Congress on Education, the general and later Presi-dent outlined his vision for the future of Turkish education:

Although we are at war and our resources must be spent on the war effort, we should still try to formulate carefully a national educational policy for the post-war period. By this policy I mean a culture fitting for our national and historical character, completely separate from all influences from both the East and the West, and far from influences foreign to our own character. (Cumhurbaşkanları, başbakanlar ve millî eğitim bakanlarının millî eğitimle ilgili söylev ve demeçleri 1946: 4)

The aim, according to Mustafa Kemal, was “to create a Turkish youth strong enough to battle other nations” (Yücel 1993: 118). Three years after the Congress, Kemal expounded this approach and dismissed “religious education” and “inter-national education” in favor of “national education”. The latter would be adopted to create “a new generation in the New Turkish Republic” (Atatürkçülük 1988: 290). In the 1920s and 1930s, Mustafa Kemal launched himself and was presented as the “Chief Teacher” who would lead the nation into learning. In one classic photograph Kemal is seen wearing a dark suit, pointing at a blackboard to ex-plain the new Latin alphabet to the locals.

Whether out of true conviction or political appeasement, in reality Atatürk had trusted followers who supported his enthusiasm and echoed his cries for a nationalist education. İsmet İnönü (1884–1973) repeated his superior in a speech about his interpretation of “national education”. Speaking of Turkish society, he




complained that “the people still do not exhibit the appearance of a monolithic [mütecanis] nation”. A new generation needed to be molded in order for “the political Turkish nation fully to become a cultural, mental and societal Turkish nation”. According to İnönü, “foreign cultures need to melt into this mono-lithic nation . . . there can be no other cultures in this nation . . . if we are to live, we shall live as a monolithic nation. That is the general aim for the system we call national education”. (Yücel 1994: 25) If this was not clear enough lan-guage, veteran Interior Minister Şükrü Kaya (1883–1959) clarified this approach in parliament:

No matter what happens, it is our obligation to immerse those living in our society in the civilization of Turkish society and to have them benefit from the prosperity of civilization. Why should we still speak of the Kurd Mehmet, the Circassian Hasan or the Laz Ali. This would demonstrate the weakness of the dominant element. . . . If anybody has any differ-ence inside him, we need to erase that in the schools and in the body politic, so that man will be as Turkish as me and serve the homeland. (Türkiye Büyük Millet Meclisi zabıt ceridesi 1934: Volume 2, 71, 249)

As the message trickled down through the bureaucracy, it seemed to radicalize. Soon lower-level officials of the Ministry of Education personnel were seen de-livering lectures in which they formulated theories that “the Turkish school is obliged to transform every Turkish child into a thoroughly useful Turkish citizen who has fully grasped the psychology and ideology of the Republic, the Turkish Nation, and the Turkish Republic” (Yücel 1994: 364–365). The ostensible homoge-neity and purity of the nation was not a conjunctural but a structural aspect of Young Turk ideology. Well into World War II, the Minister of Education called for educational policies that would create “a Turkish youth in a . . . homogeneous nation” (mütecanis bir millet), crafted by an elite that would “govern the country with the exact sciences” (Dağlı and Aktürk 1988: 105). This combination of watch-words represented Young Turk social engineering in its purest form: scientism needed to usher population policies towards the long-term fantasy of total ethnic homogeneity.

How could this policy of homogenization be reconciled with the realities of diversity in the eastern provinces? The Young Turks dismissed the reality of eth-nic difference and promised harsh action against non-Turkish cultures, but again left the detailed planning of population policies in the eastern provinces to dili-gent Young Turk social engineers. The reports written by these social engineers in 1925 provided ample reference to the ends education would aspire in the eastern provinces. Parliament chairman Mustafa Abdülhalik Renda pleaded for “special relevance to be accorded to education . . . in order to teach Turkish and revert in-clinations to Kurdishness back to Turkishness”. Interior Minister Cemil Uybadın



132 U. U. Üngör

wrote in a 1925 report that “a strong national organization and educational propa-ganda will wipe out notions of Kurdishness”, and stipulated that Kurds needed “to be forced to become Turkish”. His report continued relentlessly in advocating cultural measures in the East:

In reality the Eastern territories are under the influence of Kurdism. . . . The people are very attached to their language and nationality. The intellectuals are all Kurdish nationalists. We need to work consciously for the destruction of . . . the ideal and movement of Kurdism in the eastern territories and prevent it from effusing into the area west of the Euphrates. . . . [The region] has to be turned into a battlefield with a strong organization of education, [Turkish] Hearths, sports, and youth groups, and through the press, schools, theatre, na-tional and general plays the national sentiments of the people and Turkish traditions need to be invigorated. (Bayrak 1993: 460, 462, 468, 475–476)

The martial metaphors suggesting that the eastern provinces were a theater of war were not simply rhetorical. During the Kurdish revolt led by Sheikh Said in 1925, İnönü predicted: “Nationality is our only instrument of adhesion. The other elements are not vested with any power in the face of the Turkish majority. It is our duty to render Turkish everybody in the Turkish homeland, no matter what. We will cut out and throw out the elements that oppose Turks and Turkism” (Vakit, 27 April 1925). Moral or practical protests against these ambitious plans were easily dismissed. When during a cultural congress a delegate argued that coercing people to become Turkish would meet with resistance, delegate Besim Atalay waived the objections away and answered: “Nations are like organisms. They cannot live if they do not eat. You have to kill, you have to kill in order to live”. Delegates chanted approvingly (Türk Ocakları 1928 Senesi Kurultayı Zabıtları 1929: 123–124, 151–152).

4 Linguistic persecution of KurdsThe Young Turk cultural revolution gravitated for a large part around lan-guage. Already in 1908, a prominent nationalist had written in the party organ: “How can any nationality that does not know Turkish, be enthused for and give trust to the Ottoman Empire? Therefore, it is incumbent that Turkish be-comes mandatory” (Tanin 25 September 1908). The imposition of the Turkish language under the Young Turks was not simply a demonstration or shadow of their power. A significant ideological factor was involved. Interwoven through-out much of the Young Turks’ writings was the belief that Turkish is the lan-guage of civilization, administrative rationalism, and cultural enlightenment – and that the non-Turkish peoples operated on a lower cultural plane. In this




illusion, linguistic assimilation was a civilizing mission. But the disparity be-tween Young Turk ideology and societal reality was enormous. When they launched their campaigns for cultural and linguistic homogenization, languages such as Kurdish, Circassian, Arabic, Syrian-Aramaic, Zazaki, Laz and others were widely spoken in the eastern provinces. Kurmancî was a lingua franca in the countryside east of the Euphrates, Dersim spoke largely Zazaki and Tur Abdin Syrian-Aramaic, and on the streets of cities such as Siirt and Mardin a dialect of Arabic was spoken. In other words, on entire swaths of land Turkish was hardly understood.

The tone for Young Turk linguistic policy was set by Mustafa Kemal Atatürk, in a famous declaration: “One very obvious element of nationality is language. A person who claims to be of the Turkish nation, first of all must unconditionally speak Turkish. It would not be correct to believe anyone who claims allegiance to Turkish culture and society but does not speak Turkish” (Atatürk’ün Söylev ve Demeçleri 1945: 362). In this broad-brush equation of language with loyalty, mil-lions of people who did not speak Turkish were branded as problematic from a state security view. The solution was provided by İnönü, who, during a party con-gress gave a speech emphasizing that “from now on we will not keep quiet. All citizens who live with us will speak Turkish” (CHP Dördüncü Büyük Kurultayı Görüşmeleri Tutulgası 1935: 149). The sociologist Mehmed Ziyâ Gökalp (1876–1924) was perhaps the most prominent Young Turk party ideologue who advocated the “Turkification” of the Ottoman Empire by imposing the Turkish language and culture on all the citizenry and constructing a Turkish nation state (Bo-zarslan 2001: 314–319). A substantial part of Gökalp’s writings was about how population policies towards the eastern provinces should take shape. In one of his articles he set the trend by arguing that when two peoples lived side by side, “the dominating nation will assimilate the captive nation”. Gökalp named this process “dénationalization”, and argued that it had proven efficient in the French government’s campaign to suppress the use of German in Alsace Lorraine (Gökalp 1923: 4–7).

Gökalp was not alone in his ruminations. A wide spectrum of Turkish intel-lectuals discussed the language problem in nationalist clubs called the “Turkish Hearths” (Türk Ocakları). The Hearths dotted the country with its headquarters in Ankara. Delegates in these clubs would typically be members of parliament or local intellectuals with some credentials in the Young Turk party. For them, lin-guistic Turkification was not proceeding quickly enough and therefore frustrating the “national unity” (vahdet-i millî) of the country. In 1926, 1927, and 1928, a series of congresses were held in which the fate of the Kurdish language was discussed, and ultimately sealed. These were carefully orchestrated by the regime and at-tended by Young Turk satraps who glorified the regime. Hardly any of them



134 U. U. Üngör

had any academic distinctions or linguistic qualifications, and most were party apparatchiks. During the congresses, the attitude of these regime intellectuals towards Kurdish was only ambivalent in their quasi-scholarly judgments on tech-nical, linguistic issues. Whatever the roots and characteristics of that language were, the language had to be “eradicated”.

The anti-intellectual tone was set in the 1926 congress by delegate İshak Refet (Işıtman), who diagnosed: “There are people in Turkey who use languages other than Turkish. This situation is pervasive in the eastern provinces. . . . These peo-ple continue their own traditions even though they have been living in Turkey for more than half a century” (Türk Ocakları Üçüncü Kurultayı Zabıtları 1927: 177–178). Bursa MP Rüştü Bey spoke: “Everybody is Turkish. In many villages . . . many foreign languages are spoken like cosmopolitan people. . . . These things are no good. In order to nationalize them it is necessary to follow a special policy and radical reforms. For this, I demand a more drastic and definite process” (Ergüven 1937: 84). Delegate İzzet Ulvi (Aykurt) pointed out that the main problem of as-similation was language and approved of practices of fining for those who spoke languages other than Turkish; he had also applied for such a law to be passed at Parliament. He noted that non-Turkish Muslim minorities such as Circassians, Bosnians, and Kurds should be prohibited from living together and prohibited from using their languages and wearing their distinct clothing. According to Ulvi the government was in agreement with all of these proposals and would support the actions (Türk Ocakları Üçüncü Kurultayı Zabıtları 1927: 198). Still speaking of the non-Muslim minorities, Ulvi wished “to extinguish their linguistic charac-teristics” (lisanî hususiyetlerini izâle etmek) (Türk Ocakları Üçüncü Kurultayı Zabıtları 1927: 210). The Young Turk journalist Ahmet Emin Yalman argued that the true citizen spoke Turkish and added that, “let those who did not want to be Turkish citizens by means of their languages and deeds, those who did not want to adopt the public life of the country, show their colors and exclude themselves from the whole like a foreign element”. He added that citizens must be freed from “the perils of cosmopolitanism” (Aktar 2000: 122–124).

In the same session of the 1926 congress, İshak Refet (Işıtman) continued his speech, underlining that areas populated with Kurds were to be met with “special treatment” and that “Turkifying” Kurds would be easier than previously thought. He then explained his vision about Kurdish for the audience:

The Kurd has no history. The history of the Kurd has merged with the history of the Turk. There is no language called Kurdish. Nowadays the Kurdish language consists of 8000 words. Three thousand of these are Turkish. About two thousand are Kurdified Arabic. The remaining 2500 words then are old and new Persian. Gentlemen, there are a mere three hundred Kurdish words in the Kurdish language. And the strangest aspect is that this lan-guage has no verbs. The measure for the existence and identity of languages are verbs.




These people do not even have verbs. Since the Kurd has no language, history, or tradition and only has plenty savagery, how significant is it to assimilate the Kurd? It will be quite easy to assimilate the Kurds. (Türk Ocakları Üçüncü Kurultayı Zabıtları 1927: 225)

After widespread exclamations of approval, the 1926 plenum was concluded. De-spite the unanimity of the delegates on the broad direction of language policy, the regime still lacked concrete and precise directives for dealing with minority lan-guages, in particular Kurdish. The year after, during the 1927 congress, the pre-sidium of the Turkish Hearths again identified the main problem of the country to be language. The members advocated the idea that Turkish should be the only language taught and spoken in the country and agreed on lobbying the govern-ment for it. Siirt delegate Hüseyin İbrahim, for example, resented the fact that all over the eastern provinces even state officials spoke Kurdish in daily practices and wanted the governors to intervene by resorting to “most severe measures” in rooting out non-Turkish languages (Türk Ocakları 1927 Senesi Kurultayı Zabıtları 1928: 277–280). During the second meeting of the 1928 congress, again the pre-sidium regretted that “unfortunately in some regions foreign languages offend our national essence, and some citizens even deliberately refrain from using Turkish and thereby violate our national honor”. To remedy this “problem”, it concluded, “the exclusive use of Turkish among citizens has to be enforced by state laws” (Türk Ocakları 1928 Senesi Kurultayı Zabıtları 1929: 84–85). Seeping through the discussions was a vocabulary of coercion and force that would soon be unleashed on the Kurdish population. The 1920s were concluded with the es-tablishment of a consensus of linguicide among Turkish intellectuals and the marginalization and exclusion of anyone arguing otherwise.

To grant these discussions a certain quasi-scholarly legitimacy, the regime on 12 July 1932 established the Turkish Language Society (Türk Dil Kurumu, TDK). It was assigned with researching and spreading the Turkish language and played an important role in the marginalization and trivialization of the Kurdish lan-guage in Turkey, up to today. The TDK was also in charge of the dubious task of purging the Ottoman Turkish language of Arabic and Persian influences and re-shaping it into a modern Turkish. Its researchers carried out quasi-scholarly na-tionalist research on minority languages; that body of ethnographic knowledge would serve to excise those languages from Turkish society. For example, regime apologist and Young Turk ideologue Hasan Reşit Tankut presented an account of his work and noted that the main objective of linguistic research should be to impose the use of Turkish language and culture, especially in districts formerly or currently inhabited by Armenians and Kurds. He had conducted “ethnographic research” in the districts of Xerzan, Palu, Karaköse, Silvan, Iğdır, Birecik, Muş, Siirt, Çemişgezek, Van, Mardin, Malatya, and argued that the Diyarbekir region



136 U. U. Üngör

especially was of such importance that it required special attention by the gov-ernment (Üstel 1997: 332–333).

But sterile discussion and Young Turk exegesis in the academic ivory tower did not and could not cause linguistic change. Only policy could. The Young Turk regime therefore launched the propaganda campaign “Citizen Speak Turkish!” (Vatandas Türkçe Konuş!). The campaign “targeted mainly the non-Muslims and non-Turkish Muslims to make them speak the new Turkish” (Çolak 2004: 81). What seemed like an open invitation to speak the Turkish language and achieve emancipation, promote integration, and further upward social mobility, in prac-tice escalated into an often brutal and senseless process of linguistic persecution marked by large-scale prohibitions and cultural coercion. Following the Turkish Hearths congresses, TDK publications, and the “Citizen Speak Turkish!” cam-paign, the use of all non-Turkish languages was soon prohibited. The government order read: “Anyone who uses languages other than Turkish in provincial and district centers, government and municipal offices and other institutions and or-ganizations, in schools, bazaars and markets . . . will be punished for the crime of resistance against government orders” (Bayrak 1993: 486). This sweeping decree was immediately supplemented with a more precise order: “It will be strictly pro-hibited for Kurds, who have settled in a dispersed manner in parts of provinces east of the Euphrates, to speak Kurdish” (Bayrak 1993: 487).

These individual decrees were embedded in a broad and influential 1934 de-cree that foresaw a large-scale colonization of the eastern provinces. In the law, the Young Turks prescribed that “those who are not related to Turkish culture . . . will have to be settled as single families in villages of Turkish culture, with the aim of having them speedily forget their mother tongue and mingle with Turks”. The flipside of this assimilative policy would be subtractive and coercive. The law stipulated punishment of “anyone speaking any language other than Turkish, in bazaars and markets, squares, in any other public places”. It would apply first in these public places, but then also in houses and ultimately “everywhere”; the law explicitly prohibited different languages, different traditional clothes, different folklore, different songs, “in short, the elimination of anything amounting to dif-ference (başkalığa delâlet eden herşeyi kaldırmak)” (Türkiye Büyük Millet Meclisi zabıt ceridesi 1934: 23: IV: 3: I65: c1).

The discussions in parliament leading up to the promulgation of this law were held by the highest ranking Young Turks. Veteran social engineer Şükrü Kaya (1883–1959) lied in parliament that there were only 100,000 citizens in Tur-key who did not speak Turkish, and “once this mass will eventually understand the deficiency of their language and realize that speaking Turkish will not only bring pleasure and pride, but at the same time also further their political and economic interests, they will naturally begin speaking Turkish. Anyway, nowa-




days there are no different races in our motherland who do not speak Turkish” (Ergüven 1937: 83–84). At this point, the sociological imagination fails us: accord-ing to the Young Turks, there was nothing hyperbolic or paradoxical that in these theories, Kurds were the purest and most original Turks in the first place, but still needed to be “Turkified” through re-education; Kurdish was a dialect of Turkish, but still needed to be strictly prohibited (the regime thus prohibited a dialect of its own national language). The specter of European linguistic nationalism loomed large in the Young Turk historical conscience. The journalist Ayaz İshakî wrote an article titled “How can we Turkify the non-Turks?” in the party mouthpiece news-paper Cumhuriyet:

The methods of cultural action implemented in Alsace-Lorraine in the first half of the cen-tury by the French against the Germans, and in the second half by the Germans against the French is known. . . . For this, the most civilized, simple and legal way would be to educate minority children in the majority language, to instill in them the superiority of the majori-ty’s cultural civilization. (İshakî 1928)

This would mean that education in the Kurdish-populated territories was much more than the simple transmission of knowledge. Destruction and construction supplemented and fortified each other in Young Turk language politics. The spread of Turkish was not foreseen as something that would complement existing languages but replace them. In the regime’s vision future generations of Kurdish children would be brought up with Turkish and Turkish only. Thus, the educa-tional policies were not indirectly subtractive but deliberately so.

5 Kurdish responses to Young Turk language policies

How did ordinary Kurds perceive Young Turk education? The scholarship on this subject is sparse, partly because the Ministry of Education archives in Ankara have not been made available and accessible for researchers yet. Some scholars of peasant societies have argued that peasants are often suspicious towards edu-cation imposed by the state. Education is often perceived as “a foreign element, coming from the outside. It limits the family’s rearing influence, tears the child from the harmonious system of work and life and introduces into its conscious-ness patterns which are dissonant with, and values foreign to or impossible of realization within that system: hence the resistance confronted by the school in rural areas” (Gałęski 1971: 195). For many villagers, it was much more important to have their children continue working on the farm. Conversely, if one takes



138 U. U. Üngör

government propaganda at face value, linguistic assimilation was an unequivo-cal success in the eastern provinces. Both postulates, of total rejection and of total success, lack sufficient evidence to allow the conclusion that the linguistic assimilation of Kurds failed or succeeded as a result of Young Turk social engi-neering. Careful analysis is required to assess the full implication of the material at hand.

The party correspondence sheds some light on how ordinary Kurds in cities, towns, and villages underwent these measures and linguistic policies. For exam-ple, according to one report, in Diyarbekir’s northeastern district of Kulp, 28 girls and 77 boys were going to school. Here, too, many in the town were reported to understand Turkish, but in the villages everyone spoke Kurmancî. The report put the proportion of the Turkish-speaking population at only two percent, per-haps the lowest in the entire province in 1941. Of all the social problems in the district, the local Young Turk inspector identified the crucial challenge to be the spread of the Turkish language among the local Kurds. This process had been lag-ging behind because of the local district governor’s corruption: the man had been using the allocated funds for his personal ends and the Turkish-language litera-ture sent to the town was known to be in his private possession. Every time he was asked for commentary he happened to be “organizing the census in the country-side” (Başbakanlık Cumhuriyet Arşivi 490.01/996.850.1 1941). In the Lice district, with its population of 6160, a total of 85% of Lice’s men and 60% of its women could understand Turkish, though few children spoke it. In the villages practi-cally nobody spoke Turkish but Zazaki, which was portrayed by Young Turk offi-cials as “a mix between Asiatic Turkish and coarse Persian”. Here, the reports were ubiquitously positive about the people’s stance on the government’s linguis-tic policies. “All of them”, the local party official wrote to Ankara, “reported their gratitude and indebtedness to the Republican government and our elders . . . and said they were always at their command” (Başbakanlık Cumhuriyet Arşivi 490.01/996.850.1 1941). Overly enthusiastic reports such as these were most likely attempts to excite and satisfy superior officials. Suggestions for improvement of education in Lice ranged from involving more women in activities to offering villagers the opportunity to have letters and petitions written for them, for free (Başbakanlık Cumhuriyet Arşivi 490.01/996.850.1 1941). All in all, it seems that the stance of the Kurdish population was a mix of some enthusiasm, some suspicion, but mostly indifference. More research is needed in this field to reach definitive conclusions.

An insight into personal experiences of education is provided by the noted Kurdish author Mehmed Uzun (1953–2007), a native of a village west of Diyarbekir city. Uzun’s account of his first day in school is so vivid one is justified in quoting him at length:




The first lesson goes back to 1960, the year I was seven. On a hot, clear day at the end of summer, the very day on which, dressed in new clothes from head to foot, I was beginning grammar school, I received a violent slap in the face in the guise of a lesson on the impor-tance of language and words. I had been born and raised in the shelter of a Kurdish tribe. My family possessed no books except for the Koran, which hung on the wall, and had neither a radio nor a television set. In this enormous house, its garden planted with some pomegran-ate trees and an equal number of peach trees, the garden where roses bloomed, there was nothing besides my father’s bilur (shepherd’s pipe), the stories and legends told by my grandfather, and the beautiful strans (traditional songs) that my grandmother sang in the Zaza dialect of Kurdish. It was a universe forged in the feelings, ideas, norms, and values of the Kurdish language. I was seven years old and loved this universe that I was part of. But from the first hour of the first day that I set foot in school I was instructed by a slap in the face, ineradicably engraved in my memory, that my universe was meaningless, useless, primitive, and taboo, and that I had to leave it. While I was joining the ranks of my class-mates in the yard of the grammar school, which was named after the poet Ibrahim Rafet, the teacher, who came from central Anatolia and was fulfilling his civil service, called me to order by a violent slap because I was speaking with a classmate in my maternal tongue. “It is forbidden to speak Kurdish!” (Uzun 2003)

Another Kurdish man from the northern Lice, who later became a teacher himself, argued that the children’s treatment depended on how nationalist their teacher was. His primary school teacher, a fervent nationalist, would frequently beat the children “for even whispering a single word in Kurdish”.1 Physical vio-lence was also exercised against speakers of Greek and Arabic: in one Turkish Hearth, raging discussions about the prohibition of Greek escalated into a large fight and in the end a judge was killed. In the town of Tarsus, Turkish Hearth members monitored the streets; people who were caught speaking Arabic were severely beaten (Üstel 1997: 333). One anthropologist documented the most radi-cal example of linguistic oppression: an old Kurdish man had his tongue cut out by the army for speaking Kurdish (Houston 2004: 403–414). Another scholar, a musicologist researching the Kurdish oral tradition, once met a man named Seyidxan Boyacı, a bard singing traditional songs and laments in Diyarbekir. Boyacı was once threatened by the authorities that if he sang inside the city walls his tongue would be cut out.2 Examples such as these possibly suggest that cultural and linguistic policies could not and did not achieve the desired ends in the East on a short term. But the Young Turk legacy outlived the Young Turks

1 Interview conducted with Amed Tîgrîs in Stockholm, 18 May 2005.2 Interview conducted with Seyidxan Boyacı by Emrah Kanısıcak in the summer of 2007. http://sazny.blogspot.com/2008/05/nightingale-of-amed-new-edit.html (accessed 16 July 2012).



http://sazny.blogspot.com/2008/05/nightingale-of-amed-new-edit.html

140 U. U. Üngör

themselves, for their curricula and methods remained in effect after the end of Young Turk rule in 1950.

The Young Turks not only assaulted the verbal use of Kurdish, but also cate-gorically outlawed Kurdish literature. During their rule, the regime outlawed, confiscated, and destroyed innumerable books, manuscripts, and other texts in non-Turkish languages. Having entered the age of information, the Young Turks acknowledged the power of knowledge and realized that certain bodies of know-ledge had to be produced and others had to be destroyed (İçimsoy and Erünsal 2008). Concordant with national guidelines, the destruction and construction of memory involved the “reorganization” of existing bodies of knowledge in the pe-ripheries. Besides continuing the Young Turk practice of confiscating and de-stroying Armenian libraries and collections, the Kemalists attacked and banned all texts that were either non-Turkish or “non-Turkifiable” – i.e. unfit to be cast retrospectively as “Turkish”. This policy continued unabated and was pursued relentlessly. During the sixth Turkish Hearths congress in 1928, Hasan Reşit Tan-kut presented an account of his work as “Eastern Inspector”, which meant the conduct of “ethnographic research” in the eastern provinces, including Diyar-bekir. He was emphatic in pointing out that he had “confiscated many books writ-ten in foreign languages”. This included minority languages as Kurmancî, Zazaki, Syrian-Aramaic, Arabic and Armenian (Türk Yurdu 1930). During those same tours through the eastern provinces, in autumn 1940 Tankut passed through Bitlis, home town of the sixteenth-century Kurdish chronicler Sharaf Khan, and reported with content that his book the Sharafname was not read anymore among Kurds: “I believe that the pages of the Sharafname and its Kurdish sagas are not read any more or are read with less excitement than before” (Başbakanlık Cum-huriyet Arşivi 490.01/1015.916.4 1940). Tankut’s attack on Sharaf Khan’s classic was matched by practical intervention in the field: during the 1920s and 1930s the dictatorship confiscated and destroyed copies of the book. For the sake of intelli-gence, the Hearths gathered lists of other books on Armenians and Kurds as well (Üstel 1997: 215, 333, 394).

Among the hundreds of books prohibited and confiscated by the regime figured: Kamuran Ali Bedir-Khan and Herbert Örtel, Der Adler von Kurdistan (Potsdam: Ludwig Doggenreiter, 1937); Sureyya Bedir Khan, The case of Kurdistan against Turkey (Philadelphia: The Kurdish Independence League, 1928); Cigerx-wîn, Dîwana Yekem: Prîsk û Pêtî [First collection: lightning and fire] (Damascus: n. p., 1945); a 1932 booklet on the Circassian alphabet published in Syria; Abdu-laziz Yamulki, Kürdistan ve Kürt İhtilalleri [Kurdistan and the Kurdish insurrec-tions] (Baghdad: n. p., 1946); Kamiran Alî Bedir-xan, Xwendina Kurdî [Kurdish studies] (Damascus: Çapxana Tereqi, 1938), and many others (Başbakanlık Cum-huriyet Arşivi 030.18.1.2). These books were literary, linguistic and historical stud-




ies, as well as outright nationalist pamphlets. What they had in common was their language, Kurdish, and topic, Kurds. Besides these books, all Kurdish-language periodicals were individually identified and categorically banned.

Although the eastern provinces were a peasant society where illiteracy fig-ures were as high as 80%, the official texts were not only the first ones the popula-tion would read, they were often the only ones available to the population. The organization of a hegemonic canon through exclusion and inclusion aimed at the formation of a closed circuit of knowledge. This act precluded the possibilities of a participatory memory and identity formation in the eastern provinces. The re-gime warded off both external penetration and internal criticism of their belief system by banning and destroying texts on a scale perhaps only matched by the Soviet dictatorship’s censorship agency Glavlit. The information dam erected around the country by the Young Turk regime rarely leaked, and “Turkishness” was measured by the level of exposure to that new, hermetically sealed body of knowledge. The damage this policy caused to Kurdish and other minority lan-guages is difficult to assess, but that it was profoundly detrimental to Kurdish culture in Turkey is beyond reasonable doubt.

6 Boarding schools: the battle against Kurdish children

The policy of “Turkification” through schooling manifested itself in the eastern provinces most identifiably in the boarding schools. The 1925 reports on “Reform in the East” had vaguely sketched that the policy of assimilation would be carried out through boarding schools. In regions where Kurds and Arabs lived, Turkish Hearths and schools needed to be opened and “most importantly, all sacrifices need to be endured to establish girls’ schools and ensure that the girls enroll. . . . By opening boarding schools, the region can be saved from getting involved with Kurdism . . . and girls’ schools can induce women to speak Turkish” (Bayrak 1993: 487). These ideas would materialize when during the interwar campaigns, the Turkish military elite understood that assimilating the Kurds could not only be a matter of destroying and dispersing the armed resistance. Education was redis-covered as a complementary and vital method of social assimilation. The Chief of Staff solicited the government for the foundation of an educational institute that could accelerate the “Turkification” process (Hekimoğlu 1938: 32–33). Precise plans for the establishment of boarding schools in the East were formulated by Interior Minister Şükrü Kaya. In a confidential report, he deplored especially the fact that during their lives, most Kurdish women stayed in their native regions and spoke “a perverted Persian called Kurmanci”. He continued: “Thus, they do



142 U. U. Üngör

not speak even a word of Turkish, and therefore cannot teach their children Turk-ish” (Mazıcı 1996: 233). On 4 June 1937, Kaya sent the Ministry of Culture a top se-cret circular about the need to enroll Kurdish children in boarding schools:

Boarding schools for girls and boys need to be opened and girls and boys from the age of five need to be brought into these schools for education and upbringing. These boys and girls need to be married to each other and settled dispersedly on property inherited from their parents where they can establish a Turkish Nest so that Turkish Culture can be thoroughly implanted [in the region]. . . . Therefore . . . it is necessary and essential that small children be placed in this type of boarding schools. (Mazıcı 1996: 233)

This is the single most important decree documenting the Young Turk assault on Kurdish children. According to Kaya, girls in particular needed to be placed in the schools since mothers were seen as the carriers of the Kurdish culture that needed to be exorcised from their minds. This order had come from Atatürk himself, who had expressed determination to pursue a policy leaving no place for mothers to raise their children with languages other than Turkish. The aim was to drive a cultural wedge between generations in order for Kurds to become “future Turks”. The road to the nation was as coercive as it was gendered: women were seen as carriers of national reproductivity, vessels of national identity, and transmitters of culture.

The first boarding school in the eastern provinces was established in Elazığ in 1937 (Yeşil 2003). Although it mostly aimed at schoolgirls from the Dersim dis-trict, it also drew students from the regions north of Diyarbekir (Koçak 2003: 101). There were pupils from Çermik, Ergani and Palu – the latter being a district of Elazığ province by that time. A relatively young and idealistic teacher from Istan-bul named Sıdıka Avar was appointed as director. Avar’s private archive and its distillate, her memoirs, offer a rare and valuable source of insight into the official perspective on educational policies in the East, and into how the Young Turk policy-makers organized the transformation process of the children from “primi-tive Kurds” to “civilized Turks” (Avar 1986). From the authorities’ point of view, this “civilizing process” required a twofold assault on Kurdish children’s identi-ties. On the one hand, the school needed to strip away all outward signs of the children’s identification with tribal and rural life, that is to say, their “savage” ways. On the other hand, the children needed to be instructed in the principles, values, ideas and behaviors of Turkish “civilization”. These twin processes – the tearing down of the old selves versus the building of new ones – were to be car-ried out simultaneously. As the “savage” Kurdish selves gave way, so the “civi-lized” Turkish selves would emerge.

From the moment of her assignment Avar began travelling in the countryside on horseback, searching for girls to enrol in her boarding school. On arrival in a




village, she would approach the locals and explain to those who understood Turkish what her objective was. In some villages she was received cordially, in others with outright hostility. After taking girls from a village, each one was pho-tographed on arrival. These “before the school” photos would later be contrasted with the “after the school” photos to demonstrate the transition to “civilization”. The girls would be put in quarantine for two weeks and only then began attend-ing classes. The curriculum in the boarding school was obviously nationalist and patriarchal. A standard program for the Kurdish village girls was three years and provided training with a special curriculum at elementary school level. Forty-four hours of class were taught in a week, and clear priority was given to Turkish lan-guage classes. Other classes were civics and math, but also child-care, house-keeping, cooking, embroidery and sewing, which, Avar argued, were “indispens-able for a housewife” (Yeşil 2003: 96–98).

Despite these difficulties, for Kemalist philanthropists the journey of Kurdish children to the boarding school was that first step out of the darkness of “sav-agery” into the light of “civilization”. The official discourse was euphoric in the example of Avar’s boarding school. One major official explained at length that the objective of the school would be “speaking in the national language” for students who would be “told that they are Turks in their feelings and in their lifestyles”. These students would be raised “as conscious citizens and educated mothers committed to the revolution, the national ideal, the country”, then to be sent back to their villages. There, they were expected to “indoctrinate [aşılamak] their chil-dren to protect and maintain . . . the works of civilization brought by the Republic to their region”, for “only in this manner, the civilization brought by the Republic will not remain as a veneer that disappears at the slightest strain, but will leave profound traces in the deepest corners of the public spirit that cannot be rubbed out by any force” (Hekimoğlu 1938: 32–33). The veteran Young Turk journalist Ahmed Emin Yalman called Avar a “first degree Turkish-nationalist raider [akıncı]” who had promoted “cultural unity in our eastern provinces” by intro-ducing “Turkish civilization” in that region. A secondary school teacher wrote a letter to Avar, praising her for “elevating the children to the level of civilized people by teaching them our language” (Avar 1986: 341). The observations of the mayor of the small town of Karlıova are at least as thought-provoking:

You know the story of how in America a cow enters [a factory] at one end and a sausage exits at the other. Here in Elazığ we possess such a factory for “civilized people”. In the Girls’ Institute, the most primitive and savage young girls are taken in from all villages. . . . Madame director gathers children like Janissaries. . . . Yes indeed, squalid, ragged, savage-natured, stubborn and ill-tempered children with no language skills are going to school now. It is difficult to believe that the jovial, civilized child that offers you coffee two, three years later and speaks fluent Turkish is the same girl. (Avar 1986: 233)



144 U. U. Üngör

This equation of “civilization” with “Turkishness” was expressed by Avar herself as well. In one of her annual reports to the Ministry of Education she wrote that her school was engaged in a “war for the sake of Turkishness” in which she claimed to be facing “a populace that does not welcome us with good will but always perceives us with suspicion and hesitation . . . and therefore needs to be indoctrinated with the Turkish ideal” (Avar 1986: 255).

The boarding school, especially in the eastern provinces, was the institu-tional manifestation of the government’s determination to restructure the Kurds’ minds and personalities completely. To understand how it functioned in this re-gard one must attempt to understand how Kurdish students actually came to know and experience it. And this effort must necessarily begin at that point in time when Kurdish youth left behind the familiar world of tribal and rural ways for the unfamiliar world of the state’s school. The girls’ immediate physical trans-formation included the cutting of hair, the changing of dress, and the changing of names. The first transformation, the cropping of their hair, was a rite de pas-sage which symbolized their initiation into “modernity” and “civilization”. For many Kurdish village girls long hair was traditionally seen as a symbol of beauty and femininity, and the cutting of it was perceived as humiliating. The girls felt it made them look boyish (Yeşil 2003: 111–114). Although the short-hair policy was rooted in considerations of controlling the problem of head lice, the reason went deeper than cleanliness. At the heart of the policy was the belief that the children’s long hair was symbolic of “savagism”, and removing it was central to their new identification with “civilization”. The changing of dress was anoth-er policy that stripped the children of their past culture. The traditional baggy trousers (şalwar) worn in the villages were prohibited in favour of school uni-forms (Avar 1986: 389). Since the Kemalists saw in Kurdish given names symbols of Kurdish ethnicity, many students’ names were forcibly changed on arrival. Many of the orphan girls snatched from the countryside had their names changed (Baran 1986: 12–15). As another graduate remembered, “When I arrived at the school, my Kurdish name was changed into a Turkish one. But I never forgot it: Delale”.3

Finally, and most importantly, the assault on Kurdish ethnic identity diverged into the absolute prohibition of speaking Kurmancî and Zazaki on the one hand, and the practising of Islam on the other. Fatma Demir, according to Avar one of her favourite students, remembered the total ban on the Kurdish language well: “Some of my friends spoke Kurdish among themselves because they did not understand Turkish yet and they were punished severely. They were not

3 Interview conducted with Şemsiye Gezici (born 1936), in Bursa, 15 June 2002.




given dinner, they were beaten on the palms of their hands with a ruler, and had to stand on one foot for a long time”.4 Another girl responded, “We would never speak in Kurdish among ourselves. None of us. Who can dare to speak in Kurdish? There is no such possibility. There were watchmen and others. . . .” Speaking Kurdish or bad Turkish entailed corporal punishment, as one girl re-membered: “We did not like the Turkish language classes because our former teacher was scolding us, beating us with a ruler”. Religion was another factor. It was no surprise that the Kemalist state, a secular dictatorship, prohibited all ex-pressions of any religion in the schools. Students bitterly remembered the prohi-bition of the prayer, the veil, and various fasting episodes, important pillars of Islamic faith. Avar defended the measures with the argument that praying was “unscientific” and fasting bad for “a healthy brain” (Yeşil 2003: 115–116, 118, 132–133).

7 ConclusionEthnocide is often defined as the assault on ethnic markers and (im)material culture (Palmer 1992). The Young Turk regime in this definition was responsi-ble for a large-scale ethnocidal policy pursued against non-Turkish ethnic groups, in particular Kurds. It outlawed and criminalized languages, prohibited and de-stroyed books, mutilated and silenced literature for no other reason than the purely ideological construct of demographic and cultural homogeneity. But if ethnicity is a relational phenomenon, then the forced disruption of those ethnic relations and ties should also be included and problematized in any definition of ethnocide. It was for a reason that Article 2e of the 1948 Convention on the Prevention and Punishment of the Crime of Genocide recognized that the forced transfer of children from one group to another was an element of geno-cide. Perhaps we should, most of all, focus on the dissolution of ethnic rela-tions as a form of ethnocide. It then appears that the ethnocidal nature of the Young Turk dictatorship is materialized perhaps nowhere more clearly than in the boarding schools for Kurdish children. The Ministry of Education held the children in its powerfully assimilationist embrace, designed to carry out the mis-sion of “Turkification” as a classical example of a “total institution” (Goffman 1961).

But the children started their education in Turkish at the age of seven to twelve, too late to socialize children from scratch. Psychologists have argued that

4 Interview conducted with Fatma Demir (1925–2007), in Istanbul, 10 June 2002.



146 U. U. Üngör

by that age, a child will have passed “the capacity for full cultural acquisition” of one single national culture (Fromkin et al. 1974; Singleton 1995). Sociologists, too, emphasize that “identities which are established this early in life – selfhood, human-ness, gender, and . . . kinship and ethnicity – are primary identities, more robust and resilient to change in later life than other identities” (Jenkins 1996: 21). Partly for this reason, the boarding school experience for many initially was trau-matic, and in the long run alienating. At the time, the practice was presented as promoting the welfare of individual Kurdish children, because Kurdish cultural identity was seen as an insurmountable obstacle to the capacity to take a “nor-mal” part in “modern” Turkish social life. As such, contemporary officials main-tained that the overall effect was beneficial, and that the intentions were good. Similar to American, Canadian, and Australian educational policies towards Na-tive Americans, First Nations, and Aboriginals, this was interpreted as a “civilis-ing mission” (Van Krieken 2002). Up to this day, this argument is maintained in Turkey.

Inasmuch as the Young Turks allowed difference and cultural initiatives in the civil society, these were only tolerated if they served “Turkishness” as defined by the regime. Whatever the ideas of the Young Turks were on language (reform), they tallied neither with the local Turkish dialects of the eastern provinces nor the locally dominant varieties of Kurdish, Circassian, Arabic, Syrian-Aramaic, Zazaki, Laz and other languages. At no point were these languages respected or taken seriously as regionally dominant languages. In non-Turkish languages, no texts were published, no music was played and sold, no plays were performed, and no programs broadcast in the public space. The prohibition of non-Turkish culture in the public space may have been rigorous, but ordinary people, sur-rounded by omnipresent “Turkishness”, created within its strictures space to live their lives, and non-Turkish culture could function invisibly, that is, not visible in the public domain. One expert has argued that by 1960, “there were quite a few cases of successful assimilation”, but adds that this was an urban phenomenon (Van Bruinessen 1997: 8–9). But decades after the end of Young Turk rule, mil-lions of peasants living in the countryside continued speaking their own lan-guages until a new wave of nation formation prohibited the private use of Kurdish (Bozarslan 1997: 84). Urbanization, as a result of labor migration from the 1960s on, or the destruction of villages during the war between the Kurdistan Workers’ Party (PKK) and the Turkish army (1984–1997), contributed more to the spread of Turkish language and culture among these people than the massive campaigns of nation formation in the Young Turk era ever did. (This, however, is beyond the scope of this article.)

Therefore, one can perhaps argue that Young Turk forced linguistic as-similation in the eastern provinces largely failed, not because it was Turkish, but




because it was totalitarian and violent. The high levels of coercion behind the linguistic and cultural policies were violent forms of political expression the re-gime deployed in order to retain its sovereignty over the region and the popula-tion. It was the culture of prohibitions, impositions, coercion, and violence that deeply upset many people. For many, too much violence had been at the foun-dation of Turkish nation formation to be truly attractive. Instead, the death of Kurdish has to be sought elsewhere: the liberalizing and democratizing process in Turkey after 1950 achieved probably more in terms of damage to the use of Kurdish than Young Turk social engineering did. As Kurds moved to cities, they participated in the urban economies, became immersed in Turkish-language cul-ture and media, and sent their children to state schools. Within a generation, Kurdish was hardly spoken any longer among this large segment of the Kurdish population.

This suggests that De Swaan’s model of language abandonment as emancipa-tion and upward social mobility is partially applicable to the Young Turk case. There seems to be a deeply ambivalent process at play here. In the face of the Young Turk regime’s racist discourse and colonial attitude towards Kurds, it is hardly credible to claim that the regime intended to impose the Turkish language on the Kurdish population in order to stimulate the Kurds’ social mobility and socio-economic position. Neither was this an immediate outcome of their poli-cies. Skutnabb-Kangas’ argument for linguistic destruction also applies here: the majority of Kurds continued speaking Kurdish in Kurdistan well into the 1980s, until the Turkish army’s counterinsurgency war destroyed the social fabric of rural Kurdistan. It is true that, had it been up to the Young Turks, the Kurdish language would have been fully extinguished. But the linguistic policies pro-duced only limited results due to the regime’s weakness, the tribal nature of rural Kurdish society, and the utterly unrealistic objectives of full linguistic assimila-tion on that scale. It seems that this debate will continue unabated until more research can provide a sharper image of the problem at hand.

ReferencesAktar, Ayhan. 2000. Varlık vergisi ve ‘Türkleştirme’ politikaları [The wealth tax and the

“Turkification” policies]. Istanbul: İletişim.Atatürk’ün söylev ve demeçleri [Ataturk’s speeches and announcements]. 1945. Istanbul: Türk

İnkılâp Tarihi Enstitüsü Yayınları.Atatürkçülük (birinci kitap) Atatürk’ün görüş ve direktifleri [Ataturkism (first book): Ataturk’s

views and directives]. 1988. Istanbul: Milli Eğitim Basımevi.Avar, Sıdıka. 1986. Dağ çiçeklerim (anılar) [My mountain flowers (memoires)]. Ankara:

Öğretmen.



148 U. U. Üngör

Baran, Aziz. 1986. You must give a Kurdish baby a Turkish name. Kurdish Times 1(1). 12–15.Başbakanlık Cumhuriyet Arşivi [Republican Archives, Ankara]. Fond 490.01.Bayrak, Mehmet (ed.). 1993. Kürtler ve ulusal-demokratik mücadeleleri: Gizli belgeler –

araştırmalar – notlar [The Kurds and their national-democratic struggles: secret documents – research – notes]. Ankara: Özge.

Beşikçi, İsmail. 2004. International colony Kurdistan. London: Parvana.Bozarslan, Hamit. 1997. La question kurde: états et minorités au Moyen-Orient. Paris: Presses

de Sciences Po.Bozarslan, Hamit. 2001. M. Ziya Gökalp. In Ahmet İnsel (ed.), Modern Türkiye’de siyasi düşünce

[Political thought in modern Turkey], 314–319. Istanbul: İletişim Yayınları.Çapar, Mustafa. 2006. Tek parti dönemi: Milli eğitim, milli dil ve Türkleştirme politikaları [The

single-party era: national education, national language and Turkification policies]. In Derya Demirler & Mert Kayhan (eds.), Türkiye’de azınlık hakları sorunu: Vatandaşlık ve demokrasi eksenli bir yaklaşım [The question of minority rights in Turkey: an approach from the perspective of citizenship and democracy], 83–88. Istanbul: TESEV.

CHP dördüncü büyük kurultayı görüşmeleri tutulgası [Minutes of the CHP’s fourth great congress meetings]. 1935. Ankara: n. p.

Çolak, Yılmaz. 2004. Language policy and official ideology in early republican Turkey. Middle Eastern Studies 40(6). 67–91.

Cumhurbaşkanları, başbakanlar ve millî eğitim bakanlarının millî eğitimle ilgili söylev ve demeçleri [The speeches and announcements of presidents, prime ministers and ministers of national education about national education]. 1946. Ankara: Türk Devrim Tarihi Enstitüsü.

Dağlı, Nuran & Belma Aktürk (eds.). 1988. Hükümetler ve programları [The governments and their programs], Vol. 1, 1920–1960. Ankara: TBMM Basımevi.

De Swaan, Abram. 2001. Words of the world: the global language system. Cambridge: Polity.Ergüven, Ekrem (ed.). 1937. Şükrü Kaya: Sözleri – yazıları 1927–1937 [Şükrü Kaya: his words

– writings 1927–1937]. Istanbul: Cumhuriyet Matbaası.Fromkin, Victoria, Stephen Krashen, Susan Curtiss, David Rigler & Marilyn Rigler. 1974. The

development of language in genie: a case of language acquisition beyond the critical point. Brain and Language 1. 81–107.

Gałęski, Bogusław. 1971. Sociological problems of the occupation of farmers. In Teodor Shanin (ed.), Peasants and peasant societies, 180–201. Harmondsworth: Penguin.

Goffman, Erving. 1961. The characteristics of total institutions. In Amitai Etzioni (ed), Complex organizations: a sociological reader, 312–340. New York: Holt, Rinehart & Winston.

Gökalp, Ziya. 1923. Şehir medeniyeti, köy medeniyeti [Urban civilization, village civilization]. Küçük Mecmua 30. 4–7.

Hassanpour, Amir. 1991. State policy on the Kurdish language: the politics of status planning. Kurdish Times 4(1–2). 42–85.

Hekimoğlu, Nuriye D. 1938. Kız enstitüsü niçin açıldı [Why was the girls’ institute opened]. Altan: Elaziz halkevi dergisi 33–34–35. 32–33.

Heper, Metin. 2007. The state and Kurds in Turkey: the question of assimilation. London: Palgrave Macmillan.

Houston, Christopher. 2004. Creating a diaspora within a country: Kurds in Turkey. In Melvin Ember, Carol R. Ember & Ian Skoggard (eds.), Encyclopedia of diasporas: immigrant and refugee cultures around the world, Vol. 2, 403–414. New York: Kluwer Academic.




İçimsoy, A. Oğuz & İsmail E. Erünsal. 2008. The legacy of the Ottoman library in the libraries of the Turkish republic. Libri 58(1). 47–57.

İshakî, Ayaz. 1928. Gayrı Türkler nasıl Türkleştirilebilir? [How can the non-Turks be Turkified?]. Cumhuriyet, 28 April, no. 1426.

Jenkins, Richard. 1996. Social identity. London: Routledge.Koçak, Cemil. 2003. Umûmî müfettişlikler (1927–1952) [The inspectorates-general (1927–1952)].

Istanbul: İletişim.Lewis, Bernard. 2001. The emergence of modern Turkey. Oxford: Oxford University Press.Lieven, Dominic. 2002. Empire: the Russian empire and its rivals. New Haven: Yale University

Press.Mazıcı, Nurşen. 1996. Celal Bayar: Başbakanlık dönemi 1937–1939 [Celal Bayar: his period as

prime minister, 1937–1939]. Istanbul: Der.Palmer, Alison. 1992. Ethnocide. In Michael N. Dobkowski & Isidor Wallimann (eds), Genocide

in our time: an annotated bibliography with analytical introductions, 1–6. Ann Arbor, MI: University of Michigan Press.

Singleton, David. 1995. The age factor in second language acquisition: a critical look at the critical period hypothesis. Clevedon: Multilingual Matters.

Skutnabb-Kangas, Tove. 2000. Linguistic genocide in education or worldwide diversity and human rights?. Mahwah, NJ: Erlbaum.

Tanin, 25 September 1908.Tunçay, Mete. 1995. Türkiye cumhuriyetinde tek parti yönetiminin kurulması 1923–1931 [The

establishment of single-party rule in the Turkish republic 1923–1931]. Ankara: Yurt Yayınları.

Türk Ocakları 1927 Senesi Kurultayı Zabıtları [Minutes of the 1927 Congress of the Turkish Hearths]. 1928. Ankara: n.p.

Türk ocakları 1928 senesi kurultayı zabıtları [Minutes of the 1928 congress of the Turkish Hearths]. 1929. Ankara: Türk Ocakları Matbaası.

Türk ocakları üçüncü kurultayı zabıtları [Minutes of the third congress of the Turkish Hearths]. 1927. Istanbul: Kader.

Türk yurdu, 4/24:29/223, May 1930.Türkiye Büyük Millet Meclisi zabıt ceridesi. 1934.Uzun, Mehmed. 2003. Diyarbakir: the slap in the face. International Journal of Kurdish Studies

1(1). January.Üstel, Füsun. 1997. İmparatorluktan ulus-devlete Türk milliyetçiliği: Türk ocakları (1912–1931)

[Turkish nationalism from empire to nation state: the Turkish hearths (1912–1931)]. Istanbul: İletişim.

Vakit, 27 April 1925.Van Bruinessen, Martin. 1997. Race, culture, nation and identity politics in Turkey:

some comments. Paper presented at the Annual Turkish Studies Workshop Continuity and change: shifting state ideologies from late Ottoman to early Republican Turkey, 1890–1930, Department of Near Eastern Studies, Princeton University, 24–26 April.

Van Krieken, Robert. 2002. Reshaping civilization: liberalism between assimilation and cultural genocide. Amsterdams sociologisch tijdschrift 29(2). 215–247.

Winter, Michael. 1984. The modernization of education in Kemalist Turkey. In Jacob M. Landau (ed.), Atatürk and the modernization of Turkey, 183–194. Boulder, CO: Westview Press.



150 U. U. Üngör

Yeşil, Sevim. 2003. Unfolding Republican patriarchy: the case of young Kurdish women at the girls’ vocational boarding school in Elazığ. Unpublished MA thesis, Middle East Technic University, Department of Gender and Women’s Studies.

Yücel, Hasan Ali. 1993. Milli eğitimle ilgili söylev ve demeçler [Speeches and announcements about national education]. Ankara: T. C. Kültür Bakanlığı Yayınları.

Yücel, Hasan Ali. 1994. Türkiye’de orta öğretim [Secondary education in Turkey]. Ankara: T. C. Kültür Bakanlığı Yayınları.

Zeydanlıoğlu, Welat. 2007. Kemalism’s others: the reproduction of orientalism in Turkey. Unpublished Ph.D. thesis, Anglia Ruskin University.

Zürcher, Erik-Jan. 1992. The Ottoman legacy of the Turkish Republic: an attempt at a new periodization. Die Welt des Islams 32. 237–253.



DOI 10.1515/ijsl-2012-0053 IJSL 2012; 217: 151 – 180

Ergin ÖpenginSociolinguistic situation of Kurdish in Turkey: Sociopolitical factors and language use patterns

Abstract: This article aims at exploring the minority status of Kurdish language in Turkey. It asks two main questions: (1) In what ways have state policies and socio-historical conditions influenced the evolution of linguistic behavior of Kurdish speakers? (2) What are the mechanisms through which language maintenance versus language shift tendencies operate in the speech community? The article discusses the objective dimensions of the language situation in the Kurdish re-gion of Turkey. It then presents an account of daily language practices and perceptions of Kurdish speakers. It shows that language use and choice are sig-nificantly related to variables such as age, gender, education level, rural versus urban dwelling and the overall socio-cultural and political contexts of such uses and choices. The article further indicates that although the general tendency is to follow the functional separation of languages, the language situation in this con-text is not an example of stable diglossia, as Turkish exerts its increasing pres-ence in low domains whereas Kurdish, by contrast, has started to infringe into high domains like media and institutions. The article concludes that the preva-lent community bilingualism evolves to the detriment of Kurdish, leading to a shift-oriented linguistic situation for Kurdish.

Keywords: Kurdish; Turkey; diglossia; language maintenance; language shift

Ergin Öpengin: Lacito CNRS, Université Paris III & Bamberg University. E-mail: [email protected]

1 IntroductionKurdish in Turkey is the language of a large population of about 15–20 million speakers. Yet, it has rarely been the subject of formal sociolinguistic description. The existing literature (Hassanpour 1992; Akin 1995; Haig 2004; Skutnabb- Kangas and Fernandes 2008; Coşkun et al. 2011) has discussed in detail the modality of restrictions on the private and public usage of Kurdish in Turkey,

Brought to you by | Penn State - The Pennsylvania State UniversityAuthenticated


152 E. Öpengin

naming state attitudes and policies towards Kurdish under terms like “lingui-cide”, “glottophagie” and “invisibilization” practices. However, what remains relatively absent is research on the speakers’ perspectives about the vitality of their language and on issues of language maintenance and shift. To address this research gap, this study is informed by two questions: (1) In what ways have state policies and socio-historical conditions influenced the evolution of linguis-tic behavior of Kurdish speakers? (2) What are the mechanisms through which language maintenance versus language shift tendencies operate in the speech community?

This article thus tries to establish the complex array of interrelations be-tween language use and choice patterns of Kurdish speakers in Turkey and socio-political factors of the language situation to the extent that they relate to the “speaker” and “language”. The existing literature has mostly backgrounded these factual aspects as mere setting/context, whereas in this study these fac-tors are foregrounded as crucial elements behind the current linguistic situation. Here the key constructs related to minority language situations such as diglossia, language vitality and language maintenance and shift are discussed. A back-ground section presents an overview of the objective/factual conditions of the language situation in Turkey. I then analyze the linguistic behavior and percep-tions of the speakers in respect to the domains of language use, interlocutors, and speakers’ generation, gender, and level of formal education. Finally, the extent to which the factual aspects of the linguistic context are reflected in the linguistic behavior and perceptions of the speakers is assessed. The article con-cludes that the mainly unfavorable sociopolitical conditions have led, and continue to lead, to community bilingualism evolving to the detriment of Kurd-ish, a shift-oriented linguistic situation. Nevertheless, I suggest that the phenom-enon is not a wholesale one and that there are mechanisms through which the language shift tends to be reversed, for instance through media and cultural and political activism.

2 Conceptual framework

Minority language situations are often studied by referring to a set of socio-political factors supposedly influencing the maintenance and shift of minority languages (Kloss 1966; Giles et al. 1977; Edwards 1992). Language maintenance is a speech community’s use of its first language in a number of domains in a contact situation (Yağmur 1997: 18); it regulates within-community communi-cation (Fase et al. 1992) and protects against external attitudes (Grenoble and



Sociolinguistic situation of Kurdish in Turkey 153

Whaley 2006: 13). In reverse, language shift refers to a stage in the relationship of the community to its language during which the members of the community have either partially or completely abandoned the usage of the native language (Winford 2003: 14–17). Language shift is typically a gradual process and it nec-essarily implies changes in societal norms. Thus, as implied in Fase et al. (1992: 7), it can be best understood by studying the mechanisms that govern soci-etal norms and changes in them. A number of models have been developed to understand those overarching mechanisms. The construct of “ethnolinguistic vitality” (EV henceforth), proposed in Giles et al. (1977), aims at exploring and systematizing the role of socio-cultural factors on language maintenance, lan-guage shift and language loss. The concept is defined as the socio-structural factors that make a group likely to behave as a distinctive and active collective entity in intergroup situations (Giles et al. 1977: 307). The EV of a group is made up of three cover factors, namely status, demographic and institutional support and control. A group with high values on these factors will have a high ethno-linguistic vitality, thus tending to maintain its language and preserve its dis-tinctive group characteristics; conversely, a group with low values on most will end up with a low EV; thus it will tend to assimilate and consequently may cease existence as a distinctive collective identity (see Landry and Allard 1994). Fur-thermore, in order to systematize the role of individual perceptions of the socio-structural factors on language behavior, Bourhis et al. (1981) proposed a Subjec-tive Ethnolinguistic Vitality Questionnaire (SRVQ), designed to obtain the subjective assessment of group members on the vitality factors both for their own group and for one or multiple other groups coexisting in the context. The EV has been used in many countries and with many language groups, and developed substantially further.

In addition to the typological models, constructs such as “diglossia” and “do-main analysis” help to better understand the individual and interactional bases of linguistic behavior. Domain is conceived as “cluster of social situations typi-cally constrained by a common set of behavioral rules” (Fishman 1972 [1968]: 263). Linguistic behavior of members of a speech community is assumed to be governed by such sets of situational constraints. Diglossia, in turn, refers to a relatively stable stage in which the languages of the contact situation are func-tionally separated across a set of social domains and communicative situations (see Ferguson 2003 [1959]; Fishman 2003 [1967]). In this configuration, the variety that is used for more prestigious functions such as education, media and market is the high variety, while the variety used in more intimate domains and functions such as intra-familial communication, friendship and neighborhood is the low variety. When the functional separation of the languages is no more respected, the low language inevitably follows the path to shift (Fishman 2003 [1967]: 360)



154 E. Öpengin

and bilingualism tends to be temporary and transitional (Edwards 1994: 85). However, a number of authors working around “conflit linguistique” (Jardel 1982; Kremnitz 1991; Boyer 1997) take issue with the “harmonizing” interpretation of diglossia. For these scholars, the existence of a high and a low variety within a community inevitably presupposes conflictual relationships, characterized by concurrence, dominance, and violence (Boyer 1997: 6–14). The present study il-lustrates the difficulties with stable functional separation of languages in that, as it will be seen, not only the high language Turkish has started to exert its presence in low domains but also the low language Kurdish aspires to high domains and functions.

In addition to diglossia, domain analysis and ethnolinguistic vitality, I em-ploy Edwards’ (1992) typological model, which proposes eleven categories/perspectives: demography, sociology, linguistic, psychology, history, politics, geography, education, religion, economy and media. Each factor is analyzed according to the “speaker”, “language” and “setting”. For instance, the geo-graphical classification of linguistic situations makes three basic distinctions: (1) “unique minority” vs. “non-unique minority” depending on whether the language is unique to one state or is spoken in several; (2) “adjoining” vs. “non-adjoining” depending on whether members of speech community are geographi-cally connected or not; (3) “cohesive” vs. “non-cohesive” depending on the ex-tent of internal spatial cohesion among speakers of a speech community within a state (Edwards 1992: 39). The assumption is that minority strength will vary according to the three dimensions of the model. Thirty-three items formulated as specific questions also guide the analysis (Edwards 1992: 50) and provide comprehensive data and insights into aspects such as history and background of speech community, numbers and concentrations of speakers, degree and type of language transmission, the nature of maintenance or revival efforts, degree of autonomy or “special status” of the area, the relationship between language and economic success, and the association between language and identity. The model was found especially useful for the analysis of non-migrant situations (Clyne 2003: 45) and developed to include a further category of “litera-cy” (Grenoble and Whaley 1998). In what follows, I also discuss some of these categories.1

1 The “Language vitality and endangerment” model (UNESCO 2003) is another useful typology for assessing the vitality of a given language. For a concise presentation of the model see Grenoble and Whaley (2006: 3–12). For the model see: http://www.unesco.org/new/fileadmin/MULTIMEDIA/HQ/CLT/pdf/Language_vitality_and_endangerment_EN.pdf (accessed 6 February 2012).



http://www.unesco.org/new/fileadmin/MULTIMEDIA/HQ/CLT/pdf/Language_vitality_and_endangerment_EN.pdf

http://www.unesco.org/new/fileadmin/MULTIMEDIA/HQ/CLT/pdf/Language_vitality_and_endangerment_EN.pdf


3 The factual dimensions of Kurdish in Turkey

3.1 Geographical and historical contextualization of the community and the language

The sociolinguistic context treated in this paper, Turkish or Northern Kurdistan, corresponds to the east and southeast regions of Turkey with predominantly Kurdish concentration. It is bordered by Kurdish regions of Iran, Iraq and Syria (see Figure 1). The region occupies around 30% of Turkey (TESEV 2006: 34). It is mountainous with vast plains in certain parts. The Tigris and Euphrates rivers and mines are of geopolitical value (Yildiz 2006: 65–77). The European Commis-sion (2004: 39) estimated the Kurdish population in Turkey to be between 15–20 million, which amounts to 20–25% of the population of Turkey. As for the popula-tion living in Kurdish region, TESEV (2006: 34) states it as 15% of the total popula-tion. In 2003, the birth rate was estimated in the Eastern Anatolia region as 3.8, and in the south-eastern Anatolia region as 4.2, whereas in all of Turkey it is 2.3 (TESEV 2006: 36). Despite the high birth rate, the rate of population growth in the Kurdish region is thought to be under the average of Turkey because of migration to western parts of the country.

Until the 1950s, Kurds inhabited predominantly rural areas, but the advent of mass production in the agricultural sector has led to a constant flux of migra-tion. The most important reason for the Kurdish migration, though, has been the armed conflicts between the PKK (Kurdistan Workers’ Party) and the Turk-ish army. This resulted in the evacuation of 3,438 villages and the deportation and migration of 4 to 4.5 million Kurdish-speaking citizens between 1989–1999 (Göç-Der 2001: 8; Bozarslan 2009: 70–71). Kurdish population, thus, has been increasingly concentrated in urban areas of the region and in western cities of Turkey.

The modern history of the region is shaped by a categorical denial of Kurdish identity after the founding of Turkish Republic in 1923. The republican project of creating a homogenous Turkish nation imposed Turkish identity to all subjects (see Bozarslan 2009: Ch. 2). Thus, the existence of Kurds was officially occulted (Yeğen 1999) or invisibilized (Haig 2004) and all the references to Kurdishness were banned and stigmatized (Akin 1995). Under such circumstances, Kurdish reaction often mobilized as a number of armed revolts between 1921–1938, which mostly ended up with brutal repressions and the deportation of large groups of Kurds (Bozarslan 2009: 39–43). As a result, the political and public sphere be-came more and more sensitive and closed to any form of Kurdishness, stigmatiz-ing references to Kurds.



156 E. Öpengin

Fig.

1: K

urdi

sh a

nd O

ther

Nat

ive

Lang

uage

s Spo

ken

in Tu

rkey

(Has

sanp

our 1

992:

4)




As for language, with the spread of Islam in Middle East starting from the 7th century AD, Arabic grasped most of the high functions of languages of the region. Under the Kurdish principalities of 17th century Kurdish was valued among the governing elites. This led to the establishment of Kurdish literary tradition (Bru-inessen 1989: 43). However, the secondary status of Kurdish even during this era can be inferred from several facts (see also Bruinessen 1988). For instance, in his Kurdish romance Mem û Zîn, written in 1695, Ehmedê Xanî complains that Kurd-ish is not granted the value it deserves (Khani 2010: 35). Furthermore, Kurdish had the status of an auxiliary or transitory language for the learning of Arabic within the medrese-based education system (Zinar 1998). Rich (1836) makes simi-lar observations on the status and functions of Kurdish during his two years stay (1819–1820) in Kurdistan:

Kurdish language prevails over the entire country from Armenia on the North to region of Baghdad on the South, and from the Tigris on the West to Azerbaijan on the East. . . . The Kurds commonly use Persian or Turkish in their written communications. In the schools which they have here and there, a little Persian and Arabic is taught, but not the smallest portion of their vernacular tongue. (Cited in Edwards [1851: 121–123])

From the second half of 19th century, after the collapse of all Kurdish principali-ties, Kurdish was patronized by cultural and political associations established at the turn of the century and by an important number of poets writing principally in Central Kurdish. Through this period, Kurdish retained its limited status in the medrese-based educational system. During this period the first books and maga-zines and the first newspaper in Kurdish were published in Istanbul and other central cities of Middle East. Some twenty books in Kurdish were published be-tween 1844 and 1923 (Malmisanij 2006: 18). However, following the founding of the Turkish Republic, with the reforms and revolutions promulgated in 1924 and 1925 (see Zeydanlıoğlu, this issue), any activity aiming at the usage and develop-ment of Kurdish was banned. Therefore, the development of Kurdish, including corpus planning required for adapting to the modern and urban life, had to take place outside Turkey.

3.2 Kurdish in modern times: confrontation with the urban space

The works of Hawar circle led by Celadet Ali Bedirkhan and his colleagues based in Syria and Lebanon; presented mostly in the journals Hawar (1932–1943), Ronahî (1942–1945), Roja Nû (1943–1946) and Stêr (1943–1945), mark a turning point in



158 E. Öpengin

the corpus development of Kurdish aimed at rendering the language more com-patible with the emerging requirements of new language use domains (see Matras and Reer shemius 1991). Yet, because of the illegality of Kurdish in the public sphere, the influence of their innovative efforts on the use of Kurdish remained mostly negligible until the 1980s. Thus, even if the works of the Hawar circle pro-vided the basis for codification of Kurdish in Roman script, based on the Botan variety, the lack of institutions for implementing language norms2 has in practice resulted in a partial standardization with a large array of intra-lingual variation in orthography, vocabulary and grammar. During the period after the 1980s until the turn of the last century, the Kurdish diaspora in Sweden and other western European countries has been the center for terminological modernization and material development in Kurdish (see Scalbert-Yücel 2007). The marginalization of Kurdish in Turkey shows also in the number of books published in Kurdish. In Sweden alone 657 books in Kurdish were published between 1974 and 2005 (Scalbert-Yücel 2007: paras. 24–31), whereas around 632 books were published in Turkey during more than one and a half century (1844–2006) (Malmisanij 2006: 22), with over half of them within the five years between 2001 and 2006.

With the partial abolition of the categorical ban on private and public usage of Kurdish in 1991, the venue for introducing a more functional and up-to-date linguistic code to large components of the Kurdish community was opened. De-spite the repressive context of the 1990s, in addition to Kurdish cultural centers, an institute in Istanbul and Kurdish publishing houses, the first weekly newspa-per (Welat) in Kurdish was published in this period. The start of the first Kurdish satellite TV-channel in 1995 (MED TV ) was an efficient medium for the implemen-tation of language norms elaborated throughout the century (Hassanpour 1998). This presumably also led the Kurdish speakers to develop more positive percep-tions on Kurdish by attributing new perceptual and practical functions to it in educated and widely urban daily life. This evolution in the situation of Kurdish was accelerated by the legalization of “regional languages” in private institutions in 2003. Hence, a number of other factors and developments, mostly recent, have influenced the normalization of Kurdish, such as the opening of private Kurdish courses in central cities of Turkey and Turkish Kurdistan, more than one hundred books published in Kurdish annually, the publication of over 15 literary, political and research magazines, the increase in the number of Kurdish TV-channels (see Sheyholislami 2010, 2011), a strong presence of Kurdish virtual media, and even-tually the establishment of institutes and departments at Turkish universities

2 For instance, only a total of twenty books in Kurdish appeared in Turkey from 1923 to 1980 (Malmisanij 2006: 19).




aiming at studying Kurdish language and literature, as well as the start of the public TV-channel TRT 6 in Kurdish. Nevertheless, the normalization of Kurdish is hampered also by the fact that Kurdish has almost never been a full-fledged “urban” language, or the language of administration and formal education.3 This is further reflected in the kind of incompatibility between Kurdish and the city in modern times. For instance, as my own research indicates, until the last quar-ter of the last century, Kurdish or other non-Turkish inhabitants of Diyarbekir (Diyabakır), the biggest Kurdish city today, would rarely use or value Kurdish in the city center. In the same vein, in Şemzinan (Şemdinli) district of Hakkari, a small and isolated town until half a century ago, some families immigrating from surrounding villages took themselves to be the city-dwellers as they were in con-tact with the few symbolic state institutions in the district (see Erdost 1986); thus, although Kurdish was the indispensible means of communication in the district, they were often quite reluctant to transfer Kurdish to their children. Here, despite the absence of any real contact with speakers of Turkish, solely because of the way they perceive the concept of the “city” and “public institutions”, they partly shifted to Turkish as the means of intra-familial communication. These two cases illustrate the problematic confrontation of Kurdish with the city while at the same time affirming the widely held claim that the city promotes or reinforces monolin-gualism or linguistic homogenization (see Calvet 1994).

Furthermore, despite a strong will for sorting out linguistic norms (see Zêre-van 1997; Aydogan 2006), in the absence of formal education in Kurdish and sys-temic means of installing linguistic norms, one can observe much variation in Kurdish orthography, vocabulary and grammar. This is surely not unfortunate in itself. However, the constant incorporation of new vocabulary and the increase in multiple forms for one meaning or one structural function render it difficult for ordinary readers or spectators to keep up with the evolution of the language. On the other hand, a strict application of purist linguistic norms and using a large number of neologisms and borrowed words might in the special oppressed situ-ation of Kurds instigate a state of “linguistic insecurity” among the speakers in regard to their linguistic proficiency and fluency in Kurdish (see Öpengin 2009: 59–61). Furthermore, Kurdish-Turkish language contact is maximally asymmet-ric. Thus, convergence phenomena such as code-switching, code-mixing, marked word order, calques and structural convergence, that are sometimes seen as instances of linguistic regression in the speakers’ linguistic competences (see Myers-Scotton 1992), are recurrent phenomena in daily discourse (see Güçin and

3 See Bruinessen (1988) for an evaluation of the status of Kurdish in 17th century and Celîl (2002: 10–13) for an account of a 19th century Kurdish principality where Armenian is used for education and administrative issues.



160 E. Öpengin

Öpengin 2008) and written language (see Aydogan 2007). Tan (2008: 92), reflect-ing on frequent code-switchings in ordinary speech, states that in some Kurdish cities a hybrid language (zimanekî dureh) has emerged.

3.3 Economic and institutional stigmatization of Kurdish

The Kurdish region has often been given as an example for under-underdevelopment (Jafar 1976). In spite of its important geopolitical position (see O’Shea 2004: Ch. 1), including the abundance of energy and water resources, legal regulations and infrastructure deficiencies (TESEV 2006: 21) do not allow the realization of the economic potential of the region, especially when it comes to commerce with neighbouring countries. TESEV (2006: 21) reports that 60% of the population in the region lives under the poverty threshold and that the pov-erty is systematically transmitted to the following generations. The Index of Human Development for 2004 situates Turkey on 94th rank whereas the prov-inces with a predominant Kurdish population, with their 631 index value, are ranked on 124th, similar to the index value of Morocco TESEV (2006: 16).

A detailed analysis has yet to be undertaken of the extent to which the lin-guistic profiles of Kurds relates to their economic inferiority,4 but it can be stated that the relationship between language and the economic market in Turkey grants hardly any social capital value to Kurdish: the official language Turkish is the sole linguistic medium indispensible for economic and social success. Apart from the official measures on imposition of the usage of Turkish in much of the public sphere, the repressive state politics and denigratory efforts of dominant discourses seem to have created a perceptual and practical aura in which the usage of Kurdish is associated with poverty and some sort of economical back-wardness whereas the usage of Turkish is considered the given norm. The know-ledge of Kurdish rarely yields any economic profit whereas it is practically impos-sible to do any work without a certain level of competence in Turkish, other than occupations exclusively in a village context. Hence, a recovery or reversal of lan-guage shift in this domain will surely rely on speakers’ integrative motivations (Gardner 1983: 203). This is indeed observed in the region. Most politicized circles of the Kurdish community have recently undertaken conscious efforts to render Kurdish visible in the market by practices such as using it in the work and market-place, by naming their stores in Kurdish, sporadically publishing and distribut-

4 For a study indicating a relationship between linguistic capital and income among the Kurdish and Arabic women in Turkey, see Smits and Gündüz-Hoşgör (2003).




ing users’ manuals and leaflets of municipality services in Kurdish, and opening positions in municipalities and research institutions that require good command of spoken and written Kurdish. Likewise, new TV-channels in Kurdish and the establishment of Kurdish language and literature departments in several univer-sities in the Kurdish region in Turkey may well contribute to the creation and prospects of job opportunities related to the usage of Kurdish.

As the institution par excellence for the application of the republican condi-tions, the school also has served as an efficient tool for official discourse to rele-gate Kurdish to devalued domains and to replace it with Turkish (see Üngör, this issue). At present, Article No 42 of the Constitution states that no language other than Turkish shall be taught to Turkish citizens as a mother tongue in education establishments (see Zeydanlıoğlu, this issue). Naturally no school support to Kurdish exists. Yet, within the frame of reforms to harmonize with European Union membership conditions, in August 2003, the law assured the right of private learning of “regional languages”. Following this reform, seven private courses for teaching Kurdish were opened in Istanbul and in some big cities in Kurdish region. However, because of a series of political, legal and economic con-cerns, these courses were not able to survive and in August 2005, the directors of the courses announced the closure of their establishments. This experience may seem as a defeat of private Kurdish teaching or as an indication of indifference of the speakers vis-à-vis the study of their language, yet it is equally possible to consider, as Haig (2004: 140) does, the measures imposed as attempts of a total exclusion of Kurdish from the educational system. The courses nevertheless dis-tributed certificates to 1,179 language learners, and 1,780 learners were registered to their programs when they were closed down (Akin 2007: 35).

Being excluded from education and teaching domains, Kurdish has been taught illegally since the mid 1990s by Kurdish cultural organizations in big cities. For instance, NÇM (Navenda Çanda Mezopotamya-Mesopotamia Cultural Center), a Kurdish cultural center founded in 1991 in Istanbul, has held many Kurdish courses. Moreover, the Kurdish Institute of Istanbul, founded in 1992, has constantly organized Kurdish courses to create a potential cadre of lan-guage teachers. The director of the institute declared in 2009 that there were about 4,000 learners following informal Kurdish courses5 offered by their insti-tute and TZPKurdi,6 a civil movement for Kurdish linguistic and cultural activism

5 Data retrieved from the Istanbul Kurdish Institute’s website: http://www.enstituyakurdi.org/modules.php?name=News&file=article&sid=192 (accessed 12 June 2009).6 The organization directs language and literacy teaching in Kurdish under local language associations called Kurdi-Der in a number of Kurdish towns. In 2010, it started to organize an annual “school boycott”, during the first week of the school year. The boycott condemning the



http://www.enstituyakurdi.org/modules.php?name=News&file=article&sid=192

http://www.enstituyakurdi.org/modules.php?name=News&file=article&sid=192

162 E. Öpengin

established in 2006. Despite the presence of certain activism in the domain of Kurdish teaching, it is worth reminding that most of these activities take place on a voluntary basis and that a large part of the instructors and learners spare their spare time to attend these courses. Moreover, the instructors are not formally edu-cated in teaching Kurdish; thus, they rely mostly on their own personal compe-tence in Kurdish and transpose their knowledge in their respective domains to the teaching of Kurdish. The shortage of teaching material for Kurdish is yet another obstacle. Lastly, these teaching activities do not address young learners; in gen-eral the ultimate aim is to teach literacy in Kurdish.7 It is important to keep in mind that these difficulties mostly stem from oppressive state policies and heavy consequences of using Kurdish at individual level and on daily basis (see Skutnabb-Kangas and Fernandes 2008).

3.4 The politicization of the language: double-faceted folklorization

The state politics have been investigated elsewhere (see the References). Here I shall discuss the politicization of language and its consequences within the Kurdish community. Repressive state politics and denigratory representations of Kurdish language and community by Turkish academia (see Akin 1995; Haig 2004) and media (see Erdoğan 2002) have created negative attitudes and percep-tions among the mainstream community vis-à-vis the Kurdish language. A less known manifestation of this is in the perceptions of Turkish speakers of dialects. Demirci and Kleiner (1999: 267) show that Turkish speakers from Bursa found the Turkish spoken in Eastern regions (i.e. mainly by Kurds) the least correct and least pleasing dialect of Turkish. The respondents hold the most negative percep-tions on the dialects and people of this region (i.e. they are “harsh”, “backward” and “illiterate”). Besides, the authors indicate that the negative perceptions must have been influenced by the mother tongue of the people from this region, i.e. mainly Kurdish and Arabic (Demirci and Kleiner 1999: 272).

ban on education in Kurdish at schools was widely supported in its first edition in 2010 and created debates in the public sphere, however for the following year both the participation to the event and its influence on the public debates diminished substantially.See: http://www.rudaw.net/english/news/turkey/3168.html (accessed 11 October 2011).See also: http://www.hurriyetdailynews.com/default.aspx?pageid=438&n=boycott-receives-partial-participation-in-eastern-turkey-2010-09-20 (accessed 11 October 2011).7 Information on TZPKurdi is mostly drawn from my interviews with language activists in the field, in Diyarbekir, in March 2009.



http://www.hurriyetdailynews.com/default.aspx?pageid=438&n=boycott-receives-partial-participation-in-eastern-turkey-2010-09-20

http://www.hurriyetdailynews.com/default.aspx?pageid=438&n=boycott-receives-partial-participation-in-eastern-turkey-2010-09-20


An increasing political consciousness among Kurds since the 1960s, parallel to the urbanization of large blocks of Kurdish populations, has added an inte-grative dimension to Kurdish. Thus, in the absence of capital value of Kurdish in urban contexts, the transfer of language to new generations was also moti-vated by identity-related concerns. Only recently, since the 1980s, mainstream Kurdish politics has introduced the language and a discourse on linguistic rights in its intra-community propaganda and its political agenda addressing the government. The active usage of Kurdish was, however, hardly ever promoted and it was mostly excluded from the activities of Kurdish political circles (Haig 2004). This fact of instrumentalizing the language for extra-linguistic projects supports Fishman’s (1977: 25) claim that language can serve as a departure point for activation and celebration of and call for any ethnic activity. However, the process may lead to a kind of folklorization of Kurdish, paradoxically led by Kurd-ish political actors themselves.8 Hence it may function as yet another means to create negative perceptions of Kurdish and eventually contribute to the language shift.

Since the second half of the 1990s, the increasing visibility of Kurdish in the public sphere and in political domains has added to the development of positive perceptions regarding Kurdish among its speakers and other components of soci-ety in Turkey. At this stage, Kurdish political movements primed a political dis-course around cultural rights. Kurdish became a genuine issue for diverse interest groups. Kurdish writers explicitly criticized Kurdish political circles for not using and not promoting Kurdish while, at the same time, within Kurdish society the slogans such as Zimanê me rûmeta me ye [Our language is our honor] or Dilimiz kimligimizdir [Our language is our identity] were frequently invoked. However, the problematization of the language as the principal index of identity in itself may not necessarily render a direct increase in its actual usage (Fishman 1992: 401; Fishman 1999; Scalbert-Yücel 2007). An associated language9 can be limited to the identification and political argumentation dimensions without bearing a real impact on usage. Nevertheless, the outcomes of this activism and rise in lin-guistic consciousness have resulted, among others, in the multiplication of lan-guage associations and the establishment of an organization of civil activism for the promotion of language (see Note 6). Furthermore, since the 1999 elections, pro-Kurdish politics has showed a greater presence in the domain of Kurdish cul-tural activities. After the 2009 elections, 100 municipalities in the Kurdish region

8 According to Haig (2004: Note 37), even “Öcalan [the leader of PKK] considered traditional Kurdish culture to be backward and the language, in its present state, inadequate.”9 A language that is always associated to an ethnic identity (Eastman and Reese [1981], cited in Scalbert-Yücel [2006: 134]).



164 E. Öpengin

are led by elected officials from the pro-Kurdish political party, creating a “special status” for the region as they are, in a way, a local Kurdish power. Although sys-tematically hindered and penalized (see Zeydanlıoğlu, this issue), the contribu-tion of these local powers has been manifold; the most important practical ones include the annual Diyarbekir Arts and Culture Festival, which serves as a venue for the presentation of many kind of Kurdish cultural productions, and the an-nual “Diyarbekir Literature Days,” which also gathers a two-day conference for specialists and language activists to discuss the actual problems of the language. Moreover, a number of other cultural festivals and dozens of books, especially addressing the younger readership, and proceedings published by municipali-ties, are increasingly important contributions to the language planning of Kurd-ish in Turkey.

As for those Kurds not engaged in any cultural or political Kurdish activism, the language seems to be reconciling and even imposing itself as an indispensible component of their Kurdish identity thanks to official overtures on Kurdish lan-guage and culture. This case is obviously parallel to Wurm’s argument (2002, cited in Grenoble and Whaley [2006: 27]) that a change from negative to more positive attitudes and policies at the national level can result in positive change to the vitality of local languages.

4 Survey on language use and choice, language proficiency and perceptions

The overview of the objective aspects of the language situation of Kurdish in Tur-key shows that the vitality of the language is heavily weakened by factors such as historical low language status of Kurdish and century-long measures against its public manifestations. However, the language, as the main component of cultural rights discourse, is now going through a process of valorization. Apart from being established as the main index of ethnic identity within the community, efforts of language revitalization and de facto recognition as well as the application of some cultural rights have positively influenced the position of Kurdish and the perceptions of it both within and outside the community. Yet, the brief evaluation says little on the micro-sociolinguistics of the situation, i.e. individual speakers’ and speech community’s relationship with the languages in their repertoire. Given that institutional language policies have waged a relentless campaign of Turkification (see also Üngör, Zeydanlıoğlu, both this issue), intensified by so-cial changes within Kurdish community, one would not expect high rates for maintenance-oriented language use patterns and linguistic proficiency and per-




ceptions. These untouched issues in the literature are addressed in this empirical study.

4.1 Instrument and informants

The data on speakers’ language use and choice and linguistic proficiency and perceptions were collected in a questionnaire-based survey, with theoretical underpinnings in Fishman’s concepts of diglossia and domain analysis (2003 [1967], 1972 [1968], 1991). Bourhis et al.’s (1981) subjective ethnolinguistic vitality is used to devise some ad hoc questions addressing the perceptions of the speak-ers on important recent developments.

The survey, restricted to the Kurdish region in Turkey, was conducted in three contexts: in Diyarbekir (urban); Şemzinan county of Hakkari (semi-urban); and three villages of Şemzinan (rural). The questionnaire includes 79 items, grouped in 4 sections: (a) background information on the informant and his/her family; (b) language use and choice with respect to interlocutor, topic, communicative setting, media; (c) speakers’ perceptions of recent developments; (d) speakers’ self-evaluation of their proficiency in Kurdish and Turkish (speaking, under-standing, reading and writing). The questionnaire was administered in Turkish (but items were mostly orally presented in Kurdish) to a sample of 76 speakers from 18 families. The variables such as setting, gender, and partly age and educa-tion level were controlled.

The data are analyzed in the form of frequency tables and graphics to identify the general tendencies. Chi-square tests are applied to the data to explore the nature of the relationship between identified variables of the study. However, for space concerns, only succinct descriptive statistics are presented, along with summaries of important alignments. Given the limited coverage of the survey, the summarized linguistic tendencies are meant to be indicative, rather than repre-sentative of the general language situation in the region.

4.2 Findings and analysis

4.2.1 Language use in respect to the identity of the interlocutor

The language use varies drastically according to the interlocutor, yet the exclu-sive usage of Kurdish is most frequent in the communication between grandpar-ents, parents and children. Figure 2 indicates that 95% of the respondents speak exclusively in Kurdish to grandparents. Kurdish is also the main medium of



166 E. Öpengin

Fig.

2: L

angu

age

use

acco

rdin

g to

inte

rlocu

tor (

in %

). (T

r = Tu

rkis

h; K

r = K

urdi

sh)




addressing to the mother, with 70% of “only in Kurdish” and 20% “mostly in Kurdish”, while 40% of respondents use Turkish in different degrees in their con-versations with their father. Kurdish seems to be the principal language of par-ents to raise their children: only 10% of the parents say they use Turkish more than or as much as Kurdish. However, only 39% of the respondents speak to their siblings only in Kurdish; communication between siblings favors the introduc-tion and installation of Turkish in the home domain.

Turkish is used along with Kurdish more frequently in out-of-family interac-tions. Around 50% of the communication with friends, acquaintances, neigh-bors, and relatives takes place in two languages. Turkish is especially important in conversations with friends and siblings, respectively 38% and 34%, compared to visibly higher rates of Kurdish use, respectively 55% and 66%.

Kurdish is used in a much higher degree within the immediate social environ-ment and with relatively older people whereas the usage of Turkish is reinforced in out-of-family interactions. The increasing function of Turkish in communica-tion among friends implies that Turkish has partly become the language of out-of-house socialization. It should also be noted that the use of Kurdish increases when the parents, especially the mother, speak to the children. A decline of 10% is observed when the children address their parents. It is probably caused by a code-switching by children when they speak about issues that they are more at ease when they speak in Turkish, such as school. However, the discrepancy is significant for it implies the beginning of a transformation in intergenerational communication patterns. At the same time it points to a generational and gender-related tendency in respect to language use: among older generations, women are more loyal to Kurdish than men.

4.2.2 Topic dependent language use

Although Kurdish is present in all sorts of discussion topics, it is especially widely used when it comes to daily and cultural issues. Turkish is rather pres-ent in social (40%) and, to some extent, religious topics (28%). As the modern terminology of politics, sports, education, etc. is usually introduced in Turkish via schooling and TV programs, an important part of speakers use Turkish on these themes, whereas the availability of the terminology for daily, cultural and religious issues promotes the usage of Kurdish. Furthermore, relatively impor-tant presence of Turkish in religious topics must be indicative of a transforma-tion in the relationship of language with religion rendered by the spread of urban conceptions of religion in the community. Lastly, although Kurdish is dom-inant in all topics, the usage of Turkish, especially alongside Kurdish, points that



168 E. Öpengin

functions of languages in contact are not strictly separated across discursive themes.

4.2.3 Language use across social domains

Among the social domains of language use (cf. Figure 3), the home remains the domain where the usage of Kurdish is the highest (70%) and the exclusive usage of Turkish the lowest (5%). Although remarkably less frequently used, Turkish is present for 65% of the informants in this intimate domain. The usage of Turkish probably takes place mostly among siblings.

The usage rates of Kurdish systematically decline in social domains such as neighborhood (60%), workplace (50%), and marketplace (48%). A bilingual is usually not expected to stick to the exclusive usage of one language, but in a sta-ble diglossic situation language use is supposed to be mostly determined by the given domain. Thus, low exclusive usage of Kurdish in neighborhood (22%) may mean that Turkish has exerted its presence, though in relatively lower degrees, in 78% of the communication taking place in a domain traditionally associated to the usage of Kurdish. Kurdish is slightly more prevalent than Turkish in the mar-ket, nevertheless the default language of starting a conversation in the market-place must be mostly Turkish. School with 90% and public institutions with around 70% of Turkish usage mostly exclude Kurdish. Some 15% to 20% of com-munication in these two domains is held in Kurdish. It corresponds mostly to in-formal correspondences, but at the same time it points to a violation of the func-

Fig. 3: Language use across social domains (%). (Tr = Turkish; Kr = Kurdish)




tional compartmentalization of languages in contact for, in this case, the low language Kurdish starts to make its presence felt in domains traditionally associ-ated to the usage of high language Turkish.

The general tendency is to stick to the functional separation of the languages. This is reflected in the fact that the high language Turkish is more widely used in formal domains while the low language Kurdish is preferred rather in intimate domains. This is mostly in line with Fishman’s (2003 [1967]) interpretation of di-glossia and domain analysis. However, the alternate usage of both languages is also widespread in low domains while a certain usage of the low language is seen in high domains. These facts may indicate that Kurdish follows the path of a lan-guage shift, but they also show that the unstable linguistic context allows Kurd-ish to exert some presence in high domains reinforcing the tendency to reverse the shift.

4.2.4 Language choice in media

The speakers show a mixed profile when it comes to the language in which they watch TV programs. 92% of the speakers watch TV programs in both languages, with a certain dominance of Turkish. TV and its principal function of dissemi-nating popular culture must be a significant impetus in the introduction and in-stallation of the usage of Turkish in the most intimate domain of language use. Yet, television in Kurdish is a relatively recent phenomenon. In this sense, the remarkable share of Kurdish in TV programs indicates that Kurdish has appropri-ated an important aspect of modern culture and widened its usage into a new domain.

More than half of the speakers declare that they never watch TRT6, the public TV station in Kurdish launched by the government in January 2009. Most of the informants hold a negative approach to it on such grounds that the government wants to make use of it to weaken Kurdish politics, that the channel does not use Kurdish properly, that there is too much state intervention in it. Opening TRT6 has not motivated the informants to watch more TV programs in Kurdish, yet it has substantially influenced the perceptions of the speakers on Kurdish: some 37% think that it will contribute substantially to the development of Kurdish lan-guage and culture; more than half believe that the prestige of Kurdish in the pub-lic domain has improved, that Kurdish will be better accepted in Turkish society and that Kurdish will be better transferred to the children. The ideas about and reactions to the opening of TRT6 show that perceptions of evolutions around Kurdish language and culture are highly politicized and shaped by speakers’ pol-itical affiliations and convictions.



170 E. Öpengin

Listening to radio turned out to be a negligible variable, as more than half of the informants never listen to the radio at all. Most informants prefer listen-ing to music in Kurdish, with more than half sometimes listening to music in Turkish as well. Kurdish appears to be strongly present in language choice in music. This is expected since music remains one of the most obvious fields in which Kurdish cultural and political identification is exercised and enjoyed (see Kanakis 2006).

4.2.5 Language of reading in the absence of literacy in Kurdish

Some 12% of the sample is orate (see Skutnabb-Kangas and McCarty 2008: 11) while 12% is semi-literate in Turkish. The percentages increase to 48% orate and 32% semi-literate in Kurdish. Thus, only 4 informants read newspapers also in Kurdish while 31 informants (49% of the sample) read in Turkish. Kurdish has only a symbolic usage in reading books. The predominance of Turkish in literacy and reading is explicit, reinforcing the status of Kurdish as the language of oral interactions. The scarcity of journals and books in Kurdish and their limited avail-ability have a restrictive influence on the development and spread of literacy in Kurdish. There is a parallel between the sociopolitical conditions and speakers’ language use and choice in literacy-related activities since such activities in Kurd-ish are hardly ever “rewarded” (see Fishman 1980). Hence, the absence of literacy in Kurdish caused by the ban on education and teaching of Kurdish determines from the outset that Turkish will be the preferred language over Kurdish when it comes to all forms of written activity.

4.2.6 Language proficiency

Language proficiency is shaped by a clear difference in written and oral language abilities. The speakers usually declare that they are competent in understanding and speaking rather than in writing and reading in both languages. However, while almost 80% of the speakers claim language proficiency sufficient to take up a conversation in Kurdish, only about 20% declare a relative ability in reading and writing in Kurdish. The tendency to demonstrate higher levels of proficiency in oral competences is valid also in Turkish, yet different from Kurdish, high levels of written language competences are claimed in Turkish. Only about 20% declare that they do not possess sufficient competence to conduct written activity in Turkish.




The comparison of declared relative overall language proficiency10 in Kurdish and Turkish shows that 30% of the speakers claim a better mastery of Kurdish than Turkish, whereas 10% claim a better proficiency in Turkish. Around 51% of the speakers think that they have a better command in Kurdish while 31% are bet-ter in Turkish. 18% do not see any difference between their competences in the two languages. One can infer from this comparison that the majority does not perceive a seriously low level of overall language proficiency in Kurdish. It also shows that bilingual language proficiencies are quite heterogeneous.

4.2.7 Patterns of linguistic behavior

The general tendencies and respective importance of the languages were de-scribed across a number of variables above. Here an effort is made to see whether the heterogeneity of language practices and competences has a significant rela-tionship with variables such as context of living/socialization, gender, genera-tion and education. A correlation is considered to be significant if its chi-square value is below 0.05 (i.e. p < 0.05). Once the chi-square test is applied, the observed and expected frequencies are compared and interpreted.11 Note that only some of the descriptive statistics are presented here.

Context as a determinant in language maintenance. No significant relations are revealed between the context and the language in which the informants ad-dress their mother (p = 0.33), their father (p = 0.58) or their friends (p = 0.49). These linguistic practices are thus not correlated with the context. Yet, the con-text is significantly related to the language in which the informants speak to their siblings (p = 0.038). The comparison of observed and expected frequencies shows that the semi-urban context is where the usage of Kurdish among siblings is the lowest. Turkish is more present among siblings in urban than rural contexts.

The context is especially determinant in language use across domains, with significant relations with home (p = 0.0009) (see Table 1), neighborhood (p = 0.05), and the marketplace (p = 0.02).

10 The “overall language proficiency” here refers to the linguistic proficiency that the speakers “declare” to possess. In the questionnaire, the informants were asked to evaluate and compare their own “general knowledge and ability” in Kurdish and Turkish. The methodological concern behind this was to consider speakers’ perspective by leaving it to the speakers to decide upon what counted as decisive component(s) of language proficiency.11 Note that “observed frequency” corresponds to the frequency of occurrence in the present survey whereas “expected frequency” is hypothetical and would occur if there were no influence of the independent variable on the compared dependent variable.



172 E. Öpengin

The semi-urban context differs from others by the quasi-absence of the exclu-sive usage of Kurdish in the home domain; the alternate use of the two languages at home is the most widespread here. The exclusive use of Kurdish has the highest rate in the urban context; yet compared to rural context, a slightly higher pres-ence of Turkish is seen in urban context. As for the neighborhood and the market-place, the urban context clearly differs from others by the strong presence of Turkish. The semi-urban context has more balanced language use in these do-mains while in the rural context Turkish is categorically excluded and Kurdish is prevalent.

This can be summarized as follows: Kurdish is the strongly dominant lan-guage of most of the domains in rural context; the semi-urban context remains one where the alternate usage of the two languages is widespread; and the urban context is where the exclusive usage of either one of the languages is most obvi-ous, with its distinguishing feature being the very low Turkish use at home and comparatively high use of Turkish in out-of-home social domains.

There is also a significant relation (p = 0.04) between context and linguistic proficiency in Kurdish. Rural context has the highest rates, relatively inferior pro-ficiency in urban context while the lowest rates are seen in semi-urban context. Language choice in TV programs and music is not significantly related to the con-text, which indicates that these instruments of popular culture homogenize cer-tain cultural practices across different social contexts.

Gender and language proficiency. Gender has turned out to hold signifi-cant relations with dependent variables such as relative proficiency in Kurdish (p = 0.018) and language of fluent and comfortable expression (p = 0.04). See Table 2.

Obviously the number of women who declare to have a better proficiency in Kurdish than in Turkish is much higher than men. Only half of men think that

Which language do you speak at home?

Observed frequency Expected frequency

A B C D E Total A B C D E Total

Rural 9 12 5 1 0 27 7.46 13.14 3.20 1.78 1.42 27Semi-urban 1 19 3 2 0 25 6.91 12.17 2.96 1.64 1.32 25Urban 11 6 1 2 4 24 6.63 11.68 2.84 1.58 1.26 24Total 21 37 9 5 4 76 21 37 9 5 4 76

* A = in Kurdish; B = mostly in Kurdish; C = in Kurdish-Turkish; D = mostly in Turkish; E = in Turkish; p = 0.0009: significant

Table 1: Language use in the home domain across contexts*




they have a better proficiency in Turkish. The same tendency is observed on the variable of language of more comfortable expression. However, this tendency does not hold among the new generation of women since 7 out of 9 women who consider that they have a better command and ease of expression in Turkish are informants younger than 20 years old. This result implies that Turkish also be-comes the language of close network communication among the women of the young generation. Furthermore, although Kurdish remains the language of lin-guistic security for a majority of women, it has clearly lost this communicative function for a good half of male speakers. Finally, gender did not show significant relationship with other variables such as language in which the informants speak to their neighbors (p = 0.079) or language choice while speaking of daily issues (p = 0.21).

Generation related language use and choice. The generation of the informants is clearly the most discriminatory factor, for almost all of the chi-square tests in which the generation is taken as independent variable turned out to hold signifi-cant relations. Hence the age of informants, categorized into three generations (i.e. <20, 20–40, >40), is significantly related to “the language spoken to siblings” (p = 0.00001), “the language spoken to friends” (p = 0.000), “the language choice in TV programs” (p = 0.00001) and finally to “the language proficiency in Kurdish (p = 0.000).

A comparison of actual values and expected values across three generations in Table 3 points to a familiar pattern of language shift across generations that can be summarized in three points: (1) a quasi-total exclusion of Turkish among speakers over 40 years; (2) prevalent alternate usage of the two languages with slightly higher rates for the usage of Kurdish among speakers of 20–40 years; (3) relatively higher usage of Turkish among speakers below 20 years. The usage of Kurdish in interactions with friends is radically weaker among the respondents below 20 years. This means that generational language shift is reinforced by the

How would you evaluate your relative proficiency in Kurdish compared to your proficiency in Turkish?



Men 1 4 8 7 16 36 0.95 9.95 6.63 7.58 10.89 36Women 1 17 6 9 7 40 1.05 11.05 7.37 8.42 12.11 40Total 2 21 14 16 23 76 2 21 14 16 23 76

* A = much weaker; B = weaker; C = equal; D = better; E = much better; p = 0.018: significant

Table 2: General proficiency of informants in Kurdish in respect to the gender*



174 E. Öpengin

restrictions on the usage of Kurdish in domains outside the home. As for media, all three generations watch TV programs in two languages. However, the young generation is distinguished by a general predominance of Turkish TV channels, the intermediary generation (20–40) has a more or less balanced alternate lan-guage choice with a certain weight of Kurdish, and finally, the informants from older generation rarely consult Turkish TV channels in Turkish. The same ten-dency of language shift in respect to the generation is observed also in the lan-guage proficiency of the informants: language proficiency in Kurdish is high among the speakers over 40 years, it is relatively weaker among speakers from 20 to 40 years, and the younger speakers have obviously higher proficiency ratings in Turkish than in Kurdish.

Education level and language practices. The education level of the informants also plays an important role, with many of the tested variables, such as “the lan-guage of more fluent expression” (p = 0.0004), “language use when speaking to friends” (p = 0.0006), “language use when discussing social topics” (p = 0.01). It can be inferred from Table 4 that informants with no formal education are exclu-sively fluent and comfortable in Kurdish but as the level of education increases, Turkish becomes the language of more fluent expression. Thus, informants who have completed high school or undergraduate levels are far more comfortable when expressing their thoughts in Turkish.

Informants who have not had formal education use mainly Kurdish in inter-actions with friends. As the level of formal education increases, the tendency shifts to use more and more Turkish. Yet, the tendency is counteracted by wider use of Turkish among the informants from the category of primary school. This is related also to the generation; since 6 out of the 8 informants from primary level education category who declare to have a better command in Turkish and speak to friends more in Turkish are from the younger generation (see Generation

In what language do you speak to your siblings?



<20 6 7 8 4 7 32 12.37 7.68 4.69 3.84 3.41 3220–40 7 10 3 5 1 26 10.05 6.24 3.81 3.12 2.77 26>40 16 1 0 0 0 17 6.57 4.08 2.49 2.04 1.81 17Total 29 18 11 9 8 75 29 18 11 9 8 75

* A = in Kurdish; B = mostly in Kurdish; C = Kurdish-Turkish; D = mostly in Turkish; E = in Turkish; p = 0.00001: significant

Table 3: Language choice when speaking to siblings across generations*




related language use and choice). I conclude that education and generation factors together make up a determinant dimension of language use and choice patterns of Kurdish speakers as indicators of relatively wider usage of and profi-ciency in Turkish.

5 Discussion and conclusionThe parallel analysis on the objective dimensions of minority status of Kurdish on one hand and daily linguistic practices and perceptions of speakers of Kurdish on the other hand points to a tight correspondence between sociopolitical con-ditions and the actual dealings of the community members with their heritage language in their in-group interactions. It was shown that several sociopolitical dimensions are apt to reinforce the ethnolinguistic vitality of the speech com-munity such as (1) the geographical position of the Kurdish minority as a trans-border cohesive linguistic continuum of the same speech community, (2) demo-graphical and historical majority status of the community in the region leading to a decisive association of language with the territory and reinforcing the dis-courses on cultural heritage preservation, (3) the works on the corpus planning of Kurdish that have assured partly-standardized and widely followed written Kurd-ish, albeit with its difficulties, (4) the wider liberalization and democratization processes in the last two decades that have fostered a communicative space with-in which the chances for a fair share in representation is higher (5) and finally, the decades-long cultural and political activism carried out by Kurdish people and organizations that has culminated in assuring a special status for the region where the relative and limited political autonomy could substantially contribute

In which language do you express your thoughts more comfortably


Education level Kr Tr Total Kr Tr Total

Uneducated 17 0 17 10.74 6.26 17Primary 17 8 25 15.79 9.21 25Secondary 5 3 8 5.05 2.95 8High-school 7 10 17 10.74 6.26 17Undergraduate 2 7 9 5.68 3.32 9Total 48 28 79 48 28 76

* Kr = Kurdish; Tr = Turkish; p = 0.0004598: significant

Table 4: The language of better expression across formal education levels*



176 E. Öpengin

to the language planning of Kurdish. On the other hand, there are those socio-political aspects that have led and continue to lead to the subordination of Kurd-ish to Turkish such as (1) the largely oral status of Kurdish throughout previous centuries, (2) the strict measures against its public usage and its absolute and continuing exclusion from the educational system for almost a century, (3) the changes introduced by urbanization among the Kurds leading to radical transfor-mations in societal norms, (4) and the unfavorable position of Kurdish in the lin-guistic market resulting in the almost total invisibilization of its written usage while at the same time diminishing substantially its presence in the marketplace and in the domain of economy in general.

As for the patterns of language practices and language perceptions of speak-ers, in parallel with the above points, it is shown that Kurdish is no longer the default language of communication for all of its speakers: the younger the speak-ers are, and the more formally educated and out of the immediate social net-works, the less Kurdish they use. The setting, rural vs. urban contexts, is also shown to be significant, for the higher use and proficiency rates in rural contexts decrease among urban populations, pointing to a more advanced and rapid pro-cess of language shift in the urban context. As for gender, while use of and profi-ciency in Kurdish is clearly higher among women, this tendency does not hold among younger generations of women speakers. It is further shown that the per-ceptions of speakers on recent developments relating to linguistic and cultural rights are mostly shaped in line with political tendencies. Sociopolitical condi-tions are further reflected in the symbolic rates of literacy and written activity in Kurdish, reaffirming the principally oral status of Kurdish while at the same time consolidating the role of Turkish as the language of written activity.

It is attested that, in line with the objective conditions of the language situa-tion, there is a tendency among the majority of Kurdish speakers to stick to func-tional separation of languages in respect to communicative settings and domains. However, the situation is not an example for stable diglossia because the spread and consolidation of Turkish in low domains and, in return, the robust degree of Kurdish-language consumption in high domains like media, and its emerging presence in institutions are not conform to the functional compartmentalization of languages. In this sense, the generational shift observed as the prevalent ten-dency in this study is also counteracted by new and urban modes of language maintenance and linguistic perceptions. It is evident that the language situation does not fit into a “harmonizing” conception of diglossia nor can it be seen as a simple decline along generations, since the very specific dynamics of language shift and attitudinal change are indeed apt to end up with a relative reversal of the ongoing shift and to provide further circumstances for the development of the language. This is indeed manifest in growing importance of Kurdish as the princi-




pal index of identity and its emergence in domains and communicative settings usually ascribed to the use of Turkish. Nevertheless, the state of affairs of the community-language relationship do point to a non-stable linguistic context in which the dominant language, Turkish, has imposed a considerable presence in the domains and communication situations traditionally associated with the us-age of Kurdish. In this, the situation of Kurdish in Turkey can be concluded as an example of a prevalent societal bilingualism without a stable diglossic functional separation of languages. It is also an example of a continuing process of language shift in which the prevalent community bilingualism evolves both to the detri-ment of Kurdish and to its further valorization in the public sphere. Further re-search that relies on a more representative corpus and that accounts for factors such as intra- and inter-regional migration may better grasp the dynamic nature of the language situation and help to conceptualize the rapid evolutions in the field.

AcknowledgementI would like to express my gratitude to Jaffer Sheyholislami and Tove Skuttnab-Kangas for their rich feedback and critical comments on this paper. I would also like to thank Kelda Jamison for her meticulous reading of the paper and helpful suggestions.

ReferencesAkin, Salih. 1995. Désignation du peuple, du territoire et de la langue kurdes dans le discours

scientifique et politique turc. Rouen: Rouen University dissertation.Akin, Salih. 2007. Fêrkariya kurdî ya taybet li Tirkiyeyê: Pirsgirêk û pêşbînî [The private teaching

of Kurdish in Turkey: problems and perspectives]. Zend 5. 30–38.Aydogan, Ibrahim S. 2006. Indentification des temps verbaux dans les grammaires kurdes.

Paper presented at the World Congress on Kurdish Studies, Salahaddin University, 6–9 September.

Aydogan, Ibrahim S. 2007. Kurmanciya malperên kurdî û zimanê rojnamegeriyê [The Kurdish used in Kurdish websites and journalistic language]. Çirûsk 1. 51–58.

Bourhis, R. Y., H. Giles & D. Rosenthal. 1981. Notes on the construction of a “subjective vitality questionnaire” for ethnolinguistic groups. Journal of Multilingual and Multicultural Development 2(2). 145–155.

Boyer, Henri. 1997. Conflit d’usages, conflit d’images. In Henri Boyer (ed.), Plurilinguisme: «contact» ou «conflit» de langues?, 9–35. Paris: L’Harmattan.

Bozarslan, Hamit. 2009. Le conflit kurde. Paris: Edition Autrement.



178 E. Öpengin

Bruinessen, Martin van. 1988. Les Kurdes et leur langue au XVIIe siècle: Notes d’Evliya Çelebi sur les dialectes kurdes. Studia Kurdica 1(5). 13–34.

Bruinessen, Martin van. 1989. Geleneksel Kürt toplumu ve devlet. [Traditional Kurdish society and the state]. In Uluslarası Paris Kürt konferansı 14–15 Ekim 1989 [International Paris Kurdish conference], 39–50. Istanbul: Doz Yayınları.

Calvet, Jean L. 1994. Les Voix de la ville, sociolinguistique urbaine. Paris: Payot.Celîl, Celîlê. 2002. 13 rûpelên balkêş ji dîroka gelê Kurd [13 important chapters on Kurdish

history]. Wien: Înstîtûta Kurdzaniyê.Clyne, Michael. 2003. Dynamics of language contact: English and immigrant languages.

Cambridge: Cambridge University Press.Coşkun, Vahap, Şerif Derince & Nesrin Uçarlar. 2011. Scar of tongue: consequences of the ban

on mother tongue in education and the experiences of Kurdish students in Turkey. Diyarbakır: Diyarbakır Sosyal Siyasal Araştırmalar Enstitüsü.

Demirci, Mahide & Brian Kleiner. 1999. The perception of Turkish dialects. In Dennis Preston (ed.), Handbook of perceptual dialectology, Vol. 1, 263–281. Philadelphia: John Benjamins.

Edwards, Bela B. 1851. Note on the Kurdish language. Journal of The American Oriental Society 2. 120–123.

Edwards, John. 1992. Sociopolitical aspects of language maintenance and loss: towards a typology of minority language situations. In Willem Fase, Koen Jaspert & Sjaak Kroon (eds.), Maintenance and loss of minority languages, 37–54. Philadelphia: John Benjamins.

Edwards, John. 1994. Multilingualism. London: Routledge.Erdoğan, Barış. 2002. L’Etat, la presse et la violence déployée contre les Kurdes de Turquie.

Journal of Kurdish Studies 4. 49–56.Erdost, Muzaffer I. 1986. Şemdinli röportajı [Şemdinli reportage]. Ankara: Onur Yayınları.European Commission. 2004. Regular report on Turkey’s progress towards accession.

http://ec.europa.eu/enlargement/archives/pdf/key_documents/2004/rr_tr_2004_en.pdf (accessed 5 February 2012).

Fase, Willem, Koen Jaspaert & Sjaak Kroon (eds.). 1992. Maintenance and loss of minority languages. Philadelphia: John Benjamins.

Ferguson, Charles A. 2003 [1959]. Diglossia. In Christina B. Paulston & G. Richard Tucker (eds.), Essential readings in Sociolinguistics, 345–357. Malden: Blackwell.

Fishman, Joshua A. 1972 [1968]. The relationship between micro- and macro-sociolinguistics in the study of who speaks what language to whom and when. In Anwar S. Dil (ed.), Language in sociocultural change: essays by Joshua A. Fishman, 244–268. Stanford: Stanford University Press.

Fishman, Joshua A. 1977. Language and ethnicity. In Howard Giles (ed.), Language, ethnicity and intergroup relations, 15–57. London: Academic Press.

Fishman, Joshua A. 1980. Language maintenance and ethnic mother tongue school. The Modern Language Journal 64(2). 167–172.

Fishman, Joshua A. 1991. Reversing language shift. Clevedon: Multilingual Matters.Fishman, Joshua A. 1992. Conference summary. In Willem Fase, Jaspaert Koen & Kroon Sjaak

(eds.), Maintenance and loss of minority languages, 395–402. Philadelphia: John Benjamins.

Fishman, Joshua A. 2003 [1967]. Bilingualism with and without diglossia, diglossia with and without bilingualism. In Christina B. Paulston & G. Richard Tucker (eds.), Essential readings in sociolinguistics, 359–366. Malden: Blackwell.




Fishman, Joshua A. (ed.). 1999. Handbook of language and ethnicity. New York: Oxford University Press.

Gardner, R. 1983. Learning another language: a true social psychological experiment. Journal of Language and Social Psychology 2. 219–240.

Giles, H., R. Y. Bourhis & D. M. Taylor. 1977. Towards a theory of language in ethnic group relations. In Howard Giles (ed.), Language, ethnicity and intergroup relations, 307–348. London: Academic Press.

Göç-Der. 2001. Göç araştırma raporu [The research report on immigration]. http://www.diyarbakirgocder.org/icerik_listele.php?liste_id=107 (accessed 25 May 2009).

Grenoble, Lenore A. & Lindsay J. Whaley. 1998. Toward a typology of language endangerment. In A. Lenore Grenoble & Lindsay J. Whaley (eds.), Endangered languages: current issues and future prospects, 22–54. Cambridge: Cambridge University Press.

Grenoble, Lenore A. & Lindsay J. Whaley. 2006. Saving languages: an introduction to language revitalization. New York: Cambridge University Press.

Güçin, Ergin & Ergin Öpengin. 2008. Les marqeurs de l’alternance codique dans la discours pragmatique en kurde. Etudes Kurdes 9. 47–64.

Haig, Geoffrey. 2004. The invisibilisation of Kurdish: the other side of language planning in Turkey. In Stefan Concerman & Geoffrey Haig (eds.), Die Kurden: Studein zu ihrer Sprache, Geschichte und Kultur, 121–150. Schenefeld: EB-Verlag.

Hassanpour, Amir. 1992. Nationalism and language in Kurdistan 1918–1985. San Francisco: Mellen Research University Press.

Hassanpour, Amir. 1998. Satellite footprints as national borders: MED-TV and the extraterritoriality of state sovereignty. Journal of Muslim Minority Affairs 18(1). 53–72.

Jafar, R. M. 1976. Under-underdevelopment: a regional case study of the Kurdish area in Turkey. Helsinki: Social Policy Association in Finland.

Jardel, Jean-Pierre. 1982. Le concept de « Diglossie » de Psichari à Ferguson. Lengas 11. 5–15.

Kanakis, Yiannis. 2006. Speak low, sing loud: spoken and sung “discourses of Kurdishness” in Hakkari. Paper presented at the World Congress on Kurdish Studies, Salahaddin University, 6–9 September.

Khani, Ahmed. 2010. Mem and Zin. (Trans. Salah Sadallah). Place of publication and publisher not given.

Kloss, Heinz. 1966. German-American language maintenance efforts. In Joshua A. Fishman, Vladimir C. Nahirny, John E. Hofman & Robert G. Hayden (eds.), Language loyalty in the United States, 206–252. The Hague: Mouton.

Kremnitz, Georg. 1991. Y-a-t-il des diglossie neutres? Lengas 30. 29–36.Landry, Rodrigue & Réal Allard. 1994. Diglossia, ethnolinguistic vitality and language behavior.

International Journal of Sociology of Language 108. 15–42.Malmisanij. 2006. The past and the present of book publishing in Kurdish language in Turkey.

http://www.npage.org/IMG/pdf/Turkey.pdf (accessed 10 May 2009).Matras, Yaron & Gertrud Reershemius. 1991. Standardisation beyond the state: the cases of

Yiddish, Kurdish and Romani. In Utta von Gleich & Ekkehard Wolf (eds.), Standardization of national languages, 103–123. Unesco Institute of Education Reports 5.

Myers-Scotton, Carol. 1992. Codeswitching as a mechanism of deep borrowing, language shift and language death. In Matthias Brenzinger (ed.), Language death in East Africa, 31–58. Berlin & New York: Mouton de Gruyter.



180 E. Öpengin

O’Shea, Maria T. 2004. Trapped between the map and the reality: geography and perceptions of Kurdistan. New York & London: Routledge.

Öpengin, Ergin. 2009. La situation sociolinguistique du kurde en Turquie: Une étude descriptive et interprétative. Rouen: Rouen University MA thesis.

Rich, James C. 1836. Narrative of a residence in Koordistan, Vol. 1. London: James Duncan.Scalbert-Yücel, Clement. 2006. Les langues des Kurdes de Turquie: La nécessité de repenser

l’expression «langue kurde». Langage et Société 117. 117–140.Scalbert-Yücel, Clement. 2007. La diaspora kurde en Suède: Conservation, production et

diffusion d’un savoir linguistique. European Journal of Turkish Studies. http://www.ejts.org/document771.html (accessed 1 June 2009).



Smits, Jeroen & Ayşe Gündüz-Hoşgör. 2003. Linguistic capital: language as a socio-economic resource among Kurdish and Arabic women in Turkey. Ethnic and Racial Studies 26(5). 829–853.


Skutnabb-Kangas, Tove & Teresa McCarty. 2008. Clarification, ideological/epistemological underpinnings and implications of some concepts in bilingual education. In Jim Cummins & Nancy H. Hornberger (eds.), Encyclopedia of language and education, 2nd edn., Vol. 5, 3–17. New York: Springer.

Tan, Sami. 2008. Danûstandina zimanan [The interaction between languages]. Zend 8. 87–93.TESEV. 2006. Dogu ve Güneydogu’da sosyal ve ekonomik öncelikler [Social and economic

priorities in Eastern and Southeastern Anatolia]. Istanbul: TESEV Yayınları. http://www.tesev.org.tr/Upload/Publication/d9f64ec6-3481-4f82-84b0-93b95dcdfd2a/guneydogu%5B1%5D.pdf (accessed 10 June 2009).

Winford, Donald. 2003. An introduction to contact linguistics. Oxford: Blackwell.Yağmur, Kutlay. 1997. First language attrition among Turkish speakers in Sydney. Tilburg:

Tilburg University Press.Yeğen, Mesut. 1999. Devlet söyleminde Kürt sorunu [The Kurdish question in state discourse].

Istanbul: Iletişim.Yildiz, Kerim. 2006. The Kurds in Syria: the forgotten people. London: Pluto Press.Zêrevan, Arif. 1997. Bingehên rastnivîsandina kurdiyê [The principles of Kurdish orthography].

Stockholm: Nefel.Zinar, Zeynelabidin. 1998. The medrese education in Kurdistan. Les Annales de l’Autre Islam 5.

39–58.



http://www.tesev.org.tr/Upload/Publication/d9f64ec6-3481-4f82-84b0-93b95dcdfd2a/guneydogu%5B1%5D.pdf

http://www.tesev.org.tr/Upload/Publication/d9f64ec6-3481-4f82-84b0-93b95dcdfd2a/guneydogu%5B1%5D.pdf

DOI 10.1515/ijsl-2012-0054 IJSL 2012; 217: 181 – 187

Tove Skutnabb-Kangas, Jaffer Sheyholislami and Amir Hassanpour

Concluding remarks*

Tove Skutnabb-Kangas: Åbo Akademi University. E-mail: [email protected] Sheyholislami: Carleton University. E-mail: [email protected] Hassanpour: University of Toronto. E-mail: [email protected]

The Kurdish linguistic landscape has changed greatly in this past two decades and it may change further by the time this issue of International Journal of the Sociology of Language appears – but the role of the Kurdish language (and other mother tongues in the areas) and mother-tongue-based multilingual education will remain central in building and developing societal peace. We will discuss some of these changes, and we will also make some recommendations for further research on the sociology of Kurdish.

The year 1991 is a significant turning point in Kurdish language studies. Tur-key allowed the Kurds to speak their language in the privacy of homes and in the street. In this year in Iraq, the control of the central government was removed from major parts of Iraqi Kurdistan and a Kurdish Regional Government was es-tablished. Another development was the fall of the Soviet Union, where Kurds enjoyed some support for the language and this support came to an end.

We have witnessed further changes in recent years that illustrate how ques-tions of language are intertwined with socio-cultural and political issues. In Tur-key, there has been a spread of publishing in Kurdish in spite of the serious loss of language, and the spread of broadcasting. We have witnessed the beginning of Kurdish language courses on the university level for the first time. However, the slow response of the government to the mass movement for language rights in Turkey has not been acceptable to the Kurdish population. The Civil Disobedi-ence Campaign in Turkey that “peace mothers” initiated in March 2011, with Democratic Solution Tents everywhere in Kurdistan-Turkey, has four demands:

* We would like to thank Dr. Margreet Dorleijn and Dr. Shelley Taylor for useful comments and ideas that we have used in our Concluding remarks.

Brought to you by | Dalhousie UniversityAuthenticated


182 T. Skutnabb-kangas et al.

1. Stop military and political operations immediately.2. Education in the mother tongue and providing constitutional guarantees for

using the mother tongue in the public sphere.1

3. Removing the 10% electoral threshold which hinders the representation of Kurdish people in the Assembly.

4. Releasing all political prisoners.

The campaign is the strongest one ever by Kurds; the Turkish police reactions likewise, with beatings, arrests and intimidations. However, resistance, including various kinds of mass demonstrations for democracy is spreading, not only in Turkey but in the whole Middle East.

As one can see from the demands above (and experience when attending conferences and rallies in Turkey, both in Kurdistan and in western Turkey, and following the many court cases), public use of Kurdish (including the right to de-fend oneself in Kurdish in courts) and education through the medium of Kurdish (and other mother tongues in Turkey) are as central as other political demands. If Turkey did today what the Ottoman Empire promised in 1920, even excluding the independence Article 64, many of the current constitutional and legal restrictions on the use of non-Turkish languages would disappear. Local autonomy, includ-ing financial and cultural autonomy, with minority protection, would also lead to solutions addressing the economic, human rights, educational, linguistic and re-search underdevelopment of Kurdistan.

We do not want to suggest that once Kurdish human rights, including linguis-tic rights and mother tongue-based education are, respected all the other socio-economic issues in the underdeveloped Kurdish regions including those related to language and education will immediately be solved. Nation-building and lan-guage planning are centuries-long and continuous processes. Since the Kurds started to manage their affairs in Iraqi Kurdistan in 1992 they have faced many challenges, from security, corruption, and fair parliamentary elections to educa-tion and language planning issues. Despite this, Iraqi Kurdistan became one of the rare polities in the region where what were thought to be distinct languages (e.g., Turkoman, Neo-Aramaic) started to enjoy language rights (Skutnabb-Kangas and Fernandes 2008). At the same time, there has been a continuous de-bate over which Kurdish variety should be the official and standard language in the quasi-independent Kurdistan Regional Government (KRG).

1 See also Coşkun et al. (2011), and ⟨http://khrp.org/khrp-news/human-rights-documents/briefing-papers/doc_download/286-culture-and-language-rights-mother-tongue-education-in-the-kurdish-regions.html⟩, Culture and language rights – mother-tongue education in the Kurdish regions.



http://khrp.org/khrp-news/human-rights-documents/briefing-papers/doc_download/286-culture-and-language-rights-mother-tongue-education-in-the-kurdish-regions.html



Concluding remarks 183

The Kurdish nationalist movement has been the major force in resisting lin-guicide. However, now that it has assumed power in Iraqi Kurdistan, it pursues an undemocratic approach to the conflict over the official, standard dialect and denies the bi-standard and multidialectal nature of the language. The nation-state ideology which, following the French Revolution motto, calls for the one nation = one language formula, has been pressing Kurdistan parliament to de-clare Sorani Kurdish as the sole standard and official language of Iraqi Kurdis-tan. Sorani speakers’ claim has been rejected by Kurmanji speakers who believe that their variety is also standard, and thus deserves the same status as Sorani Kurdish. In the meantime, since 2006, some Hewrami speakers have expressed their desire to have elementary education in their own variety rather than Sora-ni Kurdish alone. These issues remain unsolved. Recent remarks by a Kurdistan parliamentarian, Tarigh Jambas, at the Kurdish Academy in Hewlêr indicate that very little progress with respect to both status and corpus planning has been made in Kurdistan-Iraq during the past twenty years (Anonymous 2011). A considerable number of Kurdish lexicographers, educators, linguists, writers, translators and broadcasters have expressed their frustration with the way Kurdish is managed in Iraqi Kurdistan characterized by a salient absence of not only clear general language policies but also systematic and well-managed corpus or acquisition planning. In addition, there is an extremist purist policy which has impoverished the language; this is surely not a state policy although it is practiced on all official levels and predates the KRG; it began in the 1920s. In sum, the language planning efforts are inefficient and inadequate in Iraqi Kurdistan.

It is safe to suggest that Kurds in Turkey may have fewer status planning dif-ficulties in managing their choice of language. Iraqi Kurdistan has had a haphaz-ard language policy and planning since the establishment of the new Republic of Iraq in the early 1920s. In Turkey where Kurdish was officially banned until re-cently, the new language policy, including mother-tongue-based education when it comes, will not be built on equally shaky foundations. Moreover, Kurdish in Turkey is not as fragmented as it is in Iraq or even Iran; with the exception of about two million Zaza speakers, all northern Kurds speak Kurmanji. It might be easier to declare Kurmanji as the official or national language of Kurdistan-Turkey while granting linguistic rights to Zaza speakers.

Iranian Kurds may also be in a better position than Iraqi Kurds. If Iraqi Kurds have been heavily influenced by the linguistic properties of a very different lan-guage, Arabic, Iranian Kurds have been under the influence of Persian, the offi-cial language of Iran, which is linguistically and culturally much closer to Kurd-ish than are Arabic and Turkish. In Iraq, Syria and Turkey, a large segment of the Kurdish populations in urban centers have been alienated from their own




language not only at the phonological and lexico-grammatical levels but also at the levels of register, genre and pragmatics. Furthermore, although East or Ira-nian Kurdistan is home to all the major Kurdish varieties and dialects except Zaza, it might be easier to decide on a national and official language there (i.e, Sorani) than it has been in Iraqi Kurdistan. The reason is that except for some Hewrami and Kurmanji speakers most Kurds in Iran seem to have accepted Cen-tral or Sorani Kurdish as their common language already. Thus it might be easier to declare it as the official local language, should Iranian Kurdistan obtain polit-ical autonomy. This prospect, however, seems remote at the present time. In fact, activities connected to the maintenance and development of Kurdish in Iran have experienced a considerable setback in recent years. Whereas over two dozen Kurdish and Kurdish-Persian periodicals were published in the late 1990s in Iran, the only non-state-sponsored Kurdish periodical that has survived the state’s cen-sorship axe is the monthly periodical of Mahabad (J. H., personal communica-tion, May 3, 2011). Despite this and the non-implementation of Article 15 of the constitution, as an indication of linguicide or “let die” policy, publishing “non-political” books in Kurdish, albeit in very limited numbers, have continued. A good example is the three-volume dictionary published by University of Kurdis-tan, reviewed in this special issue.

In Syria the status of Kurdish has not changed and “[v]arious bans on the use of the Kurdish language” (Malmisanij 2006) continue. However, judging by the events reported from Syria in May and June 2011, the country appears to be in a socio-political transition which may bring about changes to the status of the Kurds in that country as well. Syrian Kurds are different from Kurds in Turkey, Iraq and Iran in that all of them are speakers of the Kurmanji variety. Therefore, they may not have to deal with the challenge of internal linguistic diversity at least in any near future. Nonetheless, any linguistic or educational transition (e.g., from Arabic only to bilingual, Kurdish and Arabic) will not be without its own chal-lenges.

Of course, inter- and intralinguistic diversity is not unique to Kurdistan. All nations are more or less diversified linguistically. These diversities are situated historically, territorially and socio-politically, which means that each community needs to find its own way to appreciate linguistic diversity through reasonable and sound language planning and policies. To do this in Kurdistan, much more work and research needs to be done. For example, some of the basic information needed for research into the language and education of the Kurds is missing, such as the number of Kurdish speakers in general and also the population of each dialect group: Kurmanji in Turkey, Iraq, Iran and Syria, or Hawrami in Iraq and Iran and so forth. Knowing the number of Hewrami speakers would, for instance, help Kurdistan authorities if they were to decide on Hewrami-based education in




Hewraman. Having more reliable statistics and numbers will also assist further research in the field.

Research is also needed in the area of corpus planning something not ad-dressed much in this issue. After Hassanpour’s (1992) book not much has been done about corpus planning for any of the dialect groups. It is no secret that for example in Iraqi Kurdistan a “poor” Kurdish is used and written in hundreds of television and radio stations, web features, and periodicals. Still, there has not been any noteworthy analysis of the existing corpus and acquisition planning in Iraqi Kurdistan. Further research into the complexity of language standardiza-tion, status, corpus and acquisition planning may raise more awareness among policy makers in Iraqi Kurdistan about the fact that remaining silent about the status of Kurdish varieties and the demands of some Hewrami speakers is not go-ing to make the language debate in that region wither away. The lack of clear language policies is not a solution to linguistic conflicts and challenges of lan-guage management (Fishman 2001). In fact, the absence of a clear policy could be even more oppressive in a multilingual society: it ignores the rights of linguistic minorities and subtly continues to naturalize and perpetuate monolingualism. As Fishman puts it, “even the much vaunted ‘no language policy’ of many democra-cies is, in reality, an anti-minority-language policy, because it delegitimizes such languages by studiously ignoring them, and thereby, not allowing them to be placed on the agenda of supportable general values” (Fishman 2001: 454). As “beloved languages” need protection and preservation, they also deserve stan-dardization, nationalization, etc., to be able to function in ways that modern life requires (Fishman 1997).

There is also need for research from a critical language policy perspective (Ricento 2007). This calls for research that focuses on language use (or lack of it) by all groups in a polity, ways of regulating and talking about language use (dis-course), and, more importantly, educational practices: media of instruction, the teaching of languages as subjects, and the medium which the students and edu-cators use orally. A more complete examination of the ecology of language must see languages within the contexts where they are banned, regulated, spoken, promoted, or taught. Language policy is not just practices carried out by policy makers from the top. It must include the speakers’ bottom-up activities and resis-tance; non-official actors are the ones who directly or indirectly contribute to what languages are, are perceived to be, and will become.

The papers in this issue have a strong focus on the suppression of Kurdish. Future research should highlight the achievements of Kurdish in relation to vital-ity. Further research into Kurdish needs to be carried out with the understanding that a new Kurdish language order has started to emerge. The opening of borders has resulted in extensive contact and exchange between Kurdistans of Turkey/




Iraq and Iran/Iraq. Disaporas used to be important as sanctuaries for publishing, but with the formation of KRG and the rise of social media they have lost much of their role even though they are still important to satellite broadcasting for all Kurds except for KRG. Satellite TV, mobile phones, and various social media are central to Kurds (and others) knowing about and being inspired by what is happening everywhere (including the millions of Kurds in diaspora in Australia, Canada, Europe, the USA, etc.; Sheyholislami [2011]). It is important and neces-sary to study what the social media and the Internet have done/are doing to the language in the context of continuing repression.

The papers in this special issue examine the social life of the Kurdish lan-guage, its troubled history, and the intricacies of undertaking research on Kurd-ish. These are viewed within the historical scope of Kurdish language issues be-fore and after the fall of the Ottoman regime, and within the demolinguistic scope of Kurdish in Turkey, Iran, and Iraq but excluding issues relevant to Kurdish in Syria, Armenia and diaspora. The volume provides detailed discussion of the top-ics listed above within the context of comparative/international case studies, which allows for similarities and differences between cases to emerge. The simi-larities outweigh the differences, and point to broader issues and concerns in the field of Kurdish studies and of general significance to the field of the sociology of language, hopefully making this special issue of interest to IJSL readers from dif-ferent areas of specialization. For instance, the focus on the harm done to indig-enous languages and speech communities as a result of the forced removal of children and their re-education in residential schools as part of a “civilizing mis-sion” closely parallels Aboriginal children’s experiences in Australia and North America, and will resonate with scholars and stakeholders working on language revitalization and reconciliation. Similarly, it will resonate with (other) Indige-nous groups internationally (e.g., Indigenous/tribal researchers in Nepal, India, and the Nordic countries) engaged in language revitalization efforts and attempts to introduce mother tongue-based instruction in the state educational system.

Turkey goes to enormous efforts all over the world but especially in North America and Europe to silence the Armenians and rewrite history, trying to con-vince people that there was no Armenian genocide a century ago (for facts about the genocide, see, e.g., Fernandes [2007]). In a similar way, Turkey tries to con-vince the world that “there is no Kurdish problem” (in fact, it is, of course, a Turk-ish problem). The history of oppression and subjugation of Kurds and Kurdish, and of their purposeful under-development, vilainization, categorization as “un-civilized” or non-existent, has been silenced or rewritten in all Kurdish areas (even if this has now changed in Iraqi Kurdistan). In the articles of this volume, there is supportive overlap coming from hugely different sources, in the context of different countries, reporting on the same historical key events and processes in-




volving Kurds and Kurdish. It will be harder now for states to discredit the issues – they cannot erase them from memory or rewrite memories emerging in different countries, from such different historical accounts. With the same “history” com-ing from such different angles as language historians in Syria, Iran, Turkey, etc., all of them reporting on the same thing, there is a huge “external validity” almost jumping off the pages of the combined papers, as one of the peer reviewers of the volume pointed out.

We hope that this volume can in a small way help in starting to remove some of the roadblocks that face Kurdish studies. The scholarly silence on the sociology of Kurdish has not been due to serendipity or a lack of importance of the language itself; rather, linguicide perpetuated by the states ruling over Kurdistan and ac-cepted by others with grave economic, military and political interests in the re-gion has played a key role in the omission of Kurdish from the scholarly literature. This special issue is also significant for its symbolic value: The Kurdish language is being highlighted in IJSL, therefore it is.

ReferencesAnonynymous. 2011. Basî ziman le êwarekorêkî Korî Zanyarî da [Discussing language in a

colloquium of the Kurdish Academy]. Hewlêr 3 May. http://www.hawler.in/news16110-8.htm (accessed 4 May 2011).

Coşkun, Vahap, M. Şerif Derince & Nesrin Uçarlar. 2011. Consequences of the ban on the use of mother tongue in education and experiences of Kurdish students in Turkey. Yenişehir/Diyarbakir: DISA (Diyarbakir Institute for Political and Social Research; www.disa.org.tr).

Fernandes, Desmond. 2007. The Kurdish and Armenian genocides: from censorship and denial to recognition? Stockholm: Apec.

Fishman, Joshua A. 1997. Language and ethnicity: the view from within. In Florian Coulmas (ed.), The handbook of sociolinguistics, 227–343. Oxford: Blackwell.

Fishman, Joshua A. 2001. From theory to practice (and vice versa): review, reconsideration and reiteration. In J. A. Fishman (ed.), Can threatened languages be saved? Reversing language shift revisited: a 21st century perspective, 451–483. Clevedon, UK: Multilingual Matters.


Malmisanij, Mehmet. 2006. The past and the present of book publishing in Kurdish language in Syria. Next Page Foundation. http://www.npage.org/article127.html (accessed 15 July 2007).

Ricento, Thomas. 2007. Models and approaches in language policy and planning. In Marlis Hellinger & Anne Pauwels (eds.), Handbook of language and communication: diversity and change, 211–240. Berlin & New York: Mouton de Gruyter.

Sheyholislami, Jaffer. 2011. Kurdish identity, discourse, and new media. New York: Palgrave Macmillan.

Skutnabb-Kangas, Tove, & Desmond Fernandes. 2008. Kurds in Turkey and in (Iraqi) Kurdistan: a comparison of Kurdish educational language policy in two situations of occupation. Genocide Studies and Prevention 3(1), 43–73.



http://www.hawler.in/news16110-8.htm

www.disa.org.tr

DOI 10.1515/ijsl-2012-0055 IJSL 2012; 217: 189 – 194

Book reviewPolitics and language ideology in Kurdish lexicographyRashid Karadaghi. /The Azadi English-Kurdish diction-ary. Tehran: Ehsan Publishing House, 2006.

Mâĵed Mardux Ruhâni. . Farhange Dânešgâhe Kordestân. Kurdistan University Persian-Kurdish dictionary. Sanandaj, Iran: Kurdistan University, (Vols. 1–2) 2007, (Vol. 3) 2008.

These dictionaries are the largest bilingual lexicographic projects in Sorani Kurd-ish, which has been, since 1991, the language of the Kurdistan Regional Govern-ment (KRG) in northern Iraq and, since 2005, one of the two official languages of Iraq. Sorani, also called Central Kurdish or Southern Kurmanji, is the name of the largest dialect group in Iraqi and Iranian Kurdistan but ranks second, after Kur-manji, within the larger Kurdish speech area which is divided among the neigh-boring countries of Syria, Turkey, Iran, Iraq and Armenia (see Introduction, this issue).

Most visible in the two dictionaries is the unequal status of the source languages, English and Persian, and the target language, Kurdish. While Eng-lish and Persian themselves are not on equal footing, both are well developed standard, state languages in contrast with Kurdish, a standardizing language with many unsettling tendencies and un-stabilized components. In standardiz-ing languages, lexicographers, whether of descriptive or prescriptive persua-sions, have to make difficult decisions about how to represent or, rather, codify the norm especially its alphabet, orthography, orthoepy and lexicon. Kurdish is under much pressure to accommodate the demands of officialization, self-rule, war, modernity, traditional and new media, internet and globalization. Al-though monolingual lexicography assumes a stronger normative role, the two diction aries under review confront the most serious challenge: matching up Kurdish vocabulary with the more elaborate lexical repertoires of English and Persian.

It would be difficult to assess the two dictionaries without delving further into this sociolinguistic context. One important landmark is the unprecedented outpour of publishing and broadcasting in Kurdish in the wake of the 1991 US war on Iraq and the formation of the KRG. This led instantly to the replacement of Arabic by Kurdish as the language of the new semi-independent government. Another outcome is what the Kurdish literati characterize as chaos, a situation

Brought to you by | Northern Arizona UniversityAuthenticated


190 Book review IJSL 217 (2012)

in which authors and translators feel free to coin words and devise their own spelling rules or ignore any idea of rule-governed writing. It is now common-place among Kurdish intellectuals to accuse almost everyone of “not knowing Kurdish”, even though there is little consensus on what constitutes “correct” or “good” Kurdish; the only consensus is on “weeding” Arabic (Iraq), Persian (Iran) and Turkish loanwords (Turkey). This is a rigid purist politics which generates more anarchy in the language of both print and broadcast media. The Kurdish language academy founded by the KRG is helpless in bringing order to ortho-graphic and lexical disorder. While this sort of chaos may eventually transform into order, it will take a relatively long time, something the language and its users can hardly afford. Under the conditions, the two dictionaries inescapably undertake the role of language planners if only because they engage in intensive lexical engineering. However, they undermine their own authority by pursuing an extremist purist policy which seriously constrains the lexical repertoire of the language.

Each work is the size of an English collegiate dictionary. Karadaghi’s 1,241 pages contain about 44,000 entries with over 4,000 “sub-entries”, or phrases and idiomatic expressions (p. II). Ruhâni has used, as a basis, a Persian mono-lingual dictionary the size of 50,000 entries. In terms of lexicographic method especially differentiation of meanings and homonyms, use of idioms and phrases, and exemplification, they match the best bilingual Sorani dictionaries such Wahby and Edmonds (1966) or the monolingual work of ‘AbdurRahman Zabihi (1977). Method, however, cannot be reduced to rules for organizing the contents of a dictionary. Equally significant are the ties that bind method to theory, i.e. the central question of meaning and equivalence. While it is unfair to expect the two dictionaries to be informed by lexicographic theory, itself a new endeavor in Western linguistics, one can see in almost every entry how the complex work of creating translation equivalence has been further complicated by purist politics.

Karadaghi’s dictionary is his lifetime work produced without institutional support whereas Ruhâni is the supervisor and editor-in-chief in a collaborative work supported by Kurdistan University, Iran. Surprisingly, Karadaghi does not provide any information on his approach to the lexicon of Kurdish as if there is no lexical or onomasiological gap between the two languages and no anisomor-phism, i.e. mismatch between the two languages due to semantic, morphological and cultural differences. By contrast, Ruhâni is conscious about the inequalities and gaps as well as the contested terrain of lexical elaboration in the target lan-guage, and informs the reader about his lexicographic strategies.

In the introduction to the first volume, Ruhâni contrasts the contemporary flourishing of Persian lexicography with the absence, in Kurdish, of “what may



Book review IJSL 217 (2012) 191

be called, in today’s parlance, a dictionary”, and warns that the Kurdish used in radio and television is so “mixed up and inefficient that it can hardly be called language” (p. 8). He is aware of the ways in which the dominant position of Persian has led to lexical amnesia among the Kurdish literati, a situation that prompts the learning of Persian concepts by way of forgetting native ones. Under the conditions, his dictionary aims at recording the Kurdish lexical repertoire that is “seriously threatened with loss”, and making it available to “Kurdish writers, poets, speakers or anyone looking for Kurdish equivalents of Persian words” (p. 9). He then notes that while bilingual dictionaries aim at providing “exact and suitable equivalents for source language entries”, the principal goal of his dictionary is to cull from diverse Kurdish dialects as many equivalents as possible (pp. 10–11). For instance, the headword for âtaš/âteš ‘fire’ is first provided with its Kurdish translation equivalents in different dialects: agir, ager, agur, êgir, awir, ar, ayer, ayir, eyir, ahir, ater, hayir, hetar, hater, awîr, ahîr, giře, yagî, teş. He insists on listing all these forms because there are no criteria for selection and “the value and credibility of a language depends on all its dialects and even phonemes” (p. 14). This strategy collapses the dis-tinction between dialect diction aries and standard language dictionaries. The array of dialect equivalents is then followed by five “explanatory equivalents”, i.e. explanations in Kurdish of the meanings of the Persian headword to-gether with examples for each meaning (for example, meaning 1 is

‘combustible material which combines with air and gives out light and heat’). Here the borderlines be-tween monolingual and bilingual lexicography disappear in the explanatory equivalents or definitions which permeate the dictionary.

Ruhâni finds his dictionary deficient in scientific terminologies because, he argues, Kurdish has not had the opportunity to develop in this area. He tries to avoid, as much as possible, coining words because, for one thing, it is the pre-rogative of a language academy, and, for another, coining may lead to “chaos” (p. 15). An appendix to Volumes 1 and 3 provides a list of terms in Persian, Arabic and English for which no Kurdish equivalents could be found; dictionary users are asked to assist in finding equivalents.

Karadaghi is not interested in recording dialectal variants in the target lan-guage though he, too, appeals to explanatory equivalence whenever (a) there are no Kurdish equivalents for English words or (b) he purges Kurdicized loan-words. For instance, ‘inner circle’ is given the following explanatory treatment:

/destey desełatdaran ya nizîk le fermanřewa; destey karbedest ya şitbedestî wiłatêk ya dezgayek, which may be translated as a ‘group of power holders or [those] close to the ruler; group of officials or power holders of a country or




establishment’. A example of (b) is the word ‘experience’, which comes with the following equivalents and explanations: (1) (2) ( -3) ( -3)/(1) beser-hat ; rûdawêk mirov têyda beşdar bûbêt; (2) hemû beserhatî řaburdûy kesêk; (3-a) şarezayî; (3-b) mawey ew şarezayîye’ which may be rendered into ‘(1) life story; an event in which a man takes part; (2) all the story of the past of a person; (3-a) expertise; (3-b) duration of this experience’. The dictionary avoids the popu-lar borrowing for ‘experience’, i.e. the Arabic/Persian loanword /tecrebe. This purist approach in Sorani lexicography is solidly anti-Arabic. Within the period of five decades between 1924 and 1973, the purist (Kurdî petî ) movement was able to reduce loanwords from 46.4% to 4.4%.1 The only Sorani dictionary, Zabihi cited above which included widely used Arabic loans, was criticized for doing so.

It would be unfair to expect the two dictionaries to act as language acade-mies and coin or find suitable translation equivalents for all the words in the source languages. This is a goal that even a language academy fails to achieve. It is reasonable, however, to critique a language ideology that treats the prob-lems of the Kurdish language as one of purity and impurity. This ideology fails to appreciate borrowing as a legitimate, useful, and inescapable source of lexical and semantic elaboration. Purism in Kurdish lexicography (except in former USSR) is a nationalist reaction to Arabization (Iraq and Syria), Per-sianization (Iran), and Turkification (Turkey). The dictionaries adopt two strate-gies which distort the meanings of lexical units in both the source and target languages.

One strategy is using Kurdish lexical units which do not impart semantic equivalence with those in the source language. For instance, Kurdish and Persian have no native equivalents for, among many other concepts, ‘abstract’,2 ‘nega-tive’3 and ‘exception’.4 Both languages have borrowed Arabic (mojarrad ), (manfi) and (estethnâ’). While Persian maintains these loans and creates many combinations and derivations for each, the Kurdish literati purge them without providing acceptable coinages or semantically extended native alterna-

1 See Abdulla (1980: 182).2 Adjective in the sense of “separated from what is real or concrete; thought of separately from facts, objects or particular examples” (Hornby 1974).3 Adjective, “(of words and answers) indicating no or not (opp of affirmative) . . . (opp of positive) expressing the absence of any positive character; that stops, hinders or makes powerless” (Hornby 1974).4 Noun in the sense of “somebody or something that is excepted (not included) . . . something that does not follow the rule” (Hornby 1974).



Book review IJSL 217 (2012) 193

tives. Thus, Karadaghi provides for the following translation equivalents that share very little if any semantic space with ‘abstract’: which mean, respectively, ‘naked, bare, plain, unadorned, absolute’; ‘unseen’; ‘not to be seen, invisible’.5 Ruhâni does not do better. The closest meanings provided for are: (the first two mean ‘separate, different’ and the third ‘at a distance, far away’). As for ‘negative’, Karadaghi provides these translation equivalents: (the first two mean ‘no’ and the third ‘against everything’). Ruhâni offers which mean ‘saying no, hostility’; ‘with adverse behavior’; ‘bad, evil’; ‘discordant and useless’. The word ‘exception’ in Karadaghi is given these equivalents:

which mean ‘difference’; ‘differentiating’; ‘subtract from’. None of the above is an acceptable equivalent.

Another strategy is the use of a single pure Kurdish word for a variety of concepts. To give one example from legal terminology which is borrowed pri-marily from Arabic, Karadaghi purges such loans by using the word /tawan or its derivatives as his first translation equivalent for ‘offence’, ‘felony’, ‘crime’, ‘transgression’, ‘accusation’, and ‘indictment’. Another dictionary, Jaf Pocket English-Kurdish Dictionary (‘Abd al-Karim 2002) goes to the extreme of using yasa ‘law’ for ‘rule’, ‘regulation’ and ‘law’. Yasa is, however, a loan from Mongolian or Turkish and the author is either unaware of the etymology or prefers it to the more widely used Arabic loanword qânû. Monolingual dictionaries simply ignore loanwords commonly used in classical literature, pre-1920s jour-nalism and, sometimes, in oral Kurdish especially in the urban environment. While lexicographers differ in their dedication to purism, it would be difficult to find, in either bilingual or monolingual dictionaries, thousands of Persian and Arabic loanwords which appear in classical literature including the composi-tions of the great Sorani poet Nali, e.g., aġaz ‘beginning’, aġûş ‘embrace’, batin ‘inner part, conscience’, bişarat ‘good news’, ĥusn ‘beauty’, ġubar ‘dust, mist’, meĥbes ‘prison’, mesken ‘dwelling, residence’, me’rûf ‘well-known, famous’, mi’yar ‘measure, criterion’, or miftaĥ ‘opening, key’. Instead of enhancing the vitality of Kurdish, this purist politics has inadvertently contributed to the state-sponsored linguicidal policy of im poverishing and de-intellectualizing the language.

University of Toronto AMIR HASSANPOUR

Correspondence address: [email protected]

5 For the meanings of Kurdish words in these examples, I rely on Wahby and Edmonds (1966).




References‘Abd al-Karim, ‘Ali Ihsan. 2002. Jaf pocket English-Kurdish dictionary. Silemani, Iraq.Abdulla, Jamal Jalal. 1980. Some aspects of language purism among Kurdish speakers. York,

UK: University of York dissertation.Hornby, A. S. 1974. Oxford advanced learner’s dictionary of current English. Oxford: Oxford

University Press.Wahby, T. & C. J. Edmonds. 1966. A Kurdish-English dictionary. London: Oxford at the Clarendon

Press.Zabihi, ‘AbdurRahman. 1977. /Qamûsî zmanî Kurdî. Baghdad: Koŕî Zanyarî

Kurd.



DOI 10.1515/ijsl-2012-0056 IJSL 2012; 217: 195 – 218

Small languages and small languagecommunities 72

Editor: Emily McEwan-Fujita

CONTINUING AND SHIFTING MULTILINGUALISM IN AN ÉMIGRÉ SITUATION: LANGUAGE USE AND AT TITUDES AMONGST IRAQI CHALDEANS AND ASSYRIANS IN MELBOURNE*

JIM HLAVAC

Abstract: This study focuses on a situation of retentive multilingualism and trans-mission of more than one language in the diaspora. The population in question are bi- or multilinguals who, in an immigrant situation, added a further language, English, to their repertoire. Sixty-six Iraq-born informants who identified as Chaldeans or Assyrians now living in Melbourne, Australia were interviewed about their nominated “first” language, situation and context of acquisition, self-declared level of proficiency and notions of attachment towards and co-identification with respective languages. Proficiency and use of Chaldean and Assyrian remain high among respondents, despite these languages’ level of en-dangerment in Iraq. Only five informants had shifted to Arabic as their “domi-nant” language. However, Arabic has become, and remains, an established code in in-group situations. Although contact with Arabic-speaking Iraqis and non-Iraqi Arabic-speakers is significant, it is the established status of Arabic as a “complementary” code in Chaldean and Assyrian in-group interactions that ac-counts for why informants view Arabic as one of their languages alongside their L1, and, to a lesser extent, English. Desire for language maintenance applies chiefly to informants’ L1, but also to Arabic, confirmed by data showing infor-mants’ practices in maintaining both codes and positive affective responses to them both. In contrast, informants express distance towards Kurdish. The con-nections between this pattern of language maintenance and attitudes and the social and political situation of the emigrants in Iraq and Australia will be discussed.

* I am grateful to the past and present IJSL SLSLC section editors for detailed comments and criticisms of two previous drafts of this paper.



196 J. Hlavac

Keywords: multilingualism; Chaldeans; Assyrians; language maintenance; lan-guage attitudes; Iraqi diaspora

Jim Hlavac: Monash University. E-mail: [email protected]

1 IntroductionFor the last twenty years since the 1991 Gulf war, and particularly since the US-led invasion in 2003, Iraq has witnessed large-scale internal migration and emigra-tion. In parts of southern and central Iraq, internal migration (including evacua-tion and forced expulsion) has led to homogenization of the ethno-linguistic landscape; in the north and north-east, it has led both to largely Kurdish homog-enization and to a further “patchworking” of the linguistic landscape as Chal-deans, Assyrians and Armenians move back to areas that many of their grand-parents had once left (O’Mahony 2004). Outside Iraq, the number of Iraqis now residing in surrounding countries has also risen. The languages of overseas Iraqis – Iraqi Arabic, Kurdish, Assyrian and Chaldean – are now émigré languages spoken by a diaspora of over two million people (Marfleet 2007: 397).

Speakers of the last two language groups are the focus of this paper, based on data from 66 Chaldean and Assyrian immigrants now resident in Melbourne, Australia. This paper firstly focuses on the circumstances of informants’ acquisi-tion of their respective languages in Iraq and their self-diagnosed proficiency levels, and then examines aspects of their language use in Melbourne and atti-tudes towards the codes that informants are acquiring, maintaining, and/or seek-ing to transmit to future generations.

The sample’s quantitative data is examined and discussed to shed light on how multilinguals’ repertoires are realigned according to ethnic, socio-political, and educational features that are mostly very different from those in Iraq, but which, in some cases, appear to replicate some homeland features. These realign-ments include the use and status of Chaldean and Assyrian as codes with strong symbolic, but also instrumental importance to the sample as “in-group” lan-guages that serve large, heterogeneous but demographically close-knit communi-ties in one part of metropolitan Melbourne. Arabic has also become an “in-group” language for many, especially younger speakers, but is deprived of the official and prestigious status that it enjoys in Iraq. The transactional and neighborhood networks of many informants means that Arabic has not only a “shared in-group” status amongst Chaldeans and Assyrians but also shared in-group status with




surrounding Iraqi Arabic-speakers and lingua franca with Lebanese Arabic-speakers, i.e. as an out-group language with some market value. This, together with shared loyalties to Chaldean/Assyrian ethnicity and the Iraqi homeland, ap-pears to lead to a desire to use and actively pass on not one but two languages to the next generation.

2 Theoretical modelsThis paper contextualizes discussion of multilingual Chaldeans and Assyrians with a brief account of historical, sociological, and linguistic features. Factors influencing the maintenance of or shift from indigenous minority languages in the Middle East have received little attention in language ecology literature (ex-ceptions are Blau and Suleiman [1996]; Ennaji [1997]). However, researchers of indigenous micro-languages will find many features of this sample familiar – geographical isolation or segregation, ethno-religious peculiarity, and high levels of endogamy.

This paper does not adopt a comprehensive application of any single model on the ecology of minority languages. Fishman’s domain-focused or sociology of language approach forms a basis for data elicitation in the first place and the following domains are distinguished in this paper: family/children, religion, and social networks. Proficiency elicitations are also presented. Elicitations and data provide discussion about the role of proficiency as a factor in language use. Atti-tudes towards maintenance and means of transmission (Fishman 1991) and co-identification with respective languages (Le Page and Tabouret-Keller 1985) are discussed as factors which may accompany and influence use preferences. All of these features are drawn on in the discussion of a multi-lingual group that has moved from a situation characterized by ethno-religious separation, political au-tocracy, ethno-political favoritism, sudden urbanization and (apparent) secular modernity to one in urban Australia which is characterized by ethnic diversity, religious plurality, a post-industrial labor market and New World multicultural-ism in which the hegemony of English is not perceived as a top-down imposition but as a consequence of the need for a lingua franca between diverse groups.

3 The language of Chaldeans and Assyrians in Iraq

Chaldeans and Assyrians see themselves as custodians of an ancient, pre- Arab civilization in an area once known as one of the “cradles of civilization”,



198 J. Hlavac

Mesopotamia. A full description of the history of Chaldeans and Assyrians in their homeland goes beyond the bounds of this article (see Yildiz 1999; O’Mahony 2004; Hanish 2008). Brief mention is made here of the historical linguistics of Chaldean and Assyrian (both varieties of north-east Neo-Aramaic), their relation-ship to other surrounding languages, language contacts with other speech com-munities, and their status in Iraq.

The terms “Chaldean” and “Assyrian” are used by informants to refer to their ethnicity and to their language, and they will also be used this way in this paper. Some informants view Chaldeans and Assyrians as “being the same people” or “belonging to the same people”. Others see both groups as closely related but still distinct. This papers views both groups as closely related, distinguished chiefly through religious affiliations: most Chaldeans are Eastern-rite Catholics who rec-ognize the authority of the Pope; most Assyrians belong either to the Assyrian Church of the East or the Syriac Orthodox Church. Figure 1 shows the historical linguistic relationships of the Chaldean and Assyrian languages within the Se-mitic language family.

The unbroken lines show Neo-Aramaic’s relationship to Aramaic and Central Semitic. Broken lines indicate adstratum (usually lexical) influence from other Semitic languages.

For particular groups, historical linguistic association can be important as a feature through which group members see themselves as “custodians” (Odisho 1993; 2004) or “owners” of a language (cf. Evans 2001: 251) independent of their

Fig. 1: Relationship of Chaldean and Assyrian to other Semitic languages




proficiency in the language. This relationship can be imputed or acknowledged by others, sometimes even with very positive consequences for language revital-ization, e.g. the role of Anglican missionaries in Iraq in the nineteenth century who assisted in the codification of contemporary use and who financially sup-ported schooling in neo-Aramaic (cf. Odisho 2004: 188).

4 Chaldeans and Assyrians in Iraq, in the diaspora, in Australia

A century ago, Christians made up around about 10% of the population in the area that is now Iraq (O’Mahony 2004). By the early 1990s, the percentage of Christians had fallen to about 3% of the population: approximately 650,000 Chaldeans and 350,000 Assyrians (Yildiz 1999; O’Mahony 2004). Since 2003, the number has fallen further but at the same time, internal migration to parts of northern Iraq has led to a re-invigoration of the Chaldean and Assyrian communi-ties there, including the re-establishment of Assyrian-language schools previ-ously abandoned or forcibly closed due to Arabization policies in 1974 (Odisho 2004).

Comparatively little is known about the maintenance of minority lan-guages in Iraq, due in part to many Western linguists’ focus on European lan-guages, and to different traditions of philological and field-work enquiry in the Middle East which are not readily accessible to Western or largely Anglophone readers. Pre vious generations of Chaldeans and Assyrians have been mono-, bi- or multi-lingual – other languages such as Persian, Turkish, Arabic, Armenian, Kurdish and English were acquired for trade, educational and/or socio-political reasons (Odisho 1993: 192–196). In the 1920s and 1930s, Assyrian- and Chaldean-Kurdish bilingualism preceded Assyrian- and Chaldean-Kurdish-Arabic trilin-gualism that was common in the 1940s and 1950s, followed by Assyrian- and Chaldean-Arabic bilingualism as many began to abandon Kurdish from the 1970s onwards.

At least since the 1970s, Arabic has been the uncontested H-language in all domains of Iraqi public life. Only recently and only in certain parts of northern Iraq has another language, Kurdish, challenged the status of Arabic as the domi-nant H-language.1 In contrast, Chaldean and Assyrian are restricted codes: on

1 Since the start of this century, the status and use of Kurdish in Iraqi Kurdistan has changed greatly. Kurdish is now the most common language of instruction and most common language used in official functions there, cf. Skutnabb-Kangas and Fernandes (2008: 46–49).



200 J. Hlavac

Giles et al.’s (1977) ethnolinguistic vitality scale, these languages have medium to high vitality weightings for the following features only: sociohistorical and language status within the group (status factors); concentration and endogamy (demographic factors); religion and culture (institutional support and control factors).

In Australia, the number of Chaldean and Assyrian-speakers has increased from 10,192 in 2001 to 23,526 in 2006. The majority of them (60%) were born in Iraq while the rest were born in Syria, other neighboring countries, or Australia. As a percentage of the Iraq-born Chaldean and Assyrian population (t = 32,520), 38% (12,240) are Catholic (mainly Chaldean, cf. Pitrus 2009) and 13% (4,310) are Assyrian Apostolic (mainly Assyrian). Chaldeans and Assyrians are dispropor-tionately over-represented amongst the Iraq-born in Australia.

A city of four million people, Melbourne is largely Anglophone but over a quarter of its inhabitants speak a language other than English (hereafter “LOTE”) at home, and the number of those who have proficiency in two or more languages is likely to be even higher (Clyne 2003: 26–28). Arabic is the sixth most widely spoken LOTE with 54,000 speakers and Assyrian the 26th most common LOTE with 7,000 speakers. (The Australian Bureau of Statistics does not list “Chaldean” as a distinct language but subsumes it under “Assyrian”.) All informants in this study live in the municipality of Hume (population 150,000) in Melbourne’s northern suburbs. In Hume, Arabic is the second most widely spoken LOTE with 9,000 and Assyrian the fourth most common language with 5,000 speakers. The immediate and wider neighborhood that the informants live in is multilingual with Arabic, Chaldean and Assyrian serving the needs of many recent immigrants and English providing the means for official, some “out-group” and most “inter-group” communication.

5 MethodologyThe majority of this sample’s informants were Chaldeans – 57 who identified as Chaldean L1 speakers and five L1 Arabic-speakers who, in an ethnic sense, co-identified as Chaldeans (children and grandchildren of Chaldeans who shifted to Arabic). The composite term “Chaldean/Assyrian” is used in this paper to refer to neo-Aramaic as a language of the informants. Only in those cases where “Chal-dean” is expressly compared as a variety separate from “Assyrian” do they appear as separate terms.

Criteria for inclusion of possible informants in this study were (a) Iraq-born and (b) non-exclusive identification as Chaldean or Assyrian. The author is not an in-group member but has had substantial, on-going contact with Chal-




deans and Assyrians. Data was collected by the author and an in-group research assistant.2

Data was elicited from informants by way of a written, paper questionnaire. As Arabic is the language in which informants have strongest literacy skills, this language was chosen for the questionnaire so that informants could fill it out independently. Other studies on Chaldeans and Assyrians in the diaspora have also employed written questionnaires in Arabic (e.g. Sengstock 1982). The ques-tionnaire mainly consisted of questions with suggested responses from which informants could select their answers. The tables presented in Section 6 contain numeric responses to suggestions and/or statements. Some questions remained unanswered by many informants and these are recorded under the “no re-sponse” category in tables. The questionnaire elicited responses on the follow-ing: declared first language, self-reported proficiency, family/children domain, media consumption, language of religious services and prayer, transactions/neighborhood, identification with language varieties, attitudes to maintenance/shift, and comments on the value of language varieties.3 Data was collected from April to July 2010.

The average age of informants was 40 (youngest 16; oldest 72). Average age of informants upon arrival in Australia was 32. All informants were adults (≥20 years old) at time of arrival in Australia, except for one who was six. Age at time of departure from Iraq was not elicited. The spread of informants across age groups is the following: 23 informants (age group 16–30 years old), 20 informants (age group 31–45 years old), 23 informants (45+ years old). Age at time of depar-ture from Iraq and age upon arrival and length of residence in Australia are fea-tures which shape informants’ social, educational and occupational biographies. For all but five of the 66 informants, primary and secondary education was com-pleted in Iraq. Data elicited about informants’ childhood language/s and school-ing relates primarily to informants’ experiences of these in Iraq. All informants arrived in Melbourne after residing in transit countries such as Turkey, Syria, or Jordan, often for some time. Range of stay in transit countries ranged from six months to six years. Informants’ average length of residence in Australia is eight years.

2 I am most grateful to my research assistant, Ms Shireen Francis for her expertise as a person of standing to the informants and for her inter-personal, translation and interpreting skills. I would also like to express my gratitude to Mrs Widad Pitrus who provided me with invaluable background information on the areas of Chaldean and Assyrian settlement in Iraq, North America and Australia.3 Data were also collected on language in the domains of media use and transactions/neighborhood, but they are not discussed here due to restrictions of length.



202 J. Hlavac

The community sampled for this study is an émigré group of comparatively recent vintage which consists mainly of immigrants who left Iraq between 1995 and 2002 with a smaller group who left after the US-led invasion in 2003. The means through which informants came to Australia varied from chain migration and government-prioritized migration intakes that targeted Christian minorities in the Middle East in the 1980s and 1990s to family reunion migration policies and refugee intake cohorts in the early 2000s. Generally, most informants have other, extended, family members in Melbourne: 33% report that all other family members are in Melbourne, 44% report that most other family members are in Australia, 18% report that some are in Australia and some elsewhere and only 5% report that most family members are in Iraq.

6 Data and discussion

6.1 Informants and their first language(s)

Informants gave the following responses in regard to self-described first lan-guage: Chaldean = 57; Arabic = 5; Assyrian = 4. All responses reflect informants’ own declarations about themselves: no diagnostic testing or verification of lan-guage use was undertaken. The five informants who list Arabic as their first lan-guage are informants who self-identify as “Chaldeans” and as “Iraqi Christians”. These five informants use Chaldean in functionally restricted domains: the reli-gious domain and prayer (along with Arabic). Designation of first language and language use during childhood is shown in Table 1.

Language with parents

Language with siblings

Language with other children

Language with other schoolchildren

Language of instruction

Chaldean 52 51 32Chaldean + Assyrian 2 1 1Chaldean + Arabic 4 2 22 9 3Chaldean + English 1 1Assyrian 4 4 4Arabic 4 6 6 48 59Arabic + English 4 2Arabic + Kurdish 1Kurdish 1

Table 1: Informants’ language use in childhood




Reported language use with parents and siblings, across informants’ life-times and before and after emigration from Iraq, shows little evidence that Arabic was more popularly used with same-age siblings than with parents. The rubric “language used with other children” refers to the language used with children with whom informants communicated in their immediate locality as children themselves. Chaldean was still the most widely-used language reported by 32 users. Twenty-two informants used Chaldean and Arabic with other children, showing that overall, most informants grew up in environments in which Chal-dean and Assyrian were commonly-used languages, used also with young peers. Very high levels of use of Chaldean and Assyrian with same-age peers outside school suggest that both languages are robust and unmarked choices for both communities.

6.2 Declarations of proficiency

Language acquisition reflects informants’ interactional and educational circum-stances which directly determine proficiency. Information was elicited about in-formants’ language use with parents and siblings. Language use with these inter-locutors strongly correlated with self-declared first language, as presented in Table 1.

Arabic was overwhelmingly the language of instruction during informants’ primary school education. Chaldean- and Assyrian-language schools were forced to close in 1974 and the almost universal use of Arabic as language of instruction reflects the Arabization policy of the Iraqi government that had commenced in the 1960s. All but one informant attended primary and secondary schools in Iraq in which the language of instruction was Arabic. Ten of the informants also re-ceived some religious-based instruction in Chaldean or Assyrian at parochial Sunday schools. Twenty-five stated that they learnt some English as a school sub-ject at both primary and secondary school. Arabic was also the language which 48 informants reported using with schoolmates. In Table 2 are self-declared esti-mations of proficiency.

Self-declarations for language use and macro-skills reflect informants’ pat-terns of acquisition and functional use for each language. Spoken proficiency in most informants’ first language, Chaldean, was high, but active literacy in the same language was low due to the paucity of formal schooling available in the language.

The figures for Assyrian reflect not only the four self-declared Assyrian L1 speakers, but also the responses from Chaldean L1 speakers who gave varied re-sponses to their Assyrian proficiency, from “excellent” to “non-existent”. These



204 J. Hlavac

statistics also reveal Chaldean L1 speakers’ perceptions of their own speech and how closely it resembles their notions of Assyrian, whether they consider this identical to Chaldean or not.

Only twelve informants reported proficiency in Kurdish. This is a low number considering that over half come from northern Iraq – anecdotal information about informants known to me and their contact with Iraqi Kurds suggest to me that some informants may have under-reported proficiency in Kurdish (see Sec-tion 6.5). Estimations of English reflect the period and circumstances of infor-mants’ contact with the language. Most encountered English to any measurable degree only after arrival in Australia. Speaking skills were rated higher than writ-ing skills for all languages.

Ethnic minority group membership and ethnic minority language speaking ability need not always be co-present (cf. Evans 2001). For example, five infor-mants had already shifted to Arabic as their dominant language but still held al-legiance to and self-associated with Chaldean ethnicity.

6.3 Family / children

The robustness of Chaldean and Assyrian is demonstrated by the fact that these languages are the first-learnt languages for 61 of the 66 informants who also re-ported that they were widely-spoken by siblings and other same-age children. Forty-six of the 66 informants had children, and Table 3 contains data on infor-mants’ reported language use with children and their children’s use with them.

There are significant differences between the responses about language use with children and the responses about language use with informants’ own par-ents (see Table 1). Over half of those with children speak Chaldean to them,

Chaldean Assyrian Arabic Kurdish English

Excellent Speaking 47 13 53 3 7Writing 4 2 46 0 3

Good Speaking 11 6 8 6 30Writing 1 1 13 0 23

Fair Speaking 2 14 3 5 19Writing 4 4 2 1 28

Poor Speaking 2 12 1 8 8Writing 1 0 1 4 7

Non-existent Speaking 4 21 1 44 2Writing 56 59 4 61 5

Table 2: Estimations of language proficiency in the macro-skills speaking and writing




but of the remaining 22 parents, 13 speak Chaldean in combination with other language(s), 10 speak Arabic, nearly always in combination with other language(s) and 17 speak English, nearly always in combination with another language. Monolingual Assyrian use was reported by two informants. This data shows that in the first place, over half of informants reported monolingually transmitting their L1 to their children.

Further, only one parent has shifted to English, but English now contributes to parent-child exchanges amongst 16 informants. Dual language use commonly signals the first sign of shift to English, although it need not do so. Of particular interest is the use of Arabic. Three of the L1 Arabic informants have children: one uses only Arabic, the other two use Arabic and English. The remaining users of Arabic are Chaldean L1 speakers, three of whom have shifted to Arabic and Eng-lish as the language used with children, while two use Chaldean with Arabic and a further two use Chaldean with Arabic and English. This small group of seven Chaldean L1 speakers is of interest because of their desire and behaviour to trans-mit Arabic in addition to or even in place of Chaldean. Table 3 reveals children’s widespread use of Chaldean with their parents. The presence of English is slight-ly greater: 17 parents use some (or only) English, 23 children use some (or only) English in interactions. Of interest is that 10 children use Arabic, usually with Chaldean or English, with their parents. Further data in Sections 6.6 and 6.7 be-low is presented that shows informants’ attitudes towards language varieties and motivations for maintenance.

6.4 Language of religious services and prayer

For a minority group that is distinguished through its ethnic and religious affilia-tions, attendance at religious services is commonly not only an act of religious

Informant with child(ren) Child(ren) with informant

Chaldean 24 15Chaldean + Arabic 2 4Chaldean + English 9 17Chaldean + Arabic + English 2 1Assyrian 2 1Arabic 1 2Arabic + English 5 3English 1 2

Table 3: Reported language use with children, children’s language use with informants



206 J. Hlavac

observance but also a setting for intra-ethnic social, educational and leisure ac-tivities. Within the municipality of Hume there is a Chaldean Catholic church with mass celebrated in Syriac and Neo-Aramaic, and a Holy Apostolic Catholic Assyrian Church of the East with mass in Syriac. Reported association with and attendance at parochial religious services was high. All four Assyrians attended the Assyrian Church of the East. The vast majority of Chaldeans attended the Chaldean Catholic church while a small number of Chaldeans also attended mass celebrated in Arabic and English as well. Three of the five L1 Arabic speakers at-tended Chaldean-language services. High rates of attendance at religious services tend to be a characteristic of recently-arrived immigrant communities (Woods 2004: 147–149).

Informants’ responses about the language(s) that they pray in reflected their responses of language of religious services attended with one exception: Arabic was reported as a language in which nearly one third of informants also pray in. Prayer can be publicly or privately expressed. Public expressions of prayer are likely to be in Chaldean or Assyrian, influenced also by the language of the liturgy, Syriac.

6.5 Identification with language(s), social networks

This section presents responses on informants’ overt claims of (co-)identification with respective languages. Table 4 shows 49 informants’ responses to descrip-tions of their relationships towards different languages. (Seventeen informants did not provide any responses to these statements.)

Table 4 shows that 41 informants selected Chaldean as their own language. Generally, there was a tendency for some of the L1-Chaldean informants (not all

Language Chaldean Assyrian Arabic Kurdish English

I view this as my language / one of my languages

41 16 22 0 15

I speak and understand this language well, but it’s not my language

4 31 24 11 19

Familiar language, but I don’t consider it my language

2 6 4 8 8

I don’t consider this one of my languages

0 0 7 21 5

Table 4: Languages and degree of co-identification




of whom provided responses to these questions) to provide responses showing that they viewed other languages, Assyrian and Arabic, as their languages as well. The reverse was not the case. There were no non-Chaldeans who viewed Chaldean as “their language”. All four Assyrian L1 speakers list Assyrian as “their language” and 12 Chaldeans also view Assyrian as another one of their languages. No Arabic L1 speakers claimed any other language other than Arabic as “their language”, but 17 Chaldean L1 speakers also claimed Arabic as theirs as well.

Among the informants who responded that they speak a language well but do not consider it their own, Chaldeans were most often likely to report this about Assyrian and Arabic, while three L1 Assyrian and one L1 Arabic report this about Chaldean. The twelve who reported understanding Kurdish are all Chaldean L1 speakers, while the 19 “good” speakers of English are from all three L1 back-grounds. Informants rarely listed other languages with which they are familiar but do not co-identify.

Ethnicity and association with the language bearing that ethnicity’s name are often presumed to be synonymous. But a relationship of association can also pertain to other languages, even, as Table 2 shows, to languages acquired in adulthood. This seems to support Clyne’s (1997: 113–120) findings about the rela-tionship that trilinguals have towards their third language as a language that can assume a place of personal importance congruent to the second and even first language. The findings here clearly demonstrate informants’ associative relation-ships towards languages, a pre-requisite for in-group membership. “Association” with multiple languages can be indicative of having membership in multiple groups, e.g. being a member of “the Chaldean-speaking community” and “the Assyrian-speaking community” and “the Arabic-speaking community”. This is congruent to Tabouret-Keller’s (1997: 319–321) application of the idea of “bound-aries, but with gaps” (albeit here referring to internal rather than cross-border linguistic minorities) and to “group affiliation as a matter of choice”, based on “. . . a network of identities, reflecting the many commitments, allegiances . . . everyone tries to handle in ever-varying compromise strategies”. Identification with multiple languages can also be indicative of single membership in one group which is itself bi- or multi-lingual. Sixty of the 66 informants reported using Chal-dean or Assyrian alone or in combination with other languages in social settings with other Chaldeans and Assyrians. Arabic and English were also employed by 41 and 28 informants respectively. Responses here refer to cumulative interac-tions with fellow Chaldeans and Assyrians and bi- or trilingual responses can refer to choice of different code depending on interlocutor or code-switching or both. Variety in the choice of languages used in intra-group social situations is reported by Stoessel (2002) and Dewaele (2011).



208 J. Hlavac

6.6 Attitudes to language maintenance and shift

Informants were asked about their attitudes towards maintenance and shift for Chaldean and Assyrian, Arabic, and Kurdish. (It should be noted that only 11 of the informants claimed proficiency in Kurdish and informants’ support for its maintenance was accordingly low.) The responses are recorded in Table 5.

Responses to these questions show strong agreement that Chaldean and As-syrian should be passed on, while support for Arabic was weaker but still consid-erable. Over half the informants expressed no opinion about future transmission of Arabic to their children. But twenty-one informants supported transmission, including all five L1 Arabic-speakers and sixteen L1 Chaldean informants, while seven reported actually speaking Arabic with their children (cf. Table 3). This is a relatively high figure. Other multilingual groups such as Hungarians from Roma-nia or Slovakia, or ethnic Chinese from south-east Asia usually transmit their L1s to their children but demonstrate little desire to pass on the majority languages of their homelands (i.e. Romanian, Slovak, Vietnamese, Malay, Indonesian etc.; see Clyne [2003]). Desired maintenance of a second homeland language in the dias-pora is generally unusual. Rare parallel examples are the desire of many diaspora Cantonese-speakers for their children to also learn Mandarin (He 2006; Li and Zhu 2010) due to the cultural prestige of this language as the H-language for most Chinese communities, or Somali-speaking parents who also pass on Arabic to their children (Clyne and Kipp 2006: 75–85), largely due to the status of Arabic as the H-language in the religious domain in most Somalis’ lives.

Table 6 shows the reasons why transmission to future generations is desir-able that were selected by informants.4 In regard to Chaldean or Assyrian, 19 of

4 The statements provided for selection in the survey that describe informants’ possible motivations for maintenance are symbolic (“our language”, “integral part of culture”), symbolic and instrumental (“maintain ties with . . .”, “understanding of . . . culture”), instrumental (“vocational/employment reasons”, “communication with older relatives”), and pedagogic/socio-economic (“bilingualism is good for the child”). These reasons are based on data samples in studies (e.g. Paulston 1994; Hlavac 2003) which located and elicited factors conducive to maintenance.

Yes Indifferent No No response

Chaldean/Assyrian 43 3 7 13Arabic 21 18 13 14Kurdish 0 0 27 39

Table 5: Responses to question, “Should this language be passed onto future generations?”




the 43 informants who agreed that these languages should be passed on selected multiple reasons for transmission.

Perhaps surprisingly, an abstract notion such as personal additive bilingual-ism was the single most often selected reason for Chaldean or Assyrian to be passed on. Most other responses relate to ethno-cultural heritage (“understand-ing culture, identity development”, “original language”) while retention of links with the homeland (Iraq) was selected as a stronger reason than communicative norms within the family (“communication with older relatives”). For Arabic, 39 responses were gained from the 21 informants who advocate its transmission. Vocational/employment reasons (cf. Bourdieu’s marketplace notion) and bilin-gualism were selected as the most common reasons, closely followed by ties with “other in-group” members, Iraqis in Australia. Seven select ties with Iraq as a motivation for acquisition of Arabic, indicating that maintenance of connections with Arabic-speaking Iraq is of importance to some.

Results of a question asking informants to select the best means through which transmission can occur are shown in Table 7.

Chaldean/Assyrian Arabic

Child to have a Chaldean/Assyrian identity 8 0It’s our original language 1 6Maintain ties with Iraq 11 7Maintain ties with Chaldeans/Assyrians 2 0Maintain ties with Iraqis in Australia 0 7Understanding of Chaldean/Assyrian or Iraqi/Arabic culture 14 1Communication with older relatives 6 0Vocational/employment reasons 3 9Bilingualism is good for the child 17 9

Table 6: Reasons for language maintenance


Speak only this language to (grand)children 16 4Speak mainly this language to (grand)children 11 4Speak some of this language to (grand)children 2 6(Grand)children attend this language church 8 4Participate in this language’s cultural activities 4 0Send (grand)children to this language’s school 3 9Watch this language on television 0 2

Table 7: Means through which Chaldean/Assyrian and Arabic can be transmitted



210 J. Hlavac

Home/family transmission is nearly the only means by which minority languages can survive as languages of daily community use (Fishman 2001). The impor-tance of the linguistic behavior of older family members was apparent to 27 infor-mants who stated that they speak only or mainly Chaldean or Assyrian to chil-dren or grandchildren. Church attendance was rated more instrumental than participation in cultural activities. School attendance was rarely mentioned, due largely to the closing of the only Chaldean school in 2002.

Language use with succeeding generations has been widely studied in Aus-tralia, through qualitative studies on individual languages (e.g., Clyne 2003; Hlavac 2003; Beykont 2010) and implicationally through “home language re-sponses” given by Australian-born children in census data collections (Clyne 2003). While only first-generation informants are included in this study, re-sponses suggest that Chaldean and Assyrian are the most widely used languages by these respondents with the second generation. Iraq-born Australians report a low level of language shift to English: nearly 50% report Arabic as the “home language”, 39% Assyrian (including Chaldean), 3.8% report Kurdish and only 3.9% English (Department of Immigration and Citizenship 2007).

In comparison with the informal means through which Chaldean and Assyr-ian are transmitted to children, more formal, external institutions support the acquisition of Arabic. Arabic is taught at two supplementary Saturday schools in the area where all informants live. Further, all L1 Arabic parents stated that they (would) speak Arabic with their children. There was some overlap as well as complementarity between the responses of “speaking mainly” or “speaking some” Arabic and similar responses about Chaldean and Assyrian amongst some informants. Anecdotally, three Chaldean-speakers revealed to me that they code-switch into Arabic with their children, even with L1 Chaldean + English, Australian-born children who have little contact with Arabic outside the family home. Despite such restricted input, they claim that their children “can also speak Arabic”.

6.7 Comments on the value of language varieties

Value judgments were elicited from informants on their attitudes towards lan-guages through their selection of suggested responses.5 Responses presented in

5 The statements listed in Table 8 are expressive of gate-keeping, symbolic and symbolic/instrumental features (cf. Note 3 above). In addition to these, aesthetic positive (“expressive language with a long oratory tradition”) and aesthetic negative (“simple, unsophisticated language”) were added (cf. Hlavac 2003).




Table 8 show that over a third of informants identified language as an integral part of Chaldean or Assyrian culture. This high response appears to support Smolicz’s (1981) notion of language as an ascertainable value within a social group, which may be a core feature of individuals’ and groups’ notions of themselves.

Each language was generally considered to be an integral part of its respec-tive culture. Expressive and linking functions were also selected. Arabic was not selected as having attributes that associate it with Chaldeans and Assyrians. Ara-bic is a language that many informants co-identified with (see Table 4) but it is characterized in the responses without mention of this associative status.

The high number of responses that selected language as an “integral part” of the culture supports Liebkind’s (1999: 144) co-dependent or reciprocal view of the relationship between ethnic identity and language: “Language use influences the formation of group identity, and group identity influences patterns of language attitudes and usage.” Aesthetic (“expressiveness”) and cultural-rhetorical dis-course mores (“oratory tradition”, “prestige”) were also selected ahead of the co-hesion or gate-keeping (“link between . . .”, “defines who is . . .”) functions. The aesthetic and cultural-rhetorical tradition afforded to language, more specifically the H-version of informants’ languages, is perhaps explained by the “double-diglossic” situation of the informants: near-extinct Syriac contrasts with neo-Aramaic as one of the informants’ H-languages while Classical Arabic contrasts with colloquial Iraqi Arabic as the other H-language.

7 Discussion: Chaldean, Assyrian, Arabic, Kurdish and English in informants’ lives

This paper has employed a single-survey approach to elicit data in a range of ar-eas that seek to account for multilinguals’ language use in an émigré situation


Integral part of Chaldean/Assyrian or Iraqi/Arabic culture 25 29Link between Chaldeans/Assyrians or Arabs 10 16Defines who is Chaldean/Assyrian or Iraqi 3 3Expressive language with a long oratory tradition 15 17Means of communication, just like any other language 10 3Simple, unsophisticated language 3 0

Table 8: Value of languages



212 J. Hlavac

where use of more than one “homeland” language remains possible. These areas are presented in Sections 6.1 to 6.7 and the implications of informants’ language use in various areas can be re-stated in descriptions of each language’s sociology. The roles and functions that each language performs for informants are summa-rized here, followed by a discussion of patterns and generalizations about this multilingual sample.

Eighty-six percent of the informants listed Chaldean as their chronologically first-acquired language and the language that they commonly used with siblings and similarly-aged children in Iraq. Declarations of oral/aural proficiency in this language were very high while very few were literate in this language due to the paucity of formal instruction and written texts available in Chaldean. Chaldean was also reported as most commonly used with the succeeding generation and current responses indicate a strong likelihood that oral proficiency in Chaldean will be maintained by the subsequent generation; acquisition of literary Chal-dean was rare and restricted to a small number. At the same time, informants’ children commonly used Chaldean with their parents – Chaldean transmission is reliant chiefly on the home/family domain. Preferred strategies for the trans-mission of Chaldean to future generations included exclusive or high-volume use of Chaldean with children and grandchildren. Chaldean L1 speakers identified Chaldean as “their” language, but at the same time often expressed the same association with Assyrian and Arabic, something that Assyrian and Arabic L1-speakers did not do in relation to Chaldean. Chaldean was commonly but not exclusively used with in-group friends. Attitudes towards maintenance were pos-itive, while reasons for maintenance were mixed: views which equate language proficiency with Chaldean culture and identity contrasted with more general views about the benefits of bilingualism and ties with Iraq.

Assyrian is a less-well represented language in this sample – only four infor-mants identified their language and ethnicity as Assyrian, but the above general-izations made about Chaldean also apply to Assyrian. Few features about Assyr-ian that emerge from the data distinguish it from Chaldean, although the small number of informants makes comparison difficult. The first contrast is that a sig-nificant number of the L1 Chaldeans-speakers (12) nominated “Assyrian” as one of their languages. This relationship was not reciprocal. “Assyrian” is also a more widely-known term. It is the only term used for any neo-Aramaic language in official Australian statistics, and the term “Assyrian” features more frequently in Westerners’ descriptions of the background and language of Iraq’s Christians. As-syrians belong to churches that are separate and independent of any other Chris-tian hierarchies. The ecclesiastical language in these churches remains Syriac and the separateness of these churches is likely to aid Assyrian language mainte-nance efforts inasmuch as informants’ association with these churches remains




constant. In this way Assyrians may be less prone to language shift in the reli-gious domain than eastern-rite Catholic Chaldeans may be if they choose to at-tend Arabic- or English-language mass.

All informants but one claimed proficiency in Arabic. Informants’ proficiency levels in spoken and written Arabic are high because Iraqi education policies pre-scribed Arabic as the sole language of instruction. Thus the socio-economically and culturally privileged status of Arabic in Iraq has had predictable and appar-ent consequences on informants’ language repertoires. Official Arabization poli-cies that included forced name changes and the marginalization of non-Arab nationalities led to some inter-ethnic resentment. Data from L1 Chaldeans and L1 Assyrians in this sample shows that whatever resentment some informants may have had towards Arabization policies, there are mixed consequences for lan-guage use such that Arabic is not abandoned as a communication code used in Chaldean and Assyrian “in-group” settings and it is used in Arab Iraqi “in-group” settings and with members of one numerically significant “out-group”, Lebanese Arabic-speakers. Other Iraqis and Lebanese Arabic-speakers are commonly neighbors and workmates and informants may negotiate their linguistic identi-ties to seek or downplay engagement with them.

In addition to the L1 Arabic-speakers, amongst this group of Chaldeans there are five who are also passing on Arabic to the following generation. This is a rare example of two transposed languages being transmitted in an immigrant situa-tion. Arabic is a commonly-used code in Chaldean and Assyrian in-group social interactions, even amongst those who do not strongly co-identify with Arabic in a personal sense. In the émigré setting of Melbourne, Arabic has some “market-place” value, especially in social networks and in the consumption of locally-produced media. There is a mix of instrumental and affective responses that ac-count for most informants’ continuing use of Arabic. On the basis of these features that encompass the same in-group, group-based theories that forecast attributes that are coterminous with the in-group (e.g. Giles et al. 1977) require modifica-tion: intra- and inter-group features are dynamic and this can lead to the lan-guage of “others” being adopted as one of one’s “own” languages. Where bilin-gualism becomes a habitualized and widespread phenomenon, it can become a group or sub-group attribute. And where differences exist, particular linguistic behaviors can be “pressed into service” to further define them – there is some statistical evidence here and anecdotal evidence from other sources to suggest that Arabic has become an in-group code for Chaldeans, but less so for Assyrians.

Kurdish is less easily quantified in this sample due to the low number of responses that report its use. Eleven of the 66 informants report various levels of spoken proficiency in it; only three report literacy skills in Kurdish. Histori-cally, Kurdish-Assyrian and Kurdish-Chaldean bilingualism preceded bi- or



214 J. Hlavac

tri-lingualism with Arabic as an additive or displacing language. Since the 1950s Arabic increasingly replaced Kurdish as the language of inter-ethnic Chaldean- and Assyrian-Kurdish communication. The re-emergence of Kurdish as an H-language in Iraqi Kurdistan is too recent for informants’ repertoires to have been realigned. Kurdish is not reported as a code used in family, personal, social and educational settings. At the same time, there appears to be under-reporting of knowledge of Kurdish: eleven informants claim proficiency in Section 6.2, but a further seven claim that Kurdish is “familiar” to them in Section 6.5.

Another circumstance that is not suggested in any of the responses, but which is an occasional feature of émigré and homeland political discourse is that of inter-ethnic antagonism between Kurds and Chaldeans/Assyrians. Tables 4 and 5 are striking for the much larger number of responses showing distance from Kurdish than the number showing distance from English, for example. Inter-ethnic discordance has been shown to lead to negative views about others’ lan-guages (cf. Ehala and Zabrodskaya 2011), but with less clear consequences for language use and admissions of proficiency. In this study, many informants have proficiency in Kurdish but this language is clearly perceived as an out-group lan-guage. But this development may be of recent vintage, and may serve as an ex-ample of macro-socio-political changes affecting inter-group features which led to the abandonment of Kurdish as a part of this ethnic group’s linguistic reper-toire. Historically, relations appear to have been different: there is diachronic language contact data to show lexical and morpho-syntactic transference from Kurdish into neo-Aramaic, which, as we know since Thomason and Kaufman’s (1988) seminal work, could only occur in conditions of long-standing bilingual-ism (Chyet 1995).

English is most informants’ newest language. But notions of English pre-date arrival even if many informants had no chance to acquire English in Iraq. English is the language of late 19th century Anglican missionaries who visited indigenous Christian communities to record their language, Aramaic. English was the lan-guage of a colonial power that occupied Iraq for most of the first half of the twen-tieth century until 1958 and the language of British oil companies for whom many northern Iraqis worked as laborers. English represented modernism and engage-ment with the Western world. Twenty-five informants had contact with English as a subject taught at school and it is likely that some informants acquired further skills in English through educational advancement rather than trade or commer-cial activities (due to trade sanctions).

Apart from the unchallenged status of English within most aspects of Austra-lian public life, informants consider acquisition of English not as a strategy to displace other codes but as an activity which is instrumental for their socio-economic engagement and advancement. This is certainly the “top-down” socio-




political message that is conveyed to recently-arrived migrants. The message “on the ground” corroborates this: English is a lingua franca used to facilitate com-mercial and occupational interaction between fellow residents who have diverse cultural and linguistic backgrounds. Use of English is not coterminous with the ethnicity of the other interlocutor and need not be indexical of identification with dominant Anglo-Australian culture. English is certainly now taking on many in-strumental functions in informants’ lives: commercial transactions, internet use and with neighbors. Although the language of the religious domain is usually seen to bear great cultural symbolism, seven informants also attend English-language mass – English is instrumental in expressing religious devotion as an addition to religious devotion expressed in Chaldean. A significant number of informants (15) view English as one of their languages. This shows that instru-mental functions commonly performed in English, together with other socio-political factors which promote this language, can lead informants to developing a close or “personal” relationship with English.

8 ConclusionAs a comparatively recent group of emigrants from Iraq, data from the informants show that only five of the 66 informants had shifted to Arabic in Iraq and that Chaldean and Assyrian, through geographical and/or socio-religious distance, remain well-maintained and robust. Familial, local and religious-based networks ensure their transmission to future generations. This suggests that individuals are embedded within certain group parameters and are greatly influenced by group norms and behavior, to which they contribute in a reciprocal sense. This description of Chaldean and Assyrian maintenance easily fits the ethnolinguistic vitality model based on “groupness” with its notion of entativity (cf. Yağmur and Ehala 2011) and individuals’ behavior being strongly guided by this. But, groups are made up of individual members and the phenomena and actions that indi-viduals experience can also determine their behavior, independent of the group. At the same time, when numbers of individuals within a group share the same changing behavior, this can diversify the group’s features leading to a greater het-erogeneity within the group, and further to the formation of sub-groups, and to re-formulations of group characteristics that include individuals’ exiting and re-entering. Linguistic choices are not only consequences of changing behavior: dif-ferent pre-existing linguistic constellations in a new environment can determine individuals’ and groups’ behavior.

Personal language use and identification with respective codes bear out the bi- and increasingly tri-lingual repertoires of informants. Proficiency, or



216 J. Hlavac

rather use, are strongly associated with situational and environmental cir-cumstances, which, in the context of Melbourne include some old and some newly (re-)established networks – Chaldeans and Assyrians from homeland lo-cality, Chaldeans and Assyrians from outside homeland locality, Iraqi Arabic-speakers, non-Iraqi Arabic-speakers – and with English as the over-arching code of “ambient Australia”, towards which many informants now claim group affiliation.

Migration to Australia and the addition of a third or fourth language to infor-mants’ repertoires show how a different socio-political environment may lead to a further re-alignment of in-group language norms and the inclusion of the new H-language English alongside the former H-language, Arabic, as an additive rather than displacing variety. This paper also shows that group memberships are dynamic and fluid and that for the situation of Iraqi Chaldeans and Assyrians in Melbourne, ethno-religious features, literacy in and the symbolic and instrumen-tal functions of their “homeland” languages form a meaningful basis and context for their maintenance and further transmission.

ReferencesBeykont, Zeynep. 2010. “We should keep what makes us different”: youth reflections on Turkish

maintenance in Australia. International Journal of the Sociology of Language 206. 93–107.Blau, Joyce & Yasir Suleiman. 1996. Language and ethnic identity in Kurdistan: an historical

overview. In Yasir Suleiman (ed.), Language and identity in the Middle East and North Africa, 153–164. Richmond, Surrey: Curzon Press.

Chyet, Michael. 1995. Neo-Aramaic and Kurdish. An interdisciplinary consideration of their influence on each other. In Izre’el Shlomo & Drory Rina (eds.), Language and culture in the Near East, 219–252. New York: Leiden.

Clyne, Michael. 1997. Some of the things trilinguals do. International Journal of Bilingualism 1(2). 95–116.

Clyne, Michael. 2003. Dynamics of Language Contact. Cambridge: Cambridge University Press.Clyne, Michael & Sandra Kipp. 2006. Tiles in a multilingual mosaic. Macedonian, Filipino and

Somali in Melbourne. Canberra: Pacific Linguistics.Department of Immigration and Citizenship. 2007. Community information summary. Iraq-born.

Canberra: Multicultural Affairs Branch & Programme Statistics and Monitoring Section of DIAC. http://www.immi.gov.au/media/publications/statistics/comm-summ/_pdf/iraq.pdf (accessed 5 March 2011).

Dewaele, Jean-Marc. 2011. Self-reported use and perception of the L1 and L2 amongst maximally proficient bi- and multilinguals: a quantitative and qualitative investigation. International Journal of the Sociology of Language 208. 25–51.

Ehala, Martin & Anastassia Zabrodskaya. 2011. The impact of inter-ethnic discordance on subjective vitality perceptions. Journal of Multilingual and Multicultural Development 32(2). 121–136.




Ennaji, Moha. 1997. Berber sociolinguisics. International Journal of the Sociology of Language 123. 23–40.

Evans, Nicholas. 2001. The last speaker is dead – long live the last speaker! In Paul Newman & Martha Ratliff (eds.), Linguistic field work, 217–281. Cambridge: Cambridge University Press.

Fishman, Joshua. 1991. Reversing language shift. Clevedon, UK: Multilingual Matters.Fishman, Joshua. 2001. From theory to practice (and vice versa). In Joshua Fishman (ed.), Can

threatened languages be saved?, 451–484. Clevedon UK: Multilingual Matters.Giles, Howard, Richard Bourhis & Donald Taylor. 1977. Towards a theory of language in ethnic

group relations. In Howard Giles (ed.), Language, ethnicity and intergroup relations, 307–348. New York: Academic Press.

Hanish, Shak. 2008. The Chaldean Assyrian Syriac people of Iraq: an ethnic identity problem. Digest of Middle East Studies 17(1). 32–47.

He, Agnes Weiyun. 2006. Toward an identity theory of the development of Chinese as a heritage language. Heritage Language Journal 4(1). 1–28.

Hlavac, Jim. 2003. Second-generation speech. Lexicon, code-switching and morpho-syntax of Croatian-English bilinguals. Bern: Peter Lang.

Le Page, Robert Brock & Andrée Tabouret-Keller. 1985. Acts of identity: Creole-based approaches to language and ethnicity. Cambridge: Cambridge University Press.

Li, Wei & Hua Zhu. 2010. Voices from the diaspora: changing hierarchies and dynamics of Chinese multilingualism. International Journal of the Sociology of Language 205. 155–171.

Liebkind, Karmela. 1999. Social psychology. In Joshua Fishman (ed.), Handbook of language and ethnic identity, 140–151. New York: Oxford University Press.

Marfleet, Philip. 2007. Iraq’s refugees: “exit” from the state. International Journal of Contemporary Iraqi Studies 1(3). 397–419.

Odisho, Edward. 1993. Bilingualism and multilingualism amongst Assyrians. A case of language erosion and demise. In R. Continiu, F. Pennachiatti & M. Tosco (eds.), Semitica, 189–200. Torino: Silvio Zamorano Editori.

Odisho, Edward. 2004. Assyrian (Aramaic). A recent model for its maintenance and revitalization. In A. Panaino & A. Piras (eds.), Melammu Symposio IV, 183–196. Milan: Mimesis Edizioni.

O’Mahony, Anthony. 2004. Christianity in Modern Iraq. International Journal for the Study of the Christian Church 4(2). 121–142.

Paulston, Christina Bratt. 1994. Linguistic minorities in multilingual settings. Amsterdam: John Benjamins.

Pitrus, Widad. 2009. Catholics: Chaldeans. In James Jupp (ed.), The encyclopedia of religion in Australia, 276–279. Cambridge: Cambridge University Press.

Sengstock, Mary. 1982. The Chaldean Americans. Changing conceptions of ethnic identity. Detroit: Center for Migration Studies.

Skutnabb-Kangan, Tove & Desmond Fernandes. 2008. Kurds in Turkey and in (Iraqi) Kurdistan: a comparison of Kurdish educational language policy in two situations of occupation. Genocide Studies and Prevention 3(1). 43–73.

Smolicz, Jerzy. 1981. Core values and ethnic identity. Ethnic and Racial Studies 4. 75–90.Stoessel, Saskia. 2002. Investigating the role of social networks in language maintenance and

shift. International Journal of the Sociology of Language 153. 93–131.Tabouret-Keller, Andrée. 1997. Language and identity. In Florian Coulmas (ed.), The handbook of

sociolinguistics, 315–326. Oxford: Blackwell.



218 J. Hlavac

Thomason, Sarah & Terrence Kaufman. 1988. Language contact, Creolization, and Genetic Linguistics. Berkeley: University of California Press.

Woods, Anya. 2004. Medium or message? Language and faith in ethnic churches. Clevedon, UK: Multilingual Matters.

Yildiz, Efrem. 1999. The Assyrians. A historical and current reality. Journal of Academic Assyrian Studies 13(1). 15–30.

Yağmur, Kutlay & Martin Ehala. 2011. Tradition and innovation in the Ethnolinguistic Vitality Theory. Journal of Multilingual and Multicultural Development 32(2). 101–110.



international journal of the sociology of language volume 2012, issue 217 (sep 2012)

Documents