Kurdish: An Indo-European Language

By Siamak Rezaei Durroei

Kurdish is a member of the Indo-Iranian language group which is a branch of the Indo-European family, the largest language family in the world.
Kurdish (like Persian) is grouped under the Western Iranian branch of Indo European languages. The Eastern branch of Iranian languages consists of languages such as Scythian and Avestan and more recently Pashto spoken by Afghans.

Indo-European | Indo-Iranian ______|______ | | Indian Iranian | ________|___________ | | | | | | Sanskrit Western Eastern ____|____ ____|____ | | | | Old Persian Median Scythian Avestan Figure from S. Karimi( 1989) Phd Dissertation
Kurdish, Ashkani (Parthian) Pahlavi, Baluchi and Mazandarani are from the western branch of the Northern group of Iranian languages, while Persian and middle Persian (Sassanid Pahlavi) belong to the south branch (see [Kalbasi83]). In old era western branch consisted of languages such as Old Persian and Median. It has been claimed that Kurdish is closer to old Persian than Median.

From another perspective, Kurdish (western) and Osetin (Eastern) form the Northern branch of Iranain languages. Osetin is spoken by around 600,000 in Turkey (Bitlis, Erzerum, Kars, Anatalya) and Georgia. There is no reliable source for the population of Kurdish speakers. Kurds are estimated from 20-40 Millions. A percentage of these are native speakers. The following, based on Hassanpour(1992) shows the estimated number of Kurds in year 2000 projected from Kurdish Nationalist references:

 Country    Population (M.)     Kurds M.(% of total)
 ----------------------------------
 Iraq[c]        23.753          6.65 (28%)
 Turkey[a]      67.748         15.58 (23%)
 Iran[b]        74.644         11.94 (16%)
 Syria[d]       17.328          1.90 (11%)
 USSR[e]                        0.73
 ----------------------------------
 Total                         36.800 million

   --  Population of Kurds in 2000  ---

  [a] Kendal (1980)      [b] Ghassemlou (1980) [c] Vanly(1980)
  [d] Nazdar (1980)      [e] Kendal(1980)
Iranian _______|________ | | Southern Northern ____|____ ____|____ | | | | West East West East | | | | Persian Pashto Kurdish Osetin

Kurdish vs. Persian

Kirmanji-Kurdi(1) dialects (Kurmanci/Sorani and Zaza/Gorani (Hawrami) among others) have preserved many features of older Iranian languages. But grammatical features such as ergativity or clitic movement exist in Kurdish which is not reported in Modern Persian.

Kurdish exhibits different aspects of tense/aspect ergativity in its dialects (Bynon 1979). Ergativity has been also reported in Eastern Iranian language, Pashto (Kalbasi 1988), but ergativity never fully developed in Persian and its former ancestors and the existance of eragtivity in Pashto might be because of language contact with Indian and Pamir languages, while in Kurdish it might be a result of influence of caucasian languages (such as hurrian) which were spoken in the region. The origin of ergativity in Kurdish needs further research (see (Pirejko79) and (Steiner79) for origins of ergativity).

The interesting thing about ergativity in Kurdish is that the two major dialects of Kurdish Kurmanji (North) and Sorani (south) exhibit different aspects of ergativity (Bulut 1995). The past transitive ergativity in Sorani in parallel to clitic movement is discussed in (Friend 1985). The (past transitive) ergativity in Kurmanji is a formal structure manifested by noun as well as verb morphology and verb agreement rules (Matras 1992). But according to (Dorleijn92,97) Ergativity in Kurmanji dialects is in danger and is eroding.

The existance of specificity-object marker "ra" in persian gives more flexible word order to Persian in contrast to Kurdish dialects which lack such a marker. Markers with analogous role exist in Turkish, Urdu and the existance of such a marker in Persian might be as a result of contact with Turkish.

Another difference between Kurdish and Persian is the existance of clitic movement phenomenon in Sorani verbal construction which is closely related to ergativity in Kurdish. To my knowledge, this feature has not been reported for old Persian.

Gender distiction has also been preserved in dialects such as Kurmanji which does not exist in Modern Persian.

Kirmanji-Kurdi Dialects

               
Kirmanji-Kurdi ____________|______________ | | Kirmanji (Modern) Hawrami (Old) | | .--------|-------. .-------------. [South] [Central] [North] [North] [South] Kirmanshahi Sorani Kurmanji Dimili(Kirmanjki) Hawrami
Among Kirmanji-Kurdi dialects, the Kurmanji has been reported as the most archaic dialect in its phonetic and morphological structures [MacKenzie61]. Garzoni, Lerch, Rhea, Chodsko, Beresin, Hornli, Peter and Fossum refer to Sorani as the purest and best Kurdish [Fossum19] (Fossum particularly refers to the group of dialects which was spoken by the Bebeh tribes in the districts of Suleimania). Soane(191?) refered to the Sorani dialect spoken by Mokri tribe (south of Lake Urmia) as the purest dialects of Kurdish in its accent and grammatical form.
Some Kurds believe that Gorani is the oldest Kurdish dialect, this is in contrast to some recent proposals that put Zaza and other subdialects of zaza-Gorani (Hawrami) group as non-Kurdish. What is certain is that all Kirmanji-Kurdi dialects belong to the North-West branch of Iranian languages (see Ethnologue on northwestern Iranian languages and languages spoken in Iran ).

The reader of Ethnologue should be aware that although Ethnologue is an invaluable source of language statistics, but the authorities for its facts are mostly Summer Institute of Linguistics (SIL) Bible translators. The following paragraph highlights the issue further:

What Ethnologue, in its various editions, has tried to do is to identify all the linguistically distinct traditional societies of the world; it incidentally includes all languages, including the 200 or so spoken by truly huge numbers of people. The user of Ethnologue should be aware of certain axes that are being ground. Ethnologue tends to exaggerate the number of languages in the world. A linguist who attends to linguistic structure only, leaving speaker attitudes aside, will recognize fewer than 50 per cent of the number of distinct languages that are recognized in Ethnologue. For example, Ethnologue doubles the number of languages that I, as a linguist who with colleagues has done a dialect survey of the whole of Guatemala, recognize for that country. I recognize about twenty-four languages including Spanish, while Ethnologue gives fifty-four. This proportion (`reality' 1 : SIL 2) may hold in other parts of the western hemisphere, and in some areas, such as Mexico and Peru, Ethnologue is even further from what I accept as reality. While we may concede that SIL is reporting on ethnolinguistic and sociolinguistic difference, we should not accept that low intelligibility of spoken and/or written language between communities is incontrovertible evidence for the existence of distinct languages. This is not a trivial disagreement, because while speaker attitudes can change and will often differ from individual to individual within a community, the linguistic structures do not change very readily and hence are the primary phenomena to be accounted for. I used to (and still do) think of myself as a `splitter, not a lumper', but I need to distance myself from `those who multiply entities needlessly', which is a tendency that is apparent in Ethnologue
[Kaufman 1994]

Number of Kurdish Speakers

According to a statistics from World Evangelization Research Center (available from Batonic World Prayer Center), the percentage of Kurds and Kurdish speakers in different countries (North/Central dialect) by the year 2000 will be:

Country   North(%)  Central(%) General(%)      Total(%)  Total(Kurds in Million)
--------------------------------------------------------------------------
Iraq        6.5%      8.5%      6.5%              21.0%           4.98
Turkey**    8.8%       -        8.0%              16.8%          11.38 
Iran*       0.4%      4.5%      6.0% + [7.1%Lori] 10.9% +[7.1]    8.13 + [5.29 Lori]
Syria       7.3%       -        7.3%               7.3%           1.27

Lebanon     4.0%       -          -                4.0%           0.1309
Armenia     2.0%       -          -                2.0%           0.0648
Georgia     0.6%       -          -                0.6%           0.0341   
Kyrgyzstan  0.33%      -          -                0.33%          0.0170
Azerbaijan  0.2%       -          -                0.2%           0.0139
Kazakstan   0.15%      -          -                0.15%          0.0273
Afghanistan   -        -        0.1%               0.1%           0.0267  

--------------------------------------------------------------------------
All Kurds  in year 2000                                =   26.07 M. + [5.29 Lori]
               
* It seems that this statistics does not include the 1-2 million
   Kurmanji (North) speakers in Khorasan (Iran).
** This figure also includes 1,165,000 Dimili Kurds and 169,400
   Alevica-Kirmanjki speakers  in Turkey in year 2000.
*** The statistics does not include Kurds in Diaspora and if we
   contrast it with the figures by Kurdish nationalists, the figure
   for each country is around 6% less (e.g. 23 - 16.8 in the case of
   Turkey) than what is claimed by nationalists. This might be as  a
   result of the difference between the number of Kurdish speakers and
   the number of Kurds (but not necessrily speaking the language).



The only official data on dialects can be based on the census in Iraq (1947). Hassanpour (1992) based on Edmonds (1957) gives this statistics for each liwa in 1947:
Liwa     Total Population  Kurds   Dialect  (% of liwa) (% of Kurds)  
-----------------------------------------------------------------------
Sulaymania   222,700      222,700    Sorani   100.0%     24.7%
Arbil        240,500      218,000    Sorani    91.0%     24.3%
Kirkuk       285,900      151,575    Sorani    53.0%     16.8%      
Mosul        602,000      210,970    Kurmanji  35.0%     23.4%
Diyala       110,200      27,360    Fayli/     65.6%      8.0%
(Khanaqin and Mandali)             Kirmanshahi
Other (baghdad) n.a.      23,400     Unknown    -         2.6%
----------------------------------------------------------------------
Kurdish Population       900,000                          100%
The following data based on Hassanpour(1992) gives further information about Kurdish speakers:

- According to official Turkish census (1935) 9.3% were Kurdish speaker by Mother Tongue. In 1965 this % was 7.1% of total population.

- According to official Iranian census (1956) 5.3% were Kurdish speaker and 5.6% Lori speaker. This contrast with 1949-53 Geograhical Dictionary in which 8.8% were Kurdish speaker and 5.2% Lori speaker. The underestimation in 1956 is political.

- According to official Iraqi census (1957) 16.44% were Kurdish speaker. This contrasts with 2.15% Turkish and 0.96% Assyrian and Kaldanian speaker.

- No official figure is available for Syria. The French mandate official gave 110,000 as moderate estimate in the late 1930s.

- Official figures in USSR put the number of Kurds+ Yazidis in 1926 as (54,661 + 14,523), the number of Kurds in 1959, 1970 and 1979 were respectively 58,7999, 88,930 and 115,858. (34,4 + 96.4), 89.9, 87.6 and 83.6% claimed Kurdish as Mother tongue respectively.

End Notes

(1) I have used a term "Kirmanji-Kurdi" to refer to all dialects which are spoken by Kurds. See Nebez (1976).

(2) Historically Lors are also considered as Kurd, but their language which is closer to Southern Kirmanji (Kirmanshahi) is claimed to be part of South-West branch, alongside Persian. In Iraq they consider themselves as Kurd.

(3) See zaza-kirmanc home page for some references on Zaza.

General References

Fossum, L. O.; A practical Kurdish grammar [Sorani], The Inter-Synodical Ev. Lutheran Orient-Mission Society, Minneapolis, Minn. 1919.

Hassanpour, Amir; Nationlaism and Language in Kurdistan, 1918-1985, Mellon Research University Press, 1992.

Kaufman, Terrance; The native languages of Latin America: general remarks, In "Atlas of the World's Languages" (eds. Christopher Moseley and R.E. Asher), pp. 31-33, 1994.

Kalbasi, Iran; Kurdish Dialect of Mahabad, Institute for Cultural Research and Study, 1983, Tehran.

Karimi. Simin, "Aspects of Persian Syntax, Specificity, and the Theory of Grammar", PhD dissertation, univ. of Washington, 1989.

Leezenberg, Michiel; "Gorani Influence on Central Kurdish: Substratum or Prestige Borrowing?", ILLC prepublication Series X-93-03, 1993.

MacKenzie, D.N.; "The Origins of Kurdish", Transactions of the Philological Society: 68-86, 1961.

Nebez, Jemal, Ziman^i Yekgirt^uy Kurd^i (Back cover: Towards a Unified Kurdish Language). Mamberg. W. Germany: National Union of Kurdish Students in Europe, 1976. Soane, E. B.; To Mesopotamia and Kurdistan in Disguise, 19??-15.

For Zaza Litreture see http://members.tripod.com/~zaza_kirmanc/index.html

Ergativity References

Bynon, T. " The Ergative Costruction in Kurdish", Bulletin of the School of Oriental and African Studies, Volume XLII,part 2, 1979

Bynon, Theodora, From Passive to Active in Kurdish via the Ergative Construction Papers from the 4th Int. Conf. on Historical Linguistics, Amsterdam: John Benjamins. PP. 151-163, 1980.

Bulut, Christiane Ergativity in Sorani/Kurmanji, University of Uppsala, 1995

Dorleijn, Margreet, Will Split-Ergativity survive in Kurmanji Kurdish spoken in Turkey? 1992 [see Leezenberg93]

Dorleijn, Margreet, The decay of ergativity in Kurmanci: Language internal or contact induced?, Studies in Multilingualism, Tilburg univ. Press, 183 p.,1997, ISBN 90-361-9567-5.

Friend, Robyn Christine. Some Syntactic and Morphological Features of Suleimaniye Kurdish, PhD thesis, UCLA, 1985 - x-19-238440-5

Friend, Robyn Christine. Bound Clitic Objects of Independent Adpositions in Suleimanye Kurdish. MESA Conf., Nov. 1988 http://www.cogsci.ed.ac.uk/~siamakr/Kurdish/Papers/friend88.html

Kalbasi, I. " Ergative in Iranian Languages and Dialects", Iranian Journal of Linguistics, Vol. 5, No. 2 Autumn & Winter 1988 (in Persian)

Matras, Yaron. Ergativity in Kurmanji (Kurdish), Orientalita Suecana XLI-XLII (1992-1993) pp. 139-154.

Matras, Yaron, Ergativity in Kurmanji (Kurdish). Notes on its use and distribution. Arbeiten zur Mehrsprachigkeit 41/1990. Hamburg: Germanisches Seminar

Matras, Yaron, Coreference, ergativity, and clause combining in Kurmanji, to appear in Studies in Language.

Pirejko, L. A. On the Genesis of the Ergative Construction in Indo-Iranian, pp. 481-488 in F. Plank(1979) Ergativity: Towards a Theory of Gramatical Relations, Academic Press.

Stiener, G., The Intransitive-passival Conception of the Verb in Languages of the Ancient Near East, pp. 185-216 in F. Plank(1979) Ergativity: Towards a Theory of Gramatical Relations, Academic Press.

Siamak Rezaei Durroei, Feb. 1998