The Dialects of Kurdish Alexander Johannes Edmonds Ruprecht-Karls-Universität Heidelberg As to the Kurdish language, so little is known of it that it has been described as a corrupt dialect of Persian and Arabic, “a kind of dog-Persian," and “a degraded old Persian dialect." It is none of these. (Soane 1910: 387) It has been more than a century since Ely Bannister Soane, a noted British orientalist, wrote these words. Sadly almost half a century later Cecil J. Edmonds, a British political officer and scholar, felt it still necessary to make practically the same clarification at the beginning of his classic work Kurds, Turks and Arabs (1957: 7). While Kurdish became an official language within Iraq a decade after Soane wrote his words and has received increasing recognition within the international community since then, it remains quite overlooked as a source of study in Europe, with a few notable exceptions, and studies of its dialect remain comparatively few in comparison to other Middle Eastern languages. In this paper an elementary sketch of Kurdish and its dialects will be undertaken before the question of language and nationhood is examined in brief. First, however, a few introductory remarks are here necessary. It must be remembered from the onset of this work that ‘Kudistan’ remains an extremely sensitive term; it will be used in two senses here: ‘Iraqi Kurdistan’ refers to the Kurdish Autonomous Region in Iraq, while ‘Kurdistan’ refers to an envisioned extent of Kurd-inhabited territory in the Middle East, including besides Iraqi Kurdistan some parts of Turkey, Syria, Iran and other neighbouring countries. The Kurdish language will be defined in the following paragraphs but it is worth briefly pausing over the question of dialect. In defining language and dialect one must consider however briefly the great Max Weinreich’s axiom that ‘a language is a dialect with an army and navy’1; while this reflection is naturally somewhat facile2, there is a grain of truth to it inasmuch as a language must possess power of some nature to be elevated from the status of dialect. For the purpose of this paper no overly complex definition is necessary and it will suffice to define a language as a distinct entity which is codified, written, spoken by a distinct population group with some measure of nationhood, and, as a nod to Weinstein, possesses some power, prestige, or influence. A dialect will in turn have few if any of these features. As will be demonstrated in the following paper, however, Kurdish and its dialects are not so clear-cut as they may initially appear. Finally, two provisos must be established before the topic of Kurdish dialects is broached. 1 ‘a shprakh iz a dialekt mit an armey un flot’, (Weinreich 1945), ironically spoken in Yiddish, a language whose own position in this regard has been debated. 2 If Iraqi Kurdistan possesses a navy then its influence is negligible 1 Firstly, it must be kept in mind that the sources used here are largely western. They are the product of western scholarship and hence to some degree an act of interpretation. The study of Kurdish is a relatively young field, starting with a treatise written by Maurizio Garzoni in 1787, and Kurdish lacks the extensive fieldwork that many other Middle Eastern languages have received from western academia, leaving our understanding of many dialects somewhat patchy. Secondly, Kurdish as a written language can only be traced back to the 16th Century, making dialectical research extremely difficult (Paul 2008). Moreover, population movements further confuse the picture. Despite this, some excellent work has been accomplished regarding the evolution of Kurdish, most notably by Mackenzie (1961). Kurdish will now be described in brief. Kurdish is an Indo-European language, belonging more precisely to the Western Iranian language family, although this has been at times challenged (cf. ibid.). It is closely related to Persian but not, as has often been claimed in the past, a dialect of the aforementioned. Estimates of the number of Kurdish speakers vary between 20 million and 40 million. The diaspora of Kurds naturally somewhat influences this. Despite this large number of speakers, the only country to officially recognize Kurdish as a national language is Iraq, which has theoretically done so since its inception in the 1920’s. The dialects of Kurdish will now be briefly examined. The three most important subdivisions of Kurdish are the Northern (Kurmanji), spoken predominantly in eastern Turkey, Central (Sorani), spoken in northern Iraq, and the Southern Group (eg. Kermanshahi) spoken for the most part in some regions of Iran (Allison 2007). The differences between the main dialects are in some cases so wide that it has been a matter of dispute over whether in fact they constitute separate languages in their own right. Kurmanji is spoken by Kurdish communities in Turkey, Syria, Lebanon, Georgia, Armenia, Azerbaijan and in the far north of Iraqi Kurdistan in the district of Badinan and in Jebel Sinjar. Hence it is often referred to as Badini or Badinani. Kurmanji has gender differences in nouns (Mackenzie 1954), something Sorani lacks, and a case system with a direct, vocative, oblique case, the last of which fulfils dative and accusative functions. It also displays more ergative tendencies than Sorani. The agent of a transitive verb occurs in the oblique case in the past tense. NW, Badini, and former USSR dialects within this group are discernible through differing grammar (Allison 2007). Sorani is named for the province from which it originated, Soran, and is also sometimes referred to as Middle Kurmanji or Central Kurdish. It is also known as Mukri. As Soane noted: The Mukri claim that their dialect is the most ancient of all, and while its antiquity is probably not greater than that of its neighbours, its excellent preservation of ancient forms gives it a claim to be considered the standard by which to compare other dialects. (1920: 375) During the early British Mandate of Iraq Sorani was promoted as the dialect of choice for the governance of Iraqi Kurdistan by figures such as Cecil J. Edmonds and Taufiq Wahby and has since largely retained this position. While Sulemani Kurdish is the most prestigious form of this dialect, in the past that of Mahabad in Iran almost overtook it in this regard (Allison 2007). Sorani has the definite noun suffix –eke and a set of pronominal suffixes, and is claimed by many to have lost its ergativity and to have become ‘accusative’ by nature. Mackenzie additionally notes the innovation of a ‘transitive past’ construction in Central Kurdish (1961). 2 There are also a number of southern dialects, such as Kermanshahi, Kalhori, Luri and Lakki. These are mostly found in Iran, although they are also spoken in Khanaqin and also in cities by the Fayli Kurds. Edmonds noted the difficulty in distinguishing these various dialects and that Sulemani Sorani seems to merge gradually into the Kermanshahi and Lakki dialects (1957: 10). Lakki’s position as a dialect of Kurdish could be debated, however, it is close enough to Kurdish that a reasonable claim can be made that it is in fact a dialect (Allison 2007). Zazaki, also known as Dimlî, is a language closely related to Kurdish but not a dialect of it as has been claimed. Gorani falls into the same category. While both languages are close to Kurdish, they have their own distinct heritage and their many phonological, phraseological and syntactic similarities are a result of their long relationship with Kurdish, during which these languages have influenced each other tremendously (ibid.). Gorani was the high literary language during the period of independence that the Ardalan family achieved from the Safavids in 16th Century and a great deal of poetry was composed in it. According to the hypothesis of Mackenzie, Gorani played a great role in the formation of the modern Sorani dialect. This has, however, been contested by Michiel Leezenberg, who claims that the influence of Gorani upon Sorani was largely in the form of prestige vocabulary and superficial in nature (1993). Gorani is also used by the Ahl al-haqq or Kakai and is the language in which their sacred texts, or kelams, are written. Gorani as a literary language can be distinguished by its syntax, which differs from that of conventional Kurdish dialects (MacKenzie 2002). It should be swiftly noted that the influence of prestige upon these dialects is considerable. Hence, we have a situation in which all of the previously mentioned languages and dialects are spoken by Kurds but not all of them in turn constitute Kurdish. Certainly within Iraq each dialect of Kurdish is mutually intelligible however it is more likely that a Kurmanji speaker will understand a Sorani speaker than vice-versa (Allison 2007). Allison proposed that when handling these dialects and languages the phrase the ‘Kurdish Languages’ should in fact be used (ibid.), not at all a bad idea, but as will be discussed next, a politically sensitive one. As discussed in the introduction of this paper, a common language is one of the defining characteristics of nationhood, and naturally a key part of the argument for any united Kurdistan in the broadest sense of the word. This has been hampered by the mutual unintelligibility of many dialects. Indeed, academic opinion on their relations to one another is often widely split or fiercely contested. For example, Philip Kreyenbroeck has contested that Kurmanji and Sorani are as far apart linguistically as English and German (1992), sparking concern among Kurdish nationalists that the case for a unified nation would be taken less seriously as a result (Allison 2007). There have been attempts to create a unified Kurdish language, most notably by Nebez (1976) but none have been entirely successful as of yet. This may be necessary if a unified Kurdistan in the broadest sense is to come about. Even within Iraqi Kurdistan standardisation of dialect has met with difficulties; the teaching of Sorani in some Badini areas was originally met with derision after 1991 (Allison 2007), now both are taught. English is now favoured as a second language and Kurds appear to be increasingly distancing themselves from Arabic. As the case of Gorani and Sorani has demonstrated, prestige is a moving force within Kurdish dialects. Making proper use of this factor should be a key part of any attempt towards a unified language, however, the cultural wealth that is dialect must also be protected 3 in such a case so as to avoid so iconoclastic a language reform as that which Turkish underwent, which has left recent generations of Turks remarkably estranged from their own heritage. Yet before a unified language can even be attempted, a unified script is a necessity. A modified PersoArabian script is still used in Iraqi Kurdistan today; however, various attempts have been made to Romanise Kurdish since the 1930’s (e.g. Edmonds 1931 & 1933; Wahby 1925 & 1929; Edmonds & Wahby ) though this has at times faced opposition, at one stage memorably put as ‘Latînî ye, Ladînî ye’ (Latinization is irreligiosity) by religious protestors (Nebez 1976:86). More recently, the movement towards Romanisation has been renewed, one excellent example being the Kurdish Unified Alphabet initiative3. With a mutual alphabet, Kurdish speakers would be much more capable of grass-roots political discourse across boundaries, especially in this modern age of the Internet, which played an astounding role in the Arab Spring only last year. The results of such discourse can only be speculated but it would only aid the Kurdish nationalist cause. In conclusion, it can be seen that within the study of Kurdish dialect there is still much to be done within the study of these dialects, as noted at the beginning of this paper. This work could only hope to briefly sketch the dialects of Kurdish, however, the factor of prestige has been singled out as a defining factor in its dominant dialects and it has been indicated that this may be a motivating force in the adoption of a single Kurdish language. Sadly it should be mentioned in brief that in the form of frequent violence in eastern Turkey Weinstein has been proven somewhat correct. It can only be hoped that a resolution to this conflict can be found without further bloodshed on either side. As a brief aside, while occidental scholarship has provided a basis, increased participation of Kurds in the study of their own language is vital; since the 1960’s this has grown a great deal (e.g. Hassanpour 1992). Importantly these studies often situate themselves within the greater field of Iranology (Karimi 2012, to quote an example). An increased interplay between western and Kurdish academics can only aid the field of Kurdish dialectology. Politically Kurdish language and dialect has never been so important an issue, with the capacity to change the political landscape of the Middle East. While the role of a unified Kurdish language or alphabet can fulfil both academic and nationalistic goals simultaneously, it is, however, imperative that the wider academic community itself remain as impartial as possible in this regard, difficult as this may be. As a final reflection, one might return to a remark, this time pessimistic, of Soane’s written almost a century ago: It must be admitted that the acquisition of any one of the many dialects is attended with great difficulties, and it is not within the powers of all of us to spend long periods in Kurdistan, nor do I think many, even linguists, would look upon such a comfortable sojourn with any feeling of pleasurable anticipation. (1913: viii) One can only hope that this coming generation of scholars will admirably prove him wrong. 3 See http://www.kurdishacademy.org/?q=node/2 4 Bibliography Allison, C., 2007 ‘‘The Kurds are alive’: Kurdish in Iraq’, in Postgate, J. N. (ed.) Languages of Iraq: Ancient and Modern, Cambridge: Cambridge University Press Edmonds, C. J., 1931 ‘Suggestions for the Use of Latin Character in the Writing of Kurdish’ The Journal of the Royal Asiatic Society of Great Britain and Ireland, No. 1, pp. 27-4 1933 ‘Some Developments in the Use of Latin Character for the Writing of Kurdish’, The Journal of the Royal Asiatic Society of Great Britain and Ireland No. 3, pp. 629-642 1957 Kurds, Turks and Arabs: politics, travel, and research in north-eastern Iraq, 1919-1925, London Edmonds, C. J. & T. Wahby, 1966 A Kurdish-English dictionary, Oxford: Clarendon Press Garzoni, M., 1787 Grammatica e Vocabulario della Lingua Curda, Rome Hassanpour, A., 1992 Nationalism and Language in Kurdistan, 1918-1985, San Francisco: Mellen Research University Press Karimi, Y., 2012 ‘The Evolution of Ergativity within Iranian Languages’, Acta Linguistica Asiatica, Vol. 2, No. 1 Kreyenbroeck, P. G., 1992 ‘On the Kurdish Language’, in Kreyenbroeck, P. G. & S. Sperl, The Kurds: An Overview, London & New York: Routledge, pp. 63-83 Leezenberg, M., 1993 Gorani Influence on Central Kurdish: Substratum or Prestiege Borrowing?, home.hum.uva.nl/oz/leezenberg/GInflCK.rtf, accessed 15/08/2012 MacKenzie , D. N, 1954 ‘Gender in Kurdish’ BSOAS 16, pp. 528-41 1961 ‘The Origins of Kurdish,’ TPhS ,. Kurdish Dialect Studies, London Oriental Series 9 & 10, Oxford: Oxford University Press, pp. 68-86 2002 ‘Gurāni’ Encyclopaedia Iranica, Vol. XI, Fasc. 4, pp. 401-403 Nebez, J., 1976 Zimanî Yekgirtûy Kurdî "Towards a Unified Kurdish Language", Bamberg: National-Union Kurdischer Studenten in Europa Paul, L., 2008 ‘Kurdish Language I: History of the Kurdish Language’ Encyclopaedia Iranica, http://www.iranicaonline.org/articles/kurdish-language-i, accessed 15/08/2012 Soane, E. B., 1910 To Mesopotamia and Kurdistan in Disguise, Boston: Small, Maynard & Co. 1913 Grammar of the Kurmanji or Kurdish Language, London: Luzac & Co. 5 Wahby, T., 1925 ‘Kurdiyekeman be con hirûfék ú con binûsîn?’ “In what character and how should we write our Kurdish?”, Diyarî Kurdistan, No. 5, pp. 5-6, No. 6, pp. 5-6. 1929 Destúrî Zimanî Kurdî “Grammar of the Kurdish Language”, Baghdad: Dar al-Tiba'a al-Haditha. Weinreich, M., 1945 ‫דז ר‬ ‘Der YIVO un di problemen fun undzer tsayt (‫דר‬ )’ in YIVO Bleter, Vol 25, No. 1 ‫ד‬ ‫ר‬ 6