Harvard Ukrainian Studies a8, no. (aoo6): a;.

Te Turkic Etymology of the Word Qazaq Cossack

Orvii Pvi:s
Cocvvio :nv Tuvic v:vroioov of the word Qazaq, there is no
consensus among scholars. While the historians (Hruevskyj I9u9, 76, Stkl
,, ) and the etymologists of the Slavic languages (Berneker ,a, ,6,
Brckner ,;, a6a, Vasmer and Trubaev ,6;, ,, as well as Melnyuk ,8,
,,6) accept without any reservationbut also without proofthe Turkic
etymology of this word, some leading Turkologists have their doubts. Rsnen
(,6,, a), in his etymological dictionary of the Turkic languages, gives no
etymology in his entry qazaq, and Doerfer (,6;, 6a68) states atly: eine
sichere Etymologie fr das Wort [qazaq] existiert nicht. Te same opinion
was expressed by Menges (,;,, ,6, nos. , a).
From the formal point of view the word qazaq can be easily explained. It is
a deverbal noun in i-AKi from the verb qaz-, as are ksk piece from ks- to
cut, jatak bed from jat- to lie down, qonaq palace, guest from qon- to pass
a night, and srk runner from sr- to run (Zajczkowski ,a, 66).
As we can see, the su x i-AKi forms nouns expressing the result of the
action, instruments and the actor. Te problem is that the verb qaz-, which is
also attested in the older Turkic literary languages, has the meaning to dig, to
dig out. Also the noun qaz-aq theoretically must have had the meaning the
digger (Clauson ,;a, 68o). And in fact such meaning is attested to in ,
(Sreznevskij 8,, cols. ;;, cf. Doerfer ,6;, 68). It means that there
existed some Turkic languages where the said meaning was the basic one.
It appears especially in Eastern Europe. But I shall not dwell on it here since
Larysa Pritsak dealt with it in a recent paper (aoo6).
But I would like to propose the following hypothesis concerning the word
qazaq. Te given meaning of the word qazaq entered the given Turkic language
at the time when it commonly used that particular meaning. It became exis-
tent when the literary language of the time included it in its vocabulary.
Hence one Turkic linguistic group preserved the words qazaq and qaz- in
the meaning the digger, to dig out, and this meaning of qaz- has been pre-
served in the majority of the old and new Turkic literary languages.
a8 vvi:s
Annemarie von Gabain (,6o) has convincingly proven that the verb qaz-
had in the Old Turkic language of the Orkhon inscriptions (8th century) the
same meaning as the verb qazan- to gather, conquer. Tis meaning had
remained in the Eurasian steppe, but since it was connected with the nomads,
who were replaced as rulers by the sedentary dynasties of the Ujurs (;o8o)
and thereafter with the sedentary Karakhanids (8oaao), it was not included
into the Turkic literary language of that time. We have an excellent comparative
dictionary of the Turkic languages from o;;, but it neglects the languages of
the nomads. It was compiled by the Karakhanid prince Mah, md al-Kar
(,8, ). Hence the meaning of qaz- and qazaq- from the nomadic era was
not included into his Dvn lut at-Turk.
In the Muslim world there were two types of slaves. Te rst type was called
abd (pl. ibd), slave, son of the slaves,
I
who were usually black and were
customarily used for heavy work and as domestic servants. Te second were
mamlks (purchased slaves), purchased from the slave market. Tey were chil-
dren of free white people who originated from the Polovcian (Qipaq) steppe.
Ethnically they were Polovcians, Cherkes, or the ancestors of the Ukrainians.
Tey were purchased at a young age and kept in special barracks where they
were instructed in military arts. Te Muslim rulers who employed the mamlks
were the dynasties of the Ft, imids (,o;;) and Ayybids (6,ao). In
ao the mamlks dethroned the last Ayybid and their commander took for
himself the rulers title sult. n and ruled over Syria and Egypt. Te rule of the
mamlks lasted until ;, when the Ottoman sultan Selm conquered both
Syria and Egypt.
In order to communicate with the mamlks, the Arabs of Egypt and Syria
compiled several Arabic-Polovcian glossaries. One of the oldest such glossaries,
written in n 6io a, was published by Martin Houtsma in 8,. Tere
is one lexeme qazaq which is translated into Arabic as al-muj arrad, i.e. free.
2

Unfortunately both Hruevskyi (,o,) and Stckl (,) had no knowledge
about the existence of the Arabic-Polovcian dictionary, published in 8,, and
even contemporary historians repeat that the lexeme qazaq rst occurred in
the Codex Cumanicus of o.
Concerning the date of the Codex Cumanicus, one has to mention the
special study of the Hungarian Turkologist Gyrgy Gyry (,a), which also
remains unknown to historians. Gyry established that the Codex Cumanicus
was a collection of several texts, all from the th and th centuries (ibid.,
o). Te rst of them (where the word Cosac appears) was written in a,i,.
It is a Middle-Age Latin, Persian, Polovcian glossary. It was probably written
in Solchat in the Crimea.
3
In o a copy of it was made, probably in the mon-
astery of St. John in Sarai, the capital of the Golden Horde. Between oo and
o a new copy of it was made that was ultimately owned by the compilers
of the Codex Cumanicus, who included it in the collection at the beginning.
:uvic v:vroioov or qazaq a,
Tus on fol. ov, l. the Medieval Latin gloss guayta guard has the Persian
correspondence naobat guard and the Polovcian correspondence Ghasal
Cosac. Te rst word (an attribute) has not yet been deciphered, the second
is Cossack.
4
In New Persian literature of the period of the Golden Horde (ca. aooo)
there can be found a Turkic loanword qazaq. I should quote here just one case
of its usage. I am quoting from the work of the Persian writer Nat, anz (the
Anonymous of Iskander), written ca. a.
3
He wrote: dar n navh. i dar s.urti
qzq me-gardad, in this region he roamed in the manner of a qzq. Tese
three examples clearly testify that the Polovcians (Qipaqs) had and used the
institution of qazaq. But since the Polovcian literary language never developed,
the Polovcian lexeme qazaq was not included into the vocabulary of the Turkic
literary languages, and did not exist until the sixteenth century.
But the word qazaq and its original meaning came down to us in the nota-
tions of foreigners: Arabs in a, in the Latin of the Genoese a,a,, and
in the New Persian literary language which, on par with the Chinese, was the
o cial language of the Mongolian empire (ao6oo).
Bbur (d. a6) the great Cossack, an excellent statesman, the conqueror
of India as well as a beautiful man of letters who wrote his memoirs (Bbur-
nme) even during a battle became the leading classic of the Turkic Chaataj
literary language.
6
Tis literary language often used the lexeme qazaq in its
meaning as freebooter, one who takes possession. I quote here three passages
from his Bbur-nme: Tulun xvj a molni eki jz qazaq jigitlr bil ilar
ajirdk [We sent the Mongol Tulun with two-three hundred of young qazaq
on the raid] (:,, ll. ,o), mni bil qazaqliqlarda v mihntlrd bolan
kiiler bil jaman ma qila kiriti [we have undertaken Cossack raids with
the people who were with me suering and in want of means of sustenance]
(:o, ll. ,ao), and zm bil qazaqliqlarda bil bolup klgn bglrg v
jigitlrg azs kent v jatal dk brildi [To those princes and warriors who
were with me on the Cossack raids were given cities as well as possessions]
(:aa, ll. o). Te analytical method in the etymologization of Turkic words
was elaborated in the rst third of the aoth century in Berlin by two Ger-
man Turkologists, Prof. Willy Bang and his student (and my teacher) Prof.
Annemarie von Gabain.
In ,6o Prof. Gabain published a very important study dealing with the
Cossacks. I shall not discuss the sociological part of her theory, since this
has already been done by Larysa Pritsak (aoo6), but I will limit myself to the
philological theory. According to Gabain the word qazaq already existed in the
language of the Old Turkic inscriptions of the eighth century. But it suered
a change after the denominal verbal su x iai was added and its voiceless q
changed into voiced so that it became qazaa-. Te deverbal reexive su x
ini was added to this so that the form became qazaan. With the Turkic Mit-
ao vvi:s
telsilbenschwund it became qazan. Some Turkic languages, e.g. Polovcian
(Qipaq and Chaataj), retained the simple form qaz-, whereas the others,
e.g. Orkhon-Turkic, preferred the reexive form qazan-, apparently because
they regarded it as more expressive. Te new lexeme became attractive and
took the meaning of the word qaz-. Te new verb qazan- now received the
meaning to be engaged in raids, to conquer.
Tis word qazan- now played an important role in the political structure
of Old Turkic political life. Te excerpt from the inscription of the Bismarck
of the Steppe, the minister Bilg Tonuquq, from ;6 testies to it. It runs as
follows:
Iltri qaan qazanmasar, joq rti rsr, bn zm, bilg Tonuquq,
qazanmasar, bn joq rtim rsar,
Qapaan qaan Trk sir budun jirint bod jm, budun jm, kisi jm
idi joq rti rti.
Iltri qaan, bilg Tonuquq qazantuq n Qapaan qaan Trk sir
budun joryduqy (bu)
Trk bilg qaan, Trk sir budunu Ouz budunu igid olurur.
(Malov ,, 6)
(N) If Il-teri Qaan [68a6,] had not won, or if he had never existed,
and if I myself, Bilga Tonuquq had not won, or if I had never existed.
(Na) in the political organizations of Qapaan Qaan [6,;6] and
in the land of Trkic and Sir (Sogdian) there would have been neither
tribes nor political body nor human beings at all since Il-teri Qaan and
Bilg Qaan have won the political organization of Trk and Sir of the
Qapaan Qaan has ourished this much. (N) Trkic Bilg Qaan (;6-
;) presently rules and is taking care of the political organizations of
Trk, Sir and also Ouz.
Te lexeme qaz - qazan came into being in the territory of the Old Turks
around o, at the beginning of the new Trk dynasty, which was originally a
body of the Qazaq type. After the fall of the Old Turkic nomadic empire, their
successors, the Manichaean Ujurs (;o8o) and the Muslim Karakhanids
(8oaao) became sedentary states and the designation qazaq was no longer
in use.
And only after Bbur (d. a6) established his empire of the Qazaq type
did the word Qazaq reenter the Turkic literary languages, beginning with his
Chaataj.
Te history of the Turkic word qazaq presents an example of the inclu-
sion of a given lexeme into the literary language provided there is real need
of including it in the lexicon. In this way, the word qazaq from ;o until the
:uvic v:vroioov or qazaq a
6th century was beyond the boundary of a Turkic literary language, so that
even the modern Turkologists presented the view that the word did not exist
in Turkic.
Cociusio
Let me now summarize the results of this investigation. Tere were in nomadic
Proto-Turkic at least two verbs designating to be engaged in raids, to conquer:
qaz- and qazan. Te Polovcians (Qipaqs), a nomadic group active in Eurasia
since the twelfth century (Golden ,,a, a68), used the simple verb qaz-.
Te same was the case with the semi-nomadic Chaataj who emerged in the
fteenth century and became sedentary over the course of the sixteenth and
seventeenth centuries (ibid., o,;). On the other side the Imperial Orkhon-
Turks, as well as the sedentary Central Asian Karakhanids (Clauson ,;a,
68a8), preferred to use the more attractive reexive variant of the same verb,
qazan-, for to be engaged in raids, to conquer. Hence there can be no doubt
that the East Slavic etymon kazak > kozak was borrowed from the Polovcian
qazaq (< qaz-) some time after the fourteenth century. In this way the etymon
kazak in Eastern Europe is of nomadic Turkic origin.
No:vs
I. Concerning Arabic abd, see Islam: nciklopedieskij slovar I99I, 67.
2. See Wehr I966, I2u. Houtsma (I894) translates the Arabic word al-mujarrad as
Landstreicher [=vagabond], but in our context it is not accurate, see Doerfer I967,
462.
3. Te Genoese occupied Kaa in I26I, see O. Pritsak I99I.
4. I am quoting the newest edition by Vladimir Drimba (2uuu), fol. 3uv, dipl. ed., p.
IuI [=commentary p. 22I]. Doerfer (I967, 46667) rejects any connection between
the notation in CC cosac with the Turkic qazaq, because in the notations rst
syllable the vowel is ioi, a form under the inuence of the Ukrainian kozak, and
not iai, i.e. qazaq. But a sporadic change of iai into ioi is attested in the text of
the CC, see Gabain I949, 3I. Te same sporadic change appears in some old and
new Turkic languages, see Rsnen I949, 6u, 79, Brockelmann I934, 49, 6I.
3. 8I68I7 = o I4I3I4I4, see Muntaxab at-tavrx I336iI937, 87.
6. I am quoting the newest and the fullest edition of Bbur-nme (I993), prepared
by my former student Eiji Mano. It consists of four parts in ve volumes: part I,
the Turkic text of Bbur-nme, part 2 (vols. 23), an index of all words in all forms
appearing in the text, part 3 (vol. 4), a Japanese translation, part 4 (vol. 3), Japanese
commentary.
aa vvi:s
Rvrvvvcvs
al-Kshar, Mah, md. I983. Compendium of the Turkic Dialects (Diwn Lut
at-Turk). Ed. Robert Danko. Pt. 3. Cambridge, Mass.
Bbur-nme. I993. Ed. Eiji Mano. Kyoto.
Berneker, Erich. I924. Slavisches etymologisches Wrterbuch. Vol. I.
Heidelberg.
Brockelmann, Carl. I934. Osttrkische Grammatik der islamischen Litteratur-
sprachen Mittelasiens. Leiden.
Brckner, Aleksander. I937. Sownik etymologiczny jzyka polskiego.
Warsaw.
Clauson, Sir Gerald. I972. An Etymological Dictionary of Pre-Tirteenth-
Century Turkish. Oxford.
Doerfer, Gerhard. I967. Trkische und mongolische Elemente im Neupersischen,
unter besonderer Bercksichtigung lterer neupersischer Geschichtsquellen,
vor allem der Mongolen- und Timuridenzeit. Vol. 3. Wiesbaden.
Drimba, Vladimir. 2uuu. Codex Comanicus: Edition diplomatique avec fac-
similes. Bucharest.
Gabain, Annemarie von. I949. Die Sprache des Codex Cumanicus. Philologiae
Turcicae Fundamenta I. Wiesbaden.
. I96u. Kasakentum, eine soziologisch philologische Studie. Acta
Orientalia Academiae Scientiarum Hungaricae II:I6I67.
Golden, Peter B. I992. An Introduction to the History of the Turkic Peoples:
Ethnogenesis and State-formation in Medieval and Early Modern Eurasia
and the Middle East. Wiesbaden.
Gyry, Gyrgy. I942. Autour du Codex Cumanicus. Bibliotheca Orientalis
Hungarica 3. Budapest.
Houtsma, Martin Teodor. I894. Ein trkisch-arabisches Glossar. Leiden.
Hruevskyj, Myxajlo. I9u9. Istorija Ukrany-Rusy. Vol. 7. Kyiv and Lviv.
Islam: nciklopedieskij slovar. I99I. Moscow.
Malov, Sergej E. I93I. Pamjatniki drevnetjurkskoj pismennosti. Moscow.
Melnyuk, Oleksandr Savy. I983. Etymolohinyj slovnyk ukransko movy.
Vol. 2. Kyiv.
Menges, Karl Heinrich. I979. Vostonye lementy v Slove o polku Igoreve.
Leningrad.
Muntaxab at-tavrx. 6i,;. Ed. Jean Aubin. Tehran.
Pritsak, Larysa. 2uu6. Heneza fenomenu ukranskoho kozactva. Ukrainian
Historian 43, nos. I2:I4762.
Pritsak, Omeljan. I99I. Kaa. In Oxford Dictionary of Byzantium, 2:Iu8889.
Oxford.
Rsnen, Martti. I949. Lautgeschichte der Trkischen Sprachen. Helsinki.
:uvic v:vroioov or qazaq a
. I969. Versuch eines etymologischen Wrterbuchs der Trksprachen.
Helsinki.
Sreznevskij, I. I. I893. Materialy dlja slovarja drevne-russkago jazyka po
pismennym pamjatnikam. Vol. I. St. Petersburg.
Stkl, Gnther. I933. Die Entstehung des Kosakentums. Munich.
Vasmer, Max, and O. N. Trubaev. I967. timologieskij slovar russkogo jazyka.
Vol. 2. Moscow.
Wehr, Hans. I966. A Dictionary of Modern Written Arabic. Ithaca, N.Y.
Zajczkowski, Ananiasz. I932. Sufiksy imienne i czasownikowe w jzyku
zachodnio karaimskim; przyczynek do morfologji jzykw tureckich.
Cracow.
xii contributors
John-Paul Himka is professor of history, University of Alberta.
Daniela S. Hristova is temporary teaching associate, University of Cam-
bridge.
Norman W. Ingham is professor emeritus of Slavic languages and literatures,
University of Chicago.
Vyacheslav V. Ivanov is professor of Slavic languages and literatures and Indo-
European studies, University of California, Los Angeles.
Laura A. Janda is professor of linguistics, University of Troms.
Jay H. Jasano is Diebold Professor of Indo-European Linguistics and Philol-
ogy, Harvard University.
Daniel H. Kaiser is professor of history and Joseph F. Roseneld Professor of
Social Studies, Grinnell College.
Edward L. Keenan is Andrew W. Mellon Research Professor of History, Har-
vard University.
Valerie A. Kivelson is Arthur F. Turnau Professor of History, University of
Michigan.
Emily Klenin is professor of Slavic languages and literatures, University of
California, Los Angeles.
Nancy Shields Kollmann is William H. Bonsall Professor in History, Stanford
University.
Claire Le Feuvre is matre de confrences of the comparative grammar of
Indo-European languages, Universit de Strasbourg.
Ingunn Lunde is professor of Russian, Department of Foreign Languages,
University of Bergen.
Donald Ostrowski is research advisor in the social sciences and lecturer in
Extension Studies, Harvard University.
Maureen Perrie is emeritus professor of Russian history, Centre for Russian
and East European Studies, University of Birmingham, UK.
Riccardo Picchio is professor emeritus of Slavic literatures, Yale University.
Serhii Plokhii is Mykhailo S. Hrushevskyi Professor of Ukrainian History,
Harvard University.
Omeljan Pritsak (19192006) was Mykhailo S. Hrushevskyi Professor of
Ukrainian History, Emeritus, Harvard University.
Kylie Richardson is lecturer in Slavonic linguistics and philology, Department
of Slavonic Studies, University of Cambridge.
Robert A. Rothstein is Walter Raleigh Amesbury, Jr., and Cecile Dudley Ames-
Reproducedwith permission of thecopyright owner. Further reproductionprohibited without permission.