ActaOrientalia 52

ACTA
ORIENTAIT A
ACADEMIAE SCIENTIARUM
HUNGARICAE
VOL. 52 (1999) NO. 1
AKADÉMIAI KIADÖ, BUDAPEST

ACTA ORIENTALIA
A JOURNAL OF THE HUNGARIAN ACADEMY OF SCIENCES
FO U N D ED
BY
L. LIGETI
EDITORIAL BOARD
Ildiko E csedy
Päl Fodor
Éva J eremias
Andräs R öna -T as, Editor-in-Chief
Robert S imon
Katalin U. K öhalmi
Gyula WOJTILLA
ADVISORY BOARD: Gy. HAZAI, Z s . K a k u k , Gy. KARA, M . MAROTH, F. TÖKEI
TECHNICAL EDITORS: Éva CsÂKi and Imre HAMAR
ACTA ORIENTALIA (HU ISSN 0001-6446) publishes original papers on oriental philology and
culture in English, German, French and Russian. In addition it publishes reviews o f current books,
brief accounts on past and coming events in oriental studies. ACTA ORIENTALIA is published in
yearly volum es o f four numbers by AKADÉMIAI KIADÖ. H 1 117‫ ־‬Budapest. Prielle Kornélia u. 4.
http://www.akkrt.hu
Subscription information: Orders should be addressed to AKADÉMIAI KIADO, H-1519 Budapest,
P.O. Box 245. Fax: (36-1) 464-8221, E-mail: kiss.s@akkrt.hu
Subscription price for Volume 52 (1999) in four numbers is USS 156.00 including normal postage.
airmail delivery USS 20.00.
E ditorial correspondence: Manuscripts and correspondence with the editors should be sent to
ACTA ORIENTALIA. House o f Professors. Room 303, H-l 146 Budapest. Ajtösi Dürer sor 19-21.
ACTA ORIENTALIA is a refereed journal. It is abstracted in Sociological Abstracts.
© Akadémiai Kiadô, Budapest 1999
PRINTED IN HUNGARY
Akadémiai Nyomda, Martonvâsâr
AOrient 52 (1999) 1
ACTA
ORIENTALIA
HUNGARICAE
VOL. 52 (1999) NO. 1

CONTENTS
A ndräs RönA-Ta s , Chuvash and historical morphology .................................................................. 1

YOSHIZO Itabashi, On cognateship of Old Japanese i/si with Old Korean i/si .............................. 17
À rpâd B erta, On the Turkic background o f the Hungarian verb dv ‘to defend, protect, keep,
guard’ ............................................................................................................................................... 45
C laus Schönig, The internal division of Modern Turkic and its historical implications ............ 63
Miscellanea
B éla Bartdk’s Anatolian folk music collection (J. S ipos) .................................................................... 97

The eighth seminar o f the IATS in Bloomington (T. P0RC1Ö) .......................................................... 99
Review article
H. Franke-H-L. Chan, Studies on the Jurchens and the Chin Dynasty (H. Franke) ................... 101
Reviews
Hugh Richardson, High Peaks, Pure Earth (A. Röna -Tas ) ................................................................ 105
Denis Sinor, Studies in Medieval Inner Asia (A. R öna-T as ) ........................................................... 106
Gyula Décsy, The Turkic Protolanguage: a Computational Reconstruction (A. Röna -T as ) ...... 107
Présence arabe dans le Croissant Fertile avant l’Hégire (I. O rmos) .................................................... 110
Gotthard Strohmaier, Von Demokrit bis Dante: die Bewahrung antiken Erbes in der arabi-
sehen Kultur (I. O rmos) ............................................................................................................... 111
W olfgang Reuschel, Aspekt und Tempus in der Sprache des Korans (I. Ormos) ......................... 113
André Raymond, Le Caire; Robert Mantran, Histoire d’Istanbul (I. O rmos) ................................ 116
Ewald Wagner, Islamische Handschriften aus Äthiopien (I. O rmos) .............................................. 117
Seishi Karashima, A Glossary of Dharmaraksa’s Translation o f the Lotus Sutra (I. Hamar) .... 118
Donald D. Leslie, Jews and Judaism in Traditional China (P. VÄMOS)........................................... 119
Lambert Schmithausen, Maitri and Magic: Aspects o f the Buddhist Attitude toward the Dan-
gerous in Nature (G y . W0JT1LLA) .............................................................................................. 120
Books and journals received ..................................................................................................................... 123
Addresses of the au thors ............................................................................................................................ 125

Acta Orientalia Academiae Scientiarum Hung. Volume 52 (1), 1-15 (1999)
CHUVASH AND HISTORICAL MORPHOLOGY1
A ndräs R ö n a -T as
(Szeged)
The history o f suffixes is one of the most difficult fields in historical linguistics. This paper shows
an approach o f reconstructing the history of the suffix forming ordinal numerals in Chuvash. The
present suffix +mes is relatively young, but its constituents, +m, the possessive suffix +« and the +s,
are very old. In fact, the final - s in Chuvash +mës is a clear case where Proto-Turkic Is/ was pre-
served after Ini. The author claims that the Turkic suffix +nc goes back to +ns, and this is also pre-
served by Yakut, Dolgan, Tofalar and Tuva. The regional feature enables us to give a chronology
both to the historical feature and to the migration o f the Chuvash people.
Key words: Historical morphology, ordinal suffix, Chuvash, lambdacism
In the case of the Chuvash language we can dispose of the following categories:
1. Ancient Turkic morphological features which have been preserved by Chuvash,
but lost in all other Turkic languages, 2. Chuvash innovations which are peculiar
only to Chuvash, 3. Chuvash features which are Ancient Turkic and show the “nor-
mal” Chuvash changes, 4. Copies2 or borrowings from other languages, among them
from Turkic or Finno-Ugrian. In the case of Chuvash it is extremely difficult to dif-
ferentiate among features pertaining to one of the four categories. For instance in
Chuvash the possessive suffix third person singular is +*i, while in all other Turkic
languages it is after stems with vocalic finals +sP and with consonantal finals +i.
' I read the original o f this paper at a symposium in Göttingen, 13th—15th, November 1997,
organised jointly by the Seminar für Turkologie und Zentralasienkunde o f the Georg-August-
Universität and the Department of Altaistics of the Attila Jözsef University, Szeged, Hungary. I owe
sincere thanks to the Humboldt Stiftung for making possible these studies and my stay in Göttin-
gen with a research-prize.
2 Lars Johanson suggested to use the term “to copy” instead o f “to borrow”. No doubt this
is more exactly reflecting the situation, since the original which is copied, remains in the language,
if we borrow something we take it away. Similarly the term loan word can also be replaced with
the term copy. On the other hand copying is a conscious action while linguistic borrowing is in
most cases not. Metaphoric use of terms does not disturb, if we agree in the rules o f their use.
3 Already Grpnbech (1936/1979, p. 93) has clearly seen that the suffixes +i and +si had
originally only front vocalic forms. Chuvash has preserved this. We find ut ‘horse’, ucë ‘his horse’
(< ati), but utäm ‘my horse’ (< at'im).
0 0 0 1 1 9 9 9 © 5.00 $ /99/ 6446‫ ־‬Akadémiai Kiadô, Budapest

2 A. RÖNA-TAS
Theoretically it would be possible that Chuvash preserved an older stage, and +si is
a secondary development in all other Turkic languages. This is, however, as we shall
see later, not the case. All Turkic languages with the exception of Chuvash have the
historical phonemic oppositions Izl : Irl, Ш : /I/4. Some scholars claim that Chuvash is
preserving an old feature. This is not the case and the more “simple” Chuvash situa-
tion is secondary. On the other hand the nominative and the oblique stem of the per-
sonal pronouns first and second persons respectively are in Chuvash *bi T (> epe) :
*bän (> man), *si ‘you’ (> esë) : *sän (> san), while in all other Turkic languages
we find only *bän, *sän. The generalisation of the oblique stems is secondary in the
non-Chuvash languages, and only Chuvash preserved the old form of the stems per-
taining to the nominative case.
In some cases the situation is more complex. In the following I shall try to
demonstrate by which procedures and on which arguments can we decide such a ques-
tion. I selected the Chuvash suffix of the ordinal numerals: +mës.
1. Synchronic analysis
For understanding the status and function of the suffix of the ordinal numerals first
we have to look at the cardinal and collective numerals.
Cardinal numerals may have absolute and contextual forms (see Greenberg
1978, pp. 286-287). The first are used in simple counting as “one, two, three” or in
reference as “the numeral two” or as predicate “two times two is four”. The con-
textual forms are used as quantifiers “one chair, two men, three children etc.”. In
Hungarian for “two” the absolute form is kettö, the contextual form is ke't. There does
exist a third category, namely the definite cardinal numerals. E.g. “The (aforemen-
tioned) two of the children”, “the three of the Ten Commandments (which refer to
God)”. For the contextual Chuvash uses two forms5, for the absolute function a third
(on the definite forms see below):
Contextual Absolute
1. per, (përë) përre
2. ik ,ikë ikkë
3. vis, visé vissé
4. tävat, tävatä tävattä
5. pilëk pillëk
6. ult, ultä ulttä
4 In fact Chuvash has [z] as an allophonic variant of Isl while in the synchronic system Isl
and III are independent phonemes. See below on the secondary /5/ in Chuvash.
5 The two forms ik, ikë, and vis, visé respectively are used in free variation. The form përë
is rare and can only be reconstructed, see below.
Acta Orient. Hung. 52, 1999
CHUVASH AND HISTORICAL MORPHOLOGY 3
7. sic, sicë siccë

8. sakär sakkär
9. täxär täxxär
10. vun, vunä vunnä
50. al, ala alla6
80. sakärvun, sakärvunä sakärvunnä
The origin of these two series is a debated question. According to Clauson

(1959) the Chuvash numerals with long consonants preserved an old feature. Ac-
cording to Ashmarin (1898) and others, the lengthening was due to the stressed posi-
tion in the absolute form. Phonetically the opposition is not of short and long conso-
nants but of weak and strong or lenis vs fortis pronunciation. In fact we have /saGär/:
/sakär/ which is realised by the orthography as <sakär> and <sakkär>. The /с/ is
a weak unvoiced and the /к / a strong unvoiced guttural stop6 7. In stressed position the
fortis pronunciation was preserved. This is important, because we have no “long con-
sonants” in Chuvash. It is another question why is an old strong consonant preserved
in intervocalic position in numerals, while in other cases it became a weak one8. This
is surely due to the fact that numerals in the stressed and unstressed positions have
different functions and the differentiation needed a phonemic marker. It is also im-
portant that in the Chuvash monosyllabic numerals a second vowel appeared in form
of a reduced vowel (visé, vissé, cf. Turkic iic).
Collective numerals. The collective numerals single out a group from a smaller
group of people: “we two”, “you three”, “they three” etc. According to Andreev
(1957, pp. 117-121) and Pavlov (1965, pp. 181-183) Chuvash language has the fol-
lowing collective numerals:
We You They
definite indefinite
two iksëmër iksër ikës, ikësë, ikkës(ë) ikkën
three vissëmër vissër vissesë vissën
four tävatsämär tävatsär tavattàsë
five pilëksëmër pilëksër pillëkësë pillëkën
six ultsämär ultsàr ulttasë
6 The three forms al, alä and alia are now only present in the Viryal dialect. In the Anatri
dialect and in the literary language the form alia has been generalised and is the only one used.
7 On the weak and strong opposition o f consonants see Johanson (1984-1986) where Chu-
vash is not dealt with.
8 On the secondary weakening of obstruents see Johanson (1998, p. 98).

4 A. RÖNA-TAS
seven siësëmër siësër siëcësë

eight sakkarâsë sakkärän
nine tàxxàràsë
ten vunasë
twenty sirëmës sirëmën
thirty vâtârâs(ë)
forty xerexes(ë)
fifty allâsë
There is a discussion on the nature of these numerals. Ashmarin (1976, p. 44)

did not enumerate them as collective numerals and remarked only that the numerals
can have possessive suffixes as the nouns may have, but only plural forms occur with
the possessive suffixes. Andreev, Pavlov, and others think that here we have to deal
with collective numerals. Levitskaja (1976, p. 51) suggested to call them collective-
definitive or collective-selective (sobiratel’nyj-vydelitel’nyj) numerals. According to
Benzing (1959, p. 732) the forms iksëmër and iksër evolved on analogy to kiltëmër
‘we came’, kilter ‘you came’. Räsänen (1957, p. 22) suggested to find here two pos-
sessive suffixes. Levitskaja (1976, p. 122) accepted the opinion of Räsänen, but
remarked in a footnote that there does exist a view that +semer comes from the
plural, i.e. ikë+sem+ëmër (= iki-ler-imiz9)■ This is the only possible solution, because
the -s- in iksëmër cannot be a possessive suffix which is +s(e) if it is present at all, as
we shall see later. The form iksëmër ‘we two’ is in this case ‘two’ + the marker of
the plural (sem)+ the suffix of the first person plural (+ ётёг/ämär). The form for
‘you two’, iksër, is from iki+sem+r, that is ‘two’+pl+Px2pl and ‘they two’, ikkësë is
4wo’+Px3s. The model is thus the following: =iki+ler+imiz, =iki+ler+siz, -iki+si.
It is important to remark that these forms are definite forms in contrast to the non
definite forms with +än/en. In Chuvash ikkësë means ‘they two’, ikkën ‘both, two to-
gether’, e.g. epir kilte ikkënëx ‘(Only) two are at home with us’, ulttànran ikkënex
tärsa yultämär ‘out of six (only) two of us remained’. Ashmarin (1898, p. 184) quotes
the following example: Atâl xerrincë alla xur, allâsë te xura xur ‘On the bank of the
Volga there are fifty geese, all fifty of them are black’. Allâsë ‘(all) fifty of them’
corresponds to elig-i in most Turkic languages.
Chuvash has a definite form ‘the first of the already counted units’. This is
formed with the possessive suffix which joins in the case of ‘one’ the longer con-
textual form, and is péri (përë+i)'0. Thus we have ikkëren peri ‘one of both, the one
9 In the following I shall denote with the sign o f equality the Turkic analytic equivalent o f
the given Chuvash morphemic structure, a kind o f “morphemic translation” into Common, i.e. non
Chuvash Turkic.
10This needs some comments. In Chuvash if the possessive suffix o f the third person joins
a word which ends in a consonant, the suffix has the form +ë: xër ‘the girl’, xërë ‘o f the girl’
(*qi'z+i). If it joins a word which ends in a full vowel it is +i: êne ‘the cow ’, ëni ‘o f the cow ’
(* ëne+i < ineg+i). If the word ends in a reduced vowel, the preceding consonant changes to
of the two’. In contrast to the contextual and the absolute forms, which have the
nominal declension, this form has the possessive declension. The locative case is for
the absolute form përre+re, ‘to one’ (=bir+de)" while for the definite form it is
përince ‘to the one (of them e.g. who remained in the room)’ (-bir+i+nde). How-
ever, if the numeral ‘one’ is a member of a higher decade than it is in the contextual
form sirëm për ‘2 Г and the locative case is përte, e.g. Esë misé sulta? - Epë sirëm
përte ‘How old are you? - I am 21 (I am in my twenty first)’ (=ben yigirmi birde)
and not *sirëm përrere (Andreev, 1957, p. 110).
In the above examples we could see that the numerals can be used and are in
fact used with possessive suffixes for several functions, mainly for expressing defini-
tiveness.
Ordinal numbers are always referring to a group of which the given unit is the
second, the third etc. These numerals are always in singular in contrast to the collec-
tive numerals, as ‘the second’ is in contrast to ‘two of us, these three together’. Ordi-
nal numerals have to be therefore either in some of the possessive (the fifth o f the
sons) or partitive (pjatyj iz synovjej) structures or in an attributive determinative
function (the fifth son, pjatyj syn).
In the Anatri dialect and in the literary language the ordinal suffix is +mës. It
follows the absolute form as përremës, tàvattàmës, but in the Viryal dialect the same
ordinal suffix is following the contextual forms in attributive function: visëmës кип
‘the third day’, while the suffix is following the absolute form if it is used e.g. as
predicate, as: Xâs kunce ‘on which day?’ vissëmësënce ‘ on the third’ (-ücüncünde).
Note that the ordinal number has the local suffix of the possessive declension and
once more it denotes definitiveness'2.
The suffix +mës is always front vocalic, only some local dialects and spoken
idioms are now gradually introducing the back vocalic form +mäs.
If the ordinal numeral is an attributive it is not declinated, but if it is in a geni-
tive construction, then it takes the suffixes of the possessive declension as e. g. ikkë-
mës mayän ‘of the second May (i.e. this was not of the first May month but of the
second May month of her stay)’, but mayàn ikkemësënce ‘on the second of May’
(=ikincinde).
Chuvash has also an other ordinal suffix which is +m. This is, however, non
productive and used only in certain frozen expressions as: visëmkun ‘the third day’,
visëm sul ‘the third year’, visëm-tâvatàm кип ‘in a few days’, vismine ‘after tomor-
row’, tuatmine ‘on the fourth day’, further pilekmine, ulttämine, sicmine ‘on the fifth,*12
a strong (long) one and the suffix is +i: atä ‘the boot’, atti ‘o f the boot’ (* ata+i < etiig+i). In front
o f these fact the form péri seems to be anomalous. From per we would expect père, from père
përri and from përre also përri. We know, however, a few similar cases, as utä ‘hay’, uti ‘o f the hay’
(and not *utti), sutä ‘light’, sun ‘o f the light’ (and not *sutti), pusä ‘the whip’, pusi ‘o f the w hip’
etc. (see Andreev 1957, p. 36). These are secondary, because otherwise we would find *ucë etc.
11 The Chuvash form përrere is from birede , the suffix +de remained - l after final -r , as in
pirte ‘in the linen’. The vowel -e- is an inner Chuvash secondary form, see also sirëm përte ‘in the
twenty-first’.
12 Already Grpnbech (1936/1979, pp. 9 6 -9 7 ) has seen that the possessive suffix third per-
son singular, which he called article, is expressing definitiveness.
6 A. RÖNA-TAS
sixth, seventh day’. In some Viryal dialects, e.g. in the dialect of Puskart and
Horacka i.e. Maloe Karackino and Bolsoe Karackino of the Sundir rayon in the
present administrative system, Ashmarin noted the form visen kon, that is instead of
+m, they use a suffix +n. This looks like a sandhi form: visëmkon > visënkon.
It was important to collect and analyse all the recent data, because this has not
been hitherto done, and we can move to the historical analysis only if we dispose of
all data being at our disposal.
2. Historical analysis
The first thing that is clear is that the suffix +mës cannot be very old, because it does
not follow the vowel harmony. This is not the only relatively recent suffix. The Chu-
vash suffix of the plural +sem has the same peculiarity, it has no back vocalic form in
the Anatri dialect and the literary language. In the Viryal dialects the spread of the
back vocalic +sam variant is more progressive than that of the ordinal suffix. The
suffix -sem occurred already on the Volga Bulgharian inscriptions (see the 1308 in-
scription, Hakimzjanov 1986, p. 174, line 4: ulamasemne ‘the ulamas (acc.)’, Mod-
ern Chuvash ulamsene).
The fact that the ordinal suffix +mës is relatively recent and had only a front
vocalic variant means that the vowel -ë- cannot be a connecting vowel. It can go
back either to an invariable front vocalic [г] or it can go back to an /// which had
originally front and back allophones but came under the influence of a phonetic
feature and was palatalised. This, theoretically, may have been a strongly palatalised
consonant or another front vocalic [г] which later disappeared. But in both cases the
-ë- of +mës immediately has to go back to an earlier [г]. Until now it was the general
opinion that every Isl of the Chuvash language is secondary. The consonant s can
have three different origins. It can go back to a sequence -si-13, it can feature in
words of foreign origin and it can go back to a few consonant clusters. Such a
cluster, that may interest us, is -nc-.
Among the de verbal noun formative suffixes we find the ОТ suffix -(X)nc
(Erdal 1991, pp. 275-285) which is reflected in Chuvash by -s\ ОТ sävinc > Ch
saväs ‘joy’, ОТ korkinc ‘fear’ > Ch xärus ‘fearful’, ОТ säkinc > Ch sic/ßs ‘thought’.
The foreign origin of a suffix is possible if and only if more words without
and with the given suffix have been borrowed, the speakers of the borrowing lan-
guage recognised the function of the suffix and then the suffix spread to the genuine
words. This is the case e.g. in Halaj. The intrusion of Persian numerals into Halaj has
been described by Doerfer (1988, pp. 111—114). The replacement of Turkic words by
Persian words is greater in higher numerals than in lower, but in case of giving
prices, fixing calendric days and hours it is very common also in the case of the
lower numerals. Among these Persian suffixes we find also -am as in tërtâm ‘fourth’,
13 The cluster si- may have two origins. It can go back to a cluster ls/+ short or long /('/ or it
can appear in places where Isl was originally followed by a long lâl > liai. ОТ si'rt > Chuv sari
‘bristle’, ОТ sàz > Chuv sur ‘mud’.
bësüm ‘fifth’ along with duyum, säyyum, häftum etc. Sometimes the Persian structure
is contrasted to the Turkic as häfiu’m-i kin ‘der siebte Tag’, but kiniyëtisi ‘rti ‘es war
der siebte Tag’.
According to Galkin (1964), Je. I. Kovedjajeva (1976) and following them
according to Adamovic (1996) the Chuvash ordinal suffix and thus the final -s would
be of Cheremis origin. This can be excluded, because none of the Cheremis numerals
has been ever borrowed by Chuvash.
Already Ashmarin (1898), later Andreev (1957, pp. 114-116), Benzing (1954,
p. 389; 1959, p. 731) and Levitskaja (1976) claimed that -s- of +mës goes back to the
possessive suffix -si. Adamovic tried to refute this with the reasoning that “...ein
‘Possessivsuffix -ёГ [...] es in Wirklichkeit nicht gibt, und wohl auch nicht geben
kann, da das urtürkische Possessivsuffix der 3. Person bekanntlich -si lautet” (1996,
p. 17). Further he mentioned that -si is joining only words with vocalic Auslaut and
here we have -m. None of the two arguments are acceptable. Nobody claimed that
the suffix is -ës, it was claimed that the suffix is -s-.
Chuvash is the only Turkic language where the possessive suffix is joining
both the words ending with consonants and with vowels. In modem Chuvash the ОТ
suffix -si has been preserved only in a very few cases as: amàsë ‘his/her mother’, assë
‘his/her father’, kerüsë ‘his/her son-in-law’, purta türtës(ë) ‘the back of the knife’,
alâ türtës(ë) ‘the back of the hand’ (Ashmarin 1937, vol. XIV, p. 229, Paasonen 1908,
p. 177) and as we have seen it is applied after some suffixes used just with numerals.
Most of the authors use to cite the famous Materialy (1898) of Ashmarin. Ashmarin
read thirty years later lectures at the University of Kazan in the academic year
1929/30, and the text of these lectures was published in 1976. Here (1976, p. 43) he is
more explicit and says that according to him a double possessive affix is following
the numeral with the old ordinal suffix +m: ultt-âm-ë-së. This means that according
to Ashmarin the -ë- of mës goes back to the possessive suffix -i- (-alti+m+i+si).
As I have shown above in Chuvash to the final -nc in words as sevinc, kor-
kinc, etc. a final -s does correspond. Erdal (1991, p. 275), citing Bang (1930), points
to the possibility that the Ici of Old Turkic -(X)nc might be from the formative -Xs
joining verbs ending in -n. The third vowel of e.g. sevin-Xs was syncopated and from
such -ns- developed the form -nc14, which was later generalised. Following Bang,
a similar idea was expressed by Benzing (1954, p. 388), who suggested to have an
ОТ ikinc from *ikinis. Accordingly we now have to consider whether the -s of +mës
is or is not a “preserved” form.
In Chuvash the “normal” representation of Turkic *c is the palatalised s (kas
‘late, afternoon’< käc). If we find in place of Turkic -nc- Chuvash -s, we have two
options: -nc goes back to -ns, Chuvash preserved s of the phase -ns and later dropped
-n- in this position or we have to depart from -nc which as a cluster developed into
s in Chuvash. In the first case we would now have three different scenarios of the
history of PT *s in Chuvash. In many cases it became l (the well-known lambdacism,
14 The change from ns to nc (nts) is a natural change. The infixation of the oral dental stop
t between the nasal dental stop n and the sibilant s for easing the pronunciation is a well known
feature in phonetics.

8 A. RÖNA-TAS
altmi's ‘sixty’ > utmäl), in some cases it became s (as in bas ‘head’ > pus) and after
-n- it remained s. More precisely:
1 . * s > l,
2 */ > *1é > ^
3. *-ns > s.
In the third case s has been preserved because it was “protected” by the pre-
ceding nasal. As we shall see a similar procedure occurred in Yakut, Dolgan, Tofalar
and Tuva.
Now we can move to the problem of the origin of Turkic ordinal suffix. This
is, as it is well known, +(X)nc in Old Turkic (ОТ). In most of the modem Turkic
languages we find +(X)nc(X), +m(X)nc(X), or +l(X)nc(X). In four languages we do
not find the nasal. One of them is Yakut, where the suffix is -s, see Yakut alti's
‘sixth’, onus ‘tenth’, etc. The same is the situation in Dolgan, which is a language
near to or a dialect of Yakut. Yakut final can go back to -c, or -s (in a few cases
also to -s and -г). Since -s and —z can be disregarded here, the Yakut ordinal suffix
can go back either to +(X)nc or to +(X)ns. We shall decide this question later, at pres-
ent for denoting the two possibilities we shall use the notion -nc/s.
In Yakut verbs ending in -nc as ОТ yanc- ‘to crush, to trample on’, sane- ‘to
pierce’, have the following stem variation: si's- (=OT yanc-), but the oblique stem is
sïnna-, as- (= ОТ sane-), but the oblique stem is anna- respectively15. This means
that in word final position we find -nc/s = Yak -s, but in intervocalic position -nh-,
and this points to a palatalised -n+c/s (see Schönig 1991, 1997, p. 130). To this we
can add that -nc/s became -s also before consonants: Yakut astar- is the iterative
form of as- (= ОТ sanctur-).
The Yakut representations have recently been studied by Stachowski (1994).
He added to the above a very important chronological remark. In Yakut we find
yahax ‘Unterhaltung, Spiel’ from an earlier yasax. This form goes back to an even
earlier form abincak, but this could not have occurred directly, since -nc- is here not
in word final but in intervocalic position. Thus we have to suppose that the Turkic
form abinc ‘something which gives pleasure, enjoyment’ became first in Middle
Yakut abis then it was joined by the suffix +Ak and the form *abisak became regu-
larly yasax > yahax.
Stachowski (1994, p. 179) further called the attention to the fact that beside
the “normal” ordinal suffix -s there does exist a special form with -;ini'as altïnnï 'Ос-
tober < the sixth month’ in contrast to alti's ‘the sixth’. This means that in Yakut we
find the opposition *altinci : *altinc in the form altïnnï : alti's. The difference is simi-
lar to the difference between the absolute and the contextual forms of the ordinal
number. Here the absolute form of the ordinal number has a possessive suffix *+/.
Dolgan follows the Yakut model. Along with aids ‘sixth’, onus ‘tenth’ we
find in place of ОТ sane- Dolgan as- which has the oblique stem anhV-. To this we
can add from the vocabulary (Stachowski 1993, p. 34) annï ‘Art Brechstange, ein
Fischfanggerät’ ( < saneïg).
15 Anderson (1998, p. 14) quotes also Yakut mus- ‘to gather’ and its oblique stern munna-.
In Dolgan we find mus- ‘sammeln’, munnak, munnax ‘Versammlung’, munuxun- ‘sich versammeln’.
In Tuva we have -(X)k(X) (dörtkii ‘fourth’), and in Tofalar -(X)sk(X) (dörtüskü

‘fourth’). I would leave for the time being the question of + kii < +ki aside16, but the
-s- in Tofalar (Karagas) is of great interest. According to Bang (1930, p. 19) and
Menges (1968, p. 117) the s in Tofalar is due to a false separation, from ha,ski ‘first’
{bas ‘head’+Zh >+ski). Shcherbak (1976) accepted this too. Rassadin, however,
pointed out with reason that -s- in the Tofalar suffix is the same as the Turkic -nc and
as the Yakut -s. Rassadin (1978, pp. 118-121) argued that the same happened with
verbal stems ending in ОТ in -nc and in Tofalar in -n’cz. (-ndz in our transcription).
If the ending got into absolute final position or it came into preconsonantal position
in suffixed forms, the cluster -n’cz- became fan d the preceding vowel is lengthened
and nasalised: ОТ sane- > Tofalar sen’czar ‘to stick, pierce into’ > se~es! ‘bodaj!
[butt! imperative]’, se~eskak ‘butting’, ОТ yanc- > nd~es ‘soften! (imperative)’,
nen’ezar ‘to crush, to soften’, nd'eskan ‘he softened’. The feature is present also in
some Tuva words. ОТ yanc- 40 crush, to soften, to beat’ appears in Tuva as sane-
(< sane-), its imperative is: sas\ the iterative is saskila-. Or ОТ korkunc ‘fear’ is in
Tuva korgus, but the adjectival form is korguncug.
The Tofalar, the Tuva and the Yakut features are the same, and we can formu-
late the statement that ОТ -nc is yet preserved on the surface in Tofalar and in Tuva
in the form -s, in Yakut as -5. The only difference between the Tofalar, Tuva and
Yakut on the one side and the Chuvash feature on the other, is in Chuvash s was
fully generalised and the original nasal is not coming to the surface in any of the pho-
netic positions, even not in verb stems ending in -nc.
Thus we find the following: Yakut, Dolgan -*ncV: -nn-, -*nc: -5, Tofalar, Tu-
va -*ncV: ndz, - *nc: -s, -*ncC: -s, Chuvash - *nc: -s (-).
Cluster Yakut/Dolgan Tofalar/Tuva Chuvash

+Nc# s(< s) s s
+NcC s(< s) s s
+NcV nn ndz s
Taking into account that the Yakut/Dolgan ordinal suffix -s and the Tofalar -s
in the Tofalar suffix -ski are both corresponding to ОТ -nc, we have to conclude that
the Chuvash -s in +mes is also corresponding to ОТ +nc. We have, however, to leave
open for the moment the question of the origin of the Yakut/D01gan/T0falar/Tu-
va/Chuvash feature.
The following problem is connected with the -m as a separate suffix and as
a part of +mes. The problem of the nasal consonants in general has been discussed by
Doerfer (1967, 1973), Levitskaja (1976), Hovdhaugen (1972) and me (Rona-Tas
1976, 1982). According to Doerfer beside *m, which remained m in all Turkic
languages, there was an Ancient Turkic *m2 which became n in all Turkic languages
16 On +ki in the ordinal numerals see Doerfer (1988, p. 115).

10 A. RÖNA-TAS
with the exception of Chuvash where it remained m. Levitskaja, Hovdhaugen and

I argued that the problem of the nasals in Chuvash is a result of dialectal variation,
which according to Hovdhaugen may go back to the generalisation of sandhi rules.
The question of the ordinal suffixes -m and -mes has been mentioned by
Doerfer, but only in connection with Pritsak’s opinion that -nci goes back to -inti
(see ОТ ikinti), but since Chuvash has -m this is, according to Doerfer, unlikely
“...es sei denn, man trenne tü. -n-ti und tschuw. -m überhaupt voneinander, was aber
kaum beweisbar ist und was ja auch Pritsak nicht tut” (1967, p. 62). The idea of
Benzing that the Chuvash ordinal suffix -m would be of Iranian origin is as unlikely
as the Cheremis origin suggested by Galkin, Kovedjajeva and Adamovic is (see
above on the borrowing of suffixes).
From Kumük we can cite +imci/imji+si e.g. sekisimjisi (Benzing 1949, p. 400).
In some Chagatai texts e.g. in the Bäbumäma we find forms as ikimcisi, ücümcisi,
törtümcisi, besümcisi, altïmcïsï, yetimcisi, sekizimcisi, while in other Chagatai sources
we come across besinci or besilänci (cf. Eckmann 1966, p. 109). Forms with -m- on
place of -n- in *+nci as birimji, iicümjü, onumju are quoted from Turkmen, Azeri and
Üzbek dialects by Shcherbak (1976, pp. 205-206). From a phonotactic point of view
-mci- can change into -nci-, that is a homorgan nasal can develop on place of a non
homorgan before a c, but not vice versa. That means that if -m- in the -mci data is not
due to a strong Iranian influence, then the original ordinal suffix was in Ancient
Turkic +mc and this may have developed from + *m(X)s as it was suggested by Bang
and Erdal (see above). This became +nc and then with the addition of the possessive
suffix +nc+i > +nci. The suffix has been in a further step in some Turkic languages
de-etymologised and a second possessive suffix was added +(X)nc+i+si >+ncisi.
A more complex form is +minci. This form is present in Halaj as icminji ‘third’,
altaminji ‘sixth’(Doerfer 1988, p. 115), and similar forms are cited from some Azeri
(Caferoglu and Doerfer 1949, p. 300) and Türkmen dialects (Bazin 1949, p. 313).
It is, however, very unlikely that a so complex suffix as +minci is reflecting
the original state. Thus what we see is that -m is featuring on the surface and then
it is disappearing maybe two times: -*me (> +mc+i+si) >+nc >+ nc+i > +nci (> +nci
+si) > +m+i+nc ( +i).
In a paper published in 1976 I demonstrated that in the Volga Bulgharian in-
scriptions the ordinal suffixes are in a complementary distribution. After consonantal
finals we find +(X)m and after vocalic finals we find +s(i). Twenty years later Ada-
movie (1996) came to the same conclusion apparently not knowing my paper. There
were however two points which Adamovic missed. I stressed that this distribution, as
all complementary distributions, is secondary, and that it is also relatively recent. The
Volga Bulgharian ordinal for 50 älüs ‘fiftieth’ could have developed only after the
final -g of elig ‘fifty’ disappeared, otherwise we would have *älüm < elig+im. Volga
Bulghar had the distribution -C(X)m and -Vs and this situation goes back to a form
-hminc. I supposed that the complex suffix +m+i+nc changed as follows:
1. -C+(X)minc > 2. -C+(X)mis> 3. -C+(X)ms> 4. -C+(X)m,
1. -V+(X)minc > 2. -V+mis > 3. -V+ms> 4. -V+s.

This distribution could not last long. Firstly because the finals changed as per
> përë, vis > visé, and secondly because analogy demanded a unified structure. If
this complementary distribution existed not only in Volga Bulghar but also in the
contemporary spoken Middle Chuvash, then it have changed soon. Phase two of the
above changes was the nearest applicable morphophonetic unit, thus +(X)mis was
restituted, generalised and this changed later to +(X)mës. Theoretically it is also pos-
sible that the phases 3 and 4 occurred only in Volga Bulghar, and Middle Chuvash
preserved phase 2. Chuvash is not the direct continuation of Volga Bulghar, it repre-
sents a dialect that was near to, but not identical with the language represented on the
Volga Bulghar inscriptions. If phase two was preserved, then we have to explain the
front vocalic quality of Chuvash +mës. The answer could be that this is, historically
speaking, the possessive suffix, which was originally only front vocalic, and re-
mained as such in Chuvash as in ut ‘horse’ > ucë ‘of the horse’ (< at+i), ival ‘son’,
ïvalë ‘of the son’ (< ogul+i).
Now we have to turn back to the non productive Chuvash ordinal suffix +m
and ask what its origin may have been. It may be that this is a preserved very old
feature from the Ancient Turkic period, prior to the Old Turkic -nc. This would mean
that from the change + *m(X)s > - *ms > *ns > nc we can go one step back and sup-
pose a phase *m+s of which the +*m was preserved in Chuvash. The other possibil-
ity is that the Chuvash +m may also be a conditioned remnant from the Middle
Chuvash -m that evolved after consonants. This could have happened when the nu-
merals with originally consonant finals got an additional final vowel, as ОТ üc >
Middle Chuvash vis > Modem Chuvash visé. The structure vis+ëm was restructured
as visë+m. This has been, then, generalised in counting and appeared also in nu-
merals with final vowels as in ОТ altï > ultä > ultäm, ОТ yeti > sicë > sicëm and
has been preserved in some expressions. It can be argued in favour of both hypothe-
ses, but the second solution is more probable, since the generalisation in numeric
systems is well known in all languages (see e.g. the initials of Latin quatuor ‘four’
and quinque ‘five’ < *pinkwe or Hungarian hat ‘six’, hét ‘seven’ < *ét).
Summing up: The Chuvash suffix for the ordinal numbers +mës has a compli-
cated history. It corresponds to ОТ +minc. In a similar way as in Yakut/Dolgan and
Tofalar/Tuva the final *s corresponds to ОТ -nc. We did not claim, however, that
this -s goes back to -nc. We have a very important and old isogloss that connects
Chuvash with the East Siberian Turkic languages. Chuvash has in addition two more
elements, an +m and an +/.
Now we can turn to the chronology. The fact that the Chuvash +mës corre-
sponds to Turkic +minc and Halaj and some Oghuz dialects have +minji <*minc+i
that is the same structure as Chuvash plus the possessive suffix of the third person,
could have two interpretations. It can be a shared old feature, or it can be a secondary
development in both languages. The fact that historically Yakut/Dolgan and Tofa-
lar/Tuva have in place of the ОТ final -nc the same -s as Chuvash may have also the
same two interpretations.
Schönig (1997, p. 122) collected a few other common isoglosses between
Chuvash and Yakut/Dolgan (Lena Turkic in his terminology). Such are the low vow -

12 A. RÖNA-TAS
els of the second syllable in ОТ olor- ‘to sit down, to sit’, Yakut olor-, Chuvash lar-,
low vowels in the suffix -DWR- where the other Turkic languages as a rule have high
vowels. The loss of the opposition с : у or the preservation of the verb tasiq- ‘go out’
(Yakut tayïs-, Chuvash tux-) are less clear. According to Schönig, Lena Turkic and
South Siberian Turkic formed not too long ago the North East Turkic area. Accord-
ing to him “[t]he connection between Chuvash and North East Turkic goes back to
old areal language contacts”. In favour of this claim he mentions the early contacts
between the Proto Samoyed and the Oghur/Chuvash group and the shape of the word
for ‘stirrup’. In fact in the word for ‘stirrup’ in Tuva/Tofalar (ezerji/ezetjge) and in
Yakut (irjehe) the first vowel is illabial as in Chuvash yärana < irärjä, while in most
other Turkic languages (with the exception of Hakas and Yellow Uighur) it has a la-
bial vowel (*üzerjgi). This suggests a situation in which, after the separation of the
Oghur and Common Turkic languages, the speakers of the Oghur languages re-
mained for a time in a region in South Central Siberia together with the ancestors of
Yakut, Tuva/Tofalar, Hakass and Yellow Uighur.
In another paper (Rona-Tas 1988, p. 745) I have shown that the Chuvash
word for ‘pine-tree’ xïrà is ultimately of Proto-Samoyed origin, the same word is
also present in a number of Siberian Turkic languages as Soyot xadï, Hakas hazï, Tu-
va xadï, Tofalar hadï, Nizhne Iyus xayï. This means that the Oghur people, or with
other words the ancestors of the Chuvash/Bulghar people had to live once on a terri-
tory where they were in contact both with the Proto Samoyed speakers and with the
speaker of the Central Siberian Turkic languages. These facts do not give an absolute
certainty, but make it historically very probable that the change -s corresponding to
ОТ nc is a shared old feature.
At this point we have to make a short stop. Is it certain that we have to do with
an +*nc in the ordinal suffix of Yakut/Dolgan, Tofalar and Chuvash as it was sup-
posed by Schönig and Stachowski? As we have seen in the case of the deverbal
suffix -(X)nc Erdal supposed a more ancient +(X)ns. Is it impossible that Yakut/Dol-
gan and Tofalar had ~/+ns that changed to s and later in Yakut/Dolgan to -si In this
case Yakut/Dolgan and Tofalar would, together with Chuvash, have preserved an
older stage.
Ns
Old South Siberian ns East Old Turkic nc
s Nc
Yakut/Dolgan Tuva/Tofalar Chuvash All other Turkic languages
s s s nc

Would be this hypothesis chronologically acceptable? The exact time of the

departure of the Oghur/Bulghar/Chuvash groups from the Turkic homeland in Cen-
tral South Siberia can only be guessed. But according to what we know about the
early Oghur migration, it had to be before the 4th century1718. That means that this
shared feature dates back to the times around the birth of Christ. Since Chuvash -s in
+mës did not take part in the lambdacism (altmi's ‘sixty’ > Chuvash utmäl) we have
also a terminus post quern18 for this feature. The change s > l began in the centuries
before the birth of Christ, but it did not become generalised19. In other words: /s/ be-
came At in some words, it became through tlct > /s/ in others, and it has been pre-
served after /п/ at least in two suffixes.
I cannot give a simple answer to the question I have asked at the beginning of
the paper, whether the Chuvash ordinal suffix as a whole is preserving an old feature
or is it a new one. The answer is that it reflects old elements, but as such it is a new
suffix built of old elements. Something which may be true of most of the suffixes.
We can summarise the steps of the history of the Turkic ordinal suffix and the his-
tory of the Chuvash suffix as follows:
1. +(X)m —>Chuv +(X)m ?

2.1. +(X)mc
2. +(X)m(X)s > +(X)ms > '
- 2.2 <— ‫־‬. +(X)ns
2.1. +(X)mc+i —>Türkmen, Azeri, Uzbek dial +(X)mci, Chagatai +(X)mci+si
> +(X)mcisi
2.2.1. +(X)ns —* Yakut/Dolgan +ns > +s > +s, Tofalar +ns >-s, (+ki) Tuva +ski
2.2.2. +(X)ns+i —>Yakut +nni ----------
3. +(X)m+i+ns —>Volga Bulghar *+mins > +mis >> Chuv +mis > + m es
C+(X)m ( > Chuvash +m) V+s

4. +m+i+nc+i —> Halaj, Azeri, Tiirkmen dialectal +minji
5. +(X)nci —>most Turkic languages +(X)nc(i).
The historical processes of the morphological feature can be formulated as

follows: in the historical deep structure we have to suppose three archimorphemes
which became constituents of the ordinal suffix : M+S+Рдг and we can describe the
historical phonotactic rules according which the three elements appear and disappear
17 Around 350 Oghur tribes already appeared on the Kazakh steppe, see Czeglédy 1983;
Golden 1992, pp. 9 2 -1 0 0 . We have new data on the Oghur tribes in Siberia, see R6na-Tas (1996),
Rdna-Tas (1999).
18 In other places I tried to give my reasons as to why I am convinced that the change was s
> l and not l > s which I would not repeat here. In short: it is irrelevant whether we depart from
l{ : l2 or s : /, because the opposition disappeared in Chuvash where w e have no l2. The period when
the rotacism has been finished and the lambdaism ceased to work was in the few hundred years
before Christ.
19 The idea o f the “incomplete” change in the case o f lambdacism is an idea first expressed
by Ligeti (1986), pp. 16-17.

14 A. RÖNA-TAS
on the surface in different times in different languages. It is also important that the
possessive suffix of the third person always played a great role by pointing to the
definitiveness of the numeral mainly in the absolute function.
The history of the Chuvash ordinal suffix helped us to demonstrate that the
history of a suffix is subject to complex and complicated morphophonemic proc-
esses. Assimilation, syncopation, simplification, analogical extension and de-ety-
mologisation are at work, parallel to the “normal” phonetic changes.
References
Adamovic, M. (1996), Ordinalia im Bolgarischen. Symbolae Turcologicae, Studies in honour of

Lars Johanson on his sixtieth birthday 8 March 1996, Istanbul.
Anderson, G. D. S. (1998), Historical Aspects o f Yakut (Saxa) Phonology. Turkic Languages 2,
pp. 3-31.
Andreev, N. A. (1957), Imja cislitel’noe, in: Materialy po grammatike sovremennogo cuvasskogo
jazyka. Ceboksary, pp. 107-123.
Ashmarin, N. I. (1898), Materialy dlja izsledovanija cuvasskogo jazyka, Kazan’.
Ashmarin, N. I. (1937), Slovar’ cuvasskogo jazyka. Ceboksary.
Ashmarin, N. I. (1976), Zametki po grammatike cuvasskogo jazyka: Cuvasskij jazyk, Naucno-issle-
dovatel’skij Institut, Trudy 66.
Bang, W. (1930), Turkologische Briefe aus dem Berliner Ungarischen Institut V, UAJb 10, pp. 16-20.
Bazin, L. (1949), Le türkmène, in: Philologiae Turcicae Fundamenta I, eds J. Deny et al., Aquis
Mattiacis, pp. 308-317.
Benzing, J. (1949), Das Kumükische, in: Philologiae Turcicae Fundamenta 1, eds J. Deny et al.,
Aquis Mattiacis, pp. 391-406.
Benzing, J. (1954), Tschuwaschische Forschungen 5. ZDMG 104, pp. 386-390.
Benzing, J. (1959), Das Tschuwaschische, in: Philologiae Turcicae Fundamenta I, eds J. Deny et
al., Aquis Mattiacis, pp. 695-751.
Bereczki, G. (1994, 1992), Grundzüge der tscheremissischen Sprachgeschichte I-II, Szeged / Stu-
dia Uralo-Altaica 34, 35).
Caferoglu, A. Doerfer, G. (1949), Das Aserbeidschanische, in: Philologiae Turcicae Fundamenta
I, eds J. Deny et al. Aquis Mattiacis, pp. 280-307.
Clauson, G. (1959), The Turkish Numerals, JRAS, pp. 19-21.
Czeglédy, К. (1983), From the East to West: The A ge o f Nomadic Migrations in Eurasia, AEMAe
3, pp. 25-125.
Doerfer, G. (1967), Türkisch -n tschuwaschisch /‫?מ‬, UAJb 39, pp. 53-70.
Doerfer, G. (1973), Tschuwaschisch -m < urtürkisch *-m (> gemeintürkisch -n), UAJb 45, pp. 1 7 4 -
212 .
Doerfer, G. (1988), Grammatik des Chaladsch. Wiesbaden.

Eckmann, J. (1966), Chagatay Manual. Bloomington.
Erdal, M. (1991), Old Turkic Word Formation I-II. Wiesbaden.
Galkin, I. S. (1964, 1969), Istoriceskaja grammatika marijskogo jazyka , I-II. Joskar-Ola.
Golden, P. B. (1992), An Introduction to the History of the Turkic Peoples. Wiesbaden.
Greenberg, J. H. (1978), Generalizations about Numeral Systems, Universal of Human Language,
vol. 3. Stanford, pp. 250-295.
Grpnbech, K. (1936/1979), The Structure of the Turkic Languages. Bloomington.

Hakimzjanov, F. S. (1986), New Volga Bulgharian Inscriptions, AOH 40, pp. 173-177.
Hovdhaugen, E. (1972), Some Remarks on the Development o f Nasal Phonemes in Chuvash,
UAJb 44, pp. 208-212.
Ishakov, F. G. (1956), Cislitel’noe, in: Issledovanija po sravnitel’noj grammzatike tjurkskih jazy-
kov. Cast’ vtoraja, Morfologija, Moskva, pp. 176-207.
Johanson, L. (1 9 8 4-1986 ), Zur Konsonantenstärke im Türkischen, Orientalia Suecana pp. 33-35,
195-209.
Johanson, L. (1998), The history o f Turkic, in: L. Johanson-É. Â. Csato, The Turkic Languages,
London-New York, pp. 81-125.
Kovedjaeva, Je. I. (1976), Marijskij jazyk, in: Osnovy fînno-ugorskogo jazykoznanija. Marijskij,
permskie i ugorskie jazyki, Moskva, pp. 3-96.
Levitskaja, L. S. (1976), Istoriceskaja morfologija cuvasskogo jazyka. Moskva.
Ligeti L. (1986), A magyar nyelv török kapcsolatai a honfoglalds elâtt e's az Ârpdd-korban, Bu-
dapest.
Menges, K. H. (1968), The Turkic Languages and Peoples. An Introduction to Turkic Studies,
Wiesbaden.
Paasonen, H. (1908), Csuvas szàjegyzék, Budapest.
Pavlov, I. P. (1965), X al’xi cdvas literaturâ cëlxi. Morfologi, Cheboksary.
Räsänen, M. (1957), Materialien zur Morphologie der türkischen Sprachen, Helsinki.
Rassadin, V. I. (1978), Morfologija tofalarskogo jazyka v sravnitel’nom osvescenii, Moskva.
Röna-Tas, A. (1976), A Volga Bulgarian inscription from 1307, AOH 30, pp. 153-186.
Röna-Tas, A. (1982), The periodization and sources o f Chuvash linguistic history, in: Röna-Tas,
A. [ed.], Chuvash Studies, Budapest, pp. 113-169.
Rona-Tas, A. (1988), Turkic influence on the Uralic languages, in: Sinor, D. [ed.] The Uralic lan-
guages, Description, History and Foreign Influences, Leiden etc., pp. 742-780.
Röna-Tas, A. (1996), Ugor, ogur or ugur? Remarks on the name “Finno-Ugrian” in: Ünnepi könyv
Mikola Tiber tiszteletére, szerkesztette Mészâros Edit, Szeged, pp. 265-269.
Rona-Tas, A. (1999), Hungarians and Europe in the Early Middle Ages. An Introduction to Early
Hungarian History, Budapest.
Schönig, C. (1991), Das Lenatürkische und die sprachlichen Merkmale des nordöstlichen türki-
sehen Areals, in: Altaica Osloensia. Proceedings o f the 32nd Meeting of the Permanent
International Altaistic Conference, Oslo, June 12-16, 1989, Oslo.
Schönig, C. (1997), A new attempt to classify the Turkic languages (1). Turkic Languages 1,
pp. 117-133.
Serebrennikov, B. A.-Gadzieva, N. Z. (1979), Sravnitel’no-istoriceskaja grammatika tjurkskih ja-
zykov, Baku.
[Shcherbak] Scerbak, A. M. (1976), Porjadkovye cislitel’nye v tjurkskih jazykah, UAJb 48,
pp. 205-212.
Stachowski, M. (1993), Dolganischer Wortschatz, Kraköw.
Stachowski, M. (1994), Urtürkisch *me, *nc und das jakutische Ordinalsuffix, Rocznik Orienta-
listyezny 49, pp. 177-183.
Andrâs Rona-Tas
House o f Professors, Room No. 303
H-l 146 Budapest
Ajtösi Dürer sor 19-21
Hungary

Acta Orientalia Academiae Scientiarum Hung. Volume 52 (1), 17—44 (1999)
ON COGNATESHIP OF OLD JAPANESE i/si

WITH OLD KOREAN i/si
YOSH IZO ITABASHI

(Fukuoka)
Until very recent years, few attempts to compare the Old Japanese morpheme i/si with the Old Ko-
rean morpheme i/si have been made in order to ultimately discover the origin of those Old Japanese
morphemes. Most o f the earlier attempts (e.g. S. Yoshitake 1929, pp. 889-895) were concerned
only with a few o f the extant examples of OJ i and never evolved a hypothesis on the origin o f those
morphemes. There are some recent attempts (e.g. S. Martin 1987, p. 64) engaged in connecting
Japanese with Korean and/or Altaic.
In this paper I try to explain one function o f Old Japanese i/si at a time and compare with
the corresponding Old Korean i/si. A conclusion will be reached with the help o f the result o f the
comparisons o f Japanese and Korean.
Key words: Japanese, Korean affixes, indefinite, definite, generic, specific
[1] Functions of personal pronouns
(1) 2nd person pronoun
(a) Japanese
As forms of the second person of the personal pronouns in Old Japanese, there are
nalnare, i, si, and 02 re, which are independently used. Among them, i is the only
pronoun which is used solely for the 2nd person and not for any other person.
As shown in the examples below, we have a 2nd person pronoun i among the
other 2nd personal pronouns like na, si, 0 2 re.
1] i 1)ga tukuri tukaFe2maturu 02 F0 2 t02 n 02 uti ni Fa... [K part 2, NKBT p. 157]
i qga tukuri tukaFe2maturu 0 2 F 02 t02 n02 uti ni Fa
you [gen.] build serve[hon.] great hall [pros.] [top.]
‘(you will go first) into the great hall you have built in service (to me)’
0001-6446 /99/ $ 5.00 © 1999 Akadémiai Kiado, Budapest
18 Y. ITABASHI
Here we have the second person pronoun i followed by the genitive case suf-
fix which nominalisés the whole relative clause modifying the following 02 ^ ° 202‫ז‬02‫מ‬
uti (the great hall).
(b) Ryukyu
Now we will consider the Ryukyu 2nd person singular pronouns. Ryukyu has a 2nd
person singular pronoun ya in the Mid-South Okinawa dialect. This form is used
only with the meaning of deprecation and was probably developed through the fol-
lowing series of stages (cf. Nakamoto 1983, p. 162):
ya < ?ya < *?iya < *irya < *ira < *iri + *a < *i + *re
Examples:
1] ya ga katji bitji [Okinawa dialect: Uchima 1984, p. 443]

y a ga katji bitji
you [nom.] write should
‘You should write (it)’
We find a reconstruction of the 2nd person pronoun *i among the four types
of the 2nd person pronouns:
ла-type, ore-type, und311-type, and oka-type,
most of which are honorific. On the basis of this classification, the following forms
can be reconstructed in pre-Ryukyu:
*na *ore *ire *onoga
(*na and *onoga are not treated here since they are out of scope of this paper.)
Pre-R. *ore compares with OJ 02 re\ this can be postulated for Proto-Japanese
as 0 2 re.
Pre-R. *ire is paired with OJ 02 re, in which Pre-R. *i is in opposition to OJ 02•
This pre-R.*! must be cognate with the Old Japanese 2nd person pronoun i.
This Ry *i is certainly cognate with OJ i, which is one of the two pure pro-
nouns (the other one is OJ na) as opposed to the denominal pronouns.
Both Miller (1981, pp. 183-184) and Murayama (1950, pp. 45-46) claim that
the Old Japanese 2nd person pronoun i is comparable to the 3rd person singular
pronoun i in all main branches of Altaic and can be postulated as *i for Proto-Altaic
in the function of the 3rd person singular pronoun. They also insist that the Old Japa-
nese 3rd person singular pronoun i was inherited from the Proto-Altaic 3rd person
pronoun and then it became extended in use to the 2nd person singular pronoun.
Since this kind of extension is also very common in numerous language families, this
seems to suggest that this transfer of meaning occurred latest by the end of the pre-
Japanese period. But when we consider all other functions of OJ i (to be mentioned
in the later section), we cannot possibly relate it with Proto-Altaic *1' or with i of all
ON COGNATESHIP OF OLD JAPANESE i/si WITH OLD KOREAN i/si 19
the Altaic branches because none of the Altaic languages display any of the other
functions of the OJ i. We can hardly think that the eventually supposable other func-
tions in all the Altaic branches might have been lost. Hence, we need to search an-
other language family for the origin of this element.
[2] Old Japanese and Ryukyu morphemes *i and *si
There are many morphemes with different functions in Old Japanese and Ryukyu
which are identical with the Old Japanese and Ryukyu pronouns i/si. All these mor-
phemes may not only phonetically/etymologically but also semantically be related
with one another. Almost all scholars (e.g. Miller 1989, pp. 251-253) have treated all
these morphemes separately because they thought that the functions of all these mor-
phemes were different from one another. They have never attempted to integrate
them into a larger framework that can readily handle all these morphemes including
the Old Japanese and Ryukyu noun i and the demonstrative and third person pro-
nouns si. The scope of their studies seemed to be too narrow to cover all of the func-
tions. That is why they failed to explain all the functions systematically.
When we systematically consider all the functions of these identical mor-
phemes and the nouns and pronouns mentioned above, we may be able to reason that
all these morphemes belonging to the two separate respective word families, *i would
have an underlying common function, although they appear to have different func-
tions on the surface.
As explained above, the functions of the Altaic 2nd and 3rd person pronouns
cannot explicate all the functions of the Old Japanese and Ryukyu morphemes and
pronouns, since the functions of the Altaic pronouns do not display most of the func-
tions of the OJ and Ry i and si. Thus, they must be traced into some other languages
and/or language families which are neighbours of Japanese.
In the following paragraph, we will consider all the functions of the Old Japa-
nese and Ryukyu morphemes i which are identical with the 2nd person pronoun i.
First, we will look at all the suffixes with different functions:
[A] All the functions of the morpheme i
(a) suffixes
® emphatic nominative case suffix
Japanese
1] naramaro komarora-i sakasimanaru t02 m02 gara wo FizanaFi Fikiwite...
[Sem. 19]

20 Y. ITABASHI
naramaro komaro-ra i sakasimanaru t02 m02 gara

Naramaro Komaro-[pl] [nom.] rebellious friend-[pl]
wo FizanaFi Fikiwite..
[acc.] inciting leading
‘Naramaro, Komaro, and other bad and rebellious men inciting and leading
a band of rebels...’
2] aritigata arinagusame2te2 ikamed02 m 02 iFenaru imo i obosimisemu

[Man. 3161]
aritigata arinagusame2te2 ika-me-d0 2 m02 iFe-naru imo i obosimisemu
[pil.word] consoling go-wish-[conj.] home-is wife [nom.] long for
‘Although I would like to go, consoling myself with the beautiful gloomy and
uncertain...’
Ryukyu
1] ototaru-( kimokarato... [Omoro. 14-11 (992)]
ototaru i kimo-karato
Ototaru [nom.] heart-from
‘Ototaru is (adored) very much and...’
The element i in the Old Japanese and Ryukyu examples clearly shows the
nominative case suffix, the function of which is to emphasise the nominativeness
(otherwise, the nominative as a case stands normally without any specific suffix in
Old Japanese). In Ryukyu ga appears in the same function; this Ryukyu ga is cog-
nate with the Old Japanese rjga.
® instrumental case suffix
Ryukyu
1] katanautfi-i jakuni toyomiyoware [Omoro. 1-5(5)]
katanautfi i jakuni toyomi yoware
sword-having [inst] country get famous [resp.]
T hope I, having a sword, will be famous throughout the country’
The instrumental case cannot be found in Old Japanese and is found only in
the above example of Old Ryukyu.
Both Iwa and Hokama (1981, pp. 226-227) have claimed that the i in the
example is a kakari-joshi. This kakari-joshi is correlated with the sentence-final verb
+ yoware (izenkei form) because this kakari-musubi concord is identical with ‘su’
---- ’ V + yoware’ (of which su corresponded to the Old Japanese k 02 s02 ) except for
the suffix i rather than su. This su kakari-musubi concord is well-established in the
Omorosaushi, so that according to Iwa & Hokama’s claim, this / must also be a ka-

kari-musubi concord having the same function as the su concord. However, there
remains a question: since the Omorosaushi has already had a well-established su con-
cord, why would it need some other concord which had the same function as that of
the su concord? This implies that this i concord must have had a different function
than the su concord had; this i can be regarded as the instrumental case suffix rather
than the kakari-joshi. The idea of the i concord is doubtful because this example is
the only one with their claimed i concord as opposed to the su concord which oc-
curred in many examples of the Omorosaushi.
® vocative case suffix
Ryukyu
1] otomako-i akamako-i okaruna [Omoro. 14-17 (998)]
otomako i akamako i okaru-na
my sister [voc.] my sister [voc.] be present-[ques.]
‘My sisters, are you home?’
This is the only example of the vocative case in Ryukyu and no similar use of
it can be found in Old Japanese. The i in this example clearly shows the vocative
case which serves as the emphasis on otomako and akamako, the nouns before the
suffix. It means that this suffix makes these nouns stand out as opposed to the other
elements in the sentence.
® locative case suffix
Japanese
1] kono kuni-i keu-wo Firomuru-ni... [Konkomyosaishokyo Vol. 6 (early Heian)]
kono kuni i keu wo Firomuru ni
this country [10c.] sutra [acc.] spread [temp.]
‘When I go on a mission in this country...’
Ryukyu
1] Fikamitji-i yaki-no-omoikiya matfiyori... [Omoro. 14-15 (996)]
Fikamitji i yaki-no-omoiki-ya matfi-yori
east road [loc.] Yaki-[pos.]-love-[emp.] waiting-is
‘My love of Yaki is waiting for me on the east road,’
/ functioning as the locative case is not found in any document of the early
Old Japanese period like Nara, which does not necessarily mean that the locative
case of i did not exist at that time. But rather, that case of i was probably used in Old
Japanese, but it did not appear in any document until the early Heian period. What is

22 Y. ITABASHI
more, the locative case suffix ni almost took over the locative suffix i in Old Japa-
nese (we believe i and ni were developed from the referential nominal *i) and the
locative function of i reappeared in the formal document, which tends to retain old
forms and remnants, such as Buddhist materials at a later time.
Ryukyu also has a locative case suffix i as shown in the example above. Some
scholars such as Hokama (1981, p. 225) suppose that the inscription of this letter was
‘ha’ rather than T , but in some other texts except for the Iwa text we encounter T,
thus we safely assume that the letter in question is T and not ‘ha’.
In terms of functions, there is no doubt that this Ryukyu i serves as the loca-
tive case suffix, rather than just an emphatic element on the noun before it, although
its original function must have been to give a deictic emphasis to the preceding noun.
© unbound flipping suffix (N —» N)
Japanese
1] mitumitusi kume2 n02 ko-ra- qga kubututu-/ isitutu-/' moti... [K. 10; N.S. 9]
mitumitusi kume2 n 02 ko-ra rjga kubututu i
[pil.word] К ите [gen.] soldier-[pl] [nom.] kubutsutsu [flip]
isitutu i moti
ishizutsu [flip.] having
‘soldiers in К ите having kubutsutsu and ishizutsu (swords)...’
Ryukyu
1] Jirotsumo-/ majii mitjaru... [Omoro. 20-42 (1372)]
Jirotsumo i majii mi tjaru
paddy [acc.] footpath look [perf.]
‘I have looked at the paddy and footpaths’
In my previous papers (1989, 1990), I regarded this suffix i in the above exam-
pies as the accusative case suffix in Old Japanese and Old Ryukyu, but Vovin (1997,
p. 281) pointed out that this element i in Old Japanese must be a relic form tutui (the
later form tuti) of kubu-tutu and isi-tutu. Furthermore, Kawabata (1997, pp. 26-27)
independently substantiated the similar idea that tutui was an unbound form of tuti
(rather than the relic form of tuti),which is found in the Shindai(ge) or Shinko of the
Nihonshoki. Thus, it would be best to describe this suffix as an unbound flipping
suffix.
There is also another problem with my previous view that the element was an
accusative suffix, which Vovin did not mention; if this suffix had been in the accusa-
tive case, then that suffix with the same function could never have appeared twice
within a clause or a simple sentence in Old Japanese because of the Old Japanese
syntactic rule. Thus, this suffix cannot be an accusative case suffix.

There is also another reason to doubt that this suffix is the accusative case
suffix: this is the one and only example found in any documents of the Old Japanese
period, and this text above appears repeated several times in identical form.
The Ryukyu example shows the suffix i, but most linguists believe that this
should be interpreted as ri in orthography which is part of the preceding noun, as in
the example above, firotsumo. Thus, this example may not indicate an accusative use
of i, so that we may exclude this Ryukyu example due to this doubt.
Japanese
1] saka- ‘sake’+ / > sake2 ‘sake’
2] ama- ‘rain, heaven’ + i > ame2 ‘rain’
Ryukyu (Omoro.)
1] saka- ‘sake’+ i > sake ‘sake’
2] ama- ‘rain, heaven’ + i > ame ‘rain’
As in the examples immediately above, the function of the suffix i is deic-

tic/emphatic; by flipping the bound nominal form to the unbound nominal form: the
noun itself becomes more definite and specific. Saka- itself is not used alone but is
used with other nouns like -ya ‘shop’, the combination of which becomes a word di-
rectly related to sake, whereas sake2 directly refers to the liquid itself. The relation-
ship of ama- to ame2 is similar to the first example: ame2 is the specific material
falling from the ama-, which is not normally employed separately.
© verbal suffixes
1) predicate introducing suffix
Japanese
1] *kak- + i > kaki- ‘to write’
2] *tat- + i > tati- ‘to stand up’
Ryukyu
1] *kak- + i > katji- ‘to write’ [Ryuka]
2] *num- + i > numi- ‘to drink’ [Ryuka]
This kind of suffix i appears only in four-step, upper one- and two-step, and ir-
regular conjugation verbs in Old Japanese. The same suffix appears only in one- and
two-step conjugation verbs in Ryukyu.
As in both Japanese and Ryukyu examples, this suffix i is thought to serve as
a nominalising element for verbs. When a Ryukyu verb has a final [k] in the stem, it

24 Y. ITABASHI
becomes palatalised into [tf] as shown in Ryukyu example 1] above. On the other
hand, when a Japanese verb has the same consonant in the stem, it never becomes
palatalised. However, this phonological difference never affects the meaning of the
word.
Furthermore, this suffix is also considered as a predicating suffix if the predi-
cate contains the final (shushikei) form with -i of ra- line irregular verbs (like ar-i
and wor-i). This -i suffixed predicate seems to be more of a basic form than the other
form ending in -u of four-step verbs (such as tat-и ‘to stand’ and tug-u ‘to tell’). That
is, this suffix -i can end a sentence. The predicating function contradicts the previous
tenet of the nominalising function. We take the latter view because we regard some
other derived conjugations mentioned above and below as part of a unified verbal
system; those conjugations can be explicated in a lot more systematic way than the
ordinary view of the nominalising function (cf. Matsumoto 1995, pp. 162-165).
2) conditional(izenkei) forming suffix
Japanese
1] *kak-a- (indefinitive [mizenkei]: vt) + i > kake2‫‘ ־‬write (vt)’
2] *tat-a- (indefinitive [mizenkei]: vi) + i > tate2‫‘ ־‬stand up (vi)’
Ryukyu
1] *ik-a- (indefinitive [mizenkei]: vi) + i > *ike2‫> ־‬ike- ‘go(vi)’
2] *mat-a- (indefinitive [mizenkei]: vt) + i > *mate2‫ > ־‬mate- ‘wait for(vt)’
The izenkei form e2 of four-step conjugation verbs is derived from the combi-
nation of the mizenkei suffix a and the izenkei forming suffix /. This suffix i is
probably the same as that of © (verbal suffixes) above: the deictic/emphatic suffix.
The relationship of the mizenkei form, for instance, kak-a- to the original form of the
izenkei and meireikei forms kak-a-i > kake2 is that of the bound form ama- to the
unbound form ama-i > ame2. The main function of the forms of four-step conjuga-
tion verbs (e.g. kak-a-) and of the bound form of nouns (e.g. ama-) is characterised
by ‘indefinite’ or ‘generic’, whereas that of the forms of verbs (e.g. kaka-i > kake2)
and nouns (e.g. ama-i > ame2) is characterised by ‘definite’ or ‘specific’. Thus, the
main function of the izenkei form is deictic/emphatic such as ‘iriFi sasinure2 (since
the setting sunshone,...)’ [Man. 135], where the sasinure2 itself makes the condi-
tional clause without a correlative emphatic word ko2so2. This implies that the origi-
nal function of the izenkei form was probably to make verb stems independent and
emphatic just as unbound nouns do (Matsumoto 1995, p. 166).

3) imperative (meireikei) forming suffix
Japanese
1] oshaku-ni maira-f to no onoFosenari [Otogisaushi; Kokurie]
oshaku ni maira i to no onoFose-nari
pouring tpurp.] coming [imp.] [quo.] [pos.] command-[cop.]
‘He commanded, ”Come to pour (sake into my cup)”
Although only one example is cited from the early Edo period (17th c.), it may
be the case that the use of this suffix in Old Japanese may not have been proven.
However, there are some dialects where this function of the suffix is found; for in-
stance, the Sendai dialect (Miyagi Prefecture) has the same suffix i(n) to make an im-
perative (meireikei) conjugation:
1] oga- ‘place’ [mizenkei] + i(n) > ogai

2] hashira- ‘run’ [mizenkei] + i(n) > hashirai
3] mi-ra- ‘look’ [mizenkei] + г(п) > mirai
4] uda- ‘strike’ [mizenkei] + г(п) > udai
We do not know the function of the final nasal /-n/ of /-i(n)/, but it may be
some kind of a suffix. In fact, this final /-n/ can be omitted, but it seems that the older
generations have this nasal or a nasalised vowel /-1/, in which case it may have been
a fusion of the vowel and the following nasal /-n/.
Of course this kind of dialectal form does not prove any existence of this suf-
fix in Old Japanese, since the listed forms in this dialect may possibly be an internal
development. However, the possibility of the internal development is extremely low;
the function of that dialect must be a remnant of the function of Old Japanese be-
cause the phonological fusion of -a-i before the Old Japanese period would not have
taken place in that pre-dialect and remained intact, while the ordinary imperative
(meireikei) form is the result of that fusion, namely, -«2 (< -a‫־‬i)> which may have
taken place in Old Japanese. These dialectal examples infer that at least this suffix is
not so rare and give support to our supposition of the existence of this suffix before
the Old Japanese period.
The function of this suffix must be the same as that of the izenkei forming suf-
fix, that is, the deictic/emphatic suffix, since the meireikei form is employed to spe-
cifically address the second person. As its name shows, the suffix of the imperative
(meireikei) form is a stronger realisation of the deictic/emphatic function than that of
the izenkei form because the action indicated by a verb is specifically directed
toward the 2nd person.

26 Y. ITABASHI
4) resultant-state forming suffix
Japanese
1] *tata- (indefinitive; vi) + i > tate2‫‘ ־‬to stand (vt)’
2] *nura- (indefinitive; vt) + i > nure2‫‘ ־‬to become wet (vi)’
3] *02 t02 ‫( ־‬nominal adjective: ‘inferior’) + i > 02 ti- ‘to fall (vi)’
4] *y02 k02 (nominal: ‘peripheral area’) + i > y02 ki- ‘to avoid/detour (vi)’
Ryukyu
1] *tata-(indefinitive; vi) + г > *tate2‫ > ־‬täte- ‘to stand (vt)’
2] *tuka- (indefinitive; vi) + i > *tuke2‫ > ־‬tuka- ‘to stick (vt)’
3] *taka- (nominal adjective: ‘high’) + i > *take2‫ > ־‬take- ‘to be superior (vi)’
4] *aka (nominal: ‘red (n)’) + i > *ake2 ‫ > ־‬ake- ‘to dawn (vi)’
In Japanese examples 1 and 2, the function of the suffix i is to change a transi-

tive to an intransitive verb and vice versa. The suffix / of the ren’yokei form of four-
step conjugation verbs, for example, tat-i is originally the same as that of lower two-
step conjugation verbs tata- i (> tate2), which infers that the original function of the
suffix i to verbal stems is to predicate them (Matsumoto 1995, p. 162).
In Ryukyu examples 1 and 2, the same can be said about the original function
of the suffix i, although there seem to be fewer examples in Ryukyu than in Japa-
nese. Thus, the suffixed verbs in the first two examples of Japanese and Ryukyu can
be called ‘resultant state verbs’. The fact that the intransitive verbs changed their
function to the transitive verbs is that the original resultant state was reinterpreted
into their actor-action motion (Matsumoto 1995, p. 165). For instance (Matsumoto
1995, p. 165):
*tata nama-i
shield line up
The possible meanings are:
® shields form in line

® shields are being lined up.
® someone forms shields in line.
The original meaning of the sentence was probably either Ф or @, but then
that resultant state was restructured into the actor-action motion, which the verb indi-
cates, name2- to mean @. There seem to be many examples of re-interpretation of
verbs into actor-action verbs or agent/situation verbs in which case the original verbs
are transitive. Therefore, it is safe to claim that each ren’yokei form of four- and two-
step conjugation verbs is a predicating form and a derived predicating form (e.g. tat-i
A da Orient. Hung. 52, 1999

(vi) and tata-i (vt)) respectively, so that these two verbs were originally one single
verb, which came to conjugate separately (Kawabata 1997, p. 337; Matsumoto 1995,
pp. 165-166).
Japanese examples 3 and 4 show that the function of the suffix i is to change
an adjective or a nominal to a verb. We can state from this observation that the origi-
nal function of this suffix i is to change from a simple state to the resultant state (to
make inchoative verbs) which the verbal stems or nouns indicate, rather than refer-
ring to the simple state or nominal form. The same principle is applied to the third
and fourth Ryukyu examples, although the number of the examples may be fewer
than those of Old Japanese.
To sum up, we can state with certainty that the original function common to
all the suffixes i is deictic/emphatic; some of those suffixes appear as markers to de-
note special emphasis.
(b) prefixes
® emphatic prefix (to nouns, verbs and adjectives)
Japanese
1] komoriku-n02 hatuse-no? kaFa-n02 kami-tu se-ni г-kuFi-wo uti, ...
[Man. 3236] "
komoriku n02 hatuse n02 kaFa n02 kami tu se ni
[pil.word] [gen.] Hatsuse [gen.] river [gen.] upstream [10c.] rapid [loc.]
i kuFi wo uti
[emp.] peg [acc.] strike
‘striking the pegs in the rapid of the upstream of Hatsuse River, ...’
2] umi-ni г-detaru sikama kaFa... [Man. 3605]

umi ni i detaru sikama kaFa
ocean [dir.] [emp.] going-out Shikama river
‘Shikama River running into the ocean...’
Ryukyu
1] JiForaJi 1-kotoba-ya... [Ryuka. 224]
JiFora/i i kotoba ya
splendid [emp.] words [excl.]
‘splendid words...’
2] to/i-ka mitose г-kiyote... [Omoro. 12-7 (658)]

toji ka mitose i kiyote
year [gen.] 3rd-year [emp.] inviting
‘In the 3rd year we invited (the Goddess), ...’

28 Y. ITABASHI
3] mi-inotji-no tsuna mi-ho/i-no tsuna (-jiyoku, ... [Nantokayo (jo) 0-29-29]

mi inotfi no tsuna mi hoji no tsuna i jiyoku
[hon.] life [gen.] rope [hon.] star [gen.] rope [emp.] strong
‘The rope of life and of a star is so strong , ... ’
The prefix i is not only found in Old Japanese but also Ryukyu. There are nu-
merous verbal examples of this prefix in Manyoshu, while there are very few nomi-
nal examples of this prefix in that document. In the Omorosaushi of Ryukyu, on the
other hand, there are a number of verbal and nominal examples of this prefix.
The function of the prefix i is to emphasise either a noun or a verb after it. It
would probably be of the same origin as that of the unbound nominalising suffix i,
namely, the deictic/emphatic function. The noun or the verb to which the prefix
added becomes more definite and specific: the noun in the first example is a definite
and specific peg, which has the very similar function of the definite article ‘the’ of
English. The verb ikoFu in example 1 indicates the particular invitation of the God-
dess. Likewise, the verb, idetaru in the second example is used with a directive case
suffix which indicates a specific destination, the ocean the river runs into.
With the i in the 3rd Ryukyu example, the adjective jiyo-shi carries emphasis.
This kind of adjectival emphasis is not found in Old Japanese. It may be the result of
a separate internal development rather than an old remnant from pre-Old Japanese.
Note that no remnants of this emphasis have been found in Old Japanese. What is
more, since the distribution of this emphasis i in Ryukyu is broader than that in Old
Japanese, this may assist to indicate that the adjectival emphasis of Pre-Ryukyu i may
have been the secondary development.
(c) demonstrative pronoun ‘that’
Japanese
1] aratay02 ni t02 m02 ni aramu t02 taman02 wo n02 tayeji i imo t02 musubite si
k02 t02 Fatasazu... [Man. 481]
aratay02 ni t02 m 02 ni aramu t02 taman02 W0 n02 tayeji
new age [loc.] together exist [quo.] [pil.word] [gen.] disappear
i imo t02 musubite si k02 t02 Fatasa-zu
that wife [com.] unite [emp.] matter realise-not
T could not promise my wife to last our relationship forever, ...’
2] Farukaze-ni midarenu-i т а -ni misemu kom02 ‫־‬gam02 [Man. 1851]

Farukaze ni midare-nu i ma ni mise mu
spring wind by become disarrayed-not that interval [temp.] show [vol.]
ko m02 gam02
child [emp.] [excl.]
‘While spring wind has not yet disarrayed (it), I wish I could show (it) to
children’

This function is found only in Old Japanese and not in Ryukyu. The demon-
strative pronoun appears to be always preceded by the attributive form (rentaikei) of
the verb, tayeji- and midarenu-.
The function of this i is to emphasise the clause immediately before it in order
to relate that clause with the following clause. In other words, the i has the function
of pointing back to the previous clause: to a demonstrative pronoun. There is at least
one more example in Manyoshu for this: No. 1359.
It is also possible to regard this i as a noun meaning ‘fact’. If this is the case,
then this noun may have developed from the original indistinguishable deictic/em-
phatic function of i through the stage of the independent demonstrative pronoun; in
this case, it belongs to the category of the following section (noun indicating ‘per-
son/fact’). In either case, these seemingly different functions may have been derived
from the original deictic/emphatic function of i.
Martin (1990, p. 500 endnote 2) mentions that the demonstrative pronoun i may
perhaps be borrowed from the Chinese pronoun i ‘he, this’. Although Martin’s idea
is possible, it is unlikely because we do not find any other borrowings of this type of
basic vocabulary if we considered it as a borrowing.
Therefore, at this point we do not strongly claim that this function of i is to
emphasise the immediate clause, rather than to nominalisé that clause.
(d) quasi-free noun ‘person/fact/one’
Japanese
1] k02 re-wo tamotu-z'-Fa Fomare-wo itasi, ... [Sem. 45]
k02 re wo tamotu i Fa Fomare wo itasi
this [acc.] cherish psn/fact [top.] praise [acc.] receive
‘Those who believe in this (teaching) are to gain praise,.../believing in this
(teaching) is to gain praise, ...’
2] suturu-z-Fa sosiri wo maneki-tu. [Sem. 45]

suturu i Fa sosiri wo maneki tu
abandon psn/fact [top.] abuse [acc.] invite [cont.]
‘those who do not believe in (this teaching) are to invite abuse/not believing in
(this teaching) is to invite abuse’
It would be easier to grasp this semantic relationship between the two, if we

think of this relationship as that of mono designating ‘person, fact’.
This noun i is found only in Old Japanese, but not in Ryukyu, although the
similar noun si is found only in Ryukyu. There are very few examples of this noun
i even in Old Japanese and the only examples we have found are cited from the same
compound sentences in Semmyo. This seems to imply that this function had already
become unproductive or obsolete by the time of the early Old Japanese period. There
is another possibility that this i was even borrowed from Old Korean i meaning ‘per-

30 Y. ITABASHI
son, fact, one’ (see the later section) since the meaning of the OJ i is identical with
that of the OK i.
The noun i in these examples indicates the two separate meanings ‘person’
and ‘fact’, which are indistinguishable in interpreting the sentence, since either
meaning can be understood. We cannot determine which of the two meanings is
original: one can postulate that the meaning ‘person’ is more likely to have been used
because it has more concrete meaning than ‘fact’, which represents abstract concept.
Normally a concrete meaning is older than the abstract one. On the other hand, if the
primordial meaning was something deictic/emphatic, then the meaning ‘fact’ may
have been derived directly from the emotional concept.
[B] All the functions of the morpheme si
(a) suffixes
® emphatic suffix
Japanese
1] kimi-wo-si 02m02Feba inekatenukam02 [Man. 607]
kimi wo s i 0 2 m02 Fe ba ine katenu kam02
you [acc.] [emp.] think [cond.] sleep difficult wonder
‘as I long for you, alas I cannot sleep!’
2] F02t02t02gisu k02 yu nakiwataru к02к02Г02‫?־‬г aru rasi [Man. 1476]

F02t02t02gisu k02 yu naki wataru k02k02r02 si aru rasi
cuckoo this place [pros.] sing cross heart [emp.] exist appear
‘the cuckoo, crying, flies over this place, perhaps he has (the same) feeling (as
I have)’
Ryukyu: no data
There is only one kind of suffix si, emphatic suffix, well attested by many
examples, but only in Old Japanese, not in Ryukyu. One of the features of that suffix
(as opposed to the suffix i) is that it can follow most parts of speech: after a noun
(e.g. k02k02r02 si, hito si), verb (e.g. miraku si, yorite si), other suffixes (e.g. sika si,
wo si), and it can precede other suffixes (e.g. si m02, si zof)■ It is interesting to note
that this suffix represented by the character *L ‘this’ and its meaning is similar to
that of z02 and ko2s02 in the later language, although si cannot end a sentence, while
Z02 and k02so2 can. This evidence shows that unlike the suffix i, this suffix cannot
morphologically and syntactically affect the predicate of a sentence. Its function is
simply to put emphasis on the particular part of speech immediately before it and/or
transfer on it the dominance of a deictic element.
Acta Orient. Hung. 52. 1999

ON COGNATESHIP OF OLD JAPANESE i/si WITH OLD KOREAN 1/5« 31
(b) demonstrative pronoun ‘that’
Japanese
1] sasibu-wo sasibu-n02 ki «-qga sita-ni 02 F2tateru [Kojiki: Nintoku]
sasibu-wo sasibu-n02 ki .sf-qga sita ni 02 F2tateru
sashibu-[emp.] sashibu-[gen.] tree it-[gen.] below [loc.] stand-grow
‘A sashibu-tree, below it stands growing (a true broad-foliaged camellia-tree
with five hundred-fold branches)’
2] FaFiro yutu matubaki sj-gga Fana-n02 teri-imasi [Kojiki: Nintoku]

FaFiro yutu matubaki s«-1]ga
broad-foliage five-hundred-fold-branches true camellia it-[gen.]
Fana-n02 teri-imasi
blossoms-[gen.] shine-be
‘a true broad-foliaged camellia-tree with five hundred-fold branches, its bios-
soms shine’
Ryukyu: no data
The suffix as a demonstrative pronoun is found only in Old Japanese, but not
in Ryukyu. However, there are only two examples for its use. This is because there are
other forms of the demonstrative pronoun s0 2 , which had taken over most of this de-
monstrative pronoun si by the time of recorded history. This infers that this example
above must have been a relic of an earlier systematic use of si as a demonstrative
pronoun. As mentioned above, the 2nd and 3rd personal pronouns were expressed by
the same 51; this means that the same element si had the undivided function of the
demonstrative and personal pronouns. In this section, we will not explain those cases
again, referring only to some further places of occurrence of the 3rd person pronoun
in the documents: Semmyo 24, 28, 48; Nihonshoki: Yuryaku, 13th year; Manyoshu
4211.
(c) noun indicating honorific title
Japanese: no data
Ryukyu
1] tjiyan-no-ji Fa nei/iyari, ... [Omoro. 15-48 (1099)]
tfiyana no ]-i Fa nei Jiyari,
Jana [gen.] person [top.] sound doing
‘The person from Jana calls the tune and...’

32 Y. ITABASHI
2] tau-no-Ji namutjiya kokane motfi mitjiweru [Omoro. 15-48 (1099)]

tau no Ji namutjiya kokane motfi mitjiweru
Do [gen.] person silver gold having filled
‘The person from Do has much silver and gold’
3] kise-no-Ji-ya waka ototjiya [Omoro. 17-7 (1181)]

kise no Ji ya waka ototjiya
Kise [gen.] person [top.] my brother
‘The person from Kise is my brother’
This type of abstract noun designating an honorific title such as ‘Mr, Majestic’
is not attested in Old Japanese and is found only in Ryukyu. Old Japanese had two
types of demonstrative pronouns i and si, the latter of which had probably the same
form as the previous stage of the Ryukyu abstract noun fi. This Ryukyu abstract
noun f i became developed from the demonstrative pronoun f i ‘he, she, they (that)
person’. Many examples of this semantic development can be found in the
Omorosaushi. The element i (Old Ryukyu), *i of *ire (both pre-Ryukyu) and f i (Old
Ryukyu) may be not only identical in form but also with regard to their etymology.
In a descendent system:
*ire---- < *i [3rd person sing.] + *re [pre-Ryukyu]

f i ‘person (honorific)’ ---------- < *si [pre-Ryukyu]
Although there are many other examples of f i in the Omorosaushi, we will

consider here just a few examples, f i in example 1 indicates ‘person (honorific title)’
and the whole nominal phrase refers to the particular person (a man with a honorific
title) who came from the village called ‘Jana’ and who may be the village chief.
f i in the second example shows, just the same way as in the first example,
‘person (honorific title)’, so the entire nominal phrase indicates the particular person
who came from the village called ‘Do’. If we go one step further, this nominal phrase
in its entirety can mean a person’s name (personal name). This type of semantic and
functional derivation is common in the Omorosaushi.
The f i in example 3 also indicates that ‘person (honorific title)’, and the whole
noun phrase refers to the particular person with a honorific title fi.
All these pieces of evidence suggest that the primordial function of this si was
something deictic or emphatic; however these qualities (deicticness and emphasis)
are not always separable and distinguishable in function.
In terms of the relationship between OJ i and OJ si, there seems to be a strong
morphological and semantic/functional relationship between the two; OJ i is very
close to OJ si in that the latter was developed from Pre-Proto-Japanese (hereafter
PPJ) *s + *i, the former (*si) of which gives the meaning of indefiniteness/generic-
ness to the latter (*/) of which has definiteness/specificity.
In connection with the primordial function of OJ i, that of OJ i had a tendency
to be definite and specific, while that of OJ si tends to be indefinite and generic,

ON COGNATESHIP OF OLD JAPANESE Usi WITH OLD KOREAN i/si 33
although the demarcation is not always clear. Both primordial functions may have
been generalised into emphasis, which is found as the primary function in the Old
Japanese texts. The distribution of all the functions of OJ i is much broader than that
of OJ si primarily because there was another predominant emphatic suffix 802 , which
is much broader in distribution than OJ si.
[C] All the functions o f Old Korean morpheme i
We find that a morpheme i has such uses as nominal uses, several nominal and ver-
bal suffixal uses, and a préfixai use in Middle and Old Korean. To itemise these uses:
® demonstrative pronoun (OK)

<Dquasi-free noun ‘person/fact/one’ (OK, MK)
® emphatic nominative case suffix (OK)
® genitive and possessive case suffix (OK)
© nominalising suffix (OK, MK)
© passive and causative forming suffix (МК)?
These elements listed above are normally interpreted as a phoneme N by Ko-

rean linguists. The element i as the nominative case suffix in Middle Korean, how-
ever, becomes a semivowel y and attached to the preceding vowel to produce a diph-
thong, after a vowel other than i or a semivowel y (Martin 1992, pp. 551-552).
Most of the following are examples from the Silla period of the Old Korean
texts. Citations are numbered in the system of Yang unless indicated otherwise. The
italicised portions of the texts are written in phonograms in the original, whereas the
rest are written in semantograms. The Chinese character written in the texts is indi-
cated in the parentheses.
(a) nominals
Consider the following examples:
Ф demonstrative pronoun ‘this’ (OK)
1] i her т э ki tasкга [3:6.,1-6.3] (life)

i her m3ki tasAra
this [acc.] feed govern
‘feed them this and govern’
2] i sta her рлп кокэп каПуэ [3:7.1-7.5] (JVc)

i sta her рлп ko& эй ka r tya
this land [acc.] desert and where go [fut.] [rhet.]
‘Even if you desert this land and go somewhere else, ...’

34 Y. ITABASHI
There are some other examples in the Hyangga, but two examples would be
enough to show that the morpheme i has the function of a demonstrative pronoun as
well as of a demonstrative adjective. In the first example the / is used for the demon-
strative nominal, whereas the i in example 2 is employed to display the demonstra-
tive adjectival.
This type of i may be cognate with (or the PK form may have possibly been
borrowed into pre-Old Japanese) the OJ demonstrative pronoun i as described ear-
Her, though the OJ i appears to immediately follow the attributive form (rentaikei) of
a verb. As shown above, the OK demonstrative i is not immediately preceded by
a verb unlike the case of the OJ i.
® quasi-free noun ‘person/fact/one’ (OK, MK)

1] küras hAsyaniroi [22:8.2-8.4] (§1: MC ?i)
kuras Ьл sya n i roi
that-way do [hon.] [perf.] [q.f.n.] [excl.]
‘they did in that manner’
2] hürunün i nunmuriyo, chinun i hansumira [Chapka]

hürünün i nunmuri yo, chinun i hansum ira
flowing [q.f.n] tear [excl.] coming out [q.f.n.] sigh is
‘what flows is tears and what comes out is sighs’
3] namujejurul chalhanün inun namuesô ttôrôjinta [Soktam]

namu jeju rül chalhanün i nun namu esô ttôrôjinta
tree talent [acc.] being good at [q.f.n.] [top.] tree [abl.] fall
‘the person who is good at climbing can fall from the tree’
In the first example, the morpheme i is translated into ‘fact’ or ‘thing’ as a quasi-
free noun. Unlike the case of the demonstrative pronoun above, this quasi-free noun
is always modified by a verb or other modifiers.
In the second example, both of the Middle Korean form i denote ‘thing’,
whereas in the third example, it means ‘person’. As indicated, these examples are
from Middle Korean, and so they are secondarily important to support the existence
of OK
In connection with the OJ quasi-free noun, the function of the OK quasi-free
noun is exactly the same as that of the OJ counterpart. In this respect, we can claim
that the OK quasi-free noun may be cognate with the OJ counterpart, although the
PK form may possibly be borrowed into pre-Old Japanese, just as the PK demonstra-
tive may have possibly been borrowed into pre-Old Japanese.
In Modern Korean, there are numerous types of so-called i forms denoting
‘a ... one (thing/person/creature)’, but the ones with the last two meanings (‘per-
son/creature’) are derived either from a (bound) noun or other parts of speech (Martin
1992, p. 554):

( 1) ‘a ... one (thing/person/creature)’

allok-z/allong-z ‘mottled pattem’
ccolccol-f' ‘a dogged follower, a hanger-on’
helleng-г ‘an unstable person, a twit’
mulkheng-z ‘soft stuff
( 2 ) ‘a creature that goes ... (that makes the sound...)’
kilek-г ‘a wild goose’
kkoykkol-г ‘a bush warblar’
mâym-z ‘a cicada’
(3) ‘a ... one (person/creature)’ derived from a noun or a bound noun’
saythwung-z ‘a silly person/acf
sisitek-/ ‘a silly fool’
kaksel-/ ‘a storyteller’
hölang-г ‘a tiger’
songchwung-z' ‘a pine-eating caterpillar’
Note that in group (2) the preceding stems in all the examples (kilek/kkoyk-
kol/maym) are certainly derived from the corresponding onomatopoeic words, so the
resulting words are formed by nominalising the onomatopoeia.
In connection with Japanese, we find some similar words with probably the
similar formation with a suffix -kil-gi or possibly -i, exemplified by such nouns
meaning ‘a ... creature’ as:
OJ tanuki ‘a badger’ (MK nakori ‘a badger’)

OJ usagi ‘a rabbit’ (MK t ’okki ‘a rabbit’)
OJ kasasagi ‘a magpie’ (MK kkac’i ‘a magpie’)
These OJ words may be cognate with or borrowed from its primordial forms
of Proto-Korean which are seen as the later forms of Middle Korean described
above. Note also that some Tungusic nouns for animal names are formed with a suf-
fix -kil-gi, which may be cognate with some Japanese (and possibly Korean) words
for animal names. The reason that those Tungusic nouns may not be cognate with the
Middle Korean nouns is because the morpheme boundary of the Middle Korean
words (N stem + i) is different from that of the Tungusic words (N stem + ki/gi).
® emphatic nominative case suffix (OK)
Now, let us look at the following examples:
1] karori noihi эга [5:4.1-4.2] (§*: MC ?i)

karor i noihi эга
legs [nom.] four exist
‘there are four legs’

36 Y. ITABASHI
2] kirangi cüz i isyura [4:5.1-5.3] (5Ë: MC si)

kirang i cüz i isyu ra
Kilbo [gen.] appearance [nom.] exist [excl.]
‘Knight Kilbo’s image is there!’
3] т о т i рлгак tüt’üri kamai [22:5.1-5.4] ($J: MC ?i)

тот i рлгак tüt’ür i kam ai
body [nom.] dissolve dust [nom.] going [10c.]
‘were (this) body to dissolve into dust...’
All these examples clearly show that the suffix i functions as the nominative
case. It seems that this is the only function of the nominative case suffix. Therefore
we may conclude that the Old Korean nominative case suffix i had nothing but the
nominative function and that that function has remained intact down to the Modem
Korean period, though the other nominative case suffix ka in Middle Korean began
to share the nominative function from the Middle Korean period on.
In connection with the Old Japanese nominative case suffix, the form and the
meaning are identical with that of the Old Korean nominative case suffix i. Thus, the
Old Korean nominative case suffix i may be cognate with the Old Japanese nomina-
tive case suffix.
® genitive and possessive case suffix (OK)
There are many other examples in the Hangga text, we will show some of the typical
examples as follows:
1] пуэг s a l s murskAz kantarp ’ai wovon cashurran рлгако [12:1.1-2.1] ($|: MC ji)
пуэ/ szi s murskAz kantarp’a i wovon cas hurran
long ago east [gen.] shore Gandharva [gen.] play castle [acc.]
рлга ко
advance toward [conj.]
‘... and the (Japanese armies) were advancing toward the castle on the eastern
shore where a long time ago the Gandharva used to play...’
2] samhwa i ovAmposyaor tut&0 ... [12:5.1-5.3] ( ^ : MC ji)

samhwa i orAm po syaor tut ко
three knights [gen.] climb see [hon.] hear [conj.]
‘as the (moon) heard that the three knights visited the (mountain) ...’
3] cai mAZAmi... [14:1.1-1.2] ( ^ : MC p’jwei)

ca i т л гл т i
self [gen.] mind [gen.]
‘of my mind’

The genitive case suffix in examples 1 and 2 nominalisés the clause, whereas
the third example shows the regular possessive (not genitive) case suffix. Note that
the genitive case suffix in these examples is identical in function with OJ no2 and OJ
Vga-
Although the genitive case suffix in the earlier Old Korean period does not
show a morphological dichotomy between a genitive and a possessive, it does dem-
onstrate the possessive as in example 3.
© nominalising suffix (OK, MK)
We find some examples from the Hyangga texts for this use. Here is an example
(Yang 1957, pp. 393-394):
1] syaßxr pxrki tzrai [5:1.2] ($0: MC gi)

syaßzr рлгк-i tAr-ai
capital bright-[nom i.attr.] m oon-[dir.]
‘to the bright moon of the capital...’
Middle Korean (Yi 1958, p. 106):
2] nyarümjizi < nyarumjiz-i ‘farming’ < пуэ ‘farm’ + mmjiz ‘to perform’ + i
3] sonjabi < sonjap-i ‘a handle’ < son ‘hand’ + jap ‘to grasp’ + i
4] c’uksari c c ’uksar-/ ‘life and death’ < c ’uk ‘death’ + sar ‘to live’ + i
The function of i as a nominaliser is attested in Old Korean as shown in exam-

pie 1. Note that this i in example 1 functions not only as a nominaliser but also as an
attributive form for the following noun ш ‘moon’. This claim has been accepted by
many Korean linguists such as Lee Ki-moon ( $ 2= JC), Chi Нэп-уэп (jfe 'M H ),
and Kim Chun-уэп (35z >Ü). Therefore, we can state with certainty that Old Ko-
rean indeed had a nominalising suffix i.
The p- in -jap- in example 3 and the / in -sal- in example 4 become b and r in
intervocalic position in Middle Korean respectively. The forms of these verbs are the
stems so that the combined forms of these verbal stems with i make the deverbal
nouns. Thus, the i in these examples is not a quasi-free noun but a nominalising suf-
fix. We are now almost certain that this nominalising suffix i existed probably before
the Old Korean period through the present time.
Therefore, we can almost certainly contend that the OK nominaliser i may be
cognate with the OJ equivalent.
© passive & causative forming suffix (MK)
There are basically two conflicting ideas about the primordial form(s) of the passive
and of the causative of Middle Korean:
There are several different shapes of suffixes for the passive /-hi/ with its alio-
phones [-hi (stem C: p, t, c), -ki (stem C: m, s), -’i (stem C: z, r), -i (stem C: others]

38 Y. ITABASHI
and for the causative /-hi/, /-рл/; /-h/, /‫־‬л/, (the first phoneme /-hi/ has the same alio-
phones with the passive). Each verb has its own set suffixes for the passive and the
causative (Lee 1975, pp. 168-169).
(1 ) *-hi for the passive and *-hi and *-/)/ for the causative in Old Korean, and
(2) *i for the passive and the causative and *u also for the causative
(e.g. Ch’oe Bôm Kong 1981, p. 124).
As indicated above, the reason that we touched on the forms of the passive
and of the causative is that there is an opinion that the primordial forms of the passive
and the causative would be reconstructed as *i for both.
If we take the second claim, then it would be impossible to systematically
explain the development of each form for the passive and the causative. Thus we
would rather take the first claim, which would readily explicate the ramification of
the passive and the causative forms. Thus, we will dismiss the claim that the original
forms of the passive and of the causative would be *i and *u, the second claim men-
tioned above.
[D] All the functions o f the morpheme si
It seems that there is only one morpheme si, serving as a honorific/respectful suffix.
This function is well attested not only in Old Korean but also in Middle Korean. In
the Hyangga texts, both /-si/ and /-sya/ forms are attested for this honorific function.
Yang (1957, pp. 624-625) insists that /-sya/ is the primordial form of /-si/, but there
is no evidence that this is so. But as Lee (1975, pp. 90-91) states, -si was probably
older than -sya because the latter is derived from -si (hon.) + -a (auxiliary verb)/-0
(purposive stem). Here is one example from the Hyangga:
1] kot’i syaon... [13:4.3]

kot’i sya-on
change [hon.]-[attr.]
‘would he change (his face)...’
In Middle Korean the function of the honorific was very productive, and so
some honorific verbs became developed either from that plain verbs or entirely
separate honorific verbs as suppletives. Here are some examples of these uses (Lee
1975, pp. 207-208; Ch’oe 1981, p. 128):
1] kyosi- ‘(someone superior) exists’ < куэ- ‘exist’ + si [hon.] (plain verb: iss-
‘to exist’)
2] cwasi- ‘(someone superior) eats’ < cwa- ‘eat?’ + si [hon.] (plain verb: тэк-
‘to eat’)
3] casi- ‘(someone superior) sleeps’ < ca- ‘sleep’+ si [hon.] (plain verb: ca-
‘to sleep’)

As indicated in all the Middle Korean examples, the -si- was used to make
a honorific verb, while the -sya was never employed to make a honorific verb. This
evidence assures us that -si was older than -sya.
In conjunction with OJ -si and OK -si, we do not find anything common to
both. The closest we can state would be the Old Ryukyu noun indicating a honorific
title, as discussed earlier. The relationship of the OK honorific -si with the ORyu si
appears to be coincidental since each morpheme seems to have developed internally.
Since some scholars such as Miller and Murayama claimed some of the Old
Japanese i may be cognate with some of the Altaic morpheme i, we just state that
they may not be cognate with each other as we have seen earlier. We also just men-
tion that there are also some identical morphemes si in the Altaic languages, whose
functions may never correspond to those of OJ si and OK si, and that the functions of
OJ i/si greatly outnumber those of Altaic, so OJ i/si may not be cognate with the
Altaic correspondences.
Conclusion
As seen in the previous sections, we find the following identical Japanese uses and
meanings with those of Korean: morpheme i:
® demonstrative pronoun
(1) quasi-free noun ‘person/fact/one’
® emphatic nominative case suffix
® genitive and possessive case suffix
© nominalising suffix
morpheme si:
none
Old Korean morpheme i seems to be identical in use and meaning with that in
Old Japanese. These identical forms and meanings, however, may mean that the
morphemes i in both languages are cognate with each other or borrowed from the
other. It would be unlikely that Japanese borrowed them from PK or OK since Japa-
nese greatly outnumbers Korean in use and meaning of the morpheme i, although
these PK or OK morphemes common to the OJ ones may have had a minor role in
stabilising the already existed OJ morphemes i/si.
As stated earlier, OJ i and OJ si seem to be paired with each other in function
and meaning:
(1) PJ *i must have been definite and specific, whereas PJ *si must have been
indefinite and generic. Thus, OJ i tends to be specific and deictic, which is general-
ised into emphatic; OJ si, on the other hand, has a tendency to be generic, which is
also generalised into emphasis.

40 Y. ITABASHI
(2) OJ i is very similar in form to OJ si in that OJ si may have been derived

from PPJ *s + *i, the latter of which was the original form of OJ i.
(3) All the functions of OJ i greatly outnumber those of OJ si probably because
another emphatic element so2 (as opposed to OJ si) became predominant in Old
Japanese so that OJ si was taken over by this emphatic element so2 ■
OK i is, however, not paired with OK si since OK i does not share any mor-
phological and semantic features with OK si unlike the strong relationship of OJ
i with OJ si. But OK i shares so many functions with OJ i and is outnumbered in
function by OJ i so that we cannot just regard them as loans, although we admit that
we cannot entirely exclude the possibility of borrowing; there is only one choice left
to us, i.e., cognateship of OK i with OJ i. Therefore, we point to the conclusion that
the primordial morpheme *i in Japanese must be cognate with that in Korean.
This observation leads us to further research on all the affixes, (pro)nouns and
the related aspects of some other languages and/or language families which are
neighbours of Japanese and Korean in order to find the origin of these Old Japanese
morphemes. These families are Paleo-Asiatic, Austronesian, and Austroasiatic. We
must add a final word: in conjunction with the origin of OJ i/si, we discussed it in
another paper (Itabashi 1998, 1999), which claims that OJ i/si may be cognate with
Pre-Proto-Oceanic (PPOC) i/si because OJ i/si share not only the functions but also
the diversification with PPOC i/si.
List of abbreviations
1. Reconstructions
PA: Proto-Altaic PJ: Proto-Japanese PPJ: Pre-Proto-Japanese

PK: Proto-Korean PPOC: Pre-Proto-Oceanic
2. Languages
MK: Middle Korean OJ: Old Japanese OK: Old Korean

ORy: Old Ryukyu Ry: Ryukyu
3. Grammar
abl.: ablative acc.: accusative attr.: attributive

com.: comitative cond.: conditional conj.: conjunction
cont.: continuative cop.: copula dir.: directive
emp.: emphatic excl.: exclamation fut.: future
gen.: genitive hon.: honorific imp.: imperative
inst.: instrumental 1 0 c.: locative nom.: nominative

perf.: perfect pil. word: pillow word pi: plural

pos.: possessive pros.: prosecutive purp.: purposive
q.f.n.: quasi-free noun ques.: question quo.: quote
resp.: respective rhet.: rhetric temp.: temporal
top.: topic voc. : vocative
4. Texts
K.: Kojiki Man.: Manyoshu N.: Nihonshoki

Omoro.: Omorosaushi Sem.: Senmyo
References
Japanese (written in Japanese)

Fukuda, K. (1989) Nihongo to Tsungusugo, revised edition, Yokohama: FLL.
Hashimoto, Sh. (1973) Joshi, Jodöshi no Кепкуй, Tokyo: Iwanamishoten.
Hattori, Sh. (1959) Nihongo no Keitö, Tokyo: Iwanamishoten.
Hirayama, T.-Nakamoto, M. (1964) Ryukyü Yonaguni Hôgen no Кепкуй, Tokyo: Tokyodö.
Hokama, M. (1981) Nihongo no Sekai, Tokyo: Chuökoronsha.
Hokama, M. (1988) Omorosaushi, Tokyo: Iwanamishoten.
Iida, S. (1937) Nihonshoki Sinkö; parts 1 ,2 , 3, Tokyo: Meibunsha.
Itabashi, Y. (1990) Kodaikankokkugo no Ninshodaimeishi no Kigen ni tsuite. Gengokagaku
[Univ. o f Kyushu, Journal of the Institute of Languages and Cultures] 25.
Itabashi, Y. (1998) Kodai Nihongo to Ôsutoronesiashogengo ni okeru Ichi-keitai no Dogensei (1),
Gogentankyü 6, Tokyo: Meiji-shoin.
Itabashi, Y. (1999, to appear) Kodai Nihongo to Ôsutoronesiashogengo ni okeru Ichikeitai no
Dogensei (2), Gogentankyu 7, Tokyo: Meiji-shoin.
Kawamoto, T. (1978) Minamikara kita Nihongo, Tokyo: Sanseido.
Kawamoto, T. (1980) Nihongo no Genryu, Tokyo: Kodansha.
Kuginuki, T. (1997) Kodainihongo no Keitaihenka, Tokyo: Chuöshoin.
Matsumoto, K. (1995) Kodainihongo Boinron, Tokyo: Hitsujishobö.
Miller, R. (1981) Nihongo to Arutaishogo, Tokyo: Taishükanshoten.
Miller, R. (1982) Nihongo no Kigen, Chikumashobö.
Murayama, Sh. (1950) Kodainihongo ni okeru Daimeishi, Gengokenkyu 16, pp. 4 0 -4 7 .
Murayama, Sh. (1962) Nihongo no Tsungusteki Kôseiyoso, Minzokugaku-kenkyü 26, 3, pp.
157-167.
Murayama, Sh. (1975) Nihongo no Kenkyühöhö, Tokyo: Köbundö.
Murayama, Sh. (1981) Ryukyugo no Himitsu, Tokyo: Chikumashobö.
Murayama, Sh. (1995) Nihongo no Hikakukenkyu, Tokyo: San’ichshobo.
Nakahara, Z.-Hokama, M. (1965) Omorosaushi Text, Tokyo: Kadokawa shoten.
Nakamoto, M. (1983) Ryukyu Goishi no Кепкуй, Tokyo: San’ichishobo.
Ohno, S. (1957) Nihongo no Kigen, Tokyo: Iwanamishoten.
Saigo, N.-Hokama, Moriyoshi (1972) Omorosaushi, Nihonshisötaikei 18, Tokyo: Iwanamishoten.
Tai, N. (1978) Nihongo no Gogen, Tokyo: Kadokawashoten.

42 Y. ITABASHI
Torigoe, K. (1968) Omorosaushi Zenshaku, 5 Vols. Osaka: Seibundô.

Umematsu, Y.-Ohtsuka, T. (1934) Kojiki Zen’yaku, Tokyo: Kokumin kyoikusha.
Yamada, M. (1981) Amami Yoronhögen no Taigen no Gohô, Tokyo: Daiichishobô.
Japanese (written in English)

Akiba-Reynolds, К. (1984) Internal reconstruction in pre-Japanese syntax, in: Jacek Fisiak [ed.]
Trends in Linguistics, Studies and Monographs 23, Historical Syntax, New York: Mouton
Publishers, pp. 1-23.
Aston, W. G. (1988) Nihongi, Tuttle.
Itabashi, Y. (1989) The Origin o f the Old Japanese Accusative Case Suffix i, Ural-Altaische
Jahrbücher, Neue Folge, Band 9.
Itabashi, Y. (1991) The Origin o f the Old Japanese Genitive Case Suffixes *n/na/nö/rjga and the
Old Korean Genitive Case Suffix *1 , Central Asiatic Journal 35/3-4.
Itabashi, Y. (1993) A Comparative Study o f the Old Japanese and Korean Nominative Case
Suffixes i with the Altaic Third Person Singular Pronouns, Central Asiatic Journal 37/1-2.
Itabashi, Y. (1997) Are the Old Japanese Personal Pronouns Genetically Related to Those o f the
Altaic Languages?, Acta Orientalia H. Turnus L (1 -3 ) pp. 117-146.
Itabashi, Y. (1998) The Old Japanese Personal Pronouns as an Etymological Problem, Eurasian
Studies Yearbook 70, pp. 123-154.
Martin, S. (1987) The Japanese Language Through Time, New Haven: Yale University Press.
Miller, R. (1989) Old Japanese i, in I. Hijiya-Kirschnereit-J. Stalph, eds, Bruno Lewin zu Ehren,
Festschrift aus Anlass seines 65. Geburtstages, Bd. I, Japan, Sprach- und literaturwissen-
schaftliche Beiträge, pp. 251-291.
Vovin, A. (1997) On the Syntactic Typology o f Old Japanese, Journal of East Asian Linguistics 6,
pp. 273-290.
Yoshitake, S. (1929) The history o f the Japanese particle /, Bulletin of the School o f Oriental and
African Studies 5, pp. 889-895.
Korean (written in Korean, Japanese)

Ch’oe, B. (1981) Chungsehanguga Мипрэргоп, Seoul: Tuuch’ulpansa.
Kim, W. (1991) Hyangga Hedokpapyangu, Seoul: Tehakkyo Ch’lpanpu.
Nam, W. (1993) Koasajan, Ilchogaksa.
Ogura, Sh. (1974, 1923) Hyanga no Kenkyû, Ogura Shinpei Hakushi Chosaku-shü, Vol. 1, Kyoto-
daigaku Kokubungakkai.
Yang, Ch. (1957) Kogayangu, Pakmun Ch’lpansa.
Yi, K. (1975) Kankokugo no Rekishi, Tokyo: Taishukanshoten.
Korean (written in English)

Lee, P. (1959) Studies in the Saenaennorae Old Korean Poetry, Rome: Istituto Italiano per II Me-
dio ed Estremo Oriente.
Martin, S. (1992) A Reference Grammar of Korean, Charles E. Tuttle Publishing Co.
Martin, S. (1994) Recent Research on the Relationships o f Japanese and Korean, in: S. Lamb-
E. D. Mitchell [eds], Sprung from Some Common Source, Stanford: Stanford University
Press.

Altaic Languages (written in Japanese)

Choku-Tsumagari, Toshiro-Kazama, Shinjiro (1991) Sown Kisoreibunshu, University o f Hokkai-
do, School o f Letters.
Fukuda, K. (1988) Manshügo Bungojiten FLL.
Fukuda, K: Gengogaku Daijiten, Vol. 1, Tokyo: Sanseidö.
Ikegami, J. (1971) Tsungüsugo no Hensen, in: Gengo no Keitö to Rekishi, Iwanamishoten.
Ikegami, J. (1993) Uirutago Daimeishi to sono Kakuhenka, Journal o f Sapporo University Women’s
junior College.
Itabashi, Y. (1999) Nihongo-Ryukyugo no Ninsho Daimeishi no Keisei no Кепкуй (tentative).
Kazama, Sh. (1991) nanaigo text, Tsungus Gengobunkaronshû, Otarushökadaigaku Gengo Center.
Ozawa, Sh. (1979) Chüseimökogo no Shokeitai no Кепкуй, Tokyo: Kaimeishoin.
Altaic (languages other than Japanese)

Avrorin, B. A. (1959) Grammatika nanaiskogo jazyka I, Moscow: Nauka.
Avrorin, B. A. (1961) Grammatika nanaiskogo jazyka II, Moscow: Nauka.
Benzing, J. (1955a) Lamutische Grammatik, Wiesbaden: Otto Harrassowitz.
Benzing, J. (1955b) Die tungusischen Sprachen: Versuch einer vergleichenden Grammatik,
Wiesbaden: Franz Steiner Verlag.
Cincius, V. I. (1982) Negidalskij jazyk, Moscow: Nauka.
Itabashi, Y. (1988) A Comparative Study o f the Old Japanese Accusative Case Suffix wo with the
Altaic Accusative Case Suffixes, Central Asiatic Journal 32, 3—4.
Itabashi, Y. (1989) The Origin o f the Old Japanese Accusative Case Suffix i, Ural-Altaische
Jahrbücher, Neue Folge, Band 9.
Martin, S. (1961) Dagur Mongolian Grammar, Texts, and Lexicon, Indiana University Press.
Menges, K. (1969) Die tungesischen Sprachen, Leiden.
Miller, R. (1971) Japanese and the Other Altaic Languages, Chicago: University o f Chicago Press.
Miller, R. (1977) The Altaic Accusative in the Light o f Old and Middle Koren, Proceedings o f the
19th Annual Meeting of the Permanent International Altaistic Conference Held in Helsinki,
7-11 June 1976, MSFOu 158, pp. 157-169.
Miller, R. (1993) More on the Old Japanese and Old Korean Accusatives, UAJB Neue Folge, Band
12, pp. 230-239.
Monahova, E. V.-Konstantinova, O. A. (1958) Evedu Turen, Moscow: Nauka.
Murayama, Sh. (1957) Vergleichende Betrachtung der Kasus-Suffixe im Alt-japanischen, Studia
Altaica, Wiesbaden: Otto Harrssowitz.
Ning, J. (1993) Sibe-English Conversations, Wiesbaden: Otto Harrassowitz.
Pak, E. (1973) A Study of Written Manchurian, V ols I-II, Taegu: Hyung-seul Publishing Со.
Pemrova, T. I. (1967) Jazyk Orokov, Moscow: Nauka.
Poppe, N. (1955) Introduction to Mongolian Comparative Studies, MSGOu 110, Helsinki.
Poppe, N. (1960) Buriat Grammar, Indiana University Press.
Poppe, N. (1964) Grammar o f Written Mongolian, Wiesbaden: Otto Harrassowitz.
Street, J. (1957) The Language o f the Secret History of the Mongols, American Oriental Series,
Vol. 42, New Haven: American Oriental Society.
Tekin, T. (1968) A Grammar of Orkhon Turkic, Ural und Altaic Series Vol. 69, Bloomington:
Indiana University Press.

44 Y. ITABASHI
Ternir, A. (1956) Die Konjunktion und Satzeinleitungen im Alttürkischen II, Inaural-Dissertation

zur Erlangung der Doktorwürde vorgelegt der Philosophischen Fakultät der Universität
Berlin, Berlin, pp. 233-280.
Thomason, S.-Kaufman, T. (1989) Language Contact, Creolization, and Genetic Linguistics, Uni-
versity o f California Press.
von Gabain, A.-M. (1974) Alttürkische Grammatik, Wiesbaden: Otto Harrassowitz.
Yoshizo Itabashi
Institute of Language and Cultures
University o f Kyushu
4-2-1 Ropponmatsu, Chuo-ku
Fukuoka, Fukuoka 81 0 -8 5 6 0
Japan

ON THE TURKIC BACKGROUND OF THE HUNGARIAN

VERB OV TO DEFEND, PROTECT, KEEP, GUARD’
Ä rpad B erta
(Szeged)
An overview o f previous etymological studies o f the Hungarian verb ov ‘to defend, protect; to take
care o f sy/sg; to guard sy/sg’, the article makes recommendations on its Turkic origin. As it is likely
the ultimate source o f ov, the Ancient Turkic verb *âb(ï)- ‘to console, to pamper’ and its dériva-
tives are also discussed in detail, as are possible contaminants using data that may stem from the
Ancient Turkic verb *apt- ‘to hide, cover, protect’. The article, furthermore, leaves open the possi-
bility that the Mongolian verb abura- is etymologically related to the Ancient Turkic verb *äb(i')-.
Certain verb forms are also examined (the Ancient Turkic *amX- ‘to be calm, peaceful’, the Tuvan
ovär- ‘to pay attention’ etc.), which, despite similarities in sound and meaning, have no bearing on
the Turkic etymology o f the Hungarian verb ôv.
Key words: Turkic, word formation, morphology, etymology, Hungarian
According to TESz. (2: 1105b-1106a) the Hungarian verb ov ‘to defend/protect; to

take care of sy/sg; to guard sy/sg’, first recorded in the early 13th century, is of
uncertain origin. The editors - on the basis of Karoly Rédei’s attempt1 to explain the
1 Kâroly Rédei (1969, pp. 106-108), who maintained that the Hung, verb ov is o f Ugrian
origin, having sorted out and rejected earlier etymological explanations (Budenz, Mikola) criti-
cally, was o f the opinion that the etymological counterparts o f the Hung. 6v are to be found in the
Ob-Ugrian languages. Rédei - having accepted the findings o f Andrea P. Hidvégi’s exemplary short
study (1956, pp. 3 04-306) - points out that the original verbal base of the Hung, ov was d, and that
this Hung, form 6 may have resulted either from an Old Hung, *uu-, *ou- or *au- < Ancient Hung.
*UW3-, * 0 W 3 -, *aw3-, Rédei reconstructs the Ugrian form as *wyj'3-, but this form presents various
phonetic problems. The original meaning o f the reconstructed Ugrian verb *wy/3- may have been
‘to see’, and Rédei suggests that the change in meaning from ‘to see’ —» ‘to defend/protect’ must
have taken place in the course o f the history o f the Hungarian language. To illustrate this probabil-
ity, Rédei lists Uralic and Finno-Ugric examples (cf. also Rédei 1988, p. 572). The fact that, de-
spite Rédei’s careful etymological explanation, the editors o f TESz. consider the verb 6 > ov to be
o f uncertain origin justifies our suggesting a Turkic etymology, completely free o f phonetic prob-
lems and with less questionable semantical ones, instead o f the Ugrian etymology full o f phonetic
as well as lexical obstacles. We must note that later Rédei himself also questions the link between
the Hung, verb ov and the Ugrian form (cf. Rédei 1988, p. 588) and suggests another etymology
(1988, p. 901), though he himself considers it unlikely at the same time. To the best o f my knowl-
edge, the verb ov has never been suggested as being of Turkic origin.
46 Â. BERTA
origin - do mention that the verb ov may be identified with the Ob-Ugrian forms
(Vog. T wä ‘will see’, So. ß ä y ‘to see’; Osty. V uùta ‘to see’, etc.) but point out that
certain semantic and numerous phonetic problems arise if the verb is considered to
originate from Ugrian.
The etymon of the Hungarian verb ov can be found in the Turkic languages.
The most significant historical forms are gathered in Clauson’s and Sevortjan’s ety-
mological dictionaries and in Rohrbom’s dictionary of Old Uyghur2. Further data can
be found in Räsänen’s etymological dictionary3 and certain other forms of uncertain
origin in Radloff’s large-scale work4.
In Käsgari’s famous compendium of 11th-century Turkic dialects the follow-
ing corresponding data can be found:
Käs. 79 (Dankoff-Kelly 1:154) awi'n-: mafia awïndï1He was friendly with me’;
Käs. 110 (Dankoff-Kelly 1:196) awi'n-: ol mafia awïndï ‘He was friendly to
me’, the aorist form of the verb awi'n- is awnur, to which the kasra diacritical mark
was later added, thus modifying the form to awïnur (‘AWINUR);
Käs. 136 (Dankoff-Kelly 1:226) awi'n-: yigitldrig ïslatu / yïyàc yemis i'ryatu /
qulän käyik awlatu / badram qïlïp awnalïm Describing a festival ‘We’ll put the youths
to work / we’ll have them shake trees and fruits / we’ll have them hunt the wild ass
and other wild beasts / while we occupy ourselves with merry making for a few days’;
Käs. 79, 637 (Dankoff-Kelly 1:154, 2:380) awi'nc ‘Friendliness or familiarity
with a thing’;
2 Cf. the following entries in the dictionaries mentioned:

Cl. 7ab: D av'it- Caus. f. o f *avi- ‘to comfort (e.g. a crying child Acc.)\ to amuse or distract
{someone Acc.)', cf. here: awut- ‘to comfort (sägala) a crying child’ (id.); uwut- (?) ‘to comfort,
console’ (CCG);
Cl. 12b: D avi'n- Refl. f. of *avf-; ‘to enjoy oneself, be happy; to be friendly, kind {to sy
Dat.): to take pleasure in (sg Dat.)’;
Cl. 12b: D av't'nc N .Ac. in -c fr. avïn-, ‘friendliness, kindliness’;
Cl. 13a: D av'tncu Dev. N. in -ей (Intrans.) fr. avi'n-: lit. ‘sg which gives pleasure’.
Sev. 1:66: avun- ‘razvlekat’sja; utesat’sja; uspokoit’sja’ also avut- ‘razvlekat’, utesat’,
kacat’ ( ljul’ku)’.
UW 278b-279a: avi'n-, Var. au[m]- (Br) TT VIII C 15. ‘vergnügt sein, fröhlich sein’ (?);
UW 279a: avi'rda [nach Räsänen EtymWb. 2b vielleicht zu —> avi't- ‘{Kind) beruhigen’],
Var. ä wi rdhä (Br) TT VIII D 11. ‘Ammen-, von Ammen’. Cf. also: Käs. 339 (Dankoff-Kelly 2:
35) awurta ‘wet-nurse’: 01 awurta tutundï ‘He took on a wet-nurse’; Khwar. awurta ‘wet-nurse’,
awurtga ‘id.’ (Boesch.). Unlike Räsänen and Röhrborn, who was considering Räsänen’s idea, we
must separate these data from the Turkic verbal derivatives examined here. It is probably a loan-
word in the Turkic languages.
UW 279a: avi't- ‘{ein Kind) beruhigen, trösten, hätscheln’;
UW 279a: avilit- ‘gehätschelt werden’.
3 Cf. REW 2b: data listed under the unrelated entry “mtü.” abï- [recte: apt-} ‘verbergen’.
4 Cf. Uig. ab't'n- ‘prisoedinjat’sja, privjazat’sja, uspokoit’sja, horoso cuvstvovat’ sebja -
sich an etwas anlehnen, anpassen, an etwas machen, sich beruhigen, behaglich fühlen’ (Radloff
1:626); Uig. abi'nc ‘utesenie, spokojstvie, horosee raspolozenie duha - Trost, Ruhe, Behaglichkeit’
(Radloff 1:626); Uig. abït- ‘vozstanovit’, utésit’, uspokoit’ - aufrichten, trösten, beruhigen’ (Rad-
loff 1:628-629). The meanings listed in the entry in Radloff ‘gotovit’ - bereiten, herrichten’ do not
belong here; Chag. auttïr- ‘uspokoit’, radovat’ - beruhigen, ergötzen’ (Radloff 1:79).
ON THE TURKIC BACKGROUND OF THE HUNGARIAN VERB ÔV 47
Käs. 79 (Dankoff-Kelly 1:156) aw ih cu nan ‘Something with which one enjoys

pleasure and familiarity’. Thus, slavegirls may be named awihcu.
Most recently, Erdal has taken the Old Turkic corpus into consideration5, and
stated that the Turkic verbal base does not even appear in Old Turkic written sources;
however, its derivatives have been widespread as of the Old Turkic period through
the Middle Turkic6 era up to Modem Turkic languages.
The Turkic verbal base in Ancient Turkic may have been * ä b (i)-, meaning ‘to
console; to pamper’. Upon reconstructing the Old Turkic form *äb(i')-, three points
are well worth considering: the length of the vowel in the first syllable, the number
of syllables in the verbal base and, if the verbal base had two syllables, the labial or
illabial quality of the vowel in the second syllable.7
Based on the Khalaj forms ( ävu n - ‘mit dem Weinen aufhören’ (Doe-Tez);
ä vu t- ‘das Weinen jemandes stillen’ (Doe-Tez)), we can assume that the initial vowel
of the Ancient Turkic verbal base must have been long.8
5 See Erdal p. 287: av'inc abstract ‘pleasure-seeking’, avi'ncu is very concrete ‘something
with which one enjoys pleasure and familiarity, hence slavegirl-concubine’; p. 591: avi'n- ‘to enjoy
oneself, take pleasure (in something)’■ , p. 654: avït-ïl- ‘to get distracted’; p. 763: av'it- ‘to comfort’.
6 We will provide only a brief selection o f the data on the language sources of the Middle
Turkic period:
Xak. awun- ‘byt’ raspolozennym; privjazyvat’sja’ cf. in the follow ing example: ibrahim
yalawac anar awund'i ol färistä ärdti ‘poslannik Ibragim stal к nemu raspolozen, to byl angel’
(Tefs.), awut- ‘uspokaivat’, utesat” cf. in the following example: yusufga qudug icindd es bolmïs
yusufnïawutmis ärdi ‘byl drugom Jusufu v kolodce, uspokoil Jusufa’ (Tefs.);
Khwar. awunca ~ awùïnca ‘comfort; joy, delight’ (Boesch.), avïncqa ‘id.’ (Boesch. - this
item is questionable, and its value is further reduced by the added publisher’s note: “looks like a
slip o f the pen”).
For further Khwar. data see the following entries in Nadzip’s dictionary: awnu, awutmaq,
awun, awunja and awi'ngu.
Kip. uvut- (aor.: uvuti'r) ‘beruhigen, besänftigen’ (CCG), avut- ‘aglamaktan çocugu vaz
geçirtmek’ (id. 25); avundu ‘Bir §eyle me§gul olarak (çocuk) aglamaktan vaz geçti’ (id. 25).
7 The reconstruction o f the intermediate consonant requires no special consideration: on the
basis o f historical and contemporary Turkic data it must have been *-b-. The Ancient Turkic bila-
bial voiced explosive had lost its explosive quality by the Old Turkic period and was represented
by a corresponding fricative which was either [ß] or [v],
8 Neither the verb nor derivative(s) from it which have been recorded are present in Turkmen
- as indicated in the TurkmRSl. - which maintained the original length present in Ancient Turkic.
(The source of the Turkmen form ävun- ‘avunmak [= to be consoled]’ is unclear in Tekin (1995,
p. 66), while in his Turkmen-Turkish Dictionary [see Tekin et al. 1995, p. 37] only one item can
be found: ävun- ‘aci duymak, sizi duymak [= to feel pain]’. However, if the Turkmen item in Tekin
(1995, p. 66) is real, the Turkmen initial long vowel ä- would indicate an originally etymologically
long vowel.) In Yakut, the other language considered to play a key role in the reconstruction o f
original length, present-day representatives o f the forms *äbi- and *abï- would be identical (i'a ~ a)
(cf. Stachowski 1993, pp. 20, 40, 45), and therefore the Yakut derivatives (see below) are not con-
elusive here. Upon examining the quantity, we are forced to rely exclusively on the Khalaj data,
and, even if other linguistic data fail to bear out what they tell us, etymological length can be

48 Ä. BERTA
The verbal base was most probably disyllabic. It is true that only its reflexive
derivatives formed by the suffix -(X)n- and causative derivatives formed by -(X)t- as
well as further derivatives of these are known to exist in the Turkic languages, and
these forms *abi'n- and *abi't- enable us to segment not only *abi'-n- and *abï-t-, but
also *ab-ih- and *ab-ït-, but this is rather only a matter of theoretical consideration
and has no bearing on the practical. According to one of the underlying notions of
Johanson’s 1979 work, ascending tonality may have been a significant feature in
Ancient Turkic word structure, and we can hardly suppose an absolute final explo-
sive in reconstructing the Ancient Turkic verb form. The causative forms with -(X)t-*9
play a significant role in clarifying this question. In Erdal’s monograph on the Old
Turkic corpus, as he summarises the appearance of the causative suffix -(X)t- (p.
799), he concludes that in 84% of the cases the suffix -(X)t- was added to polysyl-
labic verbs ending in a vowel or -r-. This phenomenon is similar to the appearance of
the suffix -(X)t- in the Middle Turkic era (Cf. Berta 1996, pp. 616-620) as well as in
Modern Turkic languages (Cf. Berta 1996a, pp. 23-29).
The vowel in the second syllable of the Ancient Turkic verbal base was
most probably [- rounded]. The Old Turkic forms bear this out. The rounded forms
which appear in the Middle and Modern Turkic era may be secondary and can be ex-
plained by the labialising impact of -v-/-w-, following the spirantisation of the origi-
nal *-b-.
It was important to overview the steps that resulted in the reconstruction of the
Ancient Turkic form *âbï- in order to clarify the correlation between the various
Turkic forms although this has no real bearing on the phonetics of the Turkic etymol-
ogy of the original Hungarian verb *o, which I will now discuss. On the basis of the
reconstructed Ancient Turkic form *âbï- and its derivatives (avi'n- and avi't-) having
appeared as of the Old Turkic period, we can reconstruct with certainty an Old Turkic
form *awï- / *awu- as the base borrowed for the Old Hungarian form *au- < Ancient
Hung. *aw3-.
As I have already mentioned, the reflexive -(X)n- and causative -(X)t- dériva-
tives of the Ancient Turkic verbal base *âbï- can be traced as of the Old Turkic pe-
riod as can their derivatives, which are widespread in Modern Turkic languages. We
will survey these forms based on the suffixes mentioned above:
reconstructed with great probability. The written form of the Brâhrnî äwinil- ‘sich wohl fühlen’
(TT VIII: 31 [D 13], 36) cannot be used as an argument supporting the etymological length o f the
initial vowel in the Ancient Turkic *äb(i')-, as we are o f the opinion that the length indicated in the
Brahmi written form is only orthographic (cf. similarly, Röhrborn 1977, p. 7; Erdal 1991, p. 15;
and also see Tekin’s corresponding examples: 1995, pp. 94-96).
9 In the description o f the Old Turkic system of suffixes, the symbol -(X)t- is used to indi-
cate the causative suffix, in which X represents the vowel which follow s vowel harmony, and
indicates four possible sounds which may have been [i], [V], [Ü] and [u] - although some scholars
maintain that these may have been [ё], [ä], [ö] and [6]. The Ancient Turkic form must have been
*-//-, which is related to the Mongolian suffix -ci-, (Cf. Johanson 1991, p. 170.)

Old Turkic avih- < *àbï-(X)n-

New Turkic:
Kipchak: uvun- ‘utesat’sja, uspokaivat’sja’10 (Kar. T, H, C); avun- [!] ‘uspo-
kaivat’sja, utesat’sja’ (Kar. C);
Turki: avun- ‘zabavljat’sja; uspokaivat’sja’ (NUygh), ovun-" ‘zabavljat’sja;
uspokaivat’sja, utesat’sja’ (NUygh); ovun-'2 ‘perestavat’ plakat’, perestavat’ kapriz
nicat’; (peren.) utesat’sja, uspokaivat’sja; (peren.) zabavljat’sja, razvlekat’sja’ (Uzb);
Oghuz: avun- ‘nahodit’ utesenie, utesat’ sebja, razgonjat’ tosku; tesit’ sebja na-
dezdami; byt’ stel’noj (o korové)''1' (Turk); cf. ovunmaz ‘bezutesnyj; bezutesno’ (Az)14;
Khalaj: ävun- ‘mit dem Weinen aufhören’ (Doe-Tez).15
Numerous derivatives of the reflexive form avih- can be found in Modem
Turkic languages.16
The verb form with the suffix -(X)n- is also the direct base of the Old Turkic
word avïnc, which meant ‘friendliness, kindness’17, but it may also have meant
‘search for happiness or beauty’18. Among Modem Turkic languages, there are only
111The meaning ‘szalit’sja’ as indicated in KRPS1. is not a part o f this group. ?Cf. Kzk. obal
‘obidno’, obalsi'n- ‘szalit’sja’.
11 The appearance o f the initial o- in the NUygh. form is not related to the labialisation fol-
lowing the spirantisation o f -b- > -w-, as it does in the Kipchak languages, but is a result o f regres-
sive labial assimilation, which is well documented in New Uyghur.
12 The initial Uzbek o- must be separated from both the Kipchak and New Uyghur initial la-
bial, since w e have observed a tendency in Uzbek that suggests that о developed from historical *a.
13 The Turk, avun- meaning ‘byt’ stel’noj (o korove)’ definitely belongs here and can be in-
terpreted as a development from the meaning ‘to be in a happy, joyful state’. From a semantic point
o f view it bears mentioning that the Hung, word boldog ‘happy’ also has the meaning ‘pregnant’ as
dated to a later period.
14 In the Azerbaijani form and in other derivatives of the given base, the initial o- is a result
of secondary labialisation. The segment ov probably developed from an earlier segment o f av < ab,
cf. Azerbaijani ov ‘hunting’ < äv < ab (see Cl. 3b). This phenomenon in the Azerbaijani language
is well known; for further examples see: Caferoglu-Doerfer (1959, p. 287). The secondary labialisa-
tion can be considered a tendency in the Azerbaijani dialects. From the point o f view o f phonetics,
this is identical to similar tendencies in the Kipchak languages.
15 Cf. also Doerfer’s comments: “< T äbm-, später avïn- ‘to enjoy oneself, to be happy’,
auch chwar. ‘to be comforted, consoled’, altosm. ‘to take comfort’.”
16 Cf. *abi-(X)n-(X)l-: Kipchak: uvunulmagan ‘bezutesnyj’ (Kar. C);
*abi'-(X)n-tUr-: Turki: ovuntir- Caus. o f ovun- ‘uspokaivat’ {placuscego); (peren.) uspokai-
vat’, utesat’; (peren.) zabavljat’, razvlekat” (Uzb); Oghuz: avundur- ‘utesat’, uspokaivat’; obnade-
zivat” (Turk); ovundur- ‘utesit’, uspokoit” (Az);
*abi'-(X)n-tUr-A gerund: Oghuz: ovundura-ovundura ‘utesaja, uspokaivaja’ (Az);
*abï-(X)n-tUr-mA: Oghuz: ovundurma ‘utesenie, uspokaivanie’ (Az);
*ab'1-(X)n-tUr-(X)gcl: Oghuz: ovunduruju ‘utesitel’nyj’ (Az);
*abi'-(X)n-Xs: Turki: avunus ‘uspokoenie, utesenie’ (NUygh), ovunus ‘zabava; uspokoenie,
utesenie’ (NUygh);
*abi'-(X)n-tU (?): Oghuz: avuntu ‘utesenie, uteha’ (Turk).
17 Cf. Cl. 12b: ‘friendliness, kindliness’.
18 Cf.Erdal 1991, p. 287.
50 Ä. BERTA
scarce data available in the Oghuz and Kipchak branches meaning ‘consolation’, but
other derivatives can be found elsewhere as well.
Cf. Old Turkic avi'nc (< *äbi'-(X)nci9 < *àbï-(X)n-Xs)
New Turkic:
Kipchak: uvunc ‘utesenie’ (Kar H)20, uvunc ‘utesenie, uteha’ (Kar T, C)
Oghuz: avutiç ‘utesenie’ (Turk)
*abï-(X)nc+Ak
Turki: ovuncaq ‘zabava, uteha’ (NUygh), ovuncoq ‘uteha, utesenie; radost’,
otrada; zabava’ (Uzb)
Yakut i'ahax ‘Unterhaltung, Spiel’ (Stach.)21, Cf. also i'asax22 ‘zabava, zanja-
tie’, as well as numerous derivatives, e.g. i'asaxtan- ‘zabavljat’sja, razvlekat’sja, zani-
mat’sjacëm’ (Рек. 3:3751-3752)
*abï-(X)nc+lXk
Kipchak: uvanclik ‘utesenie’ (Kar. H), uvuncluk, uvunclux ‘utesenie’ (Kar. T),
uvuncluq ‘radost” (Kar. C)
Turki: avuncuq ‘utesenie; zabava’ (NUygh)23
*abï-(X)nc+lXg
Kipchak: uvunclu ‘utesitel’nyj’ (Kar. H); uvunclu ‘utesitel’nyj’ (Kar. T)
19 On the secondary Old Turkic suffix -(X)nc see most recently Erdal’s opinion (pp. 275-
277). If Erdal, who had developed Bang’s idea further, is right, the source of the Old Turkic pro-
ductive suffix -(X)nc may have been the compound formant *-(X)n-Xs. The vowel in the second
part o f the compound suffix - although it was dominant - dropped after °n# when added to a poly-
syllabic base, and the segment °n s, which could not remain in this form for phonotactic reasons,
becam e °nc. Erdal’s explanation may not be weakened by the fact that newer derivations (cf. the
N Uygh. forms avunus and ovunus mentioned in footnote 16) do not indicate this phonetic change.
However, it is interesting that besides the NUygh. data mentioned above, the form ovuncaq ‘zaba-
va, uteha’ may also be found with an identical meaning, which must represent a much older dériva-
tion granted that we accept the rationale above.
2,1The second vowel in the form uvanc [!] ‘utesenie’ found in the Halic dialect o f the Ka-
raim language is “irregular”, most probably explained as being a result of an analogous effect o f the
Kar. H word yubanc / T yubanc / yuvanc which is identical in meaning. The reverse effect can also
be found in the Halic dialect: the second vowel in the form yuvunc ‘utesenie’ replacing the “regular”
vow el -a- can be explained as being a result of the influence o f the word uvunc ‘utesenie’ found in
the sam e dialect.
21 The Yakut form was borrowed into Dolgan, where the following derivatives can be found
as listed in Stachowski’s compilations (1993a, 1998): i'ahaktä- ‘scherzen’, (ahaktäk ‘lustig, witzig’,
ïahakeït ‘Witzbold, Spaßmacher’.
22 The form which appears in Pekarskij’s dictionary is the immediate precursor to the form
o f the word cited by Stachowski: Yak. i'ahax < i'asax < *abi'ncak. Stachowski (1993, pp. 20-21)
points out that the suffix +Ak may have been related to a Middle Yakut form abi's < *abi'nc, since
-s w ould replace *-nc only in the final position, and the intervocalic representation in Yakut would
be -nn-. Cf. also Rdna-Tas’s article in the present volume, pp. 1-15. Stachowski (l.c.) presupposes
that the Ancient Turkic *abï is the source for the Yak. initial diphthong i'a-. According to his inves-
tigations, the regular representation o f the Ancient Turkic * a b ï and *agi' in the Yak. is i'a (~ a).
23 The form avuncuq probably belongs here and may not be the result o f another derivation
but o f the phonetic simplification o f -ncl- > -nc-.

Old Turkic avi't- < *àbï-(X)t-

New Turkic:
Kipchak: uvut- ‘utesat” (Kar. T, H, C); auït- 1. ‘kacat’, ukacivat” , 2. ‘yuat-,
zanimat’, razvlekat” (Sib-Tat. in Tumaseva 1992), üüët- ‘kacat’, ukacivat” (Sib-Tat.
in Tumaseva 1992);24
Siberian Turkic: ät- ‘kacat’ (rebenka)’ (Tuv); ät- ‘kacat’ kolybel’; ukacivat’,
ubajukivat’ (rebenka)' (Tofa, see TRS1.), ät- ‘kacat’ rebenka’ (Tofa, see Rassadin);
abi't-25 ‘kacat’ (Ijul’ku)’ (Khak.); Tel. M- ‘kacat’, uspokoit’ rebenka —wiegen, schau-
kein, ein Kind beruhigen’ (Radloff 1:1702);
Turki: avut- Caus. of avu-, ‘utesat’, uspokaivat’; zabavljat’ (rebenka),26
(NUygh), ovut- ‘uspokaivat’, utesat” (NUygh);
Oghuz: avut- ‘zabavljat’, razvlekat’; otvlekat’ (vnimanie i t.p.); utesat’, ugova-
rivât’; unimat’, uspokaivat’; vodit’ za nos, obnadezivat’; obmanyvat’, naduvat” 27
(Turk); ovut- ‘utesit’, uspokoit” (Az);
Khalaj: ävut- ‘das Weinen jemandes stillen’ (Doe-Tez).
24Radloff (1:78) makes mention o f the following from Baraba and the Chaghatay literary
language: aut- ‘kacat’, uspokaivat’ rebenka - ein Kind wiegen, beruhigen; zanimat’, razvlekat’ -
beschäftigen, einem die Zeit vertreiben’. Radloff incorrectly divides the word into segments: au-t-.
Although he does not indicate which - au- base in his dictionary he has in mind as the first seg-
ment, but, having surveyed the material in his dictionary, we have concluded that he probably
considers the form aut- to be a derivative o f the Kir., Kas. au- ‘naklonjat’sja v ” storonu, svalit’sja,
upast’, oprokinut’sja - sich herabbeugen, herabrutschen (vom Pferde), herabfallen, Umfallen; sklo-
nit’sja к” komu - sich zu Jemand hinneigen’ (1:67) originating from the base ag-,
25 The intermediate -b- in the Khak. abi't- originated in the Ancient Turkic *-b-, not in the
Old Turkic -V -. In Khakas the regular representation o f the Ancient Turkic *-b- in an intervocalic
position is -b- (cf. e.g. ibek ‘bystryj, rezvyj (о kone)’ - Old Turkic ëvâk ‘hurry; hurrying’ Cl. 8 b -
9a; kibls ‘kover’ ~ Old Turkic käviz ‘carpet, rug’ Cl. 692b), nor did the Ancient Turkic *-b in the
final position become -v (see e.g. Khak. ib ‘yurt’). Cf. also the following form found in Koybal
which is considered to be one of the dialects of Khakas: Koib. abi't- ‘kacat’, uspokaivat’ rebenka -
schaukeln, wiegen, ein Kind beruhigen’ (Radloff 1:628-629).
26 Under the entry cited, the UjgRSl. also lists the following meanings: ‘vyrascivat’; uveli-
civat’; razmnozat’, rasplodit” . However, these do not belong here but to the NUygh. base avu-
‘uvelicivat’sja, razmnozat’sja; sporit’sja’ [cf. also NUygh. avu- ‘to increase, to teem’ (Jarring)]. Cf.
also NUygh. avus ‘narascenie, uvelicenie’. The verbal bases must by all means be separated here
even if the possibility o f the semantic relationship between the transitive verbs meaning ‘to de-
fend/protect’ and ‘to bring up, to grow’ may exist. The NUygh. intransitive avu- ‘to grow’ is a sec-
ondary form, and may have developed from an earlier form arvu-, cf. the Kirgh. arb't- ‘uvelici-
vat’sja’. See also the derivative avutus ‘vyrascivanie; uvelicenie; razmnozenie’ (NUygh.), the listing
of which together with the lexeme avutus ‘utesenie, uspokoenie’ under the same entry is an error.
See also NUygh. avun ‘abundant’, avut- ‘to cause to increase; to increase’ (Jarring). Further data
from the Turkic languages can be found in Rassadin’s collection (155) under the Tofa entry arbi'n
‘obil’nyj (ob urozae)’. The Turkic data are o f Mongolian origin.
27 It is not necessary to separate the special meanings o f ‘vodit’ za nos, obnadezivat’; ob-
many vat’, naduvat” from this verb as it covers a wide semantic range anyway. On the meaning
‘vodit’ za nos, obmanyvat” etc., cf. also the Hung, phrase hintdba iiltet ‘deceive’ and Germ, ver-
schaukeln ‘id.’.
52 Ä. BERTA
Derivatives of the verbal base with the causative suffix can also be found in
Modem Turkic languages.28 The Kar. C uvutla- 'uspokaivat’, utesat” , however, is
probably not a derivative of the verb formed by the suffix -(X)t-\ it most likely devel-
oped from the form *avï-(U)t+lA-,
Having discussed the phonetics and morphology of the supposed Turkic ety-
mon of the Hungarian verb 6 > ov as well as having introduced the corresponding
Turkic forms, I now turn to an examination of the semantics of the etymology sug-
gested.
The Turkic data have a wide semantic range and may have the following
meanings:
*âbï- with the reflexive suffix:
‘friendly to sy’ (Käs.), ‘to rejoice about sg’ (Käs.), ‘to stop crying’ (Khal;
Uzb), ‘to stop being fickle’ (Uzb), ‘to be comforted / consoled, to calm down’ (Kar,
NUygh, Uzb, Turk), ‘to have fun’ (NUygh, Uzb), ‘to make sy forget his troubles’
(Turk), ‘to console oneself with hopes’ (Turk), ‘(a cow) to be with a calf’ (Turk).
Derivatives of the form with the reflexive suffix:
‘friendliness, friendly behaviour’ (Käs.); ‘consolation’ (Kar, Turk, Uzb,
NUygh), ‘fun’29 (Kar, NUygh, Uzb, Yak), ‘happiness’ (Kar, Uzb), ‘game’ (Yak),
‘occupation, pursuit’ (Yak), ‘consoling’ (Kar);
‘nice; sy or sg that causes happiness’ (Käs).
*âbï- with a causative suffix and its derivatives:
‘to still sy’s crying’ (Khal), ‘to console’ (Kar, Turk, Az), ‘to console (a child)’
(NUygh), ‘to rock sy’ (Sib-Tat, Tofa), ‘to entertain sy’ (Sib-tat, Turk), ‘to entertain /
amuse (a child)’ (NUygh), ‘to rock (a child)’ (Tuv, Tofa, Tel), ‘to rock (a cradle)’
(Tofa, Khak), ‘to rock to sleep’ (Tofa), ‘to hush, to calm down (a child)’ (Koib, Tel,
28 *abi'-(X)t-(X)l-: Siberian: abïdïl- Pass, o f abït-, ‘kacat’sja ( v Ijul’ke)’ (Khak), Schor. abïdïl-
‘kacat’sja, byt’ kacaemym, uspokaivaemym’ - ‘gewiegt, geschaukelt, beruhigt werden’ (Radloff
1:629). Regarding the intermediate -b- of the Shor form cf. 25. n. Oghuz: avutul- Pass, o f avut-
(Turk);
*abï-(X)t-tUr-: Oghuz: avuttur- Caus. o f avut- (Turk);
*abi'-(X)t-(X)s-: Siberian: abïdïs- Coop, o f abït-, ‘pomogat’ kacat’ (ljul’ku)’ (Khak);
*abï-(X)t-Xs: Turki: avutus ‘utesenie, uspokoenie’ (NUygh), ovutus ‘uspokaivanie, utese-
nie’ (NUygh);
*abï-(X)t-(X)n: Siberian: abïdïn ‘kacanie, kolebanie’ (Oyr). The -b- in the Oyr. form must
be viewed in the same way as a representation o f the Khak. (and its Shor and Koybal dialects) -b-\
*abi'-(X)t-(X)n+IXg: Siberian: abïdïndu ‘kacajuscijsja’ (Oyr);
*abi'-(X)t-gUc: Siberian: abïtxïs dial, ‘ljul’ka (kolybel’)’ (Khak);
*abï-(X)t-mAk: Kipchak: uvutmax ‘uteha, utesenie’ (Kar. T);
* ab'i-(X)t-(X)gcl: Kipchak: uvutuvcu ‘utesitel” (Kar. H); uvutuvcu ‘utesitel” (Kar. T, C);
uvutuvci" utesitel” (Kar. C).
29 On the close relationship between the meanings ‘consoling’ and ‘having fun’, cf. the ety-
mological link between the Russian utesenie and uteha.

NUygh), 40 distract sy’ (Turk), ‘to calm sy down’ (Turk, Az), ‘to cheat, to lead sy
up the garden path’ (Turk).30
Based on these meanings, in all probability, we can assume that the verb *âbï-
originally meant ‘to console, succour, pamper, rock, or calm (babies or children)’,
while the medial/reflexive verb form *àbïn- may have meant '(for a child) to console
himself, to calm himself, and the derivative *äbi't- with the factitive / causative suf-
fix may have meant ‘to console, to pamper (a child)’ in our reconstruction.
The etymology of the Hung, verb ov (o') ‘to defend / protect; to take care of sy
or sg; to guard sy or sg’ <—Turkic *awï- / *awu- < *abï- < *âbï- ‘to console, pam-
per’ presents no phonetic problem. The semantic argument for this etymology is not
as compelling as arguments from the point of view of phonetics or historical phonol-
ogy, but there is no doubt that the meaning(s) of the Hung, ov can be understood on
the basis of the meanings ‘to succour, console’ ~ ‘to care for sy’ ~ ‘to devote atten-
tion to sy’.
The etymology discussed here may be supported further by another factor.
Based on conclusions drawn from the data of the Turkic languages, besides the
Ancient Turkic verb *âbï- ‘to console, pamper’, there was another Ancient Turkic
verb form, *apï-, meaning ‘to hide, cover, protect’, which sounded very similar to
the first verb but was of a different origin. The Ancient Turkic *âbï- : *apï- were in
opposition due to two distinctive features (vowel quantity and +/- intervocalic voie-
ing). By the Old Turkic period, these verbal bases could only have been in
opposition because of one feature, as by then the etymologically long vowels were
identical in most dialects with their etymologically short pairs. We can suppose,
therefore, that the Old Turkic opposition may have been /b/31 : /р/.
Small collections of the historical forms of the Ancient Turkic *apï- ‘to
hide, cover, protect’ can be found both in Clauson’s32 and also in Röhr-
3,1 It must be noted that the meaning ‘to rock (a baby ~ a cradle )’ appears only among
derivatives of the form with the factitive / causative suffix. However, the relationship between this
meaning and that o f ‘to be happy, to be consoled, to calm down’ o f the medial / reflexive verb
forms does not require any further explanation if we consider the semantic range between them
meaning ‘to calm (a baby ) down by rocking (him)’. The connection o f the word avi'ncu - ‘the
maiden giving happiness (pleasure) [to a man by rocking]’ to the same semantic range is also clear.
31 From a phonetic point o f view the Old Turkic intervocalic /Ь/ could equally have been
[b]/[ß]/[v] or [w], while /р/ could have developed from [p] as the basic variant as well as from [b] -
at least in some dialects.
32Cl. 6a: abï- [recte: apï-] Hap. leg. Clauson quotes Käs.'s illustrative sentence with the
following transcription: ol anïkisïdin abïdï ‘he hid him from the people’. Data appearing in Käs.
are not free o f various problems in meaning. Cf. applicable sections in the edition by Dankoff-Kelly,
who, as Erdal has already pointed out (pp. 762-763), are in error when they emend the phonetic
form apï- and end up with the verb 'opi-', since damma appears only in the derivatives api'-t- and
api'-n- and is written by a second hand even there. Erdal explains the “correction” by a second hand
as an impact o f the word yupa-n-, which had the same meaning: see Dankoff-Kelly Index (3:41)
opi- (api-7) ‘cover, hide’ ; Dankoff-Kelly 1:203 (Käs. 116): opi- - ol ani kisïdin opïdi ‘He concealed
it from the person.’ The same for anything that you hide from someone else. - ‘ABIYDIY (tasdld
over the D) [sic!]; Dankoff-Kelly 2:280 (Käs. 555): opi- - ol ani kisïdün opidi ‘He hid it from the
54 Ä. BERTA
bom’s*33 lexicographical and etymological dictionaries as well as in Erdal’s mono-

graph34 devoted to Old Turkic word formation. Although no traces of the Ancient
Turkic *apt- > Old Turkic apt- (~ [abi]) forms exist in Modern Turkic languages
(?35), there is no question that its presence in the Old Turkic period rendered the
contamination which supports the etymology presented here possible.
people’. Also of anything that one conceals from another person. - ‘ABIYDIY The forms for the
aorist and infinitive are: oplr opimäq ‘AB1YR• ‘ABIYM ’Q.
Cl. 7a: D abi't- [recte: -p -] Caus. f. o f abi- [recte: -p -] but with the same meaning. Cf.
Dankoff-Kelly 1:170 (Kâs. 88): opityän - ol är ol özin kisidän opityän ‘He is a man who habitually
hides from people’ the right form: api'tgan ‘BTT’N; Dankoff-Kelly 1:198 (Käs. I l l ) : opit- - ol özin
mändin opitti ‘He hid him self from me.’ - ‘UBATJY (tasdid over the T). The forms for the aorist
and infinitive are: opitur opitmäq ‘ABITUR ‘ABITM’Q - The following comment appears in the
footnote: “A smudged, reading doubtful.” [sic!]
Cl. 12b: D abi'n- [recte: -p —\ Hap. leg., Refl. f. o f abi- [recte: — p —\, Cf. Dankoff-Kelly
1:194 (Käs. 108): opin- - mändin opindi ‘He hid from m e’ - ‘UBANjDJY. The forms o f the aorist
and infinitive are: opinûr opinmâq ‘UBANUVR ‘UBANM ’Q.
33 UW 36a: abi'-g [recte: ~p~] ‘Verdeckung; geschützt, ruhig’. Röhrborn indicated the ver-
bal base o f the noun with an asterisk. This is not justifiable, since, although this form is not present
in Uyghur, it can be found in Käs. Erdal pointed out (1991, pp. 179, 181, 215) that from a pho-
nological point of view the correct entry would be api'g. The noun formed by the suffix -(X)g may
indicate both the action itself and the object of the action signified by the transitive verbal base.
Considering that in the Old Turkic sources written in Brähmi and Tibetan scripts, both able to indi-
cate the distinction between [b] and [p], it is the forms o f abi'g which can be found (cf. TT VIII:59
[I, 21 ]: ä bha g: ä bha g ye rlyä tyä ‘an gefahrenfreien Orten’), Erdal has no doubts that in this and
other texts this may represent the pronunciation o f [b], and appears as a phoneme variant of /р/.
UW 36a: abttilmaksiz (recte: -p-) ‘der nicht verdeckt werden kann, der Nicht-zu-Ver-
deckende’.
34 Erdal indicates the alternate forms api- / abi- in the index and discusses the following Old
Turkic derivatives in his work: abi'-g ‘covering, covered’ (pp. 179, 181, 215). On this last page,
upon discussing the word yap-i'g ‘covering’, ‘building’, Erdal refers to the source which has already
been referred to in Röhrborn’s dictionary: BuddhStab II 17: amrïlmïs enc nirvannin api'g yapigi
aci'lzun. Here, citing Zieme’s translation ( ‘das versperrte Schloß des friedevollen Nirväna’) Erdal
feels that the meaning o f yapi'g ‘Schloß’ is too specific although he also adds that it is still closer to
the meaning demanded by the context than the translation provided in UW ‘das ruhige Haus des
Nirväna’. The transcription in UW is as follows: nirvannin abi'g yapigi. According to Erdal, the
meaning of yapi'g is ‘covering’. The meaning o f abi'g in Röhrborn’s interpretation is ‘ruhig’. Erdal
maintains that the word abi'g when functioning as an attribute means ‘hidden, concealed’ or ‘con-
cealing’. Just like Röhrborn, Erdal also discusses (pp. 396, 654) the form apitilmaksiz : see Raxu
üzä apitilmaksiz ay t(ä)nri (ETS 10,90) ‘the moon which does not get hidden by Rähu’, where the
passive base api'til- can be identified. On the causative form o f the verbal base cf. Erdal (pp. 7 6 2 -
763): api'-t-.
35 The etymological status o f the lexemes abay and avay meaning ‘careful’, present in most
Modern Turkic languages, is not entirely clear. Räsänen (REW lb), building on Radloff’s idea
(1:621), connects the Modern Turkic forms abay and avay to the interjection a va known to us
through Karakhanid (QB, Käs.), but this is hardly acceptable. Here we offer a short selection o f
Modern Turkic forms, and we do not always indicate the existence of derivatives. The data listed
ON THE TURKIC BACKGROUND OF THE HUNGARIAN VERB ОV 55
Finally, some further data from the Turkic lexicon must be mentioned36 which
must be separated from the data discussed above. These are:
!.Derivatives of the Ancient Turkic base *amX-37 ‘to be calm, peaceful’.
A collection of historical data in this case can also be found in Clauson’s and Röhr-
born’s dictionaries, and an authentic analysis of these forms is provided by Erdal.38
below may be traced back to the gerund form o f a verb *apï- (or *apa-). However, the Kkalp. and
Turki (NUygh, Uzb) data indicate an intermediate *-b- in the base.
Kirgh. abay ‘ostoroznost’, osmotritelnost” (cf. also: abaysi'z ‘necajanno’; abayla- ‘byt’ osto-
roznym’; abaylagi'c ‘ostoroznyj’ etc.);
Bashk. abay ‘ostoroznyj, osmotritel’nyj; ostorozno, osmotritel’no’;
Tat. abay. abay bul- ‘ ( razg .) byt’ ostoroznym, byt’ predusmotritel’nym; byt’ пасеки, byt’
nastoroze’;
Kar. abayla- ‘uvazat’, c t i f , pocitaf, cenit’; okazyvat’ vnimanie’ (T, H. C);
Nogh. abayla- ‘iskat’, prismatrivat’, prigljadyvat’; spohvatyvat’sja; zamecat’ osibku (svoju ),
obrascat’ vnimanie na cto-1.; zamecat’, primecat’ cto-1.’;
Cr-Tat. abayla- ‘zamecat” ;
Kzk. abay: abay bol- ‘byt’ ostoroznym, osmotritel’nym, vnimatei’nym; osteregat’sja’;
Krc-Blk. ? ayab (possibly a metathetical form) ‘(berezno) ostorozno’ (RKrc-BlkSl.);
Kkalp. a v a y lï ‘ostoroznyj’;
NUygh. avay ‘ostoroznost’, osmotritel’nost’, vnimatel’nost” (cf. also: abayli- = avayli-
‘osteregat’sja, oberegat’sja; vsmatrivat’sja; byt’ vnimatel’nym, ostoroznym’);
NUygh. avayla- ‘to have regard for, to be discreet to’ (Jarring);
Uzb. avayla- ‘berec’, oberegat’; obrascat’sja berezno, ostorozno’, avay lab ‘berezno, osto
rozno’.
However, if the data listed here may be related to the Ancient Turkic verb *apï- ‘to hide,
cover, protect’, then the Kkalp. and NUygh. forms may support the supposition that the once exist-
ing verbs *âbï- and *apï- may have influenced one another. In terms o f semantics, we may safely
assume the forms abay and avay, meaning ‘careful’, are related to the meaning ‘to guard, defend’:
cf. Hung, ôv and ôvatos, Russian storoz.it’ and ostorozno.
36 All this may not be closely related to an examination of the etymology we are discussing
here, but must be taken into consideration when clarifying the Turkic background.
37 Based on the data available it cannot be determined with complete certainty if the vowel
in the second syllable was rounded or not.
38 Cl. 160b-161a: amul {Tamil) ‘quiet, mild, equable, peaceable, even-tempered’. Clauson
mistakenly considers the Old Turkic word to be a simplex and etymologically connects the form
amul through a supposed alternate form *amur to the verb amra-, about which he - correctly -
states that it is also a loanword in Mongolian. [The Old Turkic verb amra- must be separated from
the form amul {Tamil). We cannot discuss this here, but in terms o f form it seems to be a derivative
with a denominal suffix +(X)rA-, and its base may have been an ancient, onomatopoeic noun
*am(X).] Röhrborn’s comment (UW 117a-l 18b) on the word ami'l ‘Frieden, Harmonie; Friedfertig-
keit, Milde; friedlich, mild’ is that “< zu т о . amur ‘ruhig’, Monguor xamura- ‘ruhen’” is most likely
a printing error, and he does not wish to explain the Old Uyghur word on the basis of a Mongolian
form but has a reverse borrowing in mind. Röhrborn refers to Räsänen’s etymological dictionary
and one of Doerfer’s most essential methodological articles to support his argumentation. Räsänen
- in his usual manner - connects the Turkic, Mongolian and Manchu data only with the symbol ~
(REW 19a), while Doerfer (1971, p. 334) says that the Monguor xamurä- ‘ruhen’ form indicates
that the Turkic word may have been an early loanword into Mongolian. Doerfer later modified this

56 Ä. BERTA
Most of the widespread Modern Turkic data are presented here in a relatively de-
tailed manner.39
opinion (cf. here note 39) and in his latest view considers the Monguor x- a “hypercorrect” devel-
opment - although with a question mark.
Erdal (p. 330) - not commenting on the opinions above - mentions the Old Turkic form
ami! ‘calm etc.’, which could be used both as a noun or an adjective, in his discussion of the formant
-(X)l and suggests an Old Turkic base *ami'-, although not dated, as the etymological starting point.
Cl. 161b: D amulluk A.N. fr. amul-, ‘peace, tranquillity’;
Cl. 162b: ?D amrû ‘continuously’, cf. correctly UW 125ab: amru, Erdal, p. 726 (mention-
ing that the immediate base with the suffix *-(U)r- o f the converb cannot be found in the Turkic
languages either, and that Clauson was wrong to consider the usage o f the lexeme amru a char-
acteristic feature of Buddhist texts).
Cl. 163b: D amrit- (amru!-) Pass. f. o f *amur-; UW 123b-125a; Erdal, pp. 3 16-317.
Cl. 164b: D ami'rtgur- Caus. f. o f ami'rt- (amurt-); UW 118ab, Erdal, p. 748.
UW 123b: amrit;- ‘ruhig werden’, and cf. Erdal, p. 645.
UW 125a: amriltur- ‘zur Ruhe kommen’; Erdal, p. 825.
As I have already mentioned, I will not talk about the data related to the Old Turkic verb
amra- *love, wish’ - they must be treated separately at the present stage o f our investigation, al-
though Clauson and certain other scholars believed that these data belong to the same group.
39 The data from the Modern Turkic languages may not be evaluated with certainty. We are
able to provide appropriate explanations regarding Turkic data, but very often we cannot exclude
the chance of a word having been re-borrowed from Mongolian. Although I will not point it out in
the case o f each word, it must be borne in mind when looking at the following list, so 1 will pro-
vide the Mongolian data deriving from the verbal stem *amX- borrowed from Turkic before listing
the Turkic data.
Mongolian amu- v.i. ‘to rest, relax; to feel contentment or joy; to be relieved’ (L) and its
derivatives: amuyul- v. caus.; amuyulang n. and adj. ‘peace, quietude, happiness, well-being; quiet,
calm, peaceful, peaceable’; amur n., adj. and adv. ‘peace, quiet, calm, rest; leisure; pleasure; good
health, well-being; easy, not difficult; peacefully; quietly’; Cf. also: KWb. 9b: amr ‘ruhig, Ruhe,
friedlich, Friede, gesund, bequem’ etc. The Mongolian forms amura- (amara -) v.i. 'to rest, relax;
to feel contentment or joy; to be relieved’; amuray (amaray ) n. and adj. ‘love, affection; beloved,
dear, darling’ cannot be considered forms which are directly related to it etymologically; they are
contaminated forms which may be a result of the mutual influence o f the Mongolian word amu- of
Turkic origin discussed here and the Mongolian loanword (amura- ~ amara-) from the Turkic verb
amra- ‘love’. (We will see the same phenomenon in some o f the Tuvan forms listed among the
Turkic data.)
Turki: YUygh. amir ‘spokojnyj, spokojno’ petrified and became part o f the lexeme as the
historical aorist form and its derivative with the suffix +A-: ami'ra- ‘byt’ spokojnym, uspokaivat’sja’;
Siberian: Tuvan amir ‘(privetstvie ) pokoj, spokojstvie, mir; legkij’ and its derivatives,
which are concomitant with the forms derived from the historical verbal stem amra- ‘to love’:
ami'ra- ‘uspokaivat’sja, utesat’sja; radovat’sja, veselit’sja’; amiral ‘radost’; otdyh, peredyska’; ami-
raldi'g ‘blagodetel’nyj, spasitel’nyj, poleznyj; milostivyj’; ami'ranci'g ‘radostnyj’. I do not mention
tw o Tuvan items, ami'rat- ‘radovat’, veselit” and ami'rak ‘milaja, vozljublennaja’, because on the
basis o f their meaning they can be directly related to the base amra-, which is to be separated.
Oyr. amir ‘mir, pokoj, otdyh, spokojstvie, bezopasnost’, blagopolucie, soglasie; mimyj, po-
kojnyj, blagopolucnyj; spokojno, blagopolucno’ and derivatives: ami'ra- ‘otdyhat’, uspokoit’sja’;
ami'rak ‘mimyj, druzeljubnyj’ (where the length in the third syllable seems to be secondary);
2. The Tuvan verb form ovär- ‘obrascat’ vnimanie na kogo-cto-l.; zamecat’

kogo-cto-l.' with uncertain suffixation and its derivatives40;
3. Turkic loanwords41 derived from the Mongolian form abura- ‘to save,
amiral- ‘zastavit’ otdyhat” ; ami'ras ‘otdyh, uspokoenie’. Cf. also in the Oyr. dialects: Ojr.Tuba
ami'ra- ‘otdyhat” ; amiral- ‘razlec’sja, raspolozit’ svoe telo dlja otdyha, zastavit’ otdyhat” ; Ojr.Kmd.
abïr-eùncu ‘spokojno’ (where -b- may be secondary in place o f -m-, cf. e.g. tebir, debir ‘zelezo,
zeleznyj’); amir ‘mir, spokojstvie, soglasie, blagopolucie, tisina’; ami'ra- ‘usmirit’sja, uspokoit’sja’;
ami'räk ‘mirnyj, druzeljubnyj’.
Tofa amir: amir m’endi ‘dobrogo zdorov’ja! (pozelanie uezzajuscemu)’ (Rassadin). (Ras-
sadin’s Mongolian” comment is correct. It is beyond doubt that this Tofa structure maintains
a loanword from Mongolian.)
Khak. amir ‘mirnyj, spokojnyj; smirnyj, tihij; skromnyj; umerennyj’ and its derivatives:
ami'ra- ‘uspokaivat’sja, utihat’; prismiret” ; ami'rabas ‘neugomonnyj, bespokojnyj’; ami'rat- ‘uspo-
koit’ kogo-cto-l.’.
It is worth listing separately the data derived from R adloff s dictionary, which mostly con-
tains data from the Siberian Turkic languages, in which the intermediate -b- is secondary in place
o f the earlier -m-: Schon, Leb., Kir., abi'r (cf. amir) ‘spokojstvie, mir”, soglasie - die Ruhe, der
Friede; vëzlivost’ - die Höflichkeit’ (1:627) (it is uncertain if the latter meaning belongs here or
not); Schon abi'raktan- ‘uspokaivat’sja - sich beruhigen’ (1:627); Schon abi'rla- ‘uspokoit’, ostavit’
v” pokoë - zur Ruhe bringen, beruhigen, Ruhe lassen’ (1:627) and some other derivatives: Schon
abi'rlan- ‘byt’ spokojnym, otdyhat’ - sich ausruhen, ruhig sein’ (1:627); Schon abirlïg ‘zivusëij v”
miré i spokojstvii - in Ruhe und Frieden lebend’ (1:627); Schon abi'rak = abrak ‘spokojnyj, tihij -
ruhig, still’ (1:627). In R adloff s dictionary we also find data with an intermediate -m- in them:
Alt., Tel., Sag., Koib., Küär. amir [the inclusion of the Mongolian and Manchu word amuran
‘geneigt zu etwas’ should be rejected] ‘spokojstvie, mir”, soglasie, otdyh” - die Ruhe, der Frieden,
das Ausruhen, die Erholung’ (1:647), the derivatives of which are also provided in the dictionary.
We do not consider the forms Khal. havul, havul ‘gut’ (Doe-Tez) applicable here, al-
though Doerfer - despite the serious doubts he expressed- attempted to connect to the Old Turkic
data ami'l - amul ‘ruhig, friedvoll’ at several places (cf. D oe-T ez 127b; Doerfer 1985, p. 99, No.
283; and, following Doerfer, Scerbak also did the same most recently, see 1997, pp. 50, 52, 182.
They both also connect the Old Turkic ami'1 - amul data to the verb amra- ‘to love’ as well. The
inclusion of the Khalaj form encourages them to provide a totally different Ancient Turkic recon-
struction, which is *pabul (*pamul).)
40 Cf. the following derivatives: oväri'n- Refl. ‘osteregat’sja; brat’ sebja v ruki, odumat’sja;
unimat’sja’; ovärti'n- = oväri'n-; ovärt- Caus. of ovär- ‘obrascat’ c ’e-1. vnimanie; predosteregat” ;
ovàrïmca ‘ostoroznost” ; ovâri'mcalïg ‘ostoroznyj; vnimatel’nyj; ostorozno, vnimatel’no’.
41 It began to spread during the Mongolian period [cf. among the Middle Turkic data abra-
(Aor. abrar) ‘schützen, verteidigen’ (CCI), abramaq ‘defensio’ (CCI); Khwar. apra- ‘to keep, store’
(Boesch.)]. Their appearance is limited and concerns only the following languages and branches:
Kipchak (Karaim dialects and one Volga Tatar dialect)
Kar. abra- ‘hranit’, sohranjat’, berec’; oberegat’; sterec’, ohranjat’; prjatat’; spasat’, zasci-
scat” (T, H, C) and derivatives: abrav ‘zascita, spasenie’ (Kar T); abravcu ‘oberegajuscij; storoz,
straznik’ (Kar H); abravcu ‘stereguscij, storoz, straznik’ (KarT); abramak ‘ohrana, hranenie’ (Kar
H); abran- Refl-Pass. ‘berec’sja, prjatat’sja, hranit’sja, sohranjat’sja; berec’sja, uberec’sja’ (Kar T);
abrat- ‘berec’, zascicat’; strahovat” (Kar H); abratmak ‘strahovanie’ (Kar H); abratuvcu ‘zascit-
nik’ (Kar H);

58 A. BERTA
rescue, help, deliver, preserve life; to protect’ (L)*

42. However, the Mongolian abura-
is probably not a verbal base43 and most likely has its origin in the Ancient Turkic
verb *àbï-, which is the same as the etymon we defined for the Hung. dv. If this
assumption proves to be correct, the number of lexemes of identical Turkic origin in
both Mongolian and Hungarian44 increases further.
Languages and dialects
Alt. - Oyrot in Radloff

Ancient Hung. - Ancient Hungarian
Az. —Azerbaijanian, see AzRSl.
Bashk. - Bashkir, see BRS1.
Chag. - Chaghatay, see Radloff
Cr-Tat. - Crimean Tatar, see CrTatRSl.
Hung. - Hungarian
Kar. H - the Halic dialect o f Karaim, see KRPS1.
Volga Tatar dialect abra- ‘spasti, vyrucit’; hranit” (TTDS, Ngb-krs).

Siberian (Tuvan)
Tuvan avi'ra- ‘(ust.) spasti, izbavit’; okazat’ milost’, pomilovat” and its derivatives: avi'ral
rel. ‘spasenie, izbavlenie; pokrovitel’stvo; milost’, pomosc” ; amiral = avi'ral‫׳‬, avi'raldi'g: avi'raldi'g
kizi ‘celovek, pomogajuscij drugim (zabotjascijsja о drugihy, al’truist’.
Yakut and Dolgan
Yak. abrä- (abïrâ-) ‘popravljat’; prinosit’ pol’zu; milovat’ kogo, scadit’, okazyvat’ komu
m ilost’, proscat’; zasciscat” (Рек. 1:8-9) and its derivatives: abral (abi'rat) ‘pomosc’, vspomozenie
v” krajnosti, v” nuzdah”, - p o l’za; milost” (Рек. 1:9); abralläx (abi'ralläx) ‘poleznyj, vaznyj, spasi-
tel’nyj dlja kogo {Dat.y (Рек. 1:9); abrâllâxtïk, abïrallâxtik Adv. ‘polezno, spasitel’no’ (Рек. 1:9);
abran-, abi'ran- ‘polucat’ pom osc’ v” krajnosti, v ” nuzdah”, pol’zovat’sja; milovat’sja, proscat’sja’;
abrat- Kaus. von abrä- and there are some other derivatives as well.
Dolg. (Stachowski 1993a) abi'rä- ‘retten; nützlich sein, zustatten kommen’ and its dériva-
tives abi'ran- ‘sich retten’ (Stachowski 1998); *abi'rat- ‘helfen’, cf. abïratï- ‘finanzielle Hilfe der
reicheren Sippenmitglieder für einen verarmten Verwandten’ (Stachowski 1998).
Chuvash
ирга- ‘berec’, hranit” and its derivatives up rav ‘zascita, ohrana’; upran- ‘hranit’sja; sohra-
njat’sja; oberegat’sja, berec’sja’; uprattar- Caus. o f upra- (Skvorcov).
All o f this indicates that we have no data from the majority of Modern Kipchak and Sibe-
rian Turkic languages, this item is missing from the Turki and Oghuz language branches, and it is
not a part of Khalaj either.
42 Cf. also aburayci ‘saviour, rescuer; deliverer; protector’ (L); aburayda- pass, o f abura-
(L). Among Modern Mongolian data, see Khalkha avra- ‘to save, to rescue, to free’ (Kara), Buryat
abarxa- ‘osvobodit” (KRBS1). On the way the Mongolian word entered the Turkic languages, see
also REW 2b, KWb. 20a, and with some uncertainty Sev. 1:59-60 (with further literature there).
43 As for its formation, the Mongolie abura- may be a denominal derivative with the suffix
+A-, the base of which - *abi'r (*abur) - may have developed in the Mongolian language. This
*abur form, which does not exist today, may be a deverbal noun formed by the suffix -r.
44 Cf. Rdna-Tas (1977).
Kar. C - the Crimean dialect o f Karaim, see KRPS1.

Kar. T - the Trakai dialect of Karaim, see KRPS1.
Kas. - Volga Tatar in Radloff
Khak. - Khakas, see HakRSl.
Khal. —Khalaj, see D oe-T ez.
Khwar. - Khwarezmian Turkic
Kip. - Kipchak Turkic
Kir. - Kazakh in Radloff
Kirgh. - Kirghiz, see Judahin
Kkalp. - Karakalpak, see Baskakov 1951
Koib. - Koybal dialect o f Khakas in Radloff
Krc-Blk. - Karachay-Balkar, see RKrc-BlkSl.
Küär. —Kuerik dialect o f Chulym Turkic in Radloff
Kzk. - Kazakh, see KzkRSl.
Leb. - the Lebed dialect o f Oyrot, see Radloff
mo. - Mongolian in Röhrborn, К. (1977-1994)
mtü. - Middle Turkic in REW
Ngb-krs. - Naghaybek-Keräsen dialect of Volga Tatar, see TTDS
Nogh. - Noghay, see NRS1.
NUygh. - New Uyghur, see UjgRSl.
Old Hung. - Old Hungarian
Osty. V - the Vach dialect of Ostyak
Oyr. - Oyrot, see OyrRSl.
Ojr.Kmd. - the Kumanda dialect of Oyrot, see Baskakov (1972)
Oyr.Tuba - the Tuba dialect o f Oyrot, see Baskakov (1966)
Sag. - the Saghay dialect o f Khakas in Radloff
Schon - Shor in Radloff
Sib-Tat. - Siberian Tatar
Tat. - Volga Tatar, see TatRSl.
Tel. - Teleut, see Radloff
Turk. - Turkish, see TuRSl.
Turkm. - Turkmen, see TurkmRSl.
Tuv. - Tuvan, see TuvRSl.
Uig. - Old Uyghur in Radloff
Uzb. - Uzbek, see UzbRSl.
Xak. - Xakani
Yak. - Yakut
YUygh. - Yellow Uyghur, see Tenisev (1976)
Vog. So. - the Sosva dialect o f Vogul
Vog. T - the Tavda dialect o f Vogul
References
Arat, R. R. (1947-1959), Kutaclgu bilig. 1-2. Istanbul.

AzRSl. (1985), Azerbajdzansko-russkij slovar’. Ed. H. A. Azizbekov. Baku.
Baskakov, N. A. (1951), Karakalpakskij jazyk. I. Materialy po dialektologii (teksty i slovar’).
Moskva.

60 Ä. BERTA
Baskakov, N. A. (1966), Dialekt cemevyh tatar (Tuba-кЩ). Grammaticeskij ocerk i slovar'.

Moskva.
Baskakov, N. A. (1972), Dialekt kumandincev (Kumandi'-kizi). Grammaticeskij ocerk, teksty, pere-
vody i slovar’. Moskva.
Berta, Â. (1996), Deverbale Wortbildung im Mittelkiptschakisch-TUrkischen. (Turcologica 24),
Wiesbaden.
Berta, Ä. (1996a), Zur Frage der Kausativsuffixe im Kiptschakischen. Symbolae Turcologicae.
Studies in Honour o f Lars Johanson on His Sixtieth Birthday 8 March 1996. Edited by
A. Berta, B. Brendemoen and C. Schönig. (Swedish Research Institute in Istanbul. Trans-
actions 6 ), istanbul, pp. 23-29.
B oesch. see Boeschoten, H. E.-Vandamme, M .-Tezcan, S. (1995)
Boeschoten, H. E.-Vandamme, M.-Tezcan, S. (1995), Al-Rabghiizi. The Stories o f the Prophets.
Qisas al-Anbiyä’. An Eastern Turkish Version. Vols 1-2. Leiden-New York-Köln.
Borovkov, A. K. (1963), Leksika sredneaziatskogo tefsira XII-XIII vv. Moskva.
BRS1. - Basqortsa-russa hiidlek. Ed. Z. G. Uraqsin. Mäskäü, 1996.
Caferoglu, A. (1931), Abû-Hayyân. Kilâb al-tdrâk li-lisân al-Atrâk. istanbul.
Caferoglu, A.-Doerfer, G. (1959), Das Aserbeidschanische. PhTF 1. pp. 280-307.
CC see Grpnbech, K. (1936) and (1942)
CCG - the German part o f the Codex Cumanicus, see CC
CCI - the Italian part o f the Codex Cumanicus, see CC
Cl. see Clauson, G. Sir (1972)
Clauson, G. Sir (1972), An Etymological Dictionary o f Pre-Thirteenth-Century Turkish. Oxford.
CrTatRSl. - Krymskotatarsko-russkij slovar’. Eds S. A. Asanov et al. Kiev, 1988.
Dankoff-K elly see Dankoff, R .-K elly, J. (1982-1985)
Dankoff, R.-Kelly, J. (1982-1985), Mahmud al-Käsgari. Compendium o f the Turkic Dialects
{Diwan Luyät at-Turk). 1-3. Edited and Translated with an Introduction and Indices by R.
Dankoff in collaboration with J. Kelly. (Sources o f Oriental Languages and Literatures 7),
Harvard University Printing Office.
Doerfer, G. (1971), Bemerkungen zur Methodik der türkischen Lautlehre. OLZ 64:7/8, pp.
325-344.
Doerfer, G. (1985), Mongolo-Tungusica. (Tungusica 3), Wiesbaden.
Doerfer, G.-Tezcan, S. (1980), Wörterbuch des Chaladsch (Dialekt von Charrab). (Bibliotheca
Orientalis Hungarica 26), Budapest.
D oe-T ez see Doerfer, G.-Tezcan, S. (1980)
Erdal see Erdal, M. (1991)
Erdal, M. (1991), Old Turkic Word Formation. A Functional Approach to the Lexicon. 1-2.
(Turcologica 7), Wiesbaden.
EtymWb. see REW
Gabain, A. von (1954), Türkische Turfan-Texte. VIII. (Abhandlungen der Deutschen Akademie der
Wissenschaften zu Berlin. Klasse für Sprachen, Literatur und Kunst. Jahrgang 1952. Nr. 7),
Berlin.
Grpnbech, K. (1936), Codex Cumanicus. Cod. Marc. Lat. DXLIX. In Faksimile herausgegeben mit
einer Einleitung von ... (Monumenta Linguarum Asiæ Maioris. 1), Kopenhagen.
Gr0 nhech, K. (1942), Komanisches Wörterbuch. Türkischer Wortindex zu Codex Cumanicus. (Mo-
numenta Linguarum Asiæ Maioris. Subsidia. Vol. 1), Kopenhagen.
HakRSl. - Hakassko-russkij slovar’. Ed. N. A. Baskakov. Moskva, 1953.
id. see Caferoglu (1931)

Jarring, G. (1964), An Eastern Turki-English Dialect Dictionary. (Lunds Universitets Ârsskrift.

N.F. Avd. 1. Bd. 56. Nr. 4), Lund.
Johanson, L. (1979), Alttürkisch als ‫״‬dissimilierende Sprache“. (Abhandlungen der Geistes- und
Sozialwissenschaftlichen Klasse. Jahrgang 1979. Nr. 2.) Mainz-Wiesbaden.
Johanson, L. (1991), Die Ersetzung der türkischen -t-Kausativa, in: Johanson, L. Linguistische Bei-
träge zur Gesamtturkologie. (Bibliotheca Orientalis Hungarica 37), Budapest, pp. 170-197.
Judahin, К. K. (1965), Kirgizsko-russkij slovar’. Moskva.
Kara Gy. (1998), Mongol-magyar kéziszôtdr [Mongolian-Hungarian Dictionary] Budapest.
Käs. see Dankoff, R.-Kelly, J. (1982-1985)
KRBS1. - Kratkij russko-burjatskij slovar’. Eds C. B. Cydendambaev, M. N. Imehenov. Moskva,
1962.
KRPS1. - Karaimsko-russko-pol’skij slovar’. Eds N. A. Baskakov, A. Zajonckovskij, S. M. Sap-
sal. 1974, Moskva.
KWb. see Ramstedt, G. J. (1935)
KzkRSl. - Kazahsko-russkij slovar’. Ed. G. Musabaev. Alma-Ata, 1954.
L see Lessing, F. D. (1973)
Lessing, F. D. (1973), Mongolian-English Dictionary. Bloomington.
Nadzip, E. N. (1979), Isloriko-sravnitel’nyj slovar’ tjurkskih jazykov XIV. veka. Na materiale
“Xosrau i sirin’’ Kutba. Kniga I. Moskva.
NRS1. - Nogajsko-russkij slovar’. Ed. N. A. Baskakov. Moskva, 1963.
OyrRSl. - Ojrotsko-russkij slovar’. Ed. N. A. Baskakov. Moskva, 1947.
Рек. see Pekarskij (1907-1930)
Pekarskij, È. K. (1907-1930), Slovar’ jakutskogo jazyka. 1-3. St. Petersburg, Leningrad.
P. Hidvégi A. (1956), '()v - oltalom’ [On the Hungarian words ov and oltalom ]. Emlékkônyv Pais
Dezsff hetvenedik születe'snapjdra. Irtdk tisztelâi, baràtai és tam'tvdnyai. [Commemorative
Volume on the 70th birthday o f Dezsô Pais. Written by his admirers, friends and students.]
Eds Géza Bärczi and Loränd Вепкб. Budapest, pp. 304-306.
PhTF 1 - Philologiae Turcicae Fundamenta. Eds J. Deny, K. Grpnbech, H. Scheel, Z. V. Togan.
Tomus primus. [Wiesbaden] 1959.
QB - Qutadgu bilig ‘Wisdom that brings good fortune’, Yüsuf o f Balasagun’s poem (Xlth
century) see Arat (1947-1959)
Radloff see Radloff, W. (1893-1911)
Radloff, W. (1893-1911), Versuch eines Wörterbuches der Türk-Dialecte. 1-4. Sanktpeterburg.
Ramstedt, G. J. (1935), Kalmückisches Wörterbuch. (Lexica Societatis Fenno-Ugricae 3), Hel-
sinki.
Räsänen, M. (1969), Versuch eines etymologischen Wörterbuchs der Türksprachen. (Lexica Socie-
tatis Fenno-Ugricae 17,1), Helsinki.
Rassadin, V. I. (1971), Fonetika i leksika tofalarskogo jazyka. Ulan-Udé.
Rédei К. (1969), Szofejtések [Etymological Notes], NyK 71, pp. 101-112.
Rédei, К. (1988), Uralisches etymologisches Wörterbuch. Vols 1-2. Wiesbaden.
REW see Räsänen, M. (1969)
RKrc-BlkSl. - Russko-Karacaevo-Balkarskij slovar’. Eds H. 1. Sujuncev, I. H. Urusbiev. Moskva,
1965.
Röhrborn, K. (1977-1994), Uigurisches Wörterbuch. Sprachmaterial der vorislamischen türki-
sehen Texte aus Zentralasien. Lieferung 1-5. Wiesbaden.
Röna-Tas A. (1977), A magyar-bolgar-török érintkezés jeliege [The nature o f Hungarian and
Bulgarian Turkic relations]. In: Bartha A., Czeglédy К., Röna-Tas A. (eds), Magyar östör-
te'neti tanulmdnyok. [Studies in Hungarian Ancient History.] Budapest, pp. 267-275.

62 Ä. BERTA
Röna-Tas, A. (1999), Chuvash and historical morphology. Acta Orientalia Hung. 52, 1-15.
Scerbak, A. M. (1997), Rannie tjurksko-mongol’skie jazykovye svjazi (VIII-XIV vv.) Sankt-
Petersburg.
Sev. see Sevortjan, É. V. (1974)
Sevortjan, È. V. (1974), Ètimologiceskij slovar’ tjurkskih jazykov. 1. Moskva.
Skvorcov, M. I. (1982), Cävasla-vyräsla slovar’. Muskav.
Stach, see Stachowski, M. (1993)
Stachowski, M. (1993), Geschichte des jakutischen Vokalismus. (Rozprawy habilitacyjne Nr. 264),
Uniwersytet Jagiellonski. [Krakow],
Stachowski, M. (1993a), Dolganischer Wortschatz. (Prace Jçzykoznawce. Zeszyt 114). Kraköw.
Stachowski, M. (1998), Dolganischer Wortschatz. Supplementband. Kraköw.
TatRSl. - Tatarsko-russkij slovar’. Moskva (1966).
Tefs. see Borovkov, A. К. (1963)
Tekin, T. (1995), Türk dillerinde birincil uzun ünlüler. (Türk Dilleri Araçtirmalan Dizisi 13),
Ankara.
Tekin,T. et al. (1995), Türkmence-Türkçe Sôzlük. Ankara.
Tenisev, É. R. (1976), Stroj sarïg-jugurskogo jazyka. Moskva.
TESz. - A magyar nyelv tôrténeti etimolôgiai szôtdra. 1967-1984. [Etymological Dictionary o f the
Hungarian Language] 1-4. Editor in Chief: L. Benkö. Budapest.
TRS1. - Rassadin, V. I. Tofalarsko-russkij slovar’. Russko-tofalarskij slovar’. Irkutsk, 1995.
TT VIII see Gabain, A. von (1954)
TTDS - Tatar tëlënën dialektologik süzlëgë. Kazan, 1969.
Tumaseva, D. G. (1972), Slovar’ dialektov sibirskih tatar. Izdatel’stvo Kazanskogo Universiteta.
TurkmRSl. - Turkmensko-russkij slovar’. Eds N. A. Baskakov-B. A. Karryev-M. Ja. Hamzaev.
Moskva, 1968.
TuRSl. - Turecko-russkij slovar’. Eds A. N. Baskakov et al. Moskva, 1977.
TuvRSl. - Tuvinsko-russkij slovar’. Ed. A. A. Pal’mbah. Moskva, 1955.
UjgRSl. - Vjgursko-russkij slovar’. Ed. É. N. Nadzip. Moskva, 1968.
UW see Röhrbom, К. (1977-1994)
UzbRSl. - Uzbeksko-russkij slovar’. Ed. A. K. Borovkov. Moskva, 1959.
Arpad Berta
JATEВТК
H-6726 Szeged
Egyetem u. 2.
Hungary

THE INTERNAL DIVISION OF MODERN TURKIC

AND ITS HISTORICAL IMPLICATIONS
C l a u s S c h ö n ig
(Gießen)
In this article I try to establish a model for the linguistic internal division o f Modern Turkic. The
model is based on synchronic data, but like previous models it tries to integrate the diachronic
dimension. Besides phonetical data, it also features data from the fields o f morphology, lexicology
and syntax. The distributional patterns of these features are compared with each other under dia-
chronic and geographical aspects. As a result we can define some subgroups o f Turkic o f different
hierarchic rank. Furthermore these subgroups can be o f different origin. Some o f them are tied to-
gether by common linguistic features which seem to be transmitted within the languages o f this
group from one time level to another and constitute their common genetic heritage. Such groups
I call genetic (sub-)branches, e.g. the Oghuz branch, the Kipchak branch, the Bulghar Turkic branch
etc. Other subgroups of Turkic are clearly composed o f languages from different branches, which
have formed new territorial units and went into areal interaction. Such an areal group is, e.g., South
Siberian Turkic. In the second part o f the article I try to put processes o f internal differentiation o f
genetic branches into subbranches and the formation and dissolution o f areal groups in temporal
relation to historical facts and developments.
Key words: Proto-Turkic, areal features, classification
1. Common Turkic and Norm Turkic
A feature is called Common Turkic if it is attestable in all Turkic languages, or if its

absence in one, some, or most of them is explicable, e.g. the verb *al- ‘to take’, the
numerals from one to ten, or the preterite in *-DI. Some other items which were
elided or considerably changed in some modern languages can also be reconstructed
as Common Turkic, e.g. pronominal -n, the forms of the tens as found in Old Turkic
and Chuvash. It is not possible to list all the other probably Common Turkic features
here. Since Common Turkic features are not distinctive, they are not helpful for an
internal division of Turcia. Thus other parameters must be developed.
Norm Turkic is statistically defined. A feature is called Norm Turkic if it ap-
pears in a maximal group of linguistic units, i.e. not only in languages such as Tur-
kish, Azeri, Tatar, Bashkir, Kazakh, Kirghiz, Uzbek, Turkmen and Modem Uighur,
0001-6446 /99/ $ 5.00 © 1999 Akadémiai Kiadô, Budapest

64 C. SCHÖNIG
but even in quite exotic languages like Tuvan and Saryg Yugur. Norm Turkic is
bound together by a large set of common features constituting a genetic string. Norm
Turkic languages show more or less complete loss of Ancient Turkic word-initial *h-,
have kept the opposition c- : y- in word-initial position, show a nominal plural suffix
+lAr, possess a gerund in -p, display cognates of the conditional suffix -sAr, and
have a third person imperative suffix going back to -ZUn, see also table 3 on p. 89.
2. Central Turkic and Border Turkic
I consider all Turkic languages as Central Turkic that show the sound change An-
cient Turkic -d(-) > -y(-), i.e. the classic feature illustrated by the (Old Turkic) word
adaq ,foot’. Other features of Central Turkic are, e.g., forms of the personal inter-
rogative pronoun reconstructable as *kim, use of the verbs tog- ,to give birth ~ be
bom’ and ket- ,to go away’, replacement of the first person plural ending -mlz by -K
in the /-)/-preterite and the conditional, and existence of the suffix +IIK and of the
privative suffix +slz. All members of Central Turkic are Norm Turkic. Non-Central
Turkic units are called Border Turkic.
Border Turkic consists of the three non-Norm Turkic units and some Norm
Turkic ones. For Old Turkic -d(-), non-Norm Turkic Chuvash shows -r(-), Khalaj
has -d(-) and Lena Turkic (Yakut and Dolgan) -t(-). The Norm Turkic constituents of
Border Turkic are Sayan Turkic (with -d(-)), parts of Chulym and of Yenisey Turkic, 1
Saryg Yugur and Fu-yii (with -z(-)). Especially Yenisey Turkic and Fu-yii belong
closer together, see Schönig (1998b). Saryg Yugur additionally shows similarities to
Sayan Turkic, especially to Tuvan. Both the azaq- and the (h)adaq-group represent
final Old Turkic -G-sounds (tagli'g ,mountainous’) and many archaic features (see
2.2-2.2.2). Besides, in Border Turkic languages many of the above-mentioned Cen-
tral Turkic features are also missing. Especially Chuvash, Khalaj and North East
Turkic (South Siberian Turkic and Lena Turkic) have postvocalic aorist forms cor-
responding with Old Turkic -yUr or a form of the word for ,stirrup’ which seems to
be more archaic than the Old Turkic form;2 see also tables 4 and 5 on pp. 90-91.
2.1. The internal division of Central Turkic
Central Turkic shows three main branches: Oghuz, Kipchak and South East Turkic;
they are tied together internally by genetic sub-strings. Salar and Altay Turkic to-
gether with parts of Chulym and Yenisey Turkic (see note 1) belong to Central Turkic
by sharing the sound change -d(-) > -y(-), but at the same time exhibit many exotic
1 The Shor dialect o f Khakas, the Kondoma dialect of Shor and the Lower Chulym Turkic
dialects show *-d(-) > *-y(-).

2 Khalaj has -yUr (Doerfer 1988), Lena Turkic -I:r, Khakas -ir. In Shor and Sayan Turkic
the quality of the stem-final vowels determines the quality of the contraction product. For Chuvash
see Johanson (1976, p. 135).
THE INTERNAL DIVISION OF MODERN TURKIC 65
and archaic features. Thus Salar has, e.g., a genitive in +niyi, see also tables 4 and 5
on pp. 90-91. The distribution of nasality of the intervocalic consonant in the cog-
nates of *sürjök ‘bone’ and of the forms of the agent noun of the type (verbal noun)
+ cl correlate directly with the three branches; additionally they give a hint to the
Kipchak background of Uzbek, see table 1 on p. 89.
The branches can be internally differentiated by features expressed by the
classic keyword Old Turkic taglig ‘mountainous’. Its development clearly demon-
strates that Kipchak, which as a whole shows the feature tag > taw, has two main sub-
branches: Western-Central Kipchak (tawli) and Kirghiz-Kipchak (to.lu:). The distri-
bution of forms of the verb for ‘to cry’ underscores this division, see table 1 ; further-
more Kirghiz-Kipchak (Kirghiz and Altay Turkic) exhibits features in the field of
phonotactical rule sets different from those of other Kipchak languages, see 4.4.1.
Western-Central Kipchak consists of Western Kipchak, i.e. Far Western Kipchak
(Karaim and Misher Tatar), Volga-Ural-Caucasus Kipchak (Tatar, Bashkir (= Volga-
Ural K.), Karachay-Balkar and Kymyk (= Caucasus K.), and Central Asian Kipchak
(Karakalpak, Kazakh). Crimean Tatar is a Western Kipchak language. It was under
strong Western Oghuz (Ottoman Turkish) influence. Volga-Ural Kipchak was the
main Turkic contact language for Chuvash. Both have, e.g., +(I)GIz instead of
+(I)rjIz as the second person singular possessive suffix, and show palatalised forms
of sac > cac > cäc ‘hair’, see also 4.4. Central Asian Kipchak is characterised, e.g., by
*menen forms to express ‘with’ (exhibiting a tendency to become enclitic), whereas
the other Kipchak sub-branches mostly show *( +)mInAn forms (besides non-suffixal
*bi(r)lä(n) forms in all branches). Kirghiz-Kipchak took part in the development of
North East Turkic, during which it gained some special features through areal inter-
action, see 4.4.1. Mainly Central Asian Kipchak, Kirghiz-Kipchak and South Siberian
Turkic were under Oirat influence from the fifteenth until the eighteenth century.
The Siberian Tatars, who later became subject to South Siberian Turkic influence,
seem to be a mixed group of Central and Volga-Ural Kipchak type. Additionally,
there are Kipchakoid elements in South Siberian Turkic, see 4.4.2.3. Especially Lena
Turkic and Kipchak weaken p and К while they retain t in intervocalic position and
additionally exhibit some structural similarities in the field of phonotactic rule sets,
see 3.2. The development of first syllables consisting of a palatal vowel and a weak
consonant, of which at least one element is labial, additionally underscores the inner
division of Kipchak. These sound groups have normally converged in iiy and became
öy in Volga-Ural Kipchak; Far Western Kipchak shows variants with sometimes
unetymological final -w, see Berta (1989). In Kirghiz-Kipchak we find iiy besides
contracted forms with long labial vowels. Of South East Turkic, Modem Uighur and
Uzbek behave like Western-Central Kipchak languages, but Uzbek has not consis-
tently changed äv to iiy. Oghuz, especially Western Oghuz, is more conservative with
its preservation of äv; Oghuz and some Modem Uighur dialects together with the
non-Norm Turkic units exhibit a tendency to preserve nasality of the intervocalic
consonant in *sürjök ‘bone’ (Azeri siimiik, Turkmen sürjk).
As in the cases of the afore mentioned sound groups and the agent noun, the
intermediary position of Uzbek between Kipchak and South East Turkic is illustrated

66 C. SCHÖNIG
by its (literary) form tâgli, which represents a pre-modem Western-Central Kipchak

form *taglï. Other features such as the neutralisation of the opposition i : ï (présuma-
bly under Iranian influence) and loss of pronominal n connect it with South East
Turkic on the level of areal interaction. Today South East Turkic is best represented
by Modem Uighur. Unlike Uzbek, it has the ablative suffix +DIn instead of +DAn?
South East Turkic resembles Kipchak in many respects. With respect to the North-
em : Southern and Western : Eastern divisions of Turkic (see 4.3 and 4.4), South
East Turkic often takes a transitory position and shows more Eastern Turkic features.
Modem Uighur has some features in common with Khalaj, see 3.3.
In addition to the features already mentioned Oghuz - like Chuvash, see 3.1 -
has preserved old suffixes of the structure +(C)V... (e.g. a genitive suffix in +(n)Irj
as in Old Turkic), but additionally also produced new ones of the same type. Whereas
in most branches of modem Turkic case-marked 31gay-stems are used as interrogative
pronouns for places, directions and the like, Oghuz has forms like Turkish nerede
‘where’ (derived from ne ‘what’), while Azeri shows the same derivational element
+rA in hara+ on a *qa(n)-stem, see also below. Modem Uighur has nädä and qayärdä.
Oghuz can be divided into Western and Eastern Oghuz (mainly Turkmen and
Khorasan Turkic).34 Western Oghuz employs a binding consonant y in the postvocalic
forms of, e.g., the dative in +(y)A or the gerund in -(y)Ip, whereas Eastern Oghuz
Turkmen has long vowels as a result of contraction. Western Oghuz has labialised
cognates of Old Turkic bädük ‘big, great’. Turkmen has preserved long vowels,
whereas in Western Oghuz they only survived sporadically but reflexes of long vow-
els are visible on the following consonant, see 2 .2 .
Azeri has had intensive interaction with Persian, Turkish had additional inter-
action with Greek and other Indo-European languages of the Balkans, and Gagauz
with Slavic languages. As the result of such interaction we find, e.g., elaborate sys-
terns of complex conditional forms or subordinated clauses with finite predication (the
latter mainly in Azeri and Gagauz). Within Western Oghuz the transition from Tur-
kish to Azeri can be demonstrated by features such as the preservation of low vow-
els, e.g. Old Turkic bädük > Azeri böyük : Turkish büyük ‘great’, the penetration of
the 1st person plural marker -(I)K into paradigms other than those of the /}/-preterite
and the conditional in Azeri, and one also finds nasalisation of word-initial *b- in
Azeri, when a nasal consonant follows (like in most other Turkic languages). As for
originally postpositional elements of the structural type (b)i(r)lä(n) designating ‘with’
in Western Oghuz, the form ilä predominates in the literary languages, while some
Anatolian dialects also show bilä and birlän. At the same time, enclitic forms of the
type +(y)lä(n) ~ +(y)lA(n) can be found in all types of sub-standards and dialects -
as well as in neighbouring Khalaj.
Other features appear in Gagauz, Azeri and Turkmen, but are missing in Turk-
ish, e.g. the postvocalic accusative in -nl, and a negative aorist in -mAr, which has
penetrated the paradigms of these languages. Quite a number of these features are
3 The form +Dln ~ +DI 1j has also survived in Lower Chulym Turkic and the Quu dialect o f
Altay Turkic.
4 For a more detailed internal division o f Oghuz see Doerfer (1990).

common to the remaining Central Turkic languages, e.g. cognates of the Old Turkic
verb qadït- ‘to return’ and reflexive pronouns based on öz+. To express impossibility
Turkish and Gagauz have preserved the Ancient Turkic biverb -A u- in -(y)AmA-,
whereas they use -A bil- to express possibility. In Azeri (im-)possibility is expressed
by the biverb -A bil(-me-) as in Kipchak, South East Turkic and Chuvash; Turkmen
also uses forms containing the -p-gerund.
Turkmen exhibits features separating it from Western Oghuz and tying it closer
to other Central Turkic and -GAn-Turkic languages. In the case of the sound group
äv, Turkmen tends to shift the feature [-!‫־‬labial] onto the vowel, so that we often en-
counter results like oy. Unlike Western Oghuz, we sometimes find preservation of G
after the first syllable border. Furthermore Turkmen, like most -GAn-Turkic languages
(see 4.1), has replaced the participle in -mis by -An (< -GAn ?), shows postvocalic -y-
forms of the vocalic gerund, exhibits *-mAyln-forms instead of formally analogised
negative -mA-(yI-)B-forms in syntactically free use, and shows the same subject
marking strategies in relative clauses, the headword of which is not referentially iden-
tical with the subject of the relative clause. Some of these Turkmen features are made
up of Oghuz material shaped according to Central Asian Turkic patterns. Thus, al-
though Turkmen has reflexes of Old Turkic n, it shows a у-form hays'! for the attribu-
tively used Old Turkic pronoun qa.no ‘which’ instead of Western Oghuz-Khalaj ban-
forms. In Turkmen the intraterminal participle -yAn is contrastively marked against
the perfect participle -An by means of the same marker y which is used to mark the
renewed present tense form -yAr against the aorist form -(A)r. Thus Turkmen uses
the sign of intraterminality common to Oghuz and derived from *-A yori- similar to
the way Kipchak(oid) and South East Turkic languages use the marker *-A tur-. The
Old Turkic verb ï:d- and its derivation i‫־‬:du ber- ‘to send’, which have no cognate in
Western Oghuz, are represented in Turkmen by a (non-auxiliary) verb ibdr- ‘to send’
and an auxiliary verb goyber- < *qoyu ber-. Furthermore Turkmen has adopted at
least parts of the system of biverbal constructions expressing kinds of action (German
Aktionsanten) from Central Asian Turkic. All these features were preserved in or
passed over to Turkmen by means of areal interaction in the Central Asian area, see
4.2, see also Schönig (1997a-b).
2.2. Oghuz and Salar as Archaic Central Turkic
A set of archaic features mainly appears in Border Turkic and in the Central Turkic
units Oghuz and Salar. Thus, the most archaic forms of the Old Turkic word dlig have
survived with the meaning ‘hand’ in Lena Turkic, Chuvash, Fu-yii, Saryg Yugur and
sporadically in Modem Uighur, where they are still bisyllabic. Oghuz, Khalaj and
Salar have preserved a short form dl. In most New Turkic languages dlig is replaced
by qol meaning both ‘hand’ and ‘arm’. In the case of the word for ‘lip’, Oghuz, Khalaj,
Chuvash and Salar have *to.taq, Lena Turkic has uos (< agi'z [+okanie] ?), and Saryg
Yugur has dams ay in contrast to the other languages, which mostly have drin. Long
vowels are preserved as long vowels or diphthongs in Lena Turkic, Khalaj and Turk

68 C. SCHÖNIG
men; especially *ö: has survived as (ä)va in Chuvash. Western Oghuz, Sayan Turkic,
Salar and Saryg Yugur show at least reflexes of vowel length on the following con-
sonants, see Johanson (1986). The -w/f-participle is still used in Western Oghuz,
Lena Turkic, Khalaj and Salar as a finite form; functionally narrowed it has survived
in some Central Asian Turkic languages. Case suffixes of the type +(C)V... mainly
survived in Oghuz and Chuvash and can be found only sporadically in other Turkic
groups, e.g. the accusative suffix +(n)l in Lena Turkic or the dative suffix +(G)A in
Salar. See also table 4 on p. 90.
2.2.1. Common features of Oghuz and Border Turkic
Common features especially of Oghuz and Border Turkic are, e.g., survival of the
Ancient Turkic word *,hiiriiy ‘white’ (Lena Turkic, Karagas, Khalaj and Anatolian
dialects), preservation of the nasality of Ancient Turkic n (at least partly in Khalaj,
Lena-Sayan Turkic and Oghuz) and of -y- in the case of *süyök ‘bone’ (Oghuz,
Chuvash, Lena Turkic and Modern Uighur dialects). Regarding the sound group *äv
a round consonant can still be found in Western Oghuz, Khalaj, Yenisey Turkic and
Fu-yii. Turkish, South Siberian Turkic, Chuvash and Lena Turkic are connected by
absence of reflexive pronouns based on öz+ and of the verb qayt- < qadït- for ‘to
return’ as in remaining Central Turkic, see table 5 on p. 91.
2.2.2. Common features of non-Norm Turkic and Oghuz
Another set of features is absent in Border Turkic but attested in non-Norm Turkic
and Oghuz. All non-Norm Turkic and (with only a few exceptions) Oghuz languages
have preserved cognates of Old Turkic -gt/r-causatives and have not developed -GVz-
and -GVY-forms like most other New Turkic languages, see Schönig (1999a). Only
Western Oghuz and Khalaj use -DOK-participles in relative clauses regularly. Khalaj,
Turkish and Lena Turkic have preserved the verb bul- ‘to find’, which the other units
have replaced by tap-. Khalaj, Lena Turkic and Oghuz still use -mAdOK as a negative
participle, but only Gagauz and Lena Turkic still show the Ancient Turkic opposition
between the positive and negative participles -mis, -DOK : -mADOK. Only Turkmen
and Lena Turkic employ -y exclusively to designate the imperative of the second per-
son plural. Different from -G4n-Turkic (see 4.1), the cognates of Ancient Turkic tdyri
appear in Oghuz, Chuvash and Lena Turkic with back vowels, see Doerfer (1965,
pp. 577-585). Oghuz, Khalaj and Chuvash exhibit biverbal forms for ‘to begin to x’
consisting of a verbal noun in -mA(K) of the verb meaning ‘x’ and basla-, and show
a common necessitative suffix [Oghuz -mAlI, Chuvash -mAllÄ, Khalaj -mAlU(G)\,
the corresponding forms in Tatar or in Uzbek may result from Ottoman language
export.

3. Non-Norm Turkic
Non-Norm Turkic consists of Chuvash, Lena Turkic and Khalaj. Chuvash is most de-
viant from Norm Turkic. Lena Turkic and Khalaj show fewer non-Norm Turkic fea-
tures but often special representations of Norm Turkic features, see also table 3 on p. 89.
3.1. Chuvash
Chuvash is the only recent representative of Bulghar Turkic. It has a nominal plural
suffix +sem and a suffix -(t)tAr for the third person of the imperative. Instead of the
gerund in -p, a form -sA appears, which perhaps is connected with the Old Turkic
conditional suffix -sAr. It behaves like Border Turkic by showing Old Turkic -d(-) >
-r(-), survival of Old Turkic *Ö: as (a)va, preservation of the word for ‘hand’ in bi-
syllabic ah2, a *käm-ioxm of the personal interrogative pronoun, and by employing
kay- ‘to go away’ and sura- ‘to give birth’ (< *yarn-) instead of ket- and tog-, use of
the first person plural suffix +mlz in the D/-preterite, and forms of postvocalic aorists
reconstructable as -yUr. Like Oghuz and Lena Turkic it has a velar cognate turn <
tarjrï~ tagrïof tärjri ‘god; sky’. The special position of Chuvash is underscored by
numerous individual features, e.g. the verb yër- for ‘to cry’, the auxiliary verb tu- to
derive verbs from nouns, the use of tavrän- ‘to return’ (Old Turkic tägzin-), and a base
form xâ+ for the reflexive pronouns.5 See also tables 4 and 5 on pp. 90-91.
With the form yär- of the verb i':d- ‘to send’ and the representation of Old
Turkic y- as a spirant s- (< /-) Chuvash seems to belong to the Northern Turkic area.
But one must keep in mind that the latter correspondence in Chuvash may have
another source than that in the other languages of this areal group, see Rona-Tas
(1982a). Other features of Northern Turkic are missing in Chuvash, see 4.4. Like
Eastern Turkic, Chuvash has preserved a cognate of the Old Turkic negative present
tense copula ärmäz (> Chuvash mär). Chuvash exhibits Western Turkic forms for
‘tree’ (yiväs) and for ‘twenty’ (sirëm), has preserved the verb bïraq- ‘to let etc.’, and
elides the final -G of the second syllable in tagllg\ see also table 6 on p. 92.
Chuvash shows some common features with Border Turkic languages and
Oghuz such as preservation of nasality of -j- in *sürjök > sämä ‘bone’, no overt formal
connection between the opposing forms in the marking system of perfective versus
cursive participles, the word tuta for ‘lip’, preservation of the Old Turkic genitive
suffix type with (C)V...-structure, a necessitative suffix -mAllA, and -mA pusla- to
designate the beginning of an action.6 Other features are limited to Oghuz and Chu-
vash, e.g. the word for ‘navel’ kävapa < *kö:bäk. Especially Western Oghuz is tied
5 Additionally Chuvash can be individualised by typically Bulghar Turkic features like the
sound changes *-z(-), *-d(-) > -r(-), *-s(-) > -/(-), the use of ku instead o f bo/bu as the demonstra-
tive o f proximity, the meaning ‘white’ o f *sari'g instead o f ‘yellow ’, two different possessive
suffixes o f the third person, and ordinal suffixes of its own.
6 An alternative form going back to *-A basla- connects Chuvash with many Kipchak lan-
guages.
70 C. SCHÖNIG
closer to Chuvash by the existence of the verb bann- ‘to take shelter’; ‘to lodge’,
which is cognate to Chuvashpurän- ‘to live’. The words for ‘fishing pole’ in Turkish
(olta) and Chuvash (välta) may belong in this series, too.
Chuvash had intensive linguistic interaction with Kipchak. Both groups show
the development tag > taw (Chuvash tu/tav+), use of *-A turur as a renewed present
tense (intraterminal) finite form (but not in the participles), and similar biverbal con-
structions. Furthermore the Chuvash nomen agentis in -ÄvsÄ resembles Kipchak
*-Uwcl. For further common features see 2.1. Furthermore Chuvash has experienced
strong influences from Volga Finnic and Russian. It also exhibits many Tatar loan-
words, some of which are ultimately of Middle Mongol origin. Besides direct copies
from Middle Mongol, we also find Chuvash-Mongolic correspondences going back
to early Bulghar Turkic-Mongolic language contacts presumably in Southern Siberia
from at least the second century BC until the fourth century AD .7 Perhaps some cor-
respondences between Chuvash and North East Turkic, especially with Lena Turkic,
may also go back to the Siberian phase of Bulghar Turkic, too. Thus, Lena Turkic
and Chuvash show a common set of features such as preservation of the Old Turkic
low vowel of the second syllable of olor- ‘to sit down; to sit’ and in some suffixes, and
loss of the word-initial opposition c- : y-. Furthermore, the Old Turkic verb tasi'q- ‘to
go out’ has survived in Yakut as tagïs- or Chuvash tux-, i.e. in forms much closer to
Old Turkic than the form cïq- in most of the modem Turkic languages; in Karagas
we find ta”s- ‘to break its banks (river etc.)’. Chuvash has some features in common
with the Lena-Sayan Turkic area, e.g. the quite frequent sound change *a > ï in first
syllables and the absence of *-nc/-ordinals. Chuvash and Sayan Turkic do not employ
a -K-marker for the first person plural. Another important hint indicating early Bulghar-
Siberian Turkic connections is the shape of the words for ‘stirrup’, see 4.4.2.
3.2. Lena Turkic
In Lena Turkic the gerund in -p is replaced by -An, but the suffixes of the conditional
and the third person singular of the imperative -TAr and -Tin can be connected to the
Norm Turkic forms of Old Turkic -sAr and -zUn by the sound change (*z>)*s> t (as
in Buryat). Besides +lAr we find additional plural forms like +t or +ttAr < *+t+LAr
(see Schönig 1988).
Lena Turkic exhibits an array of Border Turkic features, e.g. a cognate uyuox
of *sürjök ‘bone’ with preserved -rj- and a velar cognate tarjara of Old Turkic tärjri,
which has kept the double-meaning ‘god’ and ‘sky’; for Lena Turkic’s North East
Turkic features see 4.4.2. Together with Sayan Turkic it forms the Lena-Sayan Turkic
area and exhibits closer ties to Karagas, see 4.4.2.2. Lena Turkic in the extreme
Northeast and Turkish in the extreme Southwest of the Turkic area are connected by
use of the verb *tön(ün)- ‘to turn around, return, go home’ and survival of the reflex-
ive pronoun käntü (known from Old Turkic) as kini, which is used as the personal
7 Non-Bulghar Turkic -d(-) : Chuvash, Mongolie -r(-), -z(-) : -/‫(־‬-), y- : *j- - c-, ti- : ci, or
lack o f a pronominal -n in the nominative o f the first and second person singular pronouns.
pronoun of the third person. Besides, Lena Turkic displays a wide variety of individ-
ual lexical, phonetic and morphologic specialities.89
There seem to be some special connections between Lena Turkic and Kip-
chakoid South Siberian Turkic, especially Khakas. It is still unclear whether they go
back to older connections, or came into existence only during the formation of the
North East Turkic area. Lena Turkic has some special features in common with Kir-
ghiz(-Kipchak), e.g. extensive vowel harmony, the form of the participium nondum
facti (see 4.4.1) or the additional meaning ‘castrate’ of at ‘horse’ (as in Fu-yii and the
Altay dialect). For the connections between Lena Turkic and Kipchak in general see 2.1.
Lena Turkic belongs to the Northern Turkic area. It exhibits Eastern Turkic
features as well as Western Turkic ones. As to the numerals with intervocalic conso-
nants, Lena Turkic has its distinctive profile, see 4.3. Lena Turkic is part of a young
North Eastern Siberian interactive area including mainly Tungusic and Mongolie
languages, from which it has received some strong non-Turkic impulses, see Schönig
(1988, 1990 and 1993a). See also tables 4-7 on pp. 90-93.
3.3. Khalaj
Khalaj, as the only Turkic language that has consistently preserved word-initial Ancient
Turkic h-, shows a third person imperative suffix -tA and has no syntactically free
gerund in -p in syntactically free use. But its plural and conditional suffixes are Norm
Turkic. Individual features are, e.g., a non-Norm Turkic locative in +ca, a genitive in
+(U)1,j and a negative present tense copula da:g. Khalaj had long-lasting interaction
with Oghuz. A common Modern Uighur-Khalaj set of features consists of the use of
the agent noun -GUcI, of the necessitative suffix -GUIUK, of -GAII/-GIII-forms as the
connecting element in constructions with basla- to express ‘to begin to x’, and gener-
ally doubled intervocalic consonants in numerals. Eastern Turkic features of Khalaj
such as the preservation of the final G-sounds in taglug may belong here, too. On
the other hand, the numeral for ‘twenty’ exhibits the Western Turkic form *yigirmi.
For the numerous Border Turkic features of Khalaj see 2 and 2.2-2.2.2 and
tables 4 and 5. Khalaj shares other features with Oghuz, especially Western Oghuz,
with which it developed in close areal contact. Both branches have preserved the
negative aorist in -mAz as in Old Turkic instead of today’s more common -m Asf and
present tense forms going back to *-A yon-. Especially, Western Oghuz-Khalaj inter-
action is indicated by the systematic development *b- > v- or 0 - in the words var-
‘to arrive’, var ‘exists’, ver- ‘to give’ and ol- ‘to be; to become’, 10 survival of attribu
8 It shows, e.g., köt- for ‘to fly’, a deviant word for ‘mouth’ (see Schönig 1988), survival
o f Old Turkic q'irqïn ‘(slave-)girl’ in the plural form kïrgïttar to ki':s ‘girl’, and a comitative suffix
+ll:n going back to Old Turkic +IXGX11 (see Schönig 1991).
9 Chuvash has preserved this form only in the negative present tense copula mar < ärmäz.
The Lena Turkic -ßAf-forms could also have developed after the sound change -z > -s.
10 In Saryg Yugur and Salar we sporadically find «/-forms o f *bol-, but only as one o f a
few alternatives like bol- or vol-, see Tenisev (1976a and 1976b). In Salar the sound change b- > v-
also appears with the same set of monosyllabic verbs, see Dwyer (1998, p. 61).
72 C. SCHÖNIG
tively used gan-forms going back to the Old Turkic pronoun qa:no ‘which’, and
regular use of possessive marked +D//Gparticiples as verbal cores of relative clauses.
Some archaic features are preserved mainly in Khalaj and Turkish and a few other
units, see 2.2.2. For ‘hair’ Khalaj has a Southern Turkic sac-form. For elements ex-
pressing ‘with’ see 2.1. See also tables 4—7 on pp. 90-93.
4. The interactive areas
Besides genetically related groups we find quite a number of interactive areas where
Turkic languages of different genetic strings exchanged linguistic features with each
other and with non-Turkic languages. As mentioned before, Chuvash had intensive
interaction with Volga Kipchak, in which Volga Finnic languages also were involved.
Khalaj copied a lot of material from Oghuz (mainly Western Oghuz) and Persian.
Other interactive areas contain far more Turkic languages of various genetic branches.
4.1. -GAn-Turkic
The Central Turkic sub-branches Kipchak and Southeast Turkic together with South
Siberian Turkic form the -GAn-Turkic area; Eastern Oghuz Turkmen and Salar have
a transitory position and exhibit at least some features of this grouping. Its main
feature is the use of the participle in -GAn instead of -mis in finite and attributive
positions, see also table 5 on p. 91. As the negative counterpart of the gerund in
-p and the vocalic gerund, the form *-mAy(ln) is in use. Kipchak(oid) and South East
Turkic languages use *-A tur- to mark intraterminality on participles (*-GAn :
*-A turgan) and present tense forms (-(V)r : *-A turur), whereas Turkmen applies
*-A yon- in the same way, see 2.1. To the East of this area a second wave of renewal
of intraterminal (present tense) forms by means of *-А/p yata tur- has taken place.
Some biverbal constructions expressing actionality follow identical patterns in many
-GAn-Turkic languages. Furthermore -GAn-Turkic (like Chuvash) exhibits cognates
of the Old Turkic verb i‫־‬:d- ‘to send’ or (in Kipchak, South East Turkic and Turkmen)
of its derivation ï:du ber- ‘id.’ (see table 7 on p. 93) and has palatal cognates of the
word tdrjri ‘god; heaven’.
4.1.1. Non-Oghuz -GAn-Turkic
Some features tie Kipchak and South East Turkic closer together and distinguish
them from Oghuz. One of these features is the biverbal form -A/-p al(-ma)- to
express (im-)possibility of performing an action, which also appears in Salar, South
Siberian Turkic and perhaps in Chuvash. 11 All case suffixes show a stable suffix-
11 For Chuvash -ay- < *-A al- see Benzing (1959b, p. 721) and Levitskaja (1976, pp. 54—
55). Lena Turkic has a suffix -(A:)yA-.
initial consonant by analogisation, i.e. suffixes of the structural types +(C)V... and
+(V)C... were unified by analogisation with the third type +CV... Thus we find
a genitive suffix +nI1j (Kirghiz +nln, Salar +niyi) instead of the Old Turkic type
+(n)I17, see also table 5 on p. 91.
4.2. Central Asian Turkic
The Kipchak and South East Turkic languages of Turkestan and the neighbouring
areas form the Central Asian Turkic area, which can be characterised by the para-
digms of politeness. Thus, e.g., in Uzbek, Kazakh, Kirghiz and Modem Uighur sen
designates a second person singular of the same or lower rank than the speaker, senler
is used as the plural form; siz designates a second person singular polite senler, and
has the plural sizler, see table 6 on p. 92. Some languages show comparable forms in
the paradigms of the personal and the possessive suffixes. See also Schönig (1987).
4.3. Western Turkic and Eastern Turkic
Another isogloss divides the Turkic languages into an Eastern and a Western part
(see table 6 ). It runs through the Kipchak group and the Central Asian Turkic area.
North East Turkic, Uzbek and South East Turkic mostly behave like Eastern Turkic;
Kirghiz-Kipchak, Siberian Tatar, Kazakh and Karakalpak often have a transitional
status. Interestingly, Lena Turkic often behaves like Western Turkic. It has preserved
the verb bïraq- ‘to let’ like Western Oghuz, Chuvash and Western Kipchak languages.
Western Turkic and Lena Turkic still use the personal plural marker +z. In Eastern
Turkic +z has entered into competition with the plural +lAr in the second person.
This led in Central Asian Turkic to the paradigms of politeness, whereas in South
Siberian Turkic and Salar we find total replacement of +lAr by +z. The syllable-final
G-sounds of Old Turkic taglig are both preserved in Yenisey and Sayan Turkic, Saryg
Yugur, Fu-yii and South East Turkic - but not in Lena Turkic or Salar, whereas in the
West, Khalaj fits in with Eastern Turkic, see table 4 on p. 90. Sometimes Eastern
Turkic and Chuvash are parallel to each other. A common archaic feature of Eastern
Turkic (but not Lena Turkic) and Chuvash is the survival of the negative present tense
copula ärmäz. From Western Turkic, Oghuz together with some Kipchak languages
has forms going back to tägül; Khalaj has da:g. In other cases Lena Turkic clearly be-
longs to the Eastern Turkic area, e.g. with its labial form biit- of the verb ‘to come to
an end’ as in Old Turkic, North East Turkic and Central Asian Turkic, instead of bit-
like in Western Turkic (including Salar). The Ancient Turkic word pair yïltïz : yultuz
(‘root’ : ‘star’) is preserved in Lena Turkic, some South Siberian Turkic languages,
Saryg Yugur and South East Turkic; 12 in most of the other modem units the word for
12 Uzbek has ildiz : yulduz, Modern Uigur yiltiz : zultuz, Lena Turkic sills : sulus. In Sayan
Turkic we find, instead o f forms with irregular sound change y- > s-, like Karagas sildi's (see Ras-
sadin 1971, pp. 229-230) meaning both, ‘star’ and ‘root’; in Tuvan the meaning ‘root’ seems to be
74 C. SCHÖNIG
‘root’ has vanished, see table 6 on p. 92. For ‘to return’ South Siberian Turkic, Saryg
Yugur, Salar and Modem Uighur sometimes use cognates of Old Turkic y an-, see
table 5 on p. 91. The privative suffix +slz is missing in the Eastern Turkic border lan-
guages, North East Turkic, Saryg Yugur and even Salar. In Eastern Turkic languages
the passive suffix -(1)1- can be added to verbs ending in l, while modem Western
Turkic regularly applies -(l)n- in such cases; Lena Turkic exhibits a system of its own.
Eastern Turkic has retained many features of Ancient Turkic, but also exhibits
some renewals, e.g. the personal plural sign +z, see above. Furthermore, mainly in
Central Asian Turkic and South Siberian Turkic (but not in Saryg Yugur) the form
qac ‘how much, how many’ is replaced by qanca. Sayan Turkic Tuvan does not have
qac, whereas in Karagas and Lena Turkic we find both forms. Many Eastern Turkic
languages (except Lena-Sayan Turkic) show renewed present tense forms containing
the segment -A/-p yat-.
The words for ‘tree’ show archaic forms going back to *higac > yigac only in
the “Far East” (for Khalaj hagac see Doerfer 1995). The form *(h)agac exists in
Oghuz, Kipchak and Chuvash, but also in Kipchakoid South Siberian Turkic, Fu-yii,
and Salar. Kirghiz behaves differently from Altay Turkic by having jigac (resem-
bling, e.g., Lobnor yigac or Saryg Yugur yigas). The Sayan Turkic forms (Tuvan iyas
(with nasal y), Karagas ties) also point to the Eastern protoform. Many South East
Turkic languages have a form with initial y- followed by a low vowel, like Modem
Uighur yagac, Uzbek yàgàc. The distribution of the auxiliary verbs for denominal
verb derivation leads to a comparable pattern that underscores the division of North
East Turkic into Lena-Sayanic and Kipchakoid South Siberian (see 4.4.2.3 and table
6 on p. 92). Oghuz, Kipchak, Kipchakoid South Siberian Turkic and Salar are con-
nected by the dominance of the verb et-. Mainly in Oghuz - but with, e.g., Tatar evi-
dence, too - the verb ädlä- has survived in the West and is frequently used in Azeri;
Chuvash has its own form tu-. In South East Turkic we also find the old auxiliary qil-,
which is now restricted mainly to contexts of dignity in Oghuz and Kipchak, and still
widely in use. It has kept this role in Lena-Sayan Turkic, which does not have et-;
additionally we find *qi'n- in Sayan Turkic Karagas and Lena Turkic.
The distribution of the forms of some Turkic numerals is connected with the
Western Turkic : Eastern Turkic division. The cognates of Old Turkic bi'rj ‘thousand’
have front vowels in Western Turkic; Turkmen miitj additionally is labialised. Eastern
Turkic normally has a back vowel as in mïtj (Nogay, Karakalpak, Kazakh, Baraba,
Chulym Turkic) and murj (North East Turkic except Chulym Turkic). The South East
Turkic, Salar and Saryg Yugur form mirj can also be interpreted as *bi'rj. But as shown
by Kirghiz mirj one also encounters palatal forms in the East. The numeral ‘twenty’
- if not replaced - has word-final non-high vowel in many Eastern Turkic languages
(yïgïrma ~ yigirmä ~ yägirmä ~ yägärbä);13 only Khakas cibïrgï and Tuvan cä. rbi
lost. Furthermore we find the old word for ‘root’ in Saryg Yugur yilti's, yilti's, Ktiärik yïldïs and
Kyzyl si'ltti (see Sevortjan 1974, I, p. 350), Soyot yilti's (Radloff 1905, III, p. 488), Sagay cilti's
(Radloff 111, p. 2087).
13 Palatal forms with low word-final vow els exist in South East Turkic, Baraba and Altay
Turkic. A back vowel form *yïgïrma appears in Central Kipchak, Kirghiz, Uzbek dialects and
show high word-final vowels (< *yägirmi; a similar form is reflected in the Babur-
name). The shortened form Jibir in Fu-yii can be traced back to a metathesised form
of the Khakas type. In Radloff s Karaim of Troki a form igirmä exists, which may be
an independently developed metathesised form of *yägirmi (> ägirmi in Radloff s
Karaim of Lutsk), as Crimean Karaim yägrimi and Turkmen yigrimi are. Forms with
high final vowel dominate in Khalaj, Oghuz, Crimean Tatar, Karaim, Volga-Ural
Kipchak and Salar; the Chuvash form belongs here, too. The various Saryg Yugur,
Salar and Modern Uighur sources exhibit forms with low as well as with high final
vowel (Tenisev 1976a and b; Sevortjan 1989, IV, p. 201), see table 6 on p. 92. In
*sii:rbe Lena Turkic shows an enigmatic labial vowel in the first syllable; its final
low vowel may be secondary, if the form is not metathesised. Of the numerals with
intervocalic consonants, mainly the numeral ‘fifty’ observes the Western Turkic :
Eastern Turkic borderline. It has a strong form with doubled intervocalic consonant,
*ällig, mainly in Western Turkic (Oghuz, Western-Central Kipchak), while the weak
form *älig with a single intervocalic consonant is dominant in Eastern Turkic (Täläut,
Yenisey Turkic, Chulym Turkic and Kirghiz). Other languages have developed their
numerals by means of internal analogisation. Khalaj and South East Turkic have gen-
eralised the strong forms, whereas Sayan Turkic only has weak ones. In Lena Turkic
we find weak forms for ‘eight’ and ‘nine’ and perhaps ‘thirty’ (with closed final syl-
lables) and strong forms for ‘two’ and ‘seven’ (with open final syllables). Chuvash
has two series —one weak and one strong.
4.4. Northern Turkic
The Northern Turkic area is characterised by the tendency to rearrange the system of
oppositions between (word- and sometimes syllable-)initial y-, the affricates and the
sibilants. In most of Kipchak, Modem Uighur, North East Turkic and Chuvash, Old
Turkic y- corresponds to a sibilant or an affricate. According to Radloff’s Kirghiz
data and the development of loanwords from Middle Mongol, in non-Bulghar Turk-
ic constituents this must be derived from quite recent developments, see Rdna-Tas
(1982a). Due to analogical processes we find -G- instead of -rj- in the possessive
suffixes of the second (plural) persons in Chuvash, Volga-Ural-Caucasus Kipchak,
Lena Turkic and Altay Turkic. There seems to be a common tendency of replacement
of y by G in the whole Northern Turkic area, see Schönig (1991). A feature of North-
ern Turkic (but not of Chuvash) is the form of the first person plural suffix +(I)BIz,
whereas Southern units and Old Turkic have +(I)mIz, see table 7 on p. 93. In the
same area and in Khalaj the accusative of the third person possessive suffix is +(s)In
as in Old Turkic.
Old Turkic sac ‘hair’ appears in Northern Turkic as cac - only Lena Turkic
with as exhibits a Southern .vac-form. In Volga-Ural Kipchak and in Chuvash cac is
palatalised to cäc. Salar, Oghuz, Khalaj and Standard Uzbek have Southern sac-
Chulym Turkic Küärik (but R adloffs Küärik has yigirbä) and in Caucasus Kipchak. Mixed front-
back forms exist in Kazakh, Karakalpak, Caucasus Kipchak and R ad loffs Taranchi.
76 C . S C H Ö N IG
forms. Saryg Yugur and Modem Uighur with cac- and sac-forms take an intermedi-
ary position between the North and the South, see also table 7. Karaim has sac-, cac-
and câ'c-forms. Perhaps the sac > cac-shift became active in the North after Uzbek
had left the place, but parts of South East Turkic and Saryg Yugur had fallen under
Northern influence. Sayan Turkic Karagas shows cd”s, which in Castrén’s material
still appears with a back vowel. Tuvan does not exhibit a cognate, but has bas dügü.u
The deviant behaviour of Lena Turkic (as in the case of some Eastern Turkic features)
together with the Karagas form may point to the fact that the development sac > cac
appeared relatively late, when the Lena Turks had loosened their contacts with the
remainder of Northern Turkic and the sound change c- > s- was no longer active in
Sayan Turkic. 1415
On the lexical level we may assume *qodan to be the Northern Norm Turkic
word for ‘hare’ (but Yakut has kuobax), while the South prefers forms of tabisgan.
Preservation of Old Turkic 1:d- ‘to send’ in North East Turkic, Chuvash, Karaim,
Kazakh and Kirghiz may be considered as a Northern Border Turkic feature.16 Analo-
gised forms *-mAp or *-mAylp as negative forms of the gerund in -p are very rare in
Northern Turkic, and may be called a typically Southern Turkic feature.
4.4.1. From Northern Turkic to North East Turkic
Within Northern Turkic we find an area of prominent progressive nasalisation of

suffix-initial {D, L}-morphophonemes by stem-final nasals. Progressive nasalisation
with suffixes consisting of open syllables only appears sporadically outside Lena
Turkic, whereas the ablative suffix +DAn (+Dln) ending with a nasal consonant
shows many cases of nasalisation - but not in Lena Turkic. With the plural suffix
+LAr we find nasalisation in some Kipchak languages, in Saryg Yugur and South
Siberian Turkic except Southern Altay Turkic, which behaves like Kirghiz.
Another Northern area of special phonotactic rule sets contains Bashkir, Nogay,
Karakalpak, Kazakh, Kirghiz and North East Turkic, see table 7 on p. 93. Here, suf-
fixes with initial {D}-, {L}- and {N}-morphophonemes show different initial conso-
nantal allophones in accordance to the rules of plosive dissimilation after word-final
consonants. 17 The languages differ considerably with respect to the word-final sounds
which precede this dissimilation. In North East Turkic the internal isoglosses drawn
by the rule sets do not follow patterns depicted by other features. Lena Turkic shows
some structural similarities with the Kipchak languages. The Kirghiz-Kipchak units,
14 Due to a wrong link in the database, Tuvan ca ”s ‘zenskaja kosa’ was erroneously cited in
this context in Schönig (1998a, p. 132).

15 One may argue that in the case o f cac the sound change c > s is incomplete, because it
would have produced a form with s in initial and final position of the same syllable. But there are
other syllables of the structure sVs.
16 But the latter three languages also show forms going back to the combination *ï:du ber-,
as in Kipchak and South East Turkic.

17 Plosive dissimilation means that a sequence o f two non-plosives has to be dissolved into
a sequence of one non-plosive and one plosive, see Schönig (1993b).

Kirghiz and Altay Turkic, constitute a group of their own. Bashkir plays a special
role by showing obstruent instead of plosive dissimilation, and by having unified the
suffix-initial {D}- and {N}-phonemes. All the languages concerned show plosive as-
similation plus voice assimilation after voiceless final consonants. In a sub-area of this
“phonotactic area” - in Nogay, Karakalpak, Kazakh, Kirghiz and Northeast Turkic -
suffixes with initial {M} are affected by plosive dissimilation, too. Of the languages, in
which initial {M} has become recategorised completely as {В } (i.e. not Nogay, Kara-
kalpak, Kazakh or Fu-yii), only Kirghiz and Southern Altay Turkic have not kept vari-
ants with initial m after word-final nasal - the same Kirghiz-Kipchak units for which
plosive dissimilation instead of nasalisation of the {L}-morphophoneme is attested.
The verb for ‘to cry’ shows a form iyla- < *igla- in Kirghiz-Kipchak (aftd
Bashkir), which matches South Siberian Turkic forms like Yenisey Turkic ilga- and
Sayan Turkic igla-. Fu-yii with filgï- ~ yilgï- has a (non-Kirghiz-)Kipchak-South East
Turkic form with Yenisey Turkic metathesis. Saryg Yugur shows intermediary forms
yigla- ~ igla- between Kipchak-South East Turkic and South Siberian Turkic. Lena
Turkic with i'ta:- and Chuvash with yèr- stand apart; see also table 7 on p. 93.
Within the {M} > {В }-area we find a smaller area designated by the category
of what Benzing (1959a) has called participium nondum facti. This area includes
Kirghiz, Siberian Tatar varieties, Fu-yii and North East Turkic (except Karagas). The
category is expressed by suffixes which may originate from a common root: Kirghiz
and Lena Turkic have -A elek and -A ilik, South Siberian Turkic -GAIAK and Fu-yii
-GAlAs. That Kirghiz and Lena Turkic are bound closer together may be due to the
fact that they are located at the borders of the participial area and have kept ancient
forms of this etymologically enigmatic suffix. That Altay Turkic behaves differently
from Kirghiz indicates that the final establishment of the formal marker category
happened after the dissolution of Kirghiz-Kipchak, which still may have been a unit
when the mondum facti category itself developed. The Fu-yii were still part of South
Siberian Turkic when the category was established. But because of their displace-
ment to Manchuria in the 18th century, the ultimate form of the suffix developed
slightly differently.
4.4.2. North East Turkic
South Siberian Turkic together with Lena Turkic formed another interactive area not
too long ago - North East Turkic. More or less exclusive features of this area are the
nomen actoris in +A(:)ccI, and the total replacement of the privative suffix *+slz by
constructions of the type noun ( + poss. suf.) + *yoq, which can also be found (in ad-
dition to *+slz) in other Turkic languages. As in Fu-yii and Saryg Yugur a replace-
ment of formally non-transparent Old Turkic has taken place in analogy to Old Turkic
decimal numerals sek(k)iz on ‘eighty’ and toq(q)uz on ‘ninety’ . 18 The replacement of
18 As in the case of phonotactic rule sets, units belonging to different sub-branches show
the same development. Karagas and Saryg Yugur start with ‘twenty’, Tuvan with ‘thirty’, Fu-yii,
Altay Turkic and Lena Turkic with ‘forty’. Only the Yenisey-Chulym (Siberian azag-JTurkic-group
starts as a whole with ‘sixty’. Salar has an alternative additive system starting with ‘sixty’ elli on -
Ada Orient. Hung. 52, 1999

78 C. SCHÖNIG
tog- by torn- belongs here, too. All these features have been inspired by neighbour-
ing or substrate languages, in which the tens are derived from the ones in a transpar-
ent way . 19 As to the reflexive pronouns, Lena Turkic uses bäyä, a borrowing from
Mongolie, where it originally meant ‘shape, body’ - like Old Turkic *bod, which is
used as the base of the reflexive pronouns in South Siberian Turkic. But while South
Siberian Turkic has only copied the Mongolie model, Lena Turkic has made a direct
copy. In addition to the form kindik of the word for ‘navel’ most of the South Sibe-
rian Turkic languages show a short form kin\ Lena Turkic and Karagas only have
ki(:)n. The North East Turkic protoform of the numeral for ‘thousand’ can be recon-
structed as murj, see also 4A 2.2 and tables 5-7 on pp. 91-93.
North East Turkic and Chuvash share some special Border Turkic features. The
common tendency towards devoicing of word-initial and word-final obstruents and
voicing of intervocalic ones may be of independent origin. But the shape of ‘stirrup’
can be interpreted as a hint of old direct connections with Bulghar Turkic (see Röna-
Tas 1982b). Whereas most Turkic languages show forms pointing back to a proto-
form with round initial vowel, North East Turkic, Saryg Yugur, Baraba and Chuvash
exhibit forms with unround initial vowel. Only Altay Turkic has a round vowel and
does not behave like the rest of North East Turkic.
Altay Turkic often behaves differently from North East Turkic. Beside the
form of ‘stirrup’ and the Central Turkic feature *-d(-) > -y(-), Altay Turkic has kept
the derivational suffix *+IIK, shows no consistent preservation of the first person
plural ending +mlz in the D/-preterite, exhibits no replacement of the verb *tog- ‘to
give birth; to be bom’, and has not kept postvocalic -yUr-aorists or the formal corre-
spondence between the postvocalic forms of the vocalic gerund -yU and the aorist, as
most of the other North East Turkic languages have.20 In some cases at least some
varieties of Shor and Khakas share this behaviour, see table 4 on p. 90.
4.4.2.1. South Siberian Turkic

The Border Turkic languages, Yenisey Turkic (including Fu-yii) and Sayan Turkic
together with the transitory Chulym Turkic and Altay Turkic form the South Siberian
Turkic area. It has only a few features common to every single language within it.
Thus it uses, e.g., cognates of Old Turkic bod to derive reflexive pronouns, and has
altmis, see Tenisev (1976b, p. 121). In Karachay-Balkar we find a vingesimal system inspired by
neighbouring Caucasian languages (see Pritsak 1959, p. JN 220).
19 That the formally incoherent Old Turkic system o f ones and tens is very old is proved by
the fact that Saryg Yugur, which shows a very strong tendency to decimalisation, uses in its archaic
counting system iicon for ‘thirty’, but pes oti's for ‘twenty-five’. The privative *yog-constructions
could be inspired by M ongolie constructions with *iigei ( ‘does not exist’ = *yoq). The nomen ac-
toris in -A(:)ccl goes back to the Mongolie nomen imperfecti (class.mon. -Ga, see Poppe 1955,
p. 273) in connection with the Turkic-Mongolic suffix +cl “to form nouns o f vocation” (Poppe
1954, p. 45).
2 0 The Old Turkic connection between the suffixes o f the vocalic gerund and the aorist
C-V(r); V-yU(r) is preserved in Lena Turkic and Yenisey Turkic in ...CA(r); ...I:(r) or ...i(r). For
Sayan Turkic see Schönig (1989). Altay Turkic has postvocalic -r-aorist and -y-gerund like a (Kir-
ghiz-)Kipchak language.
T H E IN T E R N A L D IV IS IO N O F M O D E R N T U R K IC 79
replaced the personal plural marker +z by +lAr in the second persons, see also 4.2.
For *kin(dik) ‘navel’ see 4.4.2. Another common feature is the Samoyed and Yenisey
(ket) substrate already mentioned by Castrén (1857); see also Menges (1955-56) and
Janhunen (1989).
Together with non-Norm Turkic Chuvash and Khalaj most South Siberian
Turkic languages (including Fu-yii and Saryg Yugur) have forms of the personal
interrogative pronoun which are reconstructable as *käm. But it is impossible to
decide whether Lena Turkic belongs to the *käm- or to the *kim-group. Sayan Turkic
gains a special profile by showing forms such as Tuvan *qïm or Karagas *qum, see
table 4 on p. 90.
The Sayan Turkic languages Tuvan and Karagas show many differences in de-
tail. Additionally they exhibit different sets of copies from Mongolie. Besides Middle
Mongol and Oirat layers Tuvan exhibits traces of strong influence from Khalkha,
whereas Karagas had interaction with Buryat. In Yenisey Turkic and Sayan Turkic,
combinations of the auxiliary verb ltd- ‘to send’ and the gerund in -p have become
suffixes. Khakas and Fu-yii exhibit metathesised forms for the word for ‘twenty’, and
have ib for ‘house’. A special feature of Yenisey Turkic are -JA^-participles to express
habituality, continuation etc. Saryg Yugur has preserved the Old Turkic counting sys-
tern. Fu-yii exhibits shortened forms bit ~ bul of the postposition bi(r)Iä(n) ‘with’
and the numeral jibir ‘twenty’.
4.4.2.2. Lena-Sayan Turkic

Lena-Sayan Turkic is mainly characterised by the absence of some widespread Cen-
tral Turkic features, e.g. forms going back to modem Norm Turkic ci'q- ‘to go out’,
cognates of the impersonal interrogative pronoun ne(md), biverbs with basla- to ex-
press ‘to begin to’, and a single renewed present tense suffix; Lena-Sayan Turkic does
not use et- for denominal verb derivation. It shares Border Turkic features with Chu-
vash, like the absence of -ncY-ordinals,21 see table 7 on p. 93, and the quite frequent
representation of Old Turkic a as Гin first syllables. Nasality of Old Turkic n is partly
preserved as in Khalaj and Oghuz. Lena-Sayan Turkic shows suffixes of the inclusive
and exclusive first plural persons of the imperative paradigm with the structure first
person inclusive = first person exclusive + second person plural}2 The cognates of
Old Turkic tag ‘mountain’ have (at least additionally) the meaning ‘mountain forest’.
Of Sayan Turkic, Karagas is more closely connected to Lena Turkic by fea-
tures only attestable in these two languages. Thus the numeral for ‘thousand’ has
been replaced by a Russian loanword,23 a verb *qïn- is used to derive verbs from
nouns, and for navel only the short form ki(:)n exists. Both units have a partitive
case, expressed by suffixes formally identical with the Old Turkic locative-ablative
suffix +DA. Furthermore, we may assume that an ordinal suffix +nc, as in Old Turkic,
21 According to Thomsen (1959, p. 566), Saryg Yugur had an ordinal suffix +nc, too. But
Tenisev (1976a, p. 74) only mentions +(l)ndzx or +(I)ndzïlïqc.

2 2 For the situation in Turkmen see Clark (1998).
23 We find Dolgan ti:касса, Yakut tihi.nca and Karagas tcsicci. The original form mug can
be reconstructed for Yakut by means o f instances such as mug älbäx ‘very much’.
80 C. SCHÖNIG
has survived here (but see Stachowski (1994)). As the only North East Turkic units
they show qac- besides the qanca-imm of the interrogative ‘how much, how many’
and use of the word iiriitj ‘white’. See also 2.2.2.
4.4.2.3. Kipchakoid South Siberian Turkic

Altay, Chulym and Yenisey Turkic together with Fu-yii exhibit some common fea-
tures different from Sayan Turkic, but also known from Kipchak. This division of
South Siberian Turkic into a Kipchakoid South Siberian Turkic and a Sayan Turkic
branch can be seen, e.g., in the use of К as a personal marker of the first person plu-
ral in Kipchakoid South Siberian Turkic mainly in the imperative paradigm; this
feature closely connects it mainly to the modem Kipchak languages and Lena Turkic.24
In contrast to Lena-Sayan Turkic (see 4.4.2.2), Kipchakoid South Siberian Turkic
shows suffixes of the inclusive and exclusive first plural persons of the imperative
paradigm with the structure first person inclusive = first person exclusive + plural.
Sayan Turkic prefers postpositional /л/à-forms to express ‘with’, whereas great parts
of Kipchakoid South Siberian Turkic show a strong tendency to use bilä(n) encliti-
cally. Furthermore, Kipchakoid South Siberian Turkic, like Central Turkic, uses et-
for denominal verb derivation and shows strategies of marking intraterminality in the
participai systems like Kipchak and South East Turkic.
By internal analogisation of suffix-final nasal consonants of the genitive +nlq
and ablative +DAn the Kipchakoid branch is more closely connected to Kirghiz,
which at the same time is separated by this feature from the remaining Kipchak lan-
guages. The final nasals became n in Kirghiz and rj in Altay and Yenisey Turkic and
in Chulym Turkic Küärik. But whereas Kirghiz has velar forms of the numeral for
‘twenty’ and exhibits at least some forms of the paradigms of politeness, Altay Turkic
has palatal forms of the numeral and only +LAr-marked forms of the second person
plural. These features - like the forms of the participium nondum facti, see 4.4.1 -
reveal the process of dissolution and reformation of Turkic genetic or areal sub-
groups: an older Kirghiz-Kipchak (Kirghiz and Altay Turkic)/Yenisey Turkic unit
split into Kirghiz and Kipchakoid South Siberian Turkic (Altay and Yenisey Turkic).
I call this group only Kipchakoid because important features of modern Kip-
chak are missing. Thus the preservation of intervocalic -t- has been given up, the fea-
ture tag > taw has not been developed, and the numerals with intervocalic conso-
nants have stabilised in forms more or less different from the modem Kipchak type.
5. Diachronic aspects
Because the data used in the proposed model all belong to Modem Turkic, mainly
the synchronic situation is reflected. But of course the various constellations through
which we progressed contain a number of diachronic implications. I shall now try
24 The -K‫־‬forms in Azeri have emerged by internal analogisations or by areal contacts with
Kipchak and South East Turkic tribes.
to summarise some major lines of possible development of the Turkic languages

which may be derived from the proposed model, and try to connect them with his-
torical data.
The existence of non-Norm Turkic or Border Turkic features in a language
does not automatically indicate that it left Proto-Turkic in earliest times. The amount
of non-Central Turkic features is only a measure of an abstract distance from Central
Turkic. Such features may have developed quite late, e.g., in relative isolation from
other Turkic languages. Archaic features indicate that a language had separated from
remaining Turkic languages in the Ancient Turkic period. But the existence of nu-
merous archaic features in a language may only indicate a more thorough separation
of its forerunner from remaining Ancient Turkic in a relatively late period. It does
not necessarily mean that the language in question separated earlier than another one
which exhibits a smaller number of such features. If we can make use of diachronic
data, we may be able to determine the terminus post quem non for the development
of a feature used in the model. However in many cases we only can guess by means
of internal reconstructions or comparative investigations at what time such features
came into existence.
That the precursors of the non-Norm Turkic languages separated quite early
from the rest of Turkic is not only indicated by their amount of non-Norm Turkic and
Border Turkic features, which in many cases consist of preservation of Ancient Turkic
features. In the framework of the Northern Turkic : Southern Turkic and Western
Turkic : Eastern Turkic divisions they additionally show features of that area to which
they do not belong geographically - or they exhibit isolated, individual forms. This
behaviour is paralleled by Salar in the “Far East” of Turcia, which due to its pre-
sumably Oghuz origin still displays some Western Turkic features. These features had
been developed in Western Turkic before the exodus of the Salars to Qinghai in the
Chinggisid period (see below). Many of the features in question also appear with ar-
chaic or special forms in Oghuz languages. This underlines the relatively early sepa-
ration of Oghuz from the remainder of Central Turkic (see below).25 The remaining
“normal” Border Turkic languages usually confirm the geographical distribution of
linguistic features. The distribution of the features in question reveals some main
lines within the internal division of Modem Turkic, see table 8 on p. 94.
The amount and the distribution of non-Norm Turkic and Border Turkic fea-
tures in Chuvash point out that its Bulghar Turkic ancestral group separated first from
the remaining Proto-Turkic ones, presumably as early as in the Hsiung-nu period in
the first centuries BC. In those times, the Bulghar Turks lived in Southern Siberia,
where they had contacts with speakers of Samoyedic and Mongolie, which is attested
by a whole set of mutual copies, see e.g. Rona-Tas (1988 and 1991). They left this
area in the fourth century to move westwards into the Volga region.
25 Such features are, e.g., preservation o f final -G-sounds in taglig, (il-)labiality o f the vowel
in bit- - biit- ‘to come to an end, fulfil’, preservation of the plural sign +z in the second person, the
forms o f the words for ‘lip’, ‘hair’, ‘hand’, ‘to cry’, ‘to go out’ and ‘to return’, o f the negative pre-
sent tense copula, o f the reflexive pronouns, o f the ordinal suffixes and the genitive suffixes, and
the verbs used for denominal verb derivation.

82 C. SCHÖNIG
Lena Turkic and Khalaj are much closer to Norm Turkic than Chuvash. Some
of their non-Norm Turkic features may have developed relatively late, but others in-
dicate an early separation from Norm Turkic. The ancestors of the Lena Turks may
be identical with the Üc Quriqan in the vicinity of Lake Baykal. But one should be
very careful to correlate historical groups, the language of which is totally unknown,
with recent groups. If at least some of the common features of modem Chuvash and
Lena Turkic go back to common roots, we may assume that ancestral groups of Lena
Turkic were independent enough from the remaining Proto-Turkic group to maintain
special relations with the Bulghar Turks. If so, this must have happened before the
Bulghar Turks left Southern Siberia in the fourth century.
The forerunners of the modem Border Turkic branches azaq-Turkic and Sayan
(ar/a<7‫)־‬Turkic also show some special features in common with Chuvash and Lena
Turkic. It is possible (but it does not seem very likely to me) that the ancestors of the
Lena Turks and even the Bulghar Turks originally spoke a language of the azaq-
Turkic type, see Räsänen (1949, p. 29); yet unlike other members of this group and
the forerunner of Sayan Turkic, the precursors of Lena Turkic and Bulghar Turkic
did not remain in very close contact with the forerunner of Central Turkic. But com-
mon features of modem Lena, Sayan and azaq-Turkic do not necessarily indicate that
their ancestors were more closely related or in direct contact. Some of these features
are archaic features which were kept by non-Norm Turkic and Border Turkic lan-
guages, but changed in Central Turkic. Other features may simply point to language
contacts between those parts of Norm Turkic whose speakers lived close to the early
Bulghar Turkic habitat in Southern Siberia. It seems that there were connections be-
tween the precursors of Sayan Turkic, azaq-Turkic Saryg Yugur and Old Uighur; in
the case of the Saryg Yugurs this is attestable by historical data.
The special common features of Sayan Turkic and Lena Turkic bear witness
to at least temporarily close areal contacts, but they may even indicate a genetically
closer connection between these branches of Turkic. The deviant representations of
some widespread Central Turkic and -GAn-Turkic features in this area together with
the assumed date of the Lena Turkic exodus to the North (see below) indicates
a phase of relative independence from the main body of Turkic from perhaps the
Late Ancient Turkic period on. This phase probably ended in the Chinggisid period.
Then - in the framework of the -GA«-Turkic area - Sayan Turkic, which was spoken
in areas closer to Central Turkic languages, began to establish more intensive con-
tacts with these groups than with the ancestral groups of Lena Turkic. The close con-
nection between Lena Turkic and Karagas indicates that the ancestral groups of the
Sayan Turkic groups, which were responsible for the Turkicisation of the South Sa-
moyed (Karagas), were in more intensive or longer lasting contact with the precur-
sors of the Lena Turks than with the ancestral groups of the modem Tuvans. That the
internal division of Sayan Turkic goes back to relatively recent constellations is un-
derscored by the fact that Karagas and Lena Turkic exhibit a layer of Buryat copies,
whereas in Tuvan we find many traces of strong Khalkha impact.
Khalaj has many features in common with Old Turkic. We may assume that it
separated from Central Turkic in the Late Ancient Turkic period. This is underscored

by the position of Arghu within the Turkic dialects in Käsgari’s dictionary - accord-
ing to Doerfer (1987) the presumed forerunner of Khalaj. It is unclear whether there
were direct connections between the forerunners of hadaq-Turkic Khalaj and adaq-
Turkic Sayan Turkic. Taking into account the features held in common with Old
Turkic in both groups, one may assume that they are indirectly connected by older -
genetic or areal - connections to groups which played a decisive role in the forma-
tion of the Old Turkic written languages. The common features of Khalaj and Mod-
em Uighur may go back to a more or less independent preservation of Late Ancient
Turkic eastern dialect features. According to Käsgari, we find among the many cor-
respondences with eastern dialects of his time an action noun in -GU in Arghu, Cigil,
Yaghma, Tukhsi, Uighur to Upper Sin, and Khakaniyya Turkic, and the nomen agen-
tis in -GUcl in Arghu, Cigil, Kashghar, Balasaghun, Barsghan, Uighur to Upper
Sin, and in the vernaculars of most of the Tiirk (Dankoff-Kelly III, p. 279).26
The special position of modem Oghuz within Central Turkic partly goes back
to the fact that at least parts of the Oghuz started moving westward as early as after
the collapse of the Second Turkic Empire in the eighth century. Thus they were able
to preserve features which in the course of time changed in the remaining group of
Turkic, or to change such features individually. Additionally, Oghuz started to un-
dergo intensive interaction with Persian. After the collapse of the Khazar empire in
the tenth century, parts of the Oghuz crossed the Volga River. Some of the Oghuz-
Chuvash correspondences may go back to this period. At the end of the tenth and
in the eleventh century a part of the Oghuz - the Seljuk confederation - moved
into Iran, thus commencing the period of Turkic immigration into Azerbayjan and
Anatolia.
The Kipchak Turks were already present in the Volga region in the eleventh
century. The intensive interaction between Kipchak and the Bulghar Turkic ancestral
groups of Chuvash may date back to those times. The interaction between these two
groups went on until recent times.
In Käsgari’s dictionary from the eleventh century we find numerous reflexes
of dialectal differentiations. It seems that preservation of the nasality of Ancient
Turkic -n- was already in this period a feature of dialects, which exhibit a whole set
of special features: Oghuz, Arghu and Bulghar.27 In fact Arghu and Bulghar exhibit
a whole set of special features not known to other Turkic dialects of this period. Thus
Käsgari’s Oghuz shows the same ambiguous behaviour as modem Oghuz does. It has
many features of the main (“central”) group of the dialects, but at the same time is
connected to dialects with deviant features - the “Border Turkic” of this period. It is
still not fully understood how one might connect modem Oghuz data with the Oghuz
data of this period. The beginnings of Oghuz-Khalaj interaction may date back to
Käsgari’s times. We may assume that in Khalaj-Oghuz (at least during the last few
centuries: Khalaj-Azeri) contacts the Khalaj speakers were on the receiving end.
The data in Käsgari’s dictionary may also point to a connection between
Kipchak and the azaq-Turkic dialects of this period. According to Käsgari the Kip-
26 But at the same time Arghu also has the “western” form -DAcI o f the nomen agentis.
27 See the instances given by Dankoff-Kelly 1982-1985, III, p. 277.
84 C. SCHÖNIG
chak together with the Yaghma, Tukhsi, Yabaqu, Tatar, Qay, Cörnül and Oghuz
change every d to y. At the same time “some of Kipchak” together with the Yemäk,
Suvar, Bulgar “and those [in the area] stretching to Rus and Rum” (Dankoff-Kelly
I, p. 85) change every d to z. This information is very vague, and we know nothing
about the sub-group called “some of Kipchak”. But if we take Kasgari’s information
seriously, there was a vernacular in the eleventh century that resembles modem Kip-
chakoid South Siberian Turkic by being “Kipchak” and at the same time showing the
feature azaq.
The question of Kipchakoid South Siberian Turkic leads to the “Kirghiz”
problem. The forerunner of the Kirghiz-Kipchak sub-group was perhaps a member
of the Ancient Kirghiz state in the tenth century and took on the name “Kirghiz” for
reasons of prestige - if they are not genetically related with the later Turkicised
Ancient Kirghiz, who were probably of Palaeo-Siberian origin. Unfortunately Käs-
gari gives no detailed information about the language of the Kirghiz of his period.
That Kirghiz-Kipchak shows some special connections to azaq-Turkic (see Schönig
1999b) points to the fact that speakers of Kirghiz-Kipchak remained in South Siberia
or neighbouring areas for a long time, forming a link between Central Turkic and
eastern Border Turkic groups.
In the Middle Turkic period some Turkic languages more or less different from
Late Ancient Turkic became dominant. The precursors of the three main branches of
modem Central Turkic are attested by Old Ottoman (Oghuz), Middle Kipchak and
Chagatay (South East Turkic). Bulghar Turkic is represented by the Volga Bui-
gharian epitaphs. The remaining branches are not attested. But we can tell by the
existence of the -GAn-Turkic area that in the political conditions of the thirteenth and
fourteenth centuries the speakers of many Central Turkic languages under direct
Chinggisid rule (Kipchak, South East Turkic, Salar and parts of Oghuz) together
with ancestral groups of South Siberian Turkic again had the occasion for contacts
and developed or preserved a whole set of common features. The Western Oghuz
were under Ilkhanid rule. Because of the political situation they were often cut off
from contacts with other Turkic groups. The Oghuz of Western Anatolia, who were
only under indirect Ilkhanid rule, may have been almost totally isolated for a while.
These Western Anatolian groups also remained outside the inner sphere of Timurid
power in the fifteenth century, and were not part of the Qara Qoyunlu and Aq
Qoyunlu statehoods after the decline of Timurid power. Due to these political
constellations, Western Anatolian Oghuz could develop quite differently from -GAn-
Turkic and preserve many archaic features, which vanished in the -GAn-Turkic area.
In contrast, the Eastern Anatolian dialects, and other Oghuz groups in Iran gained
more features in common with -GAn-Turkic and additionally entered into intensive
areal interaction with non-Turkic languages. Because most Border Turkic languages
were part of the Chinggisid realm, they exhibit some features closer to or identical
with Central Turkic and -GAn-Turkic features, whereas Oghuz sometimes - like
modem non-Norm Turkic languages - has deviant archaic forms. Khalaj, as another
language of the former Ilkhanid realm, shows at least some features in common with
-GAn-Turkic. But these may go back to earlier connections with поп-Oghuz groups.

T H E IN T E R N A L D IV IS IO N O F M O D E R N T U R K I C 85
Salar may have been part of an older Oghuz unity and perhaps is genetically
related with Oghuz. During the Chinggisid period the Salar were separated from the
Oghuz and went to Xinjiang and Qinghai. There they experienced South East Turkic
and perhaps Northern Turkic impact. Besides, there was influence from Chinese,
Mongolie and Tibetan. Some special common features of Salar and Saryg Yugur may
go back to such non-Turkic influence. Especially Saryg Yugur exhibits traces of long-
term close contacts with Mongolie groups which today bear the same designation
(Sira Yugur).
South East Turkic developed in the Ulus Chagatay; the Kirghiz-Kipchak pre-
sumably were part of this statehood, too (see below). The modem Western-Central
Kipchak languages came into existence mainly in the realms of the Golden Horde.
The internal differentiation of Western-Central Kipchak reflects the internal division
of the Golden Horde. The Horde of Batu (“Blue Horde”) is somehow represented by
Western Kipchak, whereas the White Horde is reflected by the Central Kipchak lan-
guages. Siberian Tatar probably originated in the Horde of Sibir. The many Western
Oghuz features of Crimean Tatar are due to direct Ottoman influence on the Crimean
Khanate from the fifteenth to the eighteenth century. Phenomena from contact with
Indo-European languages in all areas of grammar can be found in Far Western Kip-
chak Karaim, the precursor of which separated from the rest of Western Kipchak per-
haps in the fourteenth century in connection with the rise of Lithuanian power. Its
case is paralleled by Western Oghuz Gagauz - like Karaim, a language spoken in
geographical and cultural isolation from the remaining Turkic world for centuries.
The transitional position of Uzbek between Kipchak and South East Turkic
results from a development from a Kipchak language to a “mixed” Kipchak-South
East Turkic one. The Uzbek confederation left the Central Kipchak community and
emigrated to Mawarannahr at the beginning of the sixteenth century. The Uzbeks
settled together with speakers of Persian and Chagatay Turkic. Because of their liter-
ary prestige both languages gained much influence on Uzbek. Thus, the modem
Uzbek standard language shows both Kipchak and South East Turkic features, whereas
purely Kipchak speaking groups remained on the countryside. Uzbek does not ex-
hibit the sound change tag > taw, which is one characteristic of modem Kipchak (in-
eluding Kirghiz-Kipchak). Furthermore it does not show the typical features of the
Northern Turkic or the phonotactic area. Thus we may assume that these areas be-
came interactive only after the Uzbek had loosened their contacts to the North, i.e.
after the fifteenth century. From at least the sixteenth century on, Ottoman Turkish
had some influence on Uzbek and on Kipchak languages spoken by Muslim popula-
tions which were in closer contact with the Ottoman empire.
The Central Asian area reflects to some extent features of the prestige lan-
guage Chagatay. Taking into consideration that Kirghiz is one of its members, we
must assume that the area was still active at least during the Kirghiz immigration into
the Tien-shan in the seventeenth century. I also assume that some parallels between
Kirghiz(-Kipchak) and Kazakh date back to this relatively late period and are not due
to older closer connections.

86 C . S C H Ö N IG
The Northern Turkic area reflects some older features of early Turkic lan-
guages spoken in the north of Turcia. To a certain degree Northern Turkic is the
counterpart of the area with direct or more intensive contacts with the statehoods of
the Ottomans and Safavids from the sixteenth century on. The Northern Turkic area
contains the phonotactic area, which contains the Pre-North East Turkic area, which
itself contains the North East Turkic area. Moving from the outside to the inside of
this concentric structure we find diminishing influence from the more or less “Is-
lamic” Central Turkic languages, and at the same time an increasing impact of Mon-
golic languages, which was strongest in the North East. Mainly in the {M} > {B)-
sub-area of the phonotactic area, languages from different genetic sub-branches de-
veloped under strong Mongolie influence at least from the fifteenth century on. They
became more similar to each other through areal interaction, i.e. they developed new
and elided old features in a way unknown to most other Turkic languages. In this
respect they resemble other Turkic groups in relative isolation, e.g. Salar, Saryg
Yugur and Khalaj. Especially the Oirat conquest of large parts of Central Asia from
the fifteenth century on is not only responsible for the development of some of the
“inner” areas of Northern Turkic, but also for the presence of some Northern Turkic
features of Modern Uighur.
Southern Siberian Turkic is one of the youngest areal groups and came into
existence only after the dissolution of the North East Turkic area. The Lena Turks
began to move from the vicinity of Lake Baykal to the Lena basin perhaps in the
sixteenth century. In the same century the Kipchak element in Southern Siberia
seems to have been reinforced by immigrant groups from the collapsing Khanate of
Sibir. Of the Kirghiz-Kipchaks the Kirghiz remained at the southern margin of this
area. They withdrew southward perhaps in the sixteenth century. From then on the
Altay Turkic part of Kirghiz-Kipchak developed in closer contact with Kipchakoid
azaq-Turkic groups, of which the Fu-yii were displaced to Manchuria in the middle
of the eighteenth century. At the beginning of this century the Jungars had already
removed the ruling Ancient Kirghiz groups from Southern Siberia. There remained
relatively small Turkic groups from different genetic sub-branches living together
and entering into intensive linguistic interaction. They also had linguistic exchange
with speakers of South Samoyed and Yenisey languages, which in the course of time
became Turkicised - a development still in progress during the nineteenth century.
The Western Turkic : Eastern Turkic division may ultimately originate in the
Late Ancient Turkic period, when non-Bulghar Turkic tribes started to move west-
wards and their language became differentiated to a certain degree. Perhaps this first
division then was reinforced by the political conditions of the thirteenth and four-
teenth centuries. The many oscillations of the internal borderline may be due to the
fact that during a relatively long period the Western and Eastern Turkic areas ulti-
mately consisted of smaller interactive areas, which were more or less located West
and East of this borderline. The borderline we find today seems to be the result of
interfering isoglosses of different areas which were interactive at different times.
As we can see, at least some implications of the proposed model are in accor-
dance with more or less well-known facts from the development of Turkic and the

history of its speakers. Further investigations of Middle and Ancient Turkic will re-
veal whether other predictions derived from this model are really reliable - a decisive
test for any kind of scientific model.
References
Benzing, J. (1959a), Classification of the Turkic Languages. PTF I, pp. 1-5.

— (1959b), Das Tschuwaschische. PTF I, pp. 695-751.
Berta, A. (1989), Lautgeschichte der tatarischen Dialekte. Studia Uralo-Altaica 31. Szeged.
Castrén, M.A. (1857), Versuch einer koibalischen und karagassischen Sprachlehre. St. Petersburg.
Clark, L. (1998), Turkmen reference grammar. Turcologica 34. Wiesbaden.
Dankoff, R .-K elly, J. = Mahmüd al-Käsgan (1982-1985), Compendium of the Turkic dialects
(Diwan Luyat at-Turk). Turkish Sources VII. Parts I-III. Harvard.
Doerfer, G. (1965), Türkische und mongolische Elemente im Neupersischen. II. VOK XIX. Wies-
baden.
— (1987), Mahmud al-Käsyari, Aryu, Chaladsch. UAJb N.F. 7, pp. 105-114.
— (1988), Grammatik des Chaladsch. Wiesbaden.
— (1990), Die Stellung des Osmanischen im Kreise des Oghusischen und seine Vorgeschich-
te. Handbuch der türkischen Sprachwissenschaft. Teil I. Hrsg. Gy. Hazai. Budapest.
— (1995), Zu alttürkisch i- ~ -y't. Türk Dilleri Ara.pirmalari 5 (1995), pp. 12-18.
Doerfer, G.-Tezcan, S. (1980), Wörterbuch des Chaladsch (Dialekt von Xarrab). Budapest (1988).
Dwyer, A. M. (1998), The Turkic Strata o f Salar: An Oghuz in Chagatay Clothes? Journal of
Turkic Linguistics II, 1, pp. 49-83
lanhunen, J. (1989), On the Interaction of Mator with Turkic, M ongolie and Tungusic. Suomalais-
Ugrilaisen Seuran Aikakauskirja; Journal de la Société Finno-Ougrienne 82, pp. 287-297.
JN = Jazyki Narodov SSSR. II. Tjurskie jazyki. Moskva 1966.
Johanson, L. (1976), Das tschuwaschische Aoristthema. Orientalia Suecana XXIII-XXIV (1974—
1975), pp. 106-158.
— (1986), Zur Konsonantenstärke im Türkischen. Orientalia Suecana XXXIII-XXXV (1984-
1986), pp. 195-209.
Levitskaja, L. S. (1976), Istoriceskaja morfologija cuvasskogo jazyka. Moskva.
Menges, K.-H. (1955-56), The South-Siberian Turkic languages. I + Д. - I: CAJ 1 (1955), pp.
107-136:11: CAJ 2 (1956), pp. 161-175.
Poppe, N. (1954), Grammar o f Written Mongolian. Wiesbaden.
— (1955), Introduction to Mongolian Comparative Studies. Helsinki.
Pritsak, O. (1959), Das Karatschaische und Balkarische. PTF I, pp. 340-368.
PTF ! = Philologiae Turcicae Fundamenta. Bd. I. Ed. Jean Deny e.a. Aquis mattiacis. 1959.
Radloff, W., Versuch eines Wörterbuches der Türk-Dialekte. Bd. I-1V. St. Petersburg 1893, 1899,
1905, 1911.
Räsänen, M. (1949), Materialien zur Lautgeschichte der türkischen Sprachen. StO 15. Helsinki.
Rassadin, V. I. (1971), Fonetika i leksika tofalarskogo jazyka. Ulan-Udé: Burjatskoe Kniznoe Izda-
tel’stvo.
Röna-Tas, A. (1982a), On the History of the Turkic and Finno-Ugrian Affricates. AOH XXXVI
(1 -3 ), pp. 429-447.
— (1982b), The Periodization and Sources o f Chuvash Linguistic History. In: Chuvash Stud-
ies. Budapest, pp. 113-169.

88 C. SCHÖNIG
— (1988), Turkic Influence on the Uralic Languages. In: The Uralic Languages. Description,
History and Foreign Influences. Handbuch der Orientalistik, 8. Abteilung, Leiden, pp.
742-780.
— (1991), An Introduction to Turkology. Studia Uralo-Altaica 33. Szeged.
Schönig, C. (1987), Diachronie and Areal Approach to the Turkic Imperative Paradigm. In:
Utrecht Papers on Central Asia. Ed. by M. Vandamme and H. Boeschoten. Utrecht Turk-
ological Series No. 2. Utrecht, pp. 205-222.
— (1988), Materialien zur Stellung des Lenatürkischen unter den Türksprachen. MTurc, 14,
1988 (1990), pp. 41 -5 7 .
— (1989), Zur Normalisierung von Aorist- und Gerundialvokal im Sajantürkischen. Asiati-
sehe Forschungen, Bd. 105 (27th Meeting of the Permanent International Altaistic Con-
ference, Walberberg, 12-17th June 1984), pp. 64—80.
— (1990), Classification Problems o f Yakut. In: L ’asie centrale et ses voisins. Ed. Remy Dor.
Paris, pp. 91-102.
— (1991), Das Lenatürkische und die sprachlichen Merkmale des nordöstlichen türkischen
Areals. In: Altaica Osloensia. Proceedings of the 32nd Meeting of the Permanent Inter-
national Altaistic Conference, Oslo, June 12-16, 1989. Oslo.
— (1993a), Türkische Sprachkontakte in Ostsibirien. In: Laut, J.-P.-Röhrborn, К. (Hrsg.):
Sprach- und Kulturkontakte der türkischen Völker. Veröffentlichungen der Societas Uralo-
Altaica 1993/4. Wiesbaden.
— (1993b), Anlautvariationen von Plural- und Kasussuffixen im Türkischen. Journal of Turk-
ology 1, N 0 . 2, pp. 269-282.
— (1997a), A New Attempt to Classify the Turkic Languages. I. Turkic Languages I, 1, pp.
117-133.
— (1997b), A New Attempt to Classify the Turkic Languages. II. Turkic Languages I, 2, pp.
262-277.
— (1998a), A New Attempt to Classify the Turkic Languages. III. Turkic Languages II, 1, pp.
130-151.
— (1998b), Suffixartige und postpositionelle Ausdrücke für instrumentales und komitativisches
‘mit’ im Türkischen. In: Dogan Aksan Armagam. Ankara Üniversitesi Dil ve Tarih-Cogra-
fya Fakültesi Yayin N 0 : 366. Ed. Kâmile Imer, Leyla Suba§1 Uzun. Ankara, pp. 145-154.
— (1999a), Türkische Kausativsuffixe mit anlautendem *G ~ *K und/oder auslautendem *z.
I-III, to appear in UAJb, N.F.
— (1999b), Anmerkungen zum Fu-yü-Kirgisischen, to appear in Türk Dilleri Arafirmalari.
Sevortjan I. =, E. V. (1974), Ètimologiceskij slovar’ tjurkskich jazykov. Obscetjurkskie i meztjurk-
skie osnovy na glasnye. Moskva.
Sevortjan IV = “j, z, j” (1989). Moskva.
Stachowski, M. (1994), Urtürkisch *mc, *nc und das jakutische Ordinalsuffix. Rocznik Orientali-
styczny 49/2 (1994), pp. 177-183.
T enisev, É. R. (1976a), Stroj saryg-jugurskogo jazyka. Moskva.
— (1976b), Stroj salarskogo jazyka. Moskva.
Thomsen, К. (1959), Die Sprache der Gelben Uiguren und das Salarische. PTF I, pp. 564—568.
Claus Schönig
Institut für Orientalistik
Universität Gießen
Germany

T H E IN T E R N A L D IV I S I O N O F M O D E R N T U R K IC 89
Table 1. The internal division of Central Turkic
Oghuz Kipchak South East T.

Western Turk- Western-Central K. Kirghiz Uzbek Modem
Oghuz men -K. Uighur
Far W. VUC CA
*sürjök +nasal -nasal (+nasal)
agent n. -Щ -U w cl -G U ci
*civ äv äv ~ iiy iiy ~ iiw öy ~ iiy üy iiy ~ ii: iiy ~ ä v üy (~ üg)
‘house’
‘hair’ sac cac cäc cac sac cac ~ sas
*taglig d a g lï ta w lï to:lu: tâ g li taglïq
‘to cry’ *yïgla- - *agla- *yïgla- *ig la- *yïgla-
neg. ger. *-mAylp *-mAyln -BAy *-mAyln
Table 2. The internal division of Oghuz
Western Oghuz Eastern Oghuz
Turkish Azeri Turkmen

4 ‫־‬: du ber- 0 iber-
participles -m is : -(y)An; -DIK -An : -yAn
long vowels (reflexes) +
*bädük büyük böyük beyik
‘which’ hangi hansï h a ysï
‘where?’ nerede harada n ire de
impossibilitive *-A uma- -A bilme- -A ~ -p bilme-
b ... n > b ... n m .. . n
reflexive pronoun kendi o'z
accusative +(y)1 +(n)I
Table 3. Norm Turkic and non-Norm Turkic
Norm Turkic Chuvash Lena Turkic Khalaj

h- 0 0 0 h-
c- •‫־‬У- + 0 0 +
nominal plural +lAr +sem +LAr and others +lAr
gerund in -p + 0 0 0
conditional *-sAr + 0 + +
imper. *-zUn + 0 + 0

T a b le 4. C e n tr a l T u r k ic a n d B o r d e r T u r k ic
‘foot’ ‘moun- postvoca- ‘who?’ lst.ps.pl. past in ‘to give birth - ket- ‘to go privative in ‘stirrup’
tainous’ lie aorist +DI.mIz be born’ away’ +slz
Old T. adaq tagli'g -yUr kam, kim28 + tog- + + üzärjgü
Central T. ayaq see tab. 1 -r *kim 0 *tog- + + *iizätjgü
Salar ayax dagl'i -r *kim + *tog- 0 0
Altay T. ayaq tu.lu:, -r *kam +~0 *tog-, *törä- 0 0 iizerji
tagli'g
Saryg Yug. azaq tagli'g -r *kam + *tog- 0 0 ezenky
C. SCHÖNIG
Fu-yii azïx daxti'x *käm + *tog- 0 0
Yenisey T. *azaq tagli'g *-yUr *käm 4‫־‬ *tog-, *törä- 0 0 xks. izetje
Sayan T. adaq dagli'g *-yUr qïm, qum + *törä- 0 0 tuv. ezerjgi
kgs.
ezetjge
Lena T. *ataq t'ialatx2930 *-yUr kim < ? + *törä- 0 0 yak. irjähä
Khalaj hadaq taglug *-yUr *këmi0 0 *tog- 0 +
Chuvash ura tullâ *-yUr *käm + sural- 0 + yârana
28 The form käm appears in !‫־‬uniform script, kim in the remaining sources.
29 With the deviant meaning ‘with forest’. Furthermore, the suffix +LA:x is no direct cognate o f Old Turkic +IXG.
30 See Doerfer (1988, p. 107).
T a b le 5. P r e s e r v a tio n o f O ld T u r k ic f e a tu r e s in M o d e r n T u r k ic
‘hand’ ‘lip’ long vowels finite genitive reflexive ‘to return’ -GVz-
-mis suffix pronoun causatives
Old T. älig ärin *+ + +(n)Irj käntü+ yan-, qadi't- 0
Sayan T. *qol ärin reflexes 0 +NIrj *bod+ *yan- +

Yenisey T. *qol *ärin 0 0 + N Irj *bod+ *yan- +
Altay T. *qol ärin 0 0 +NIrj *bod+ *yan- +
Fu-yii ali'x ‘palm’ 0 0 +NI1j *bod+ *yan-
Saryg Yug. illy, elig damsay reflexes 0 +Nhj *bod+, *ö:z+ *yan- +
Lena T. ili: uos + + 0 bäyä+ tönün- 0
Chuvash alâ tuta only *ö: 0 +(n)Än xä+ tavrän- 0 (? )
Khalaj äl dudaq + + *+(U)rj31 *ö:z+ 0
Salar al dodax reflexes + +niyi *ö:z + *yan- +
Turkmen äl do. daq, erin + 0 +(n)It] *ö:z+ *qadi't- 0~(+)
Azeri äl dodag reflexes + +(n)In *ö:z+ *qadi't- 0~(+)
Turkish äl dudak reflexes + + (n )In *käntü-h dön- 0~(+)
Modern *qol, *älig kelpuk, läv 0 (+) *+nhj *ö:z. + *qadi't-, *yan- +
Uighur
remaining *qol *ärin 0 0 *+nbj *ö:z+ *qadi't- +
Centr. T.
31 With palatalised velar nasal, see Doerfer (1988, p. 79). ю

T a b l e 6 . E a s t e r n : W e s t e r n T u r k i c is o g lo s s e s
2nd ps. ‘is not’ yilti'z : 40 finish ‘thousand’32 ‘tree’ dénommai ‘twenty’
plural y ult uz etc. (intr.)’ verb et-
Old T. +z ärmäz + büt- bïrj (y)ïgac 0 y(ä)girmi y(ä)girmä
Lena T. *+z *bo.lmadoq + büt- *тир mas 0 *sü.rbe
Sayan T. *+IAr *ärmäz (+ ) büt- mu y *yïgac 0 tuv. cätrbi
Yenisey T. *+lAr nimäs ‘is not’ (+ ) *büt- mur) *(h)agac + Xks. cibïrgï Sor. cägirbä
Fu-yii *+lAr 0 murj *(h)agac + Jibir
Altay T. *+lAr *ärmäz 0 büt- mur) *(h)agac + Ji:rmâ
Saryg Yugur *+lAr *ärmäz + fyt-, p cu Jt- mïrj, murj *yïgac + Jïgïrma, yigirmi
C. SCHÖNIG
pit- etc.
Salar *+lAr *ärmäz 0 bit- mit) *(h)agac + yigirme, yigirmi
Modem Uighur +Z, *+lAr *ärmäz + püt- mît), mirj yagac, *yïgac (+) *yägirmä, zigirma
Kirghiz +z, *+lAr *ärmäz 0 büt- mirj *yïgac + Jïyirma
Kazakh +Z, *+IAr *ärmäz 0 büt- mïrj *(h)agac + zïyïrma
Chuvash *+z *ärmäz 0 pët- pin *(h)agac 0 sirëm
Tatar +z *tägül 0 bët- mer) *(h)agac + yëgërmë
Turkmen +z *tägül 0 bit- mii\j *(h)agac + yigrimi
Turkish +z *tägül 0 bit- bin *(h)agac + yirmi
Khalaj +z da:g 0 bit- miyk, min hagac + (?)33 yiyirm i
32 The Turkic numeral is replaced by a loanword from Russian in Lena Turkic (Dolgan tiihacca, Yakut tïhï.nca) and Sayan Turkic Karagas
(ti'.'sicci).
33 See ärz iet- ‘vortragen, Vorbringen’ (Doerfer-Tezcan 1980, p. 111a); perhaps a copy from Azeri.
T a b le 7. N o r th e r n T u r k ic a n d re la te d a re a s
1st ps.pl. ‘hair’ *ï:d(u ber)-'to ordinals in plosive dis- ‘to cry’ ‘navel’
possessive send’ *+(I)ncI similation
Old T. +(X)mXz sac i':d(и ber)- 0 (+) (y)igla- kindik
Chuvash *+(X)mXz siis *ï:d- 0 0 yèr- *kö:bäk

Lena T. +BIt *sac *ï:d- 0 + ïîa:- ki:n
Sayan T. +(I)vIs Kgs. cd”s *ï:d- 0 + ïgla- Tuv. xin(dik)
Kgs. hin
Yenisey T. +(I)bIs *cac *ï:d- + + ilga- *kin(dik)
Fu-yii +(I)bis sas + + Jiîgi-, yïlgï- kindik
Altay T. +(I)bIs cac *ï:d- + + i'yla- kin(dik)
Kirghiz HI )biz cac *ï:d(и ber)- + + i'yla- kindik
Tatar +(I)bIz cäc *ï:du ber- + 0 yila- këndëk
Saryg Yug. HOv saz ~saz + + (y)ïgla-
Salar [+ Ш *sac + 0 yïgla-, (yagla-) *kindik
Kazakh +(I)mIz sas *ï:d(и ber)- + + Zita- kïndîk
Modern Uighur +(I)mlz cac (lit.) *ï:du ber- + 0 zigla- kindik
~ sac (dial.)
Turkmen +(I)mI5 sac *ï:du ber- + ~ others 0 agio- *kö:bäk
Turkish +(I)mIz sac 0 + 0 agla- *kö:bäk
Khalaj *+(X)mXz sac + ~ others 0 higla- kindik
40
Table 8. Norm Turkic features, Central Turkic features, Eastern Turkic features in languages of the Western Turkic area, 40
Western Turkic features in the Eastern Turkic area and individually developed features in non-Norm Turkic,
Border Turkic and Archaic Central Turkic (the features in question are printed bold)
Western Turkic area Eastern Turkic area

Oghuz South Siberian Turkic
Chuvash Khalaj Turkish Azeri Turkmen Kipchak Modem Salar Yenisey Sayan Lena Turkic
Uighur Turkic Turkic
gerund in -p 0 + 0
adaq и га hadaq ayaq *azaq adaq * a ta q
‘to go out’ tu x - *cïq- ü n -, ta ” s- ta g ïs-

ordinal suffix + (Â )m ë s (ë ) *+(I)nc *+(I)ncI *+(I)ncI *+(I)ncl *+(l)ncl *+(l)ncl + (I )s(K I) + (I )s
I e.a. e.a. e.a.
et- yap-, et- et-, *ädlä- et- et- qïl-, et- et-
C. SCHÖNIG
denominal tu - q ïi‫־‬ q ïl-, q ïn -
verb (qtl-) (qïl-) (qïl-) (qïl-, *ädlä-) (q ïn -)
‘to cry’ y ë r- higla- a g la- *yïgla- *igla- *yïgla- yagla-, ilga- ïgla- ït a : ‫־‬
yïgla-
genitive suffix +(n)Än + (U )q + *{n)Irj *+nlrj - +NIrj +nlrj +niyi +Nlrj 0
‘is not’ *ärm äz d a :g *tägül *ärmäz, *tägül *ärmäz nimäs *ärmäz * b o ilm a d o q
taglig tullä tä ’g lu g daglï tawlï, to. lu: taglïq ta g lï taglïg daglïg tïala.x
‘hand’ a lä ai *qol *qol, *älig ai *qol ili:
‘lip’ *to.taq do.daq, erin *erin qalpük, läv dodax *erin uos
bit- - biit- *bit- bit- - but- put- b it- *but-

2nd ps.pl. sign *+z +Z ~ +7J+lAr +z/+lAr *+zlAr *+lAr +z
‘hair’ *cac *sac *cac - *sac *cac * sa c
‘to return’ * tä g z in - * tö n (ü n )- *qadït- *qadït-, *yan- *yan- * tô n (ü n )-
reflexive xä+ *ö:z+ k e n d i+ *ô:z+ *bod+ bäyä+
pronoun
Southern Turkic area Northern Turkic area
Table 9. Common Turkic
_________ ________________________________________ Common Turkic__________________________________________________

non- Norm Turkic non-
N o rm T .__________________________________________________________________ _______________________________ Norm T.
Border T . Central Turkic | _______ Border Turkic_____________
N ortheast Turkic
_________ I L ena-Sayan-Turkic _

Boigar South Siberian T urkic Lena
Turkic: ___________________________________________________________ Turkic
Chuvash Kipchak K1PCHAKOID SOUTH Sayan
Siberian T urkic Turkic Z
О
Westem-Central-Kipchak Kirghiz- Kipchak агв?-Тигкк (adaq- 5
Western Kipchak Central Asian Kipchak 1. Siberian Turkic):
Kirghiz Allay azaq-Т ш к к Tuvan, z
Volga-Ural-Caucasus- Nogay, Kazakh, Karakalpak Khakas Fu- Karagas c
Kipchak: Tatar, B a sh - ______________ _________ yü etc. § •
kir; Karachay-Balkar, Shor ° ^
Kymyk Modem Uighur Chulym Turkic___________ ■4
2. non-Siberian §
Far Western Kipchak: azag-Turkic: о
Karaim, Misher \_U zbek SarygYugur
I Southeast T.
Crimean Tatar
Turkmen Salar
H _ _ _ _ _ _ _
S -------------------- ------------------------------
_________ U Turkish, Gagauz, Khorasan Turkic

Khalaj g Azerbaijanian
^ Western Oghuz Eastern Oghuz
_________________Oghuz___________________ __________________ ________________________________________________
_________ Western Turkic [ Eastern Turkic |
Legend: capital letters (supposedly) areal groups normal letters genetic sub-groups; languages of uncertain status
bold letters genetic groups italic letters individual (standard) languages or dialect groups
MISCELLANEA
B é la B a r td k ’s A n a to lia n fo lk m u s ic collection . But now on the basis o f a large material

A r e p o r t o n a P h .D . d is s e r t a t io n d e fe n d e d collected by Janos Sipos it became possible to
o n th e 1 6 th F e b r u a r y 1 9 9 9 analyse the data and statements in Bartok’s col-
a t th e U n iv e r sity o f S z e g e d lection. The new classifications concerning
Hungarian folk music also speak in the favour
The general interest of the author is to collect, o f this revision. It seemed to be important to
analyse and compare the folk music o f different compare the three editions o f the work because
Turkic peoples in order to establish the possi- there were lots o f smaller and greater errors in
ble common musical layers if any and also to the texts and their translations.
compare these musical styles to the musical Between 1988 and 1993 Janos Sipos was
layers o f Hungarian folk music. The author is teaching at the Hungarology branch o f the
one o f the followers of the composer and eth- Ankara University, and during this period he
nomusicologist Béla Bartok, who made a short, made several field trips in various areas of
but very effective research trip to Anatolia in Turkey which resulted in some 1500 melodies
1936. collected. He studied the earlier Turkish publi-
Béla Bartdk’s Turkish folk music collec- cations containing transcriptions as well, and
tion was published after 36 years of lying ap- after a critical analysis he completed his collec-
parently forgotten in 1976, both in the US and tion with more than 3000 additional melodies.
in Hungary and somewhat later in Turkey. None The six years long residence in Turkey, famili-
o f the three editions produced a real echo, arity with the Turkish language, the consulta-
though these books are very important works in tions with the local ethnomusicologists and
the field o f ethnomusicology. first o f all the systematic collecting work, tran-
What could be the reason for this disin- scriptions and analysis o f the melodies enabled
terest? If we disregard other possible reasons, him to prepare a large and well classified ma-
there still remains a serious argument: Bartdk’s terial.
collection contains only some 100 melodies He started his field work in the area where
and this makes the formulation o f universal Bartok finished his. Later, step by step he moved
statements about Turkish folk music impossi- westward. In the end he deemed, he had suffi-
ble. Meanwhile until recently there did not ex- cient material to examine whether the state-
ist any comprehensive work which could serve ments o f Bartök’s Introduction stand the test of
as a background to Bartdk’s collection. the time, or were uncertain or perhaps proved

98 MISCELLANEA
false. The task was quite difficult as he had to The result is the critical survey o f the
take into consideration the original and the American and Hungarian editions o f Bartok’s
corrected form of Bartök’s Introduction and Anatolian collection, which assumed a perfect
transcriptions, Saygun’s and Suchoff’s notes knowledge o f Turkish and a deeper understand-
and he had to find a way to insert his own ob- ing o f Turkish folk music. But the greatest
servations too. Finally he divided the disserta- work, which enabled Sipos to put Bartök’s col-
tion into six chapters. lection into a broader context, was the consid-
In the first chapter he introduces the antece- erable field work and the transcription o f the
dents o f Bartök’s collecting trip, his field work melodies. Sipos underlines the fact that, though
and the circumstances o f the birth o f Bartök’s Turkish colleges sometimes record mourning
book. These are followed by the annotated ver- or generally parlando songs, they rarely tran-
sion o f the Introduction, the Notes to the melo- scribe them, therefore several Turkish melody
dies and Notes to the texts. types appear in his book for the first time.
In the second chapter he discusses the mu- The other important point o f the disserta-
sical classifications. First he introduces the fi- tion is the classification o f the melodies. Since
nal version of Bartok’s classification concerning in this field there had not existed this type of
Hungarian folk music, then he enumerates the comprehensive and comparative work before,
newer classifications. In the second half o f the the 1500 melodies collected, transcribed and
chapter he lists the attempts at collecting and partly published first count as filling the gap in
classifying Turkish folk music. the Anatolian folk music research, due to its
In the third chapter he introduces his own huge bulk and especially to the degree it had
collecting work. This is followed by the gen- been classified.
eral description of the collected material in the This great material enabled Sipos to enrich
follow ing order: micro tunes, scales, symmetric and in many points to modify Bartok’s com-
and asymmetric rhythms, number o f syllables, parative analysis concerning the Turkish and
forms, cadences, compass, the height o f the Hungarian folk songs and texts. He analyses in
sections and refrains. detail the similar Hungarian and Anatolian mel-
In the fourth chapter he makes an attempt ody styles as: twin-bar melody types with the
to compare Hungarian and Turkish folk music. nucleus sol-mi-re-do, laments, and “psalmodie”
He arranges the Turkish material into types, melodies. In many cases he refers to the musi-
classes and styles and examines their relations cal connection which points towards European
to the Hungarian and sometimes to other folk and non-European (e.g. Central Asian, Byzan-
music. tine etc.) folk music.
In the fifth chapter he deals with the texts The examination of the dialect background
o f the melodies. He begins this chapter with a of the texts is not complete, but the well-or-
historical overview. This is followed by the dered presentation o f the dialectal phenomena
analysis of the contemporary folk texts. He ex- made the homogeneous and consequent han-
amines with great care the logical structure of dling o f the text in the Bartok collection possi-
the stanzas in Bartök’s collection, then prepares ble. Finally the dissertation could answer some
a critical edition of the texts, and finally intro- o f Bartok’s questions concerning the relations
duces the relation between the texts and melo- between melodies and texts.
dies. Based on the recommendation o f the offi-
In the sixth chapter he examines the histori- cial reviewers ethnomusicologist Laszlo Vikar
cal background o f the comparison o f the Hun- and linguist Istvân Vâsâry, the expert commit-
garian and Turkish melody styles. He surveys tee proposed to the Ph D. committee o f the
the ancient history o f the Turks in Central Asia University o f Szeged to confer the degree on
and East-Europe and scrutinises the Turkic ele- Jänos Sipos.
ments which joined the Hungarians. J. Sipos

MISCELLANEA 99
T h e e ig h th s e m in a r o f th e IA T S A. Egyed’s paper ( Candragomi’s Treatises

in B lo o m in g to n on Music) dealt with two theoretical works on
ritual music attributed to the Indian Candra-
The International Association for Tibetan Stud- gomin (about sixth century AD). The existence
ies (hereinafter IATS) was founded at Oxford of these texts had been known only from cita-
in 1979. The main objective of the IATS is to tions in the works o f other writers, and their
support and develop the study of Tibetan cul- originals had been believed to be lost. She
ture in all its aspects and to encourage interdis- succeeded - in collaboration with Gene Smith
ciplinary exchange o f ideas and research expe- (New York City, USA) and Tashi Tsering (Dha-
rience. Perhaps the most noted event o f interna- ramsala, India) - in locating the original texts.
tional Tibetology is “The International Seminar In his paper G. Bethlenfalvy (Notes to the
on Tibetan Studies” (hereinafter “The Seminar”) Rdzong-khul Lineage) gave an overview o f the
which takes place every three or four years. important masters o f the Rdzong-khul Monas-
These conferences give Tibetan specialists from tery (in Zanskar) in the 1 8 -19th centuries. This
all over the world the opportunity to get to- monastery arose the interest of Western re-
gether and to discuss the actual advances in searchers because its abbot, Kun-dga’-chos-legs
Tibetan Studies. The latest (which was the 8th) (1770-183?, 184?) was a teacher o f Alexander
Seminar of the IATS was held at Indiana Uni- Csoma de Körös. (Kun-dga’-chos-legs is known
versity in Bloomington (Indiana, USA) from to the scholars o f the field mainly by his Dris-
July 25th to July 31st, 1998. Nearly 300 par- lan in which he answered Csoma’s questions
ticipants attended the Seminar and about 140 concerning Buddhist philosophy.) The public
papers were presented. The great number o f learnt many interesting details regarding the
valuable lectures makes it impossible to enlist Rnam-thar (biography) of Kun-dga’-chos-legs
them here. which was hitherto not available for research.
The papers were read and discussed in vari- T. Pore id (Kashmirian Religious and Cul-
ous sections, such as History; Anthropology; tural Elements in the Tibetan Versions o f Sitâta-
Philosophy; Religion; Art; Development and patrâdhâranî) attempted to establish a chrono-
Current Issues; Language and Linguistics; Ti- logical order among the different Tibetan ver-
betological Methods and Materials; Legal and sions of the Sitätapaträdhäram and to present
Political Documents; Medicine and Astrology; hints found in the newer versions indicating
Music and Material Culture and there was a spe- that the evolution o f the text took place most
cial panel: Comparative Gender roles. likely in Kashmir and was influenced by the
Hungarian Oriental Studies were represented co-existence o f Buddhism and Saivism. It was
at the Seminar by Agnes Birtalan (Budapest), just a coincidence that A. Birtalan referred in
Alice Egyed (Budapest), Géza Bethlenfalvy her paper to Marco Polo’s report in which the
(Budapest) and Tibor Pore id (Szeged). weather-makers o f the Mongolian Court were
A. Birtalan ( Tibetan Weather-Magic Rite of mentioned as “Tebet” and “Kesimur”.
a Mongolian Shaman) gave an account o f a On July 28th afternoon the participants were
weather magic ceremony that she observed in invited on an excursion and picnic outing to the
1995 in the Uws district (Western Mongolia). Brown County State Park. At the business meet-
The aim o f this ritual was to avert rainfall and ing on July 30th, Elliot Sperling, Convenor of
heavy storm. It was performed by a shaman the 8th Seminar pledged himself to edit the
who recited the Sitätapaträ (Tib. gdugs dkar proceedings o f the Seminar1. The assembled
mo can, Mong. Caga(ga)n sikürtü) -dhärani in
Tibetan. Ä. Birtalan compared also the objects 1The proceedings o f the previous Seminars
used during the ritual by the Mongolian sha- are as follows:
man with the Tibetan ones. Some parts o f the Tibetan Studies: in Honour of Hugh Rich-
ceremony were shown to the audience on video. ardson: Proceedings o f the International Semi

100 MISCELLANEA
members elected a new Board o f Advisers, and previous Seminar (among them the young Hun-
accepted the offer extended on behalf o f the garian Tibetologist Zoltän Horvâth).
International Institute for Asian Studies to host Last but not least, thanks are due to Elliot
the next Seminar in Leiden, the Netherlands, Sperling and to his assistants for all their ef-
between June 24th-30th 2000. At the meeting forts to make this conference a remarkable
w e had the sorrowful duty to remember the event o f Tibetan studies.
IATS members who had passed away since the Tibor Porciö
nar on Tibetan Studies, Oxford 1979 [Ed. by Graz 1995. 7 Vols. [General editor: Ernst Stein-
Michael Aris and Amy San Sou Kyi] - Warmin- kellner] Wien; Vol. 1-2: Tibetan Studies [Ed.
ster, 1980. by Helmut Krasser] - 1997; Vol. 3: Transmis-
Soundings in Tibetan Civilisation: Pro- sion o f the Tibetan Canon: Papers Presented at
ceedings o f the 1982 Seminar of the Interna- a Panel of the 7th Seminar of the International
tional Association for Tibetan Studies held at Association for Tibetan Studies, Graz 1995
Columbia University [Ed. by Barbara Nimiri [Ed. by Helmut Eimer] - 1997; Vol. 4: Tibetan
Aziz and Matthew Kapstein] - New Delhi: Ma- Culture in the Diaspora: Papers Presented at
nohar, 1985. a Panel o f the 7th Seminar o f the International
Tibetan Studies: Proceedings o f 4th Semi- Association for Tibetan Studies, Graz 1995
nar o f the International Association for Tibetan [Ed. by Frank J. Korom] - 1997; Vol. 5: Devel-
Studies, Schloss Hohenkammer - Munich 1985 opment, Society, and Environment in Tibet:
[Ed. by Helga Uebach and Jampa L. Panglung] Papers Presented at a Panel o f the 7th Seminar
- München, 1988. of the International Association fo r Tibetan
Tibetan Studies: Proceedings of the 5th Studies, Graz 1995 [Ed. by Graham E. Clarke]
Seminar o f the International Association for - 1998; Vol. 6: Tibetan Mountain Deities, Their
Tibetan Studies, Narita 1989. Vol. 1: 1. Bud- Cults and Representations: Papers Presented
dhist Philosophy and Literature; Vol. 2: Lan- at a Panel of the 7th Seminar o f the Interna-
guage, History and Culture [Ed. by Ihara Shoren tional Association for Tibetan Studies, Graz
and Yamaguchi Zuiho] - Narita, 1992. 1995 [Ed. by Anne-Marie Blondeau] - 1998;
Tibetan Studies: Proceedings o f the 6 th Vol. 7: The Inner Asian International Style,
Seminar o f the International Association for 12th-14th centuries: Papers Presented at a
Tibetan Studies, Fagemes 1992. 2 Vols. [Ed. Panel of the 7th Seminar o f the International
by Per Kvaerne] - Oslo, 1994. Association for Tibetan Studies, Graz 1995 [Ed.
Proceedings of the 7th Seminar of the In- by Deborah E. Klimburg-Salter and Eva Allin-
ternational Association for Tibetan Studies, g e r ]- 1998.

Acta Orientalia Academiae Scientiarum Hung. Volume 52 (I), 101-104 (1999)
REVIEW ARTICLE
H . F r a n k e - H o k - l a m C h a n (1997), Studies on were a dynasty o f conquest, many Khitan-Liao

the Jurchens and the Chin Dynasty. Ashgate institutions were taken over or adapted by the
(Variorum Collected Series CS591)' new régime. The similarities and differences
between these two alien states are evidently of
This selection o f articles on the Jurchens and great historical interest. Indeed, in 1944 Feng
the Chin Dynasty includes studies grown out o f Chia-sheng had already begun to work on the
the Chin Dynasty History Project which in turn Chin sources using the same method as for the
was a part o f the ambitious Chinese History Liao, namely, topically arranged and annotated
Project initiated by Karl-August Wittfogel. In translations of the dynastic history o f the Chin
1939 the Rockefeller Foundation granted sub- (Chin shih). In the Liao volume many refer-
stantial sums for a period of over ten years to ences can be found to the unpublished Chin
Professor Wittfogel, who was then affiliated to materials such as ‘See HCS, Chin, particularly
Columbia University in New York as a mem- VII, XIV, and XV (m s)’ (note 30 on p. 3). Feng
ber of the International Institute for Social Re- Chia-sheng, however, returned to China in the
search. Work for the project started in July summer of 1948, marking an end to work on
1939, first in New York and later in Seattle, at the Chin materials for a long time. No succès-
the University o f Washington, where the Far sor for Feng could be found, and the Rocke-
Eastern and Russian Institute under Professor feller grants expired after 1949. In 1958 Profes-
George E. Taylor assumed the administration sor Wittfogel finally handed over the existing
o f the project. During the following years sev- materials to the Far Eastern and Russian Insti-
eral prominent Chinese scholars worked in tute in Seattle.
Seattle, most notably Feng Chia-sheng, who to- A new start was made in 1964—65 when
gether with Wittfogel co-authored the monu- Herbert Franke was appointed Visiting Profes-
mental volume, History of Chinese Society, Liao sor at the University o f Washington. His inter-
(907-1125), published in 1949. est in the Chinese dynasties o f conquest led
The Liao dynasty was founded by the no- him to inquire about the Chin materials. After a
madic Khitan invaders. In Northern China they long search in the basement o f Thomson Hall, a
were succeeded by the Tungusic Jurchens who crate was discovered which contained dozens
established a dynastic state o f their own in o f folders with draft translations from the Chin
1115. This dynasty took the sinitic name Chin shih and a few explanatory notes. They were
and existed until 1234 when it succumbed to arranged according to the format o f the Liao
the Mongols. Under the Chin, who like the Liao volume, i.e., in fifteen topical sections from
0 0 0 1 1 9 9 9 © 5.00 $ /99/ 6446‫ ־‬Akadémiai Kiado, Budapest

102 REVIEW ARTICLE
‘Administrative Geography and Population’ (I) rungen: Wirtschaft und Gesellschaft unter der
to ‘Military System’ (XV). The materials were Chin-Dynastie (1115-1234), (Rheinisch-West-
in bad shape, handwritten by Feng Chia-sheng fälische Akademie der Wissenschaften, 1978).
on brittle wartime paper and in some instances The chapter on the Chin in vol. 6 of the Cam-
hardly legible. They also included a few brief bridge History o f China (1994) which also deals
signed comments and notes contributed by Karl with the alien régimes in medieval China, is to
H. M enges (who had already contributed to the a large extent based on previous work for the
Liao volume), Professor Hellmut Wilhelm and Chin Dynastic History Project. There also
‘V .B .’ (unidentified). Herbert Franke obtained exists an unpublished manuscript ‘Table of Main
Professor Taylor’s permission to work on these Events’ where, as in the Liao volume, a chrono-
materials, and Professor Wittfogel also gave his logical account from 1115 to 1234 based on
consent. Franke had the existing translations entries in the Chin shih is given. H. Chan also
revised, typed and newly annotated. From 1969 published a study o f the legitimation of the
his work was supported financially by the Ameri- Chin dynasty: Legitimation in Imperial China:
can Council of Learned Societies (of which the Discussions under the Jurchen Chin Dynasty
V ice President was Dr Gordon B. Turner) and (1115-1234) (1984), and another about the
its Committee on the Study o f Chinese Civilisa- Mongol attack on the last Chin capital in 1233-
tion. Over the years repeated grants from ACLS 34: The Fall of the Jurchen Chin: Wang E’s
have been provided to pay an honorarium to Memoir on Ts’ai-chou under the Mongol Siege
those contributing to the project. Sincere thanks (1993). In addition it must also be mentioned
are therefore due to ACLS in general and to that recently an excellent collection of Chin
Dr Jason H. Parker, its current executive asso- studies has been published in America, namely
ciate in particular, who has permitted the re- the volume China under Jurchen Rule, edited by
maining balance of a former ACLS grant to the Hoyt C. Tillman and Stephen H. West (1995).
University of Washington to be used for the pro- Over the years, both Franke and Chan real-
duction of this Variorum volume. ised that it would not be possible to emulate the
A brief report on the Chin Dynastic History Liao volume of the Chin Dynastic History Pro-
Project was published in Sung Studies News- ject, which had enjoyed the co-operation o f a
letter no. 3 (1971). At that time it was planned great number o f scholars and a large degree of
to produce a volume along the same lines as the financial foundation support. They decided
Liao volume. In 1971 Professor Hok-lam Chan therefore to be content with publishing indi-
(from the University o f Washington) was in- vidual studies on various topics of Chin history.
vited to join the project. His co-operation was A selection o f these are now reproduced in this
invaluable and has produced several learned volume.
papers, three of which are reproduced in this The homeland o f the Jurchens where they
selection. Some relevant publications by other lived as hunters, fishers and farmers was north-
scholars also have grown out o f the Chin Dy- eastern Manchuria. They spoke a Tungusic lan-
nastic History Project. Mention should be made guage closely related to that o f the Manchus,
o f Morris Rossabi, The Jurchens in the Yiian who regarded the Jurchens as their ancestors,
and Ming (Ithaca, N.Y.: Cornell University East and ruled the whole o f China from 1644 until
Asia Papers no. 27, 1982), and o f Susan Bush, 1911. Politically, from the 10th century on-
‘Archaeological Remains o f the Chin Dynasty’ ward, the Jurchens were vassals of the Khitans
in Bulletin of Sung-Yiian Studies no. 17 (1981). to whom they had to pay tribute. They did not,
H. Franke contributed the article on the Chin however, form a nation or state in the strict
dynasty to the Cambridge Encyclopaedia of sense but were organised into more or less in-
China (1982), and has published a survey of dependent clans who frequently fought against
Chin society and economy in German, Nord- each other. During the 11th century the Wan-
china am Vorabend der mongolischen Erobe- yen clan gained prominence and united the

REVIEW ARTICLE 103
other clans under their rule. Finally the Jurchen again in 1206 when Sung started an abortive
chieftain Wan-yen A-ku-ta succeeded in de- revanchist campaign against Chin. The sources
feating their Khitan overlords due to the supe- for the diplomatic relations between Chin and
rior military skills and bravery of the Jurchen Sung include a great wealth o f details and are
warriors. In 1115 A-ku-ta was proclaimed the therefore o f importance for the history o f diplo-
first monarch o f a new dynasty, the Chin. Thus macy in East Asia.
Chinese statehood was adopted together with Articles V and VI are studies of Jurchen eus-
some elements o f Chinese bureaucracy. This pe- tomary law and of the Chin legal system. The
riod o f transition from a clan society to an im- pre-dynastic Jurchens did not have written laws
perial régime is the subject of Articles I and II, but relied on orally transmitted customs. These
where many details o f the way of life among were quite different from Chinese legal tradi-
the early Jurchens are also represented. tions. Feuds between families or clans fre-
The intrusion o f Chinese cultural influ- quently resulted in manslaughter. Such cases
ences and their eventual acceptance by large led in early times to unlimited vendettas and
parts o f the ruling élite did not, however, fully thus tended to destabilise society. Vendettas
displace old Jurchen customs. Throughout the were later prevented by payment o f a compen-
Chin dynasty w e can notice an antagonism sation to the killed person’s family. Violations
between nativists and advocates of sinification. o f property by theft were punished by fines and
Although Buddhism played a great role, chiefly payment o f an indemnity, sometimes accompa-
at the court, many Jurchen customs survived. nied by harsh corporal punishment. During the
Article III describes some elements o f Jurchen 12th century the Chin gradually adopted a legal
folk religion, such as shamanism. Shamans system patterned on that o f the Chinese, and
were influential as interpreters of omens and finally a comprehensive legal code was prom-
dreams and performers o f magic in warfare. ulgated in 1201. This code was not a simple
Some rituals such as seasonal riding and arch- copy o f Chinese T’ang and Sung legislation; it
ery contests or funeral rites inherited from the preserved features of traditional Jurchen law,
Khitans survived, whereas the Chin state intro- chiefly in family and inheritance law. A gen-
duced institutions and cults of the traditional eral characteristic o f Chin law was the recogni-
Chinese state religion. tion of ethnic differences. This was a factor
Article IV is a study of the relations be- intended to stabilise the multi-national society
tween the Chin and the Chinese Sung state as o f the Chin.
reflected in the treaties concluded between Key aspects o f Chin socio-economic insti-
Chin and Sung. After the Chin armies had con- tutions and activities are treated in Articles VI
quered the Sung capital of K’ai-feng in 1127, to X. In the 12th century the Chin state adopted
harsh conditions were imposed by the victori- Chinese fiscal elements, including the bureauc-
ous invaders. The Sung had to declare them- ratised state monopolies for salt, wine, tea, and
selves vassals o f the Chin and to pay enormous other lucrative commodities. Article VI by
indemnities. There followed years of indecisive Chan treats the tea monopoly system in detail.
warfare until in 1141, when both parties agreed It appears that whereas the Chinese population
on a new treaty that gave Sung a junior status under Jurchen rule continued to drink and pro-
vis-à-vis the Chin and fixed the annual pay- duce tea subject to state levies, the Jurchen rul-
ments to be paid by the Sung. Controlled border ing élite belatedly took on the habit and height-
trade was resumed and the borderline between ened the demand for quality tea which the North
the two states demarcated. A precarious со- was unable to supply. The Chin state thus in-
existence dominated foreign politics from then creasingly drained its coffers by paying the
on for a long time. This balance was twice tern- Sung people commodities and even precious
porarily interrupted, first in 1159 when the Chin metals to import tea from the South, and the
ruler Hai-ling tried to invade the South, and situation also encouraged contraband trade. It

104 REVIEW ARTICLE
had a deleterious effect on the Chin economy. the data from the Chin shih and other content-
Article VIII by Franke is focused on the wine poraneous accounts. It identifies three classes
monopoly. It shows that the Chin state also o f labourers: compulsory labourers (i.e., statu-
adopted the Sung system o f wine monopoly tory corvée, from which one could be exempted
and discouraged private production; the reve- by paying a service remission tax), hired la-
nue from wine was considerably less compared bourers, and exempted personnel, which in-
with tea, and its enforcement by the authorities, eluded imperial nobility, ranked officials and
especially on the Jurchen people, was handled other special social groups. The administration
in a very lax and half-hearted way. This is not o f labour service, such as the registration, utili-
surprising in view o f the fact that whereas the sation, and treatment o f labourers is examined
Chinese people were moderate in their drinking in succession, followed by a detailed account
habits, the Jurchens, like other northerners, had o f several major labour mobilisation projects,
a reputation for being hard drinkers, and these such as the capital construction, water conser-
different customs undoubtedly affected the vancy, grand canal transport, military-related
monopoly system. undertakings, and others. In a nutshell, the Jur-
Article XI is a survey o f the natural ca- chen rulers evidently made considerable pro-
lamities and human disasters o f the Jurchen gress in harnessing the traditional Chinese la-
state, and the institutions and policies o f gov- bour service system to meet the timely needs of
emment relief from the beginning to the end of the state despite some flaws and shortcomings.
the Chin dynasty. A number o f calamities and By comparison, the Chin performance certainly
disasters and their frequency have been tabu- surpassed the Liao, could rank with the Sung in
lated from the Chin shih, such as drought, 10- some spheres, and was evidently superior to the
custs and pests, famine, earthquakes, floods, Mongol Yiian whose system was riddled with
storms, bad harvests, hurricanes, extreme cold, racial and ethnic inequities and discriminatory,
epidemics, and so on. The economic and politi- haphazard practices.
cal impact of these calamities and disasters has The authors would like to acknowledge
been examined and the policies and forms o f copyright and express appreciation for permis-
state relief studied in detail. It shows that by sion to reproduce the articles in this volume to
the 12th century, when the Jurchens established the following: Harrassowitz Verlag, Wiesbaden
a sinitic state in North China, the economic im- (I, II, V, VIII); The Chinese Materials Center,
pact o f natural disasters on agriculture became Taipei (III); Mme. Nicole Goulric, École des
more evident, and the Chin rulers increasingly Hautes Études en Sciences Sociales, Paris (IV);
adopted sinitic political practices in their gov- Dohosha Publishers, Tokyo (VI); Franz Steiner
ernance. The policies and forms o f state relief, Verlag, Wiesbaden (VII); The Institute o f His-
both preventive and remedial, again closely fol- tory and Philology, Academia Sinica, Taipei
lowed the T ’ang and Sung, including water con- (IX); The Harvard Journal of Asiatic Studies
servation, extermination o f locusts, tax remis- m:
sions and loan grants, establishment of state Herbert Franke
granaries, redemption o f slaves, dispensation of
medical care and others. It appears that although
the Chin could not par with the Sung in the
administration and dispensation o f state relief,
it had a better record o f performance than the
Khitans and the Mongols.
Article X is a comprehensive study of the
organisation and utilisation o f labour service in 1Republished from the preface o f the book
the Chin state, an important arena o f its socio- with slight changes. We thank Professor Franke
economic institutions and activities, drawing for his kind permission.

REVIEWS
H u g h R ic h a r d s o n , High Peaks, Pure Earth. the most exciting task from the fifties o f this
Collected writings on Tibetan history and cul- century on. A small group o f young scholars
ture. Edited with an Introduction by Michael and students under Professor Ligeti began to
Aris. Serindia Publications, London 1998, work with this material. Under the most trying
778 pp. circumstances did w e prepare our first studies.
Two o f us, Géza Uray and myself remained in
Hugh Richardson is one of the central figures the group and began to publish. It was, how-
o f Tibetan studies in the 20th century. Born in ever, only in 1958 that I dared to send my off-
1905 at St Andrew’s, fifth in a Scottish family, prints to Richardson. Since then we have had
he graduated at Oxford, then he passed the a lively correspondence. I regret that we met
Indian Civil Service exam. He represented the only once, in London on the 13th of September
British government in 1936-40 and 1946-47 in 1964, but among the letters I keep there is one
Lhasa and from 1947 until 1950 he worked for that Richardson sent the same day to me, and
the independent government o f India. He left reflected on the different topics we had talked
Tibet in the last minute before the Chinese in- about. Most o f them were o f unclear passages
vasion reached Lhasa. He brought with himself in Old Tibetan texts. I think that almost all Ti-
valuable materials, like photos and copies of betologists have had some personal contact with
the early Tibetan inscriptions, and at the same this outstanding scholar. For them and for the
time a love o f and estimation for the Tibetan younger generation these collected papers are
culture. His first publication on the Old Tibetan a real gift.
inscriptions appeared in 1949 in the Journal of The volume contains almost all papers pub-
the Royal Asiatic Society, Bengal. The ex- lished by Richardson. His books and mono-
tremely important Old Tibetan material which graphs are not included with the exception of
was found in Dunhuang and the Tarim Basin the Tibetan Precis. This description o f Tibet
gave the publication o f the inscriptions a new was first published in 1945 in Calcutta under the
perspective. Many o f the Old Tibetan Dun- heading “Secret, for official use only” in fifty
huang and Turfan materials are copies, extracts copies. According to Michael Aris only three
while the inscriptions stand in their original o f them have survived. Part One o f the book
place since the 8th to 9th centuries. The com- contains twenty-seven papers under the com-
parison and confrontation o f the inscriptions mon heading The Royal Period and Later Leg-
and the Annals, the Chronicle and the adminis- ends. In Part Two, Historical Sites and Inscrip-
trative documents o f the Old Tibetan state was tions, nine papers are included, among others
Akadémiai Kiadö, Budapest

106 REVIEWS
on early burial grounds in Tibet and decorative publication.” Since then Shoju Inaba and Hisa-
art, the Jo-khang monastery o f Lhasa, and news shi Sato published a Japanese translation in
on recently discovered inscriptions. In Part 1964 and R. A. Stein wrote a paper on a small
Three, Later History and External Contacts, 14 part o f the book in the Mélanges des Sinologie
papers deal with the history o f the Karma-pa offerts à Monsieur Paul Demiéville, Paris (1966).
sect, the Dalai Lamas, foreigners in Tibet. The An incomplete copy o f the “N ew ” Red Chroni-
Tibetan Precis has fourteen chapters. The first cle was published in 1968 by Lokesh Chandra
is a history of Tibet down to the 19th century, [The Red Annals (recensio nova from an incom-
the last one deals with the problem in retro- plete manuscript in the Library o f Rai Bahadur
spect and prospect. It is extremely interesting T. D. Densapa), Gangktok, Sikkim, New Delhi],
how Richardson has seen the future o f the G. Tucci edited the “N ew ” Red chronicle in
Tibetan-Chinese relations in 1945. W e read: 1971. We can only hope that Richardson’s notes
“A determined Chinese attack on Tibet would on the “Old” Red Chronicle will appear as soon
be successful” (p. 618). In Part Five, Testimo- as possible.
nies and Recollections, Michael Aris collected Andrés Rôna-Tas
those papers which reveal the author in his most
personal mode as a human witness to the old
order and its collapse in Tibet. The last piece D e n is S i n o r , Studies in Medieval Inner Asia.
tells us how the Dalai Lama arrived at Lhasa as Variorum Collected Studies Series. Ashgate,
a small child and what happened in the subse- Aldershot-Brookfield, 1997, X+340 pp.
quent fifty years.
Many of the papers were published in jour- This is the fourth volume which contains col-
nais or in memorial volumes not easy to obtain. lected papers of Denis Sinor. Inner Asia and its
The detailed and interesting Introduction Contacts with medieval Europe included pieces
by Michael Aris gives a biography and an over- written between 1939 and 1975, Tanulmdnyok
view on the scholarly and personal activities of (Studies) was published in Hungarian in 1982
Richardson. as vol. 110 of the series Nyelvtudomdnyi Erie-
In a paper on one o f the successors o f the kezések and his Essays in Comparative Altaic
Old Tibetan king Glang Darma, originally pub- Linguistics appeared in the Indiana University
lished in the Etudes tibétaines dédiées à la mé- Uralic and Altaic Series as vol. 143. In this
moire de Marcelle Lalou (Paris, 1971), Rich- volume nineteen papers are grouped under three
ardson mentions that the story o f the rivalry headings: The Eastern Sphere, The Western
between the two princelings makes its ap- Sphere and Civilization. A real surprise is pa-
pearance in the Hu-lan deb-tlier (the “Old” Red per V, The Uighur Empire o f Mongolia, which
Chronicle) ascribed to the year 1346. is a first publication. The text is a somewhat
In a letter written in 1964 Richardson asked updated version prepared in 1989-1900 for vol-
me to help him to acquire a paper published in ume III of Philologiae Turcicae Fundamenta.
M ongolia on the Red Chronicle. In the Fore- This is an excellent 30-page overview o f the
word to the Sikkim edition of the “Old” Red history of the Uighurs until 840. The papers of
Annals [published in Gangtok, Sikkim (1961)], Sinor always give food for learning, sometimes
Palden Thondup Namgyal, Maharajkumar of they are provoking, or simply thought provok-
Sikkim, the President o f the Namgyal Institute ing. His critical command on the great range of
o f Tibetology wrote: “An English translation of sources and his lively style makes easy the
this chronicle done by Mr. G. Tharchin with reading o f papers that deal with otherwise very
historical notes by Mr. Nirmal C. Sinha is under complicated topics. Many readers of these col-
compilation. Mr. Hugh Richardson has very lected papers may disagree with me, but I con-
kindly agreed to check the translation and notes. sider the paper Some Components of the Civili-
Translation and notes will form Part II o f this zation of the Tiirks (6 th to 8 tl1 Century A.D.) as

REVIEWS 107
the best o f the collection. Sinor gathered some the 4th to 6th centuries, we have now good
Old Turkic features, which did not seem to be reason to suppose that a smaller part remained
Turkic so far, as they are not present in the in the East. On the other hand contacts with
language and/or culture o f the later and the pre- Iranian speaking groups e.g. with the Sakas can
sent Turks. Such features are the Eastern orien- be well demonstrated. Sinor is referring to such
tation, the so called Oberstufenzählung (13 is data that can be used in the search for Ogur and
the third o f the twenties), non Turkic personal Iranian components o f the Old Turkic Empire.
and clan names of some o f the leaders o f the Papers as The Making o f the Great Khan, the
Turks in the 6th to 8th centuries, a few words Interpreters in Medieval Inner Asia, or the
which figure in the Old Turkic runiform in- Diplomatic Practices in Mediaeval Inner Asia
scriptions but are elsewhere not present in other will interest an even broader circle o f readers.
Turkic sources, three different ancestral leg- A special field where Sinor is at home is the
ends on the origin o f the Turks. In addition group o f Western sources on the people o f the
Sinor lists those data which hint at different steppe. The historical Attila, Western informa-
types o f Turks, mostly in Chinese sources. I do tion on several groups o f the Khitans, Mongols
not think that all these “non Turkic” features as seen by the West or Latin sources on the
belong to the same source. Moreover, most o f Khanate of Uzbek offer plenty o f new insights
the statements are not valid. We find the East- into Western sources and the history o f the no-
ern orientation in all Turkic groups that were mads o f the East. Sinor is one o f the few schol-
not subject to Islamisation, Sinor himself refers ars who is able to visit sources from Europe
to Yakut, Yellow Uighur and Chuvash. The ori- and from China and use information hidden in
entation o f the tombs of Turks, where they can them with the same ease. The volume will give
be identified, is also towards the East. Words the reader plenty o f new information and fun.
like sab ‘word’ do not occur only in the “sol-
emn language of their funerary inscriptions”. Andrds Röna-Tas
The word sab is present in Buddhist texts, in
the Divan o f Kashgari, in the work o f Abu
Hayyan and in Old Ottoman texts. Chuvash G y u l a D É C S Y , The Turkic Protolanguage:
sävä ‘song, verse, text’ should also pertain to a Computational Reconstruction. Eurolingua.
sab. The derivation savci ‘go-between, prophet, P.O. Box 101. Bloomington, Indiana. 1998.
messenger etc.’ is present not only in the Old (Bibliotheca Nostratica 11.), 216 pp.
and Middle Turkic sources but also in some
modern Turkic languages as the Tatar dialects It is always a kind duty and an awarding chal-
of Tobol and Tiimen, in Kazak and Halaj. Nev- lenge to review a good book. A book is good if
ertheless the idea is good and one can only sub- it brings new material, new insights, new meth-
scribe Sinor’s conclusion “I think it is safe to odological approach. A book deserves a review
say that not all the inhabitants o f the Türk state also in case if it is a textbook, a handbook that
were Türks or even Turkic speaking, and that summarises the valid knowledge o f previous re-
not all the Turks belonged to the ruling stratum searches. But what to do with a book which is
of the Türk state” (III, p. 157). The detection o f simply bad? In most cases the editor decides not
the complicated ethnogenetic processes is more to write or not to let to write a review. 1 have
fruitful than the search for the origins o f a peo- decided to write at least a warning notice for
pie. Sinor is looking to the North and is sup- several reasons. First Gyula Décsy is a person
posing groups o f Uralic and/or Samoyed origin of reputable professional career. He got his first
as constituents of the Old Turkic Empire. At scientific degree in Hungary in Slavistics. Later
the same time I would add two others. One o f he was professor o f Finno-Ugrian studies in
them is Ogur Turkic, the other is Iranian. Not Hamburg and in Bloomington. Second, the re-
all Ogur speaking groups went to the West in construction o f the Turkic Protolanguage is an

108 REVIEWS
important topic and many new attempts have dent Uigurs” is at least arbitrary. If the fore-
been made lately to clarify the way towards its bears o f ancient Uigurs spoke Old Turkic what
reconstruction. Third, the book is written in an did the ancient Uigurs speak? What was the
easy, readable language and the danger that language o f the ancient Kipchaks, or the fore-
scholars in general and comparative linguistics bears o f the ancient Kipchaks? Then we learn
w ill accept its results on its face value is great. “[theoretically, Old Turkic should be close to
The book consists o f six chapters. After an the present-day Turkic languages or dialects
Introduction, the second chapter has the head- spoken in Xinjiang(...), East Kazakhstan and
ing Phonetics/Phonology, the third Morphol- Tuva.” Under 1.3, on page 10 we read: “The
ogy, the fourth Syntax. The fifth chapter deals main source o f reconstruction of the Turkic Pro-
with Proto-Turkic Semantics. Chapter Six is a tolanguage is Old Turkic” and “Old Turkic and
Vocabulary with a Proto-Turkic-English and Proto-Turkic represent basically the same lan-
an English-Proto-Turkic Wordlist respectively. guage form.”... Later there is a distinction made
Turkic Protolanguage is, according to Décsy, between Early (pre-Islamic) and Late Old Turk-
the ancestral common form o f 27 languages rep- ic, the latter represented by Mahmud al-Kash-
resenting ca. 110 million native speakers: and gari. In the book written by Décsy “...only
the enumeration of the languages follows with Early (pre-Islamic) Old Turkic was considered
their estimated number o f speakers and the for the Proto-Turkic reconstruction”. The reason
world-ranking of the particular language. Draw- is that the Arabic Islam influenced the language
ing the borderline between a language and differ- o f Kashgari (On the Kutadgu Bilig, a work
ent dialects is always a disputable question, but written in Karakhanid Turkic a few years earlier
important languages are missing as Afshar than the Divan o f Kashgari no words are spent).
(45,000 people in Afghanistan), Yellow Uigur One would think that until the 1 1th century Old
(5000 people in China, extremely interesting lan- Turkic, whatever this means, was a pure, un-
guage for the reconstruction o f Proto-Turkic), touched language (something which does not
Kashkay (570,000 people in Iran), Khorasan exist at all). But this would exclude the Bud-
Turkic (400,000 in Iran, the latter two languages dhist, Nestorian and Manichean texts, which
known since long, but described only recently). are included in the corpus labelled by Décsy as
On the next page we get the definition of Old Early Old Turkic, based on a misunderstanding
Turkic which according to D écsy “was a lan- o f the essence o f the Alltürkische Grammatik o f
guage spoken by a group o f Oguz tribes and Annemarie von Gabain. One has the impres-
the forebears of the ancient Uigurs in the pres- sion that Décsy has never read the beautiful Old
ent-day Mongolia (Seven-Stream-Land, see 1.5, Turkic poems and proverbs, included among
Northeast of Almaty) between ca. 700 A.D. and others, in the work o f Kashgari, neither did he
1000 to 1100 A.D.”. I cannot imagine any defi- count the Arabic elements in Kashgari, a negli-
nition o f “present-day Mongolia” which would gibly small amount.
include the Seven-Stream-Land. Later the Pri- According to Décsy the “inherited” Old-
mordial Home of the Turks is defined as: “A Turkic/Proto-Turkic vocabulary comprises close
closer definition would be the area south from to 3,400 indigenous words - the largest of the
the Upper Yenisei, so-called the Seven-Stream- lexicon o f any well-established proto-language
land and the Talas Valley” (the definition fol- as proved by the following comparison (Proto-
low s more or less the introduction of von Ga- Slavic and Proto-Germanic data estimated):
bain’s Alttürkische Grammatik, cf. Gabain: “Die Uralic, around 4000 B.C., ca. 500 words, Finno-
Inschriften stammen aus der nördlichen Mon- Ugric, around 3000 B.C., ca. 1200 words,
golei, aus dem oberen Jenissei, dem Sieben- Indo-European, around 4000/3000 B.C., ca.
strom-Land und aus dem Tal des Talas”, p. 1). 2400 words, Proto-Slavic, around 800 A.D., ca.
The claim that Old Turkic was spoken “by a 2000 words, Proto-Germanic, around 500 A.D.,
group o f Oghuz tribes and the forebears of an- ca. 1800 words. The arbitrary and unfounded

REVIEWS 109
equation of Proto-Turkic with Old Turkic makes (great majority o f the Turkish speaking popu-
o f course, things relatively easy. Neither is it lation), Egypt (once had a ruling dynasty of
defined what Décsy understands under “word”. Turkic origin) and Maghreb (never under seri-
Sometimes he does not make difference be- ous Turkic influence) under the same hat is too
tween basic words and derivations. In fact he much.
remarks on p. 92 that all words were eliminated There does exist a discussion on the history
from the corpus which appeared to be of Avars in the 9th century. But on what ground
questionable (late derivatives, words denoting Décsy claims (p. 24) that the Avars or their
relatively new concepts, most loan words and groups may have retreated to the East and built
proper names). As we shall see later Décsy has up (or just enhanced) the historically well
a very scanty knowledge on Turkic word deri- known Volga-Bulghar and the Kazar “empires”
vation. The whole material is based on A. von remains enigmatic. One does not wonder that
Gabain’s Glossar to be found in her Altturki- Décsy is mixing the On Ok ‘Ten Tribes’ with
sehe Grammatik. In addition the three (now the Onogur, the ‘Ten Ogur’ with reference to
four) existing volumes o f Sevortjan’s Etymo- Golden (An Introduction to the history o f the
logical dictionary and the Drevnetjurkskij Slo- Turkic people, Wiesbaden 1992, p. 22, where
var’ was used. To demonstrate the possible pri- no trace o f such claim is to be found).
mordial homeland o f the Turks, Décsy cites the The chapter on Phonetics/Phonology (what
names o f animals in Proto-Turkic/Old Turkic, does the backslash mean?) is based on the cor-
to which “Some Late Old Turkic words were pus o f Proto-Turkic words on pp. 9 3 -1 7 2 pre-
also added.. with no reason given. Under 4.5.2 pared for computational operation. I shall come
the names of horse and donkey are listed. Here back to the wordlist, but first let us throw a
we find adghyr (I retain the transcription o f glance on the phonology. According to Décsy
Décsy) ‘stallion’, adghyrlyq ‘stud’, äshgäk ‘don- Proto-Turkic did not have closed e and vowel
key’ and as a separate item eshgek ‘donkey’. length. He found 24 occurrences o f closed e
The second item is a derivative o f the first, the (mainly from Yenissei-Runic and from Brahmi-
fourth is the same as the third. One wonders script), but according to him these are incorrect
what the numeric relationship between the dif- denotations for ä or i and were disregarded. We
ferent groups then show. I am afraid, nothing. have ample evidence that closed lei was part of
Even less can this material be compared with the phonological system o f Old Turkic as long
lists from Proto-Uralic or Proto-Indo-European. vowels were. Décsy seems to have never heard
I shall not dwell on the historical sketch about the discussion on the closed e (J. Németh,
given by Décsy. V iew s that he claims to be Zur Kenntnis des geschlossenen e im Türki-
“generally accepted” are mostly outdated if ever sehen, Korösi Csoma Archivum I. Erf. Bd.
existed. But he even does not seem to know the 1939, pp 515-531, L. Ligeti, Sur la langue des
older views e.g. that o f J. Németh, one time Afchars d ’Afghanistan, Acta Orient. Hung. 7,
the boss of Décsy in the Institute of Linguistics 1957, pp. 115-116, here quoted the earlier
in Budapest on the primordial homeland o f the works on the closed e o f K. Foy, M. Räsänen,
Turks (Probleme der Türkischen Urzeit, in: Ana- R. Arat and K. Thomsen). Though he is follow-
lecta Orientalia memoriae Alexandri de Korös ing mostly Gabain, I have the impression that
dicata. Bibliotheca Orientalis Hungarica, Buda- even her works were not read with attention.
pestini, 1942-1947, pp. 57-102). It is true that Gabain wrote in the Fundamenta (listed by
many non-Turkic ethnic groups were Turkicised. Décsy in his References on p. 216) “In den In-
But if he mentions among them “...a s the Schriften gibt es folgende Vokale: a, ä, e, '1, i, o,
perhaps most significant, all Anatolia (present- ö, и und ü. Es gab lange und auch reduzierte
day Turkey) and North Africa (Egypt, Maghreb)” Vokale” (Das Alttürkische, in: Philologiae Tur-
(p. 18), one is tempted to give up reading the cicae Fundamenta I, Aquis Mattiacis, 1959, p.
book. Three totally different cases, Anatolia 24). The etymological dictionary o f G. Clauson

ПО REVIEWS
(An Etymological Dictionary o f Pre-Thir- the demonstrative pronoun ol, which later
teenth-Century Turkish, Oxford, 1972) is some- replaced an earlier pronoun o f the third person,
times quoted, but Décsy did not realise that will remain in darkness. It is further totally
Clauson is consequently marking that both the unsubstantiated to speak about an alternation
closed e and the long vow els are quoting Old l~r, whatever this preprotolinguistic basis may
Turkic data. The discovery o f the Proto-Turkic mean. With similar tools one can derive any
long vowels by Polivanov (K voprosu о dolgih noun from any verb and any phonetic form from
glasnyh v obscetjurkskom prajazyke, Doklady any other.
AN ser. V, 1927:4), Räsänen (Über die lange The Wordlist is full o f misunderstandings,
Vokale der türkischen Lehnwörter im Ungari- improper use o f Gabain’s Glossar, ghost words
sehen, Finnisch-Ugrische Forschungen 24, etc. On p. 124 we find qara I ‘domesticated ani-
1937, pp. 246-255), and Ligeti (Les voyelles mal’, qara II ‘black, dark’, qara III ‘people’.
long en turc, Journal Asiatique 1938, pp. 177- These are o f course not three homophonous, but
204) did not impress Décsy if he has ever read different words. The first and the third are the
these papers. The reconstruction o f the Proto- same as the second. From the expression qara
Turkic vocalism by Shcherbak (Sravnitelnaja bodun ‘the black i.e. the ordinary people’ some-
fonetika tjurkskih jazykov, Leningrad, 1970), times the qualified is lacking, the ordinary
by Serebrennikov and Gadzhieva (Sravnitel’no- people are simply called “the blacks”. Décsy
istoriceskaja grammatika tjurksih jazykov, Baku forgets that Gabain gave a Glossary to Old
1979), by Tenishev (Sravnitel’no-istoriceskaja Turkic readings and not a linguistically organ-
grammatika tjurksih jazykov. Fonetika, Moskva ised wordlist. The Wordlist contains loanwords,
1984) are unknown or disregarded by Décsy, as as e.g . jinchii ‘pearl’, which is o f Chinese ori-
are the works of G. Doerfer (e.g. Proto-Turkic: gin. The Wordlist is ill organised. According to
Reconstruction Problems, Türk Dili Arastirma- Décsy he has 14 words beginning with jat, (p.
lari Yilhgi Belleten, 1975-76. pp. 1-59), M. 108), in fact only two are listed.
Erdal (e.g. Old Turkic Word Formation, Wies- Without continuing further one can only re-
baden 1991) or L. Johanson (e.g. Alttürkisch gret that such a great endeavour has been in-
als “dissimilierende Sprache", Wiesbaden 1979) vested into a work which is unreliable and su-
to mention only some o f the most important perficial. Turkologists will not use the work, but
ones. If he had ever read any o f these works he it would be good if it would not deceive the
would have not written the follow ing sentence: non-Turkologists. The name of the series “Bib-
“It is well known from Turkic historical pho- liotheca Nostratica” is a bad omen. This is the
nology that b (</;) and m are interchangable” only reason for writing this review.
(p. 26). The conclusion on p. 27 that p had as it
allophones w, b, and m is simply a factual error. Andrds Röna-Tas
I would select from the Morphology only
one paragraph. On p. 60 Décsy writes “In predi-
cative use the 3rd person singular was ol or Présence arabe dans le Croissant Fertile avant
o, and the 3rd person plural ol or olar. This l ’Hégire. Actes de la table ronde internationale
ol (plural olar) may be connected on a pre- organisée par l’Unité de recherche associée
protolinguistic basis with dr II ‘to be, to exist, 1062 du CNRS, Études sémitiques, au Collège
to take place, to concern’ (an auxiliary high de France, le 13 novembre 1993. Textes réunis
frequency word), if we accept the possibility of par Hélène Lozachmeur. Éditions Recherche
an alternation l~r and o f a vertical vowel har- sur les Civilisations, Paris 1995, 148 pp. ISBN
т о п у (и for a interpreted as o ).” If words have 2 86538 254 0
any common sense, then vow el harmony should
mean at least harmony o f two vowels. How The appearance o f Arabs and Arabic on the
would it be possible to get from the verb dr- scene o f world history is usually associated

REVIEWS Ill
with the rise o f Islam. It is true, relatively little in texts relevant to Mesopotamia from the
is known about the origin of the Arabs, their period between the 3rd с. B.C. and the 3rd c.
language and history in earlier times. The source A.D. (pp. 7 3 -79). François Bron undertakes to
material is scanty, at times difficult to interpret, trace the vestiges o f South Semitic writing in
and it lies scattered in various languages, in not the Fertile Crescent (pp. 81-91). Michael C. A.
always easily decipherable scripts. The Unite' Macdonald discusses the origin of “Saracen"
de recherche associée (URA) 1062 o f the and certain aspects o f the Rawwäfa inscription
Centre national de la Recherche scientifique from the North-Western part o f the Arabian
(CNRS), Etudes sémitiques, organised at the Peninsula (pp. 93-101). Michel Gawlikowski
initiative o f its head, Javier Teixidor, its second gives a short sketch o f Arab presence in Palmyra
International Round Table under the title “Arab and its neighbourhood (pp. 103-108), while
Presence in the Fertile Crescent before the Pierre-Louis Gatier outlines the vestiges o f Arab
Higra” at the Collège de France on the 13th of presence in Gerasa (modern Garas in Jordan)
November 1993, in which approximately 80 and the Decapolis (Southern Syria) (pp. 1 0 9 -
specialists, both Semitists and Arabists, took 118). Fiorella Scagliarini undertakes the gen-
part. The present volume contains twelve eral survey and the correction of Dedanite and
pieces which were read at the round table, all in Lihyanite chronology (pp. 119-132). On the
French. The aim o f the gathering was to join basis o f Nabatean inscriptions discovered in the
the forces o f representatives of various fields in Northwestern part o f the Arabian Peninsula,
presenting a comprehensive, up-to-date assess- Françoise Briquel-Chatonnet discusses certain
ment of the state o f our knowledge o f the sub- aspects o f cultural symbiosis: the penetration o f
ject in question. The pieces are not aimed to be the culture o f the Fertile Crescent into Arabia
exhaustive treatments o f their subjects, but lay (pp. 133-141).
considerable weight on the presentation o f the The volume presented here gives a highly
questions and problems connected with their valuable survey o f the state o f our knowledge
subjects. o f Arab presence in the Fertile Crescent before
Daniel Arnaud presents the traces o f the the rise o f Islam and the numerous problems
Arabs in Syrian texts dating from the beginning involved in this subject.
of the 2nd millennium until the Neo-Assyrian Istvdn Ormos
era (pp. 19-22). Hedwige Rouillard-Bonraisin
discusses the presence and the representations
of the Arabs in the Bible (pp. 23-35). Émile G o t t h a r d S t r o h m a i e r , Von Demokrit bis
Puech summarises our knowledge of Arab pres- Dante: die Bewahrung antiken Erbes in der
ence in the manuscripts o f the Cave o f Letters arabischen Kultur. Olms Studien, Band 43.
in the Judean Desert (Dead Sea Scrolls) (pp. Georg Olms Verlag, Hildesheim-Zürich-New
37-46). Felice Israel offers a review o f Arab York 1996. X + 558 pp. ISBN 3-487-10030-4 /
proper names traceable in Syrian and Palestin- ISSN 0175-9930
ian inscriptions. In this context the author gives
an intriguing summary o f all that can be known The present volume assembles 58 articles and
of the final wäw in proper names (cf. Arabic reviews by Gotthard Strohmaier, one o f the
cAmr[wJ), its spatial and temporal localisation, leading authorities in the field o f Graeco-Ara-
its pronunciation, origin and value, and the bica, the assimilation o f Greek heritage in
problems connected with this subject (pp. 4 7 - medieval Islam. Strohmaier - now also a hon-
57). André Lemaire discusses the ancient Ara- orary professor at the Freie Universität, Ber-
maic inscriptions in the oasis o f Taymä’ in the lin - has been on the staff o f the project Corpus
heart of Arabia (pp. 59-72). Basile Aggoula Medicorum Graecorum (CMG) at the Berlin
presents the references to Arabia and the Arabs Academy o f Sciences and Humanities since

112 REVIEWS
1958.' This project was launched in 1907 on ions, astronomy, the history o f arts and the
the basis o f international co-operation with the history of Arabic studies.
aim o f providing the scholarly world with up- The volume is divided into the following
to-date critical editions o f Greek medical texts, chapters: I. Arab testimonies o f Greek philoso-
first and foremost those o f Galen o f Pergamon, phers (8 articles, 2 reviews); II. Medical history
the greatest representative o f antique medicine. o f Antiquity (8 articles); III. The history of Arab-
Strohmaier’s task has been to cultivate the Islamic culture and religion (7 articles); IV.
availability of Arabic versions o f Galen’s works. The work o f the translators o f Baghdad (8 arti-
It is well known that within the framework of d es); V. The history o f Arab philosophy (7 arti-
a large-scale translating movement, by the 10th d es); VI. The history o f Arab medicine and
century A.D. a considerable part of Galen’s sciences (6 articles, 1 review); VII. The world
oeuvre - among other things - had been ren- o f Islam and mediaeval Europe (7 articles);
dered available in remarkable Arabic transla- VIII. The history o f Arabic studies and the edi-
tions by the circle of Hunayn ibn Ishaq. These torial tasks o f the present (5 articles).
translations deserve our interest not only in The papers contained in this volume cover
those cases where the Greek original of a cer- a wide variety of topics. What unites most of
tain work is now lost, but also when the Greek them is that in one way or another they deal
original is extant, because the Arabic transla- with the transmission o f Greek texts in Arabic
tions were made on the basis o f older, and usu- and with the reception o f antique sciences in
ally better Greek manuscripts than those extant the Islamic world. This is a much neglected field
today. In addition to his own excellent edition and Strohmaier’s pioneering activities and results
o f a formerly unknown work by Galen, De ho- have shown clearly that remarkable pieces of
moeomerum partium differentia, which is extant information concerning Greek philosophers still
only in Arabic translation, Strohmaier super- lie scattered in Arabic works and await discov-
vised the edition of Arabic translations of other ery by an expert hand. Several papers are de-
works by Galen, as well as contributing to the voted to the collections o f maxims (gnomolo-
Greek volumes the relevant data from the Ara- gia) and in addition to interesting results of
bic tradition of the texts in question. A consid- detail Strohmaier has contributed considerably
erable part of his scholarly research activities is to the theoretical problems o f the investigation
also devoted to this field o f studies. Owing to o f the material contained in relevant Byzantine
the close connection between medicine and phi- and Arabic collections.
losophy in Antiquity and in the Middle Ages, it The papers on the activities and achieve-
was only natural that Strohmaier’s interest soon ments o f Hunayn ibn Ishaq and his circle consti-
turned towards the various problems of phi- tute the central part, the core o f this volume, and
losophy, too, both in the Graeco-Roman and it was what most captured the interest o f the pre-
the Arab-Islamic world - an interest he seems sent reviewer. Questions on the history of medi-
to have retained up to the present. In addition cal terminology (e.g. Dura mater, Pia mater) are
to these two central subjects, he is of course also tackled. Strohmaier has conducted a thor-
deeply interested in the problems of textual ough investigation o f the question o f the transla-
criticism, yet the horizon o f his erudition also tors’ treatment of the names of pagan deities.
extends to areas such as the history of relig- The papers dealing with various aspects of
the activities o f al-Bïrünï and Avicenna clarify
important details of the oeuvre o f these out-
1The former Prussian Academy continued standing philosophers.‫־‬
its activities in the German Democratic Repub-
lic/East Germany under various names, and af-
ter the reunification of Germany it reformed as ‫ ־‬Strohmaier has also published a ehre-
the “Berlin-Brandenburg Academy of Sciences stomathy o f extracts o f al-BIrtini’s works in
and Humanities” in 1993. German translation with commentaries: Al-
REVIEWS 113
Five contributions deal with the widely dis- and who is at the same time an accomplished
cussed and controversial problem of the even- classical philologist.
tual Oriental connections o f Dante’s Divina “Galen o f Pergamon and the Editorial Pro-
Commedia. In contradistinction to Miguel Asin gram of the Corpus Medicorum Graecorum”
Palacios and Enrico Cerulli, Strohmaier sees (in English) is a fascinating sketch throwing
Dante’s source in Abraham Ibn Ezra’s Hay ben light on the history of this scholarly under-
Meqiz, which has passed on - perhaps without taking.
Dante’s being aware o f it - some basic tenets All the articles and reviews contained in the
o f Avicennian philosophy to the Divina Com- volume under review attest to a solid craftsman-
media. On the basis o f this insight, certain de- ship and an unusually broad horizon o f learn-
tails o f Dante’s immortal work receive a new ing. It is also fascinating to observe how the
and plausible interpretation and explanation. occupation with questions o f detail begin to
Two papers and a review are dedicated to yield to more general, theoretical questions in
the interesting and complicated figure o f the later years: having been involved for long years
founder o f Arabian studies in Germany, Johann with the truly astonishing scientific achieve-
Jacob Reiske, adding new viewpoints to the ments of mediaeval Islam, Strohmaier puts the
still unwritten biography of this outstanding, yet question why such outstanding beginnings came
most unhappy, scholar of Arabic and Greek.’ to a standstill, and why, in subsequent centuries,
“Paul Kraus and the Research o f the Gabi- the Islamic Orient was left far behind by a rela-
rean Corpus” (in German) is a relatively short tive latecomer in this field, the West.
but important article offering some interesting A complete bibliography o f Strohmaier’s
details on the troublesome life and premature publications rounds off the volume.
death o f an eminent scholar along with impor- Arabists, Islamists and historians o f medi-
tant viewpoints in a scholarly debate concern- cine are greatly indebted to Olms Publishers for
ing the dating o f the reception o f the Greek the publication o f this remarkable collection o f
legacy in Islam. articles and reviews, which appeared originally
“Textual Criticism and Editorial Technique in journals, Festschrifts and commemorative
in Arabic Philology” (in German) contains volumes, many o f which are inaccessible to
stimulating remarks about a much neglected most readers.
area o f current problems from the pen o f an Istvdn Ormos
outstanding scholar actively involved in the
practical cultivation o f the field in question,
WOLFGANG R e u s c h e l , Aspekt und Tempus in
der Sprache des Korans. Leipziger Beiträge zur
Bïrünï: In den Garten der Wissenschaft. Aus- Orientforschung, Band 6. Peter Lang, Frankfurt
gewählte Texte aus den Werken des musli- am M ain-Berlin-Bern-New York-Paris-Wien
mischen Universalgelehrten, übersetzt und er- 1996, 308 pp. ISSN 0942 2323 / ISBN 3 631
läutert. Leipzig 1988; 2. verbesserte Auflage. 47680 9
Leipzig 1991 (Reclam-Bibliothek 1228). Stroh-
maier’s acquaintance with al-Bïrûnï’s works The verb systems o f most Semitic languages
along with his familiarity with the highly dis- are based on the opposition o f two basic forms
parate subject matters dealt with by the great commonly called perfect and imperfect. It has
Khorezmian qualify him as perhaps the best
been the subject o f a long controversy what it
expert o f our time on this most original thinker
is exactly that they designate: tense or aspect.
the Islamic world has ever produced.
’ On him see in general Johann Fiick: Die The author o f the present work under review
arabischen Studien in Europa bis in den An- has undertaken the task o f investigating this
fang des 20. Jahrhunderts. Leipzig 1955, pp. question on a clearly defined corpus of Arabic,
108-124. the Quran.

114 REVIEWS
Verb aspect plays a significant role in Indo- haustive catalogue o f Quranic sentence struc-
European languages and has been the subject tures enables him to draw important general
o f extensive research especially in the Slavonic conclusions shedding light on the function o f
languages:' it may therefore be relevant that the the verb system of Quranic Arabic. His most
author has also pursued Slavonic studies (with important general conclusion is that the two
a focus on Russian) along with Arabic and basic verb forms originally denoted aspect only,
English. The intricate interrelationship o f as- but in the course of linguistic development they
pect, tense, mood and verb character (Verbal- have begun to assume a temporal meaning, too.
charakter) is well known; however, a consider- This stage is reflected in the Quran. It means
able part o f this discussion has been marred by that in certain situations/structures the verbs
an arbitrary use of the relevant terms.2 In order have retained their original grammatical func-
to avoid confusion, Reuschel straight away tion, while in others they have already begun to
gives an exact definition o f what he means by have a temporal meaning. This process is, how-
aspect, tense, mode o f action (Aktionsart) and ever, still in its early stages in the given corpus,
verb character. e.g. it is shown convincingly that the composite
Without entering into a discussion of the forms капа fa cala and капа y a f alu in the Quran
history o f the study o f verb aspect, the author have not yet fully assumed the grammatical
makes clear his point o f departure in regarding meaning so familiar to us from modern Arabic.
the tw o basic verb forms in opposition as forms Contextual factors o f course play an important
o f aspect: cursive and constative. However, since role in this extremely intricate interrelationship.
events take place in time and the hearer/reader The practical lesson that the average student of
usually experiences no difficulty in establish- Arabic will no doubt draw from this most
ing the sequence of events in time and their useful book is - even if he is not interested in
relationship to the present, the author believes the long lists o f linguistic structures with rele-
that the temporality of statements can be ascer- vant comments on the usages o f verb forms in
tained with a sufficient degree o f exactness. His them - that he has to be extremely careful in
aim in the present work is to find out the rela- the interpretation o f verb forms and shall weigh
tionship between the temporality of a state- all possibilities carefully, in addition to those
ment, its relationship to the present and the first occurring to him.
given aspectual form o f the verb occurring The investigation under review deals with
in the utterance/sentence in question. In this its subject exhaustively, within the limits de-
method the contrasting o f verb structures in fined right at the beginning. I found its state-
Arabic with those in other languages, mainly ments and conclusions to be convincing in gen-
German o f course, plays a key role. For this eral. O f course, it is not always easy to decide
purpose he investigates minutely ca. 10,000 ex- whether temporality appears from the context
amples in ca. 300 sentence structures occurring in a given sentence or if it is already signified
in the Arabic of the Quran. The behaviour of by a sufficiently developed verb form itself,
the two aspectual forms in this very nearly ex- and 1 do not think that the pitfalls of circular
reasoning have always been avoided. The re-
1 See for instance A. V. Isacenko, Die russi- peated formulations o f certain statements in
sehe Sprache der Gegenwart. Teil I, Formen- different parts o f the work are not always wholly
lehre. Halle (Saale) 1962, pp. 347-352 (§§ consistent, and thus the careful reader will do
2 0 2 -2 0 3 ), 414-418 (§220) and Yuriy Serge-
well to read the complete work from beginning
yevic Maslov, Ocerki po aspektologii. Lenin-
to end rather than merely use it as a work o f
grad 1984. Cf. also John Lyons, Introduction
to Theoretical Linguistics. Cambridge 1968, reference, picking out single statements. The
pp. 304-317. work abounds in important and illuminating
' See for instance Anatol Mirowicz, Die As- observations and the method chosen is rigidly
pektfrage im Gotischen. W ilno 1935, pp. 1-8. adhered to. I think that the correctness o f Reu-

REVIEWS 115
schel’s assumptions is proven by the fact that the transcription o f the Arabic quotations is
he has arrived at unanimous results by the rigid unacceptably excessive. My impression is that
application o f a consistent method based on the the Quranic quotations are written in Arabic
assumptions in question. It would, however, be script - probably without vowel marks - in the
interesting to carry out a “counter-proof’ by original, and it was from this that the transcrip-
using different assumptions, i.e. assuming that tion into Latin script was then made, extremely
the basic verb opposition is that of tenses in- carelessly - and perhaps with insufficient knowl-
stead o f aspects. What would the results then be? edge - , and the proofs were never checked. O f
This work will certainly not represent the course, most o f these misprints and mistakes
last word on the subject.’ Important viewpoints and some om issions o f prepositions can be
have been deliberately ignored, or taken for an recognised and easily corrected by the reader,
invariable constant in the present work: the and where this is not the case there the exact
author himself points out the importance and number o f the surah and the verse will help the
absolute necessity of future research into the reader in finding the correct version, so this
relationship between aspect/tense and the se- negligence and carelessness - luckily enough -
mantic structure/semantic components o f a given does not spoil the whole work, and a tolerant
verb (Aktionsart/Verbalcharakter), or into the reader will be able to draw considerable profit
use o f the participle, especially with reference from it. Had the book not been so interesting
to the semantic structure/semantic components and absorbing, the present reviewer would have
o f the given verb. thrown it aside after the first few dozen mis-
The present work was the Habilitations- prints and mistakes!
schrift o f the author submitted to the University The author occurring most frequently in the
of Leipzig in 1969.334 The editors state that “the references is Fleischer (with page numbers), but
present version of the book follows the original he is missing from the Bibliography. Of course,
text in all details” (... die vorliegende Fassung the full reference is Heinrich Leberecht Fleischer,
des Buches sich in allen Details an die Ur- Kleinere Schriften. Leipzig (1885-1888). (Re-
schrift hält ... p. 16). This simply cannot be print Osnabrück (1968)) Volume I. It is odd to
true: the number of misprints and mistakes in find Père Anastase listed in the Bibliography
under Père as if it was his family name: he
3 See for instance the statement of the pro- should be listed under his first name, Anastase.
found scholar, Robert Hetzron, with a wide The Introduction mentions that Wolfgang
knowledge o f Semitic languages, concerning Reuschel supervised a considerable number of
Biblical Hebrew: “It seems that the archaic sys- Ph.D. theses. I am sure that the scholarly world
tern may be reduced to a dual opposition of two
would be most grateful if Peter Lang Publish-
tenses (the traditional label ‘aspect’ for these is
unjustified and rests on indefensible argu- ers could be persuaded to publish at least some
m ents)...” In: The World’s Major Languages. o f these, especially those dealing with closely
Edited by Bernard Comrie. London-Sydney related subjects.
1987, p. 697. For the common ancestor o f the I am convinced that the work under review
Semitic languages he also posits tenses; ibid. represents an important contribution to our
pp. 660-661. understanding o f the functioning o f the verb sys-
4 A brief article o f the same subject was tern of one of the Semitic languages in a given
published in 1968 already. See Wolfgang Reu- period, while at the same time it will no doubt
schel, Wa käna llähu ‘oilman rahiman. In: Stu-
also stimulate fruitful discussions. Let us hope
dia Orientalia in memoriam Caroli Brockel-
mann. Halle (Saale) 1968, pp. 147-153 (= Wis- that similar studies will appear about other pe-
senschaftliche Zeitschrift der Martin-Luther- riods, too: they will shed more and more light
Universität Halle-Wittenberg. Gesellschafts- und on the development o f Arabic in the course o f
Sprachwissenschaftliche Reihe, Heft 2/3, Jahr- its long history.
gang XVII, 1968.) Istvdn Ormos

116 REVIEWS
A n d r é R a y m o n d , Le Caire. Histoire des Gran- detailed analyses based on an impressive array

des V illes du Monde. Librairie Arthème Fayard, o f relevant informations, in certain cases even
[Paris] (1993, 428 pp.) ISBN 2-213-02983-0; challenging the correctness or the acceptability
R o b e r t M a n t r a n , Histoire d ’Istanbul. Histoire o f certain transmitted and generally accepted
des Grandes Villes du Monde. Librairie Ar- facts. The factual richness never becomes tedi-
thèm e Fayard, [Paris] (1996, 382 pp.) ous: the attractive conversational method o f ar-
gument allows the reader to partake in reaching
In 1993 the editors of Fayard launched a new the conclusions. It may com e as a surprise to
series entitled Histoire des Grandes Villes du some readers that the Ottoman period is not
Monde, which aimed to give concise présenta- regarded as an era o f stagnation, decay and ruin
tions o f the history and development o f various in the history o f Cairo, which was abruptly ter-
world metropolises. The Middle East is well minated by the French Expedition, thus giving
represented in the series: from among the six free rein to the explosive development and
titles so far published two are devoted to two growth o f modem Egypt. The author proves, in
major centres of Islamic Civilisation: Cairo and a well-researched and documented chapter, that
Istanbul, the remaining titles being Cyril Buffet: the development o f Ottoman Cairo represented
Berlin (1993), Bernard Bressière: Madrid a direct and large-scale continuation o f trends
(1996), Serge Gruzinski: Mexico (1996), and already perceptible in Mamluk times. In view
Wladimir Berelowitch-Olga Medvedkova: of the depth o f argument, the richness of the
Saint-Petersburg (1996). It is certainly difficult factual detail, and the comprehensiveness of
to g iv e a concise account o f such a multi-fac- the present work, we may safely say that the
eted subject in a way that will fulfil the expec- work under review can now indubitably be re-
tations o f the specialist as well as those of the garded as the up-to-date, authoritative work on
educated and interested general reader - as the history o f the capital o f Egypt and one o f
seem s to be the aim o f the series, though this is the greatest centres o f the Islamic world, thus
nowhere stated explicitly. The volume dealing in many ways replacing such well-known ear-
with Istanbul seems to incline more towards the Her titles as Stanley Lane-Poole’s The Story of
expectations of the general reader. It places the Cairo (2nd edition: J. M. Dent, London (1906);
development of one o f the most enchanting cit- also available in Arabic translation: Sirat al-
ies o f the world within the framework o f the Qähira, targamat Hasan Ibrahim Hasan, cAli
main currents and events o f Byzantine and Ot- Ibrâhîm Hasan, Idwâr Halim; Maktabat an-
toman history, but without getting bagged down Nahda al-Misriya, al-Qahira (1950) or Janet L.
in tedious and tangential details - in keeping Abu-Lughod’s Cairo - 1001 Years of the City
with this method, the book has a minimum of Victorious (Princeton University Press, Prince-
commentary and footnotes. At the same time ton (1971)). Two similar titles have appeared
various lists, glossaries and a basic bibliogra- since the publication o f the present work.
phy render this work useful for the reading Though beautifully produced, Ayman Fu’äd
public. Sayyid’s At-Tatawwur a l-‘Umräni li-Madlnat
The volume on Cairo, on the other hand, al-Qähira mundu nas’atihä wa-hattä al-äna
attains the rare achievement o f keeping a per- (ad-Där al-Misriya al-Lubnäniya, al-Qähira
feet balance between the exigencies o f strict 1997/1417 A.H.) cannot be regarded as a com-
specialism and readableness, thus producing a petitor: it offers merely the outlines, a sketch o f
book that is an absorbing piece o f reading and the history o f the great Islamic metropolis. Paolo
at the same time an excellent scholarly work La Greca’s II Cairo - Una metropoli in tran-
from the pen of one of the leading experts on sizione (Officina Edizioni, Roma (1996) con-
the history of Cairo, and certainly the leading centrâtes - as its title suggests - more on the
expert on the urban history o f the Arab-Islamic problems of modern urban planning and the
world in Ottoman times. The author presents development o f contemporary Cairo - subjects

REVIEWS 117
also treated at sufficient length by Raymond - elsewhere - handwritten copies, photos and
thus the historically-oriented reader faces, in photocopies. A number o f them were commis-
fact, no difficulty o f choice. sioned by their respective owners on the spot,
The lucid and perspicuous sketch maps in others they photographed or copied in their
André Raymond’s Le Caire are an important own hands. The works are usually rather short
asset. and the manuscripts are often defective. Since,
Istvdn Ormos however, we possess rather scanty documen-
tary material in this field, the inclusion o f all
the available material is wholly justified. The
E w a l d W A GNER, Islamische Handschriften aus system o f the catalogue follow s that adopted by
Äthiopien. Afrikanische Handschriften, Teil 2. Ahlwardt in his monumental catalogue o f the
Verzeichnis der Orientalischen Handschriften Arabic manuscripts o f the Royal Berlin Li-
in Deutschland. Im Einvernehmen mit der brary:‘ works - and not manuscripts - are de-
Deutschen Morgenländischen Gesellschaft be- scribed according to subject matter following
gründet von Wolfgang Voigt. Weitergeführt Ahlwardt’s classification.
von Dieter George. Im Aufträge der Akademie I. General subjects/Public instruction (No.
der Wissenschaften in Göttingen herausgege- 1); II. The Quran/Commentaries and glosses
ben von Hartmut-Ortwin Feistei. Band XXIV, (No. 2); III. Tradition/Single traditions (No. 3);
2. Franz Steiner Verlag, Stuttgart 1997, XIX + IV. Dogmatics/The Muslim creed (Nos 4-6);
200 pp., 2 plates. ISBN 3-515-0700-0 V. Mysticism, religious orders and the venera-
tion o f saints: A. Lines o f spiritual ancestors in
The present catalogue, under the title “Islamic religious orders (Nos 7 -8 ), B. Lists o f members
Manuscripts from Ethiopia”, represents Part o f religious orders and saints (Nos 9 -11),
Two o f the volume “African Manuscripts” from C. Geographical distribution o f religious orders
the distinguished series “Register o f Oriental (No. 12), D. Mystical poetry (Nos 13-14); VI.
Manuscripts in Germany”. It is in fact a cata- Religious observances (Acts o f devotion):
logue of manuscripts in Arabic script from A. Prayers, invocations, praises and testaments
Ethiopia. It contains the description o f 199 (wasäyä) (Nos 15-98), B. Pilgrimage (No. 99);
works from 88 manuscripts in various lan- VII. Superstition and magic (Nos 100-105);
guages: 139 in Arabic, 29 in Harari, 28 in Silt’i VIII. Law: A. Court decisions (Nos 106-109),
(an East Gurage language) and one in Amharic. B. Lists o f judges (No. ПО); IX. Grammar
Two works are written in more than one lan- (No. I l l ) ; X. Belles-lettres: A. Poetry. 1. Ara-
guage. The manuscripts belong to two collec- bic poetry (Nos 112-113). 2. Harari poetry
tions: 1. The legacy o f the German Orientalist (Nos 114-115). 3. Silt’i poetry (Nos 116-121),
and diplomat Hans Martin Schlobies, the rele- B. Prose. 1. Short stories (Nos 122-138). 2.
vant items from which he seems to have ac- Edificatory literature: Collections o f proverbs
quired around 1926, and which is now pre- (Nos 139-142); XI. History: A. Auxiliary dis-
served in the Archives o f the Berlin-Branden- ciplines. 1. Genealogy (Nos 143-148). 2. Lists
burg Academy of Sciences and Humanities; 2. o f rulers (Nos 149-162), B. Universal history
The private collection o f the author o f the (No. 163), C. Local history (Nos 164—194), D.
present work under review, Ewald Wagner, Biographies (Nos 195-198).
which he acquired in Harar between 1966 and
1972, and which has been preserved in the
Berlin State Library since 1996. The manu-
1Wilhelm Ahlwardt, Verzeichnis der arabi-
scripts seem to be o f relatively late origin: all selten Handschriften der Königlichen Biblio-
the dated manuscripts were written in the 20th thek zu Berlin, I-X . Die Handschriftenver-
century. There are numerous copies among the zeichnisse der Königlichen Bibliothek zu Ber-
manuscripts copied from manuscripts preserved lin, VII-IX, XVI-XXII. Berlin 1887-1899.

118 REVIEWS
Some works in Silt’i still await precise pile, as a final result o f his investigations, a Chi-
description because in the absence o f a reliable nese Buddhist Dictionary. I wonder how many
native informant the compiler o f the catalogue volumes it would consist of, if the glossary o f
was unable to determine the exact contents of this single text extends to almost seven hun-
some manuscripts. dred pages.
In addition to detailed codicological de- First of all, it is a very handy book, for the
scriptions and ample references to identical or entries are arranged according to the order of
similar manuscripts, the location of the origi- Hanyu Pinyin, and not according to stroke
nais in the case of copies, the interested reader order as in several Buddhist Dictionaries. The
will find useful indices in the work under re- entries include 1. the Chinese pronunciation
view , in accordance with the general scheme of transliterated in pinyin with tones, 2. the Eng-
the series. Text editions - where available - are lish translation, 3 -4 . the location in the Hanyu-
also indicated. dacidian and Daikanwajiten, and the name of
This most welcome work will help to shed the oldest text (or period) cited in them, 5. the
more light upon the rather nebulous subject of location of the entry in the Taishö edition,
the general nature and the history o f Islam in 6. the sentence(s) in which the entry occurs,
Ethiopia, while also offering some interesting 7. the types of phrase, namely prose, verse or
material to Ethiopian linguists. mantra, 8. the corresponding Sanskrit word
Istvan Ormos found in the Kern Nanjio edition o f the San-
skrit version o f the Lotus Sutra, 9. the corre-
sponding Chinese word and its location in
A Glossary of Dharma-
S e is h i K a r a s h im a , Kumarajtva’s translation o f the Lotus Sutra. It
raksa ’s Translation o f the Lotus Sutra. Biblio- is misleading at the first sight that, though the
theca Philologica et Philosophica Buddhica I. mark o f the absence o f the word (-) is given,
The International Research Institute for Ad- the page numbers o f the Hanyudacidian and
vanced Buddhology, Soka University, Tokyo Daikanwajiten, where the entry should be if it
(1 9 9 8 ,1-XXXV + 698 pp.) were included, are added.
It is marked where Dharmaraksa’s transla-
With the rapid growth o f Buddhist texts avail- tion is assumed to be the result o f the trans-
able electronically on internet, the old type of later’s misunderstanding o f a Sanskrit word. He
indexes, i. e. Index to the Taishö edition of Chi- checks the entries against those found in other
nese Buddhist Canon, becomes outdated, for editions of Chinese Buddhist Canon and San-
the entries can be searched with the computer skrit manuscripts and fragments from Central
easily. Seishi Karashima’s Glossary is a good Asia. Those who are not familiar with Hanyu
example o f how to collect all information avail- Pinyin will find the four-corner system index,
able about a text with a computer in one single the radical index and the Japanese reading sys-
book. tem index very useful. The Hanyu Pinyin sys-
The author selected around four thousand tern index seems to be superfluous, as the en-
words in Dharmaraksa’s rendition o f the Lotus tries are arranged according to it.
Sutra (translated in 286) to compile this glos- As only a few publications were devoted to
sary. His main aim was to facilitate and pro- the study of vernacular in early Chinese Bud-
mote the study of neologism and vernacular dhist texts, this glossary is welcome and needed
expressions in Chinese Buddhist translations. for the further investigation o f the reason for
He focused on terms that cannot be found in vernacular elements could filtrate into Buddhist
Chinese Classics before Dharmaraksa’s transla- texts. Victor H. Mair acknowledges that the
tion. As he states in the preface (VIII), he also verbal characters o f original Buddhist scripts
plans to make similar glossaries to other earlier might have supported the spread o f vernacular
Chinese Buddhist translations in order to com in Chinese Buddhist texts, but he suggests that

REVIEWS 119
the process o f the translation played a greater The present bibliography lists primary and
role in it. He assumes that the foreign transla- secondary sources mainly about native Chinese
tors’ inability to master the wenyan led to the Jews, and the early history o f Jews in tradi-
inclusion of baihua elements into these Bud- tional China. It also includes sources in Chi-
dhist texts.1 The comparative study o f Chinese nese about Jews and Judaism, and references to
and the original Sanskrit terms can shed new China by Jewish writers. Leslie’s book does
light upon the extent to which the application not include works about Western Jewish com-
o f vernacular can be attributed to the colloquial munities in Shanghai, Harbin or Tianjin. The
features o f the Buddhist Canon. Seishi Karashi- study o f the modern history o f Jews in China
ma’s Glossary certainly facilitates the research (Sephardic Jews went to China from the Middle
of this kind. East after the opium wars in the 19th century,
Imre Hamar followed by the arrival o f Jewish refugees from
Russia and Europe in the 20th century, espe-
daily during the Nazi terror) entails such a vast
D o n a l d D . L e s l i e , Jews and Judaism in Tradi- amount o f information, that it is, as such, be-
tional China. A Comprehensive Bibliography. yond the scope of this work.
Monumenta Serica Monograph Series XLIV. The author gives a short summary o f the
Monumenta Serica Institute, Sankt Augustin - history o f Chinese Jews, then presents an over-
Steyler Verlag, Nettetal 1998, 291 pp. view o f the sources, and lists all major archives
and libraries - public and private, Chinese and
The study o f the Chinese Jews has a history of foreign - where archival materials can be found.
almost three hundred years. Following the first At the end of the preliminaries there is a list o f
visit in 1605 o f a Kaifeng Jew named Ai Tian 13 basic works which are referred to through-
to the Jesuit missionary Matteo Ricci in the out the whole book.
capital, and subsequent to Ricci’s first report to Leslie systematically makes use of short
Rome, Europeans - both religious and laymen - annotations, in the case o f primary sources he
have always been curious about Jews in China. analyses the reliability o f the source, and in the
Several travellers visited Kaifeng, the only place case o f secondary sources he evaluates the
on Chinese soil where Chinese Jews them- work, and to help the reader’s orientation, Leslie
selves left any records. Missionaries tried to marks the most important ones with a series o f
find a truly ancient Hebrew Bible, and were asterisks numbering one to four, with four being
interested in the names for God used by the the most important.
Kaifeng Jews. Many scholars - Jewish and Part One deals with the primary topics and
non-Jewish - examined their books, made cop- sources. The first and most important topic is
ies and rubbings o f their stone inscriptions and the Kaifeng community. The list o f sources in
translated their written documents concerning Section A starts with Chinese steleae and in-
the everyday life o f the community. scriptions from Kaifeng, and continues with the
This volume is a milestone in the study o f Chinese-Hebrew Memorial Book and other
the Chinese Jews. D. D. Leslie in his annotated Hebrew writings, such as books o f the Torah,
bibliography sums up the results and achieve- prayer books and other manuscripts from the
ments of Chinese and Western scholarship from Synagogue in Kaifeng. Section В lists Chinese
the first encounters in the 17th century right up sources, references to Jews and Judaism in
to the present. early Chinese writings, and works in Chinese
relating to the Kaifeng community. Section C
' See Buddhism and the Rise o f the Written sums up the sources o f Western Jewish links.
Vernacular in East Asia: The Making o f Na- The most extensive section of Part One is
tional Languages. JAS 53, no. 3 (August 1994), that o f foreign reports, from the earliest Arab
pp. 709-719. sources to Marco Polo in the 13th century and

120 REVIEWS
the Jesuit reports from the 1600s until the im- Juifs de Chine (Roma-Paris, (1980), but also
perial decree o f 1724 forbidding foreign mis- quotes the relevant section from Löczy’s book.
sionaries to leave Beijing and Canton. Other The General Index at the end o f the book
foreigners visited Kaifeng only after 1850, but was compiled by Leslie and Roman Malek,
at that time what they found was a community editor o f the Monumenta Serica Monograph
in decline. In 1900 Western Jews living in Series.
Shanghai organised a Society for the Rescue of This is an extremely useful book and a true
the Chinese Jews, but although each member summary o f the sources about those few hun-
offered at least one dollar per month for the dred Chinese Jewish families who lived in China
cause, they could not rebuild the Synagogue. for several centuries.
The information collected by visitors after Péter Vdmos
1949 can only be called “descriptions o f the de-
scendants”.
Part Two lists major bibliographies, collec- L a m b e r t S c h m i t h a u s e n , Maitrï and Magic:
tions, key journals and 166 basic works and Aspects of the Buddhist Attitude toward the
authors in chronological order. Dangerous in Nature. Wien 1997. Österreichi-
Part Three gives a comprehensive list of sehe Akademie der Wissenschaften. Philoso-
secondary works listed by author. There is a phisch-historische Klasse. Sitzungsberichte 652.
list o f books and journals in Western languages Band, 76 pp.
in alphabetical order, followed by an index by
language, and publications in Chinese and The booklet under review shows the lasting
Japanese. There is only one article mentioned interest of the author in the various attitudes o f
under Hungarian: a translation o f Edward Isaac Buddhism toward environment and in broader
Ezra’s work in the Fôldrajzi Kôzlemények sense nature. As such it is a fine piece in a
(Vol. 33, January 1905, pp. 1-12). The article series o f writings on this topic (cf. Schmithau-
mentions the visit o f the Hungarian engineer sen 1991a and 1991b). The present theme is
Jenô Cholnoky in Kaifeng in 1898. Neverthe- rather delicious: it touches upon such questions
less, this was not the first Hungarian record of as the Buddhist attitude toward poisonous
Chinese Jews. In 1886, the great Hungarian snakes as well as dangerous phenomena o f na-
geographer Lajos Loczy, a member of the tu re.
1877-1880 expedition to China led by Count The author’s way o f treating these issues is
Béla Széchenyi, in his book Description of the highly sympathetic and stands to reason espe-
Natural Relations and Countries o f the Chinese d aily if his readers feel a responsibility for the
Empire wrote about Chinese Jews supposedly preservation o f nature “in a state o f mutual
based only on the works o f William Alexander peace and friendship with and within nature.”
Parsons Martin. (Loczy Lajos: A Khinai Biro- (p. 69.)
dalom természeti viszonyainak e's orszdgainak Schmithausen presents his ideas on a very
leirdsa. Budapest (1886, pp. 270-272.) Al- wide base of primary sources of which he has a
though in Leslie’s bibliography it would not superb command. The study is written with al-
deserve a single asterisk, it is worth mentioning most a maximum o f reference material includ-
another article about the Kaifeng Jews in Hun- ing some little known texts, too. His interpréta-
garian. Pal Miklds in his article The Kaifeng tions are highly critical and at the same time
Jews (A Kaifeng-i zsidok), first published in imaginative and thought provoking. For the
1993 (in: “Ex invisibilibus visibilia" - David purpose of expounding the changing Buddhist
Katalin emlékkônyv. Budapest (1993); second attitude toward snakes he uses the Upasenasûtra,
publication in: Tus és Ecset. Budapest (1996, a part o f the Samyuktägama and the Khandha-
pp. 159-169), which relies on Dehergne-Leslie: paritta. From these texts emerges a coherent

REVIEWS 121
picture: while arhat Upasena receives snake In the exhaustive review o f all respective
bite with perfect equanimity there is an another texts I feel the need o f one source only. It is the
strand, too. According to that one may protect Garudopanisad, which is traditionally listed
himself as long as he does not violate the prin- among the Ätharvana Upanisads and dates
ciple o f not killing. This can be accomplished from the second half o f the first millennium
by the cultivation of friendliness (mettä, maitrî). A.D. (cf. Varenne 1972 and Wojtilla 1975).
Schmithausen raises the question whether it is The text contains the following remarkable sen-
understandable to Western persons who are ac- tence: “O snake go away! Happiness to you!
customed to the extirpation o f dangerous ani- Go away о great poisonous! О т salute, we
mais. He thinks that any explanation necessar- worship the nine familied snake. We praise him
ily needs further investigation for the possible who has poison-tooth. So let he who creeps
roots o f this approach. Doing so he succeeds in protect us. О т peace, peace, peace.”
pointing out the antecedents o f this approach in Altogether, I warmly congratulate Prof.
the Atharvaveda and the Satapathabrähmana Schmithausen for his study which I recommend
(pp. 25-33). The solution is magic by which for all the students o f Indology and Buddhol-
one can force animals or inanimate things to ogy as well as for those who bear the burden of
accept an offer of peace treaty or friendship responsibility for environmental ethics in East
contract (p. 33). The importance o f this kind of and West.
handling problems is strikingly relevant to our
days when magic is “supplemented or replaced
by modern techniques namely pesticides and R e fe r e n c e s
insecticides in many Buddhist countries” (p. 68).
A separate chapter is devoted to the protection Gopal, L. (1981), The Gurusamhitâ. An Ancient
from natural dangers (pp. 55-65). Among others Text on Weather-forecasting. Varanasi (1981).
Schmithausen speaks of “drastic magical vio- (Monographs o f the Department o f Ancient
lence” against animals that damage crops (p. 64). Indian History, Culture & Archaeology.
At this point I should like to call attention to No. 10.)
the Tantric mantras to be attested in practi- Schmithausen, L. (1991a), The Problem o f the
cal treatises such as the Krsiparäsara and the Sentience of Plants in Earliest Buddhism.
Vrksäyurveda o f Sürapäla since both are of Studia Philologica Buddhica, Monograph
Bengali provenance and they date from the early Series 6. Tokyo. The International Institute
medieval times when tantra flourished. These for Buddhist Studies.
are the so-called Hanumän mantras. Similarly Schmithausen, L. (1991b), Buddhism and Na-
in the respective discussions about regulating ture (Enlarged version with notes). Studia
rain, removing drought on the ground of the Philologica Buddhica, Occasional paper
Mahämäyüri, “The Great Peacock Charm” and Series 7. Tokyo. The International Institute
the (Mahä) Meghasütra The Great Cloud Ser- for Buddhist Studies.
mon evoke for me Sanskrit texts on climatology Varenne, J. (1972), The Garuda Upanisad. In:
such as the Mayüracitraka extant in various India Maior. Congratulatory Volume Pres-
readings (Gopal 1981, pp. 2 1 -2 2 ) and the Me- ented to J. Gonda. Ed. by J. Ensink and
ghamälä. One of the seven manuscripts o f the P. Gaeffke. Leiden (1972, pp. 222-231).
latter kept at the Sarasvati Bhavan Library, Va- Wojtilla, Gy. (1975), The “Longer” Recension
ranasi, as its colophon puts it, forms part of the o f the Garudopanisad. AOH 19:3, pp. 3 8 5 -
Rudrayämalatantra. It is a pity that no printed 392.
text o f the Rudrayämalatantra contains a chap-
ter on climatology (cf. Gopal 1981, pp. 23-24). Gyula Wojtilla

BOOKS AND JOURNALS RECEIVED
Bartos H.-Hamar I. (1998), Kinai-magyar szô- XVIII) Sankt Augustin, Steyler Verlag.
târ. Budapest, Balassi Kiado. HUF 3000. DM 90.
Behrens, H. (1998), Die Ninegalla-Hymne. Die Gardner, M. (1998), Science Good, Bad and Bo-
Wohnungnahme Inannas in Nippur in Alt- gus. Prometheus Books. New York.
babylonischer Zeit. (Freiburger Altorien- Kara Gy. (1998), Mongol-magyar ke'ziszötdr.
talische Studien 21.) Franz Steiner Verlag. Budapest, Terebess Kiado.
Stuttgart. DM 64. Karttunen, K. (1997), India and the Hellenistic
Birtalan A. [ed.] (1998), Öseink nyomdban Bel- World. (Studia Orientalia 83.) Helsinki.
sô-Azsidban II. Hitvilâg és nyelve'szet. (Ma- FM 200.
gyar Felsôoktatâs Konyvek 10), Budapest, Kaul, A.-Chattopadyay, S. [eds] (1999), Mani-
Nemzeti Tankönyvkiadö. HUF 1045. festation o f nature srsti vistära. Kalätattva-
Clark, L. (1998), Turkmen Reference Grammar. kosa. Vol. IV. Delhi, Banarsidass.
(Turcologica 34.) Wiesbaden, Harrassowitz. Kulke, H.-Rothermund, D. (1997 ),A History of
Clarke, G. E. [ed.] (1998), Development, Soci- India. Third edition. X, 395 pp. Routledge,
ety and Environment in Tibet. Papers Pre- London and New York. GBP 16.99.
seated at a Panel o f the 7th Seminar o f the Leslie, D. D. (1998), Jews and Judaism in Tra-
International Association for Tibetan Stud- ditional China. A Comprehensive Bibliog-
ies, Graz 1995. (Österreichische Akademie raphy. Institut Monumenta Serica, Sankt
der Wissenschaften Phil.-Hist. Kl. Denk- Augustin. (Monumenta Serica Monograph
Schriften 262. Beiträge zur Kultur- und Series XLIV) Steyler Verlag, Nettetal. 291
Geistesgeschichte Asiens 24.) Verlag der pp. DM 65.
Österreichischen Akademie der W issen- Malek, R. [ed.] (1998), "Western Learning”
schäften, Wien. ATS 566. and Christianity in China. The Contribu-
Demir, N.-Taube, E. (1998), Turkologie Heute. don and Impact o f Johann Adam Schall
Tradition und Perspektive. Materialen der von Bell (1592-1666). (Monumenta Serica
dritten Deutschen Turkologen-Konferenz. Monograph Series XXXV.) Institut Monu-
Leipzig 4-7. Oktober 1994. (Veröffentli- menta Serica, Sankt Augustin, Steyler Ver-
chungen der Societas Uralo-Altaica 48.) lag, Nettetal. DM 200.
Wiesbaden, Harrassowitz. Rédei K. (1998), Ôstorténetünk kérdései. A nyel-
Eber, I. [et al.] [eds] (1999), Bible in Modern vészeti dilettantizmus kritikàja. (Magyar Os-
China. The Literary and Intellectual Im- tôrténeti Kônyvtâr 11) Budapest, Balassi
pact. (Monumenta Serica Monograph Series Kiadô. HUF 1200.
Akadémiai Kiadô, Budapest

124 BOOKS AND JOURNALS RECEIVED
Schmidtke, S. [ed.] (1998), A Mu’Tazilite creed The Brief Catalogues to the Narthang and the
o f Az-Zamahsari (D. 538/1144). (Abhand- Lhasa Kanjur. (Wiener Studien zur Tibe-
lungen für Kunde des Morgenlandes 51,4) tologie und Buddhismuskunde 40.) Wien.
Stuttgart, Franz Steiner Verlag. DM 42. Tripathy, S. (1997), Inscriptions o f Orissa.
Schütz, E. (1998), Armeno-turcica. (Indiana Vol. 1. (Indian Council o f Historical Re-
University Uralic and Altaic Series 164) search). Motilal Banarsidass Publishers,
Research Institute for Inner Asian Studies, Delhi. Rs. 600.
Bloomington. Voprosy tjurkskoj filologii. Vyp. III. (Materialy
Schwieger, P. (1999), Tibetische Handschriften Dmitrievskih ctenij). Moskva, 1997. 132 pp.
und Blockdrucke. T.12. (Verzeichnis der (Moskovskij gos. univ. im. M. V. Lomono-
orientalischen Handschriften in Deutsch- sova. Institut stran Azii i Afriki).
land XI, 12.) Stuttgart, Franz Steiner Ver- Witz, K.G. (1998), The Su!>reme Wisdom o f the
lag. Upanisads. An Introduction. Motilal Ba-
Shimin, G.-Klimkeit, H.-J.-Laut, J. P. (1998), narsidass Publishers, Delhi. Rs. 700.
Eine buddhistische Apokalypse. Die Hol- Zekiyan, Boghos Levon (1997), The Armenian
lenkapitel (20-25) und die Schlusskapitel Way to Modernity. Armenian Identity be-
(26-27) der Hami-Handschrift der alttürki- tween Tradition and Innovation, Specificity
sehen Maitrisimit. (Nordhein-Westfälische and Universality. (Eurasiatica 49) (Armeni-
Akademie der Wissenschaften 103.) Op- ca Italica 3) Supernova, Venezia. 111 pp.
laden, Westdeutscher Verlag, Wiesbaden,
DM 86.
J o u r n a ls r e c e iv e d
Siddheswarananda, S. (1998), Hindu Thought
and Carmelite Mysticism. Motilal Banarsi- Acta Historica CIV. (Acta Univ. Szegediensis
dass Publishers, Delhi. Rs. 150. de Attila Jôzsef Nominatae).
Sueki, Y. (1998), Bibliographical Sources for Journal of Asian and African Studies 55.
Buddhist Studies from the Viewpoint of (1998).
Buddhist Philology. (Bibliographia Indica The Tibet Journal XXIII: 1, 2 (1998).
et Buddhica III.) Tokyo. Ural Altaische Jahrbücher (1997/1998) 15.

ADDRESSES OF THE AUTHORS
Ärpäd Berta
JATE ВТК
H-6726 Szeged
Egyetem u. 2
Hungary
Yoshizo Itabashi
Institute of Language and Cultures
University o f Kyushu
4-2-1 Ropponmatsu, Chuo-ku
Fukuoka, Fukuoka 810-8560
Japan
Andres Rona-Tas
House o f Professors, Room No. 303
H-l 146 Budapest
Hungary
Claus Schönig
Institut für Orientalistik
Universität Gießen
Germany
A B B R E V IA T IO N S O F JO U R N A L S A N D S E R I E S
A EM A Archivum Eurasiae Medii Aevi, Wiesbaden

A LH Acta Linguistica Hungaricae, Budapest
AM Asia Major, Leipzig
AO Acta Orientalia, Copenhagen
ArchOtt Archivum Ottomanicum, Wiesbaden
AOF Altorientalische Forschungen, Berlin
AOH Acta Orientalia Academiae Scientiarum Hungaricae, Budapest
ArOr Archiv Orientalni, Praha
ARW Archiv für Religionswissenschaft
AS Asiatische Studien, Bern
ASI Archaeological Survey of India
BAH Bibliotheca Arabo-Hispanica
BASOR Bulletin of the American Schools of Oriental Research
BEFEO Bulletin de l ’École Française d’Extrême Orient, Paris
BGA Bibliotheca geographorum arabicourm
BIFAO Bulletin de l’Institut Française d’Archéologie Orientale du Caire
BiOr Bibliotheca Onentalis
BOH Bibliotheca Orientalis Hungarica, Budapest
BSOS Bulletin of the School of Oriental Studies, London
BSOAS Bulletin of the School of Oriental and African Studies, London
CAJ Central Asiatic Journal, Wiesbaden
El, El2 Encyclopaedia of Islam (1st and 2nd edition)
ERE Encyclopaedia of Religions and Ethics
FO Folia Orientalia, Kraköw
GMS Gibb Memorial Series
GrIPh Grundriss der Iranischen Philologie
HdO Handbuch der Orientalistik, Leiden
HJAS Harvard Journal of Asiatic Studies, Cambridge, Mass.
HUS Harvard Ukrainian Studies, Cambridge, Mass.
HZ Hel Zohiol, Ulaanbaatar
IA Islam Ansiklopedisi
IA N Izvestija Akademii Nauk SSSR, Moskva
IC Islam Culture
IIJ Indo-Iranian Journal, ’s-Gravenhage
IJMES International Journal of Middle East Studies, London
IQ The Islamic Quarterly
Islam Der Islam, Berlin
JA Journal Asiatique, Paris
JAH Journal of Asian History, Wiesbaden
JAL Journal of Arabic Literature
JAOS Journal of the American Oriental Society, New Haven, Conn.
JAS Journal of Asian Studies, Berkeley, Ca.
JESHO Journal of the Economic and Social History of the Orient, Leiden
JNES Journal of Near Eastern Studies
JRAS Journal of the Royal Asiatic Society of Great Britain and Ireland, London
JSAI Jerusalem Studies in Arabic and Islam
JSFOu Journal de la Société Finno-ougrienne, Helsinki
JTS Journal of Turkish Studies, Cambridge, Mass.
KCsA Korösi Csoma Archivum, Budapest
KK Keletkutatâs, Budapest
KSz Keleti Szemle, Budapest
MEJ Middle Eastern Journal
MFOB Mélanges de la Faculté Orientale de l ’Université St. Joseph de Beyrouth
M ID E O Mélanges de l ’Institut Dominicain d’Études Orientales du Caire
MNy Magyar Nyelv, Budapest
MO Le Monde Orientale
MRDTB Memoirs of the Research Department of the Toyo Bunko, Tokyo
MS Monumenta Serica, Peking
MSFOu Mémoires de la Société Finno-Ougrienne, Helsinki
MSS Münchener Studien zur Sprachwissenschaft
NAA Narody Azii i Afriki, Moskva
N yK Nyelvtudomânyi Kôzlemények, Budapest
OC Oriens Christianus
OE Oriens Extremus, Hamburg
OLZ Orientalische Literaturzeitung, Berlin
OsmAr Osmanli Araçtirmalari, Istanbul
OZ Ostasiatische Zeitschrift, Berlin
RBS Revue Bibliographique de Sinologie, Paris
REI Revue des Études Islamiques
RE J Revue des Études Juives
RMM Revue du Monde Musulman
RO Rocznik Orientalistyczny, Warszawa
RSO Rivista degli Studi Orientali
SFo Südost-Forschungen, München
SI Studia Islamica, Paris
SM Studia Mongolica, Ulaanbaatar
SO Studia Orientalia, Helsinki
Stil Studien zur Indologie und Iranistik, Reinbek
SUA Studia Uralo-Altaica, Szeged
TD Türk Dili, Ankara
TDAYB Türk Dili Araçtirmalari Yilligi Belleten, Ankara
TDKY Türk Dili Kurumu Yaymlan
TG Toyo Gakuho, Tokyo
TP T ’oung Pao, Leiden
TS A B Turkish Studies Association Bulletin, Bloomington
UAJb Ural-Altaische Jahrbücher, Berlin-Bloomington
UAJb NF Ural-Altaische Jahrbücher: Neue Folge. Wiesbaden
UJb Ungarische Jahrbücher, Berlin
VOH Verzeichnis der orientalischen Handschriften in Deutschland
WI Die Welt des Islams, Leiden
WOr Welt des Orients, Göttingen
W ZKM Wiener Zeitschrift für die Kunde des Morgenlandes, Wien
ZA Zeitschrift für Assyriologie
ZDM G Zeitschrift der Deutschen Morgenländischen Gesellschaft, Wiesbaden
ZS Zeitschrift für Semitistik
ZSt Zentralasiatische Studien des Seminars für Sprach- und Kulturwissenschaft Zentralasiens der Universi-
tat Bonn, Wiesbaden
Z V O IR A O Zapiski vostocnago otdelenija Imperatorskago russkago arheologiceskago obscestva, Sanktpeterburg
IN F O R M A T I O N F O R C O N T R I B U T O R S
Acta Orientalia is a refereed journal which provides an international forum for original papers in the field o f
oriental studies, including Turkic, Mongolian, Manchurian, Chinese, Tibetan, Indian, Iranian and Semitic
philology, literature and history. A copy o f publishing agreement will be sent to the authors of papers accepted
by two reviewers and the board o f editors for publication. Manuscripts w ill be printed only after receiving the
signed copy o f the agreement.
Manuscripts for publication, editorial correspondence and books for review should be sent to
Andräs Rdna-Tas
Editor-in-Chief
House o f Professors, Room 303
H -l 146 Budapest
Manuscript form. Final copies should be submitted on A 4 size paper, double spaced throughout (including
footnotes, quotations and reference list), with no less than 1.5 inch margins on all four sides, printed on one
side only and with a single pagination on each side. They should be printed in two copies and preferably also
on a floppy disk in W ord for Windows ’95 or ’98. Papers not asked for w ill not be returned or kept.
Manuscripts should be written with large characters (12 points). The header o f the article is kindly requested
front the authors together with the first proof.
Text. The total length of the text (including illustrations, tables, etc.), if not agreed otherwise, should not
exceed 20 typewritten pages. The first page should contain the title o f the paper, the author’s names and
affiliations (city and state) and the abstract.
Abstract. Manuscripts o f articles should be accompanied by an abstract o f about 100 words / 15 lines
printed on a separate sheet o f paper summarising the conceptual content and the new results of the article.
Key words. The authors are kindly asked for 6 to 8 key words added to their articles.
Figures. Graphs, line drawings, photographs should be numbered in sequence with Arabic numerals. Each
figure should be identified with the figure number, its approximate location should be marked in the text.
Please indicate the top o f the figure and the name of the author o f the paper on the back. High quality half-
toned photographs should be prepared on glossy paper at their expected final size.
Footnotes. A ll footnotes must be placed on the same page, as the text and numbered continuously.
References. Full citation of literature referred to should be given in an alphabetically arranged list at the end
o f the article under the heading: References. Use the same character size and line spacing for references as for
the text. In the references use full names o f journals if they are not to be found on the last page of AO H . In the
text they should be cited by the names o f the authors and the year of publication, e.g. Doerfer and Tezcan
(1980), or if pages are referred to: Doerfer and Tezcan (1980, pp. 21 -22). In case o f publications of the same
year use letters: D oerfer and Tezcan (1980a).
Reference to books:
Doerfer, G.-Tezcan, S. (1980), W örterbuch des Chaladsch. (D ialekt von Charrab). Budapest (Bibliotheca
Orientalis Hungarica X X V I) .
Reference to articles published in book:
Gervers, M . (1997), Nomads as gatherers and creators o f world culture. The artistic sources of the great felt
hanging from Pazyryk, Kurgan 5. Sum m aries o f congress papers. 7th International Congress o f M ongolists
1 2 -1 6 August 1997, Ulaanbaatar. p. 3. Ulaanbaatar.
Reference to articles in journals:
Berta, Ä. (1994), Zum Wandel des auslautenden -G im Kiptschakischen. Journal o fT urkology 2:2, pp. 163-195.
Transcription. The authors should use the transcription internationally accepted in their respective field o f
studies. Please try to avoid the usage o f special character that is not to be found in W inW ord 95 or 98 Symbol
True Type font. Try to avoid Chinese, Arabic, etc. non Latin or Cyrillic characters. Use the Pinyin transcrip-
tion in the case o f Chinese unless you want to explain a linguistic reconstruction. I f you have to use other
special symbols, please enclose a floppy containing the fonts (in case o f Chinese preferably Big5, GB or
Chinese Star).
Proofs. The author w ill receive the first proof. Please read it carefully and return it promptly, if necessary by
air mail to the editor within one week.
Offprints. Authors w ill receive 25 bound offprints.
Reviews. General rules of scientific publication (see above) should be followed.
Addresses of the authors. W e publish the mailing addresses o f the authors as they are given in the publish-
ing agreement. Please give your e-mail address as well.
HU ISSN 0001-6446
ACTA
ORIENTAIT A
HUNGARICAE
VOL. 52 (1999) NO. 2
AKADÉMIAI KIADÔ, BUDAPEST

ACTA ORIENTALIA
FOUNDED
BY
L. LIGETI
EDITORIAL BOARD
Ildiko Ecsedy
Päl Fodor
Éva J eremiàs
Andräs R öna -T as , Editor-in-Chief
Robert Simon
Gyula W 0 J T 1 L L A
ADVISORY BOARD: Gy. H a z a i , Z s . K a k u k , Gy. K a r a , M. M a r o t h , F. T ö k e i
TECITNICAL EDITORS: É v a CSÂKI a n d Imre H a m a r
brief accounts on past and com ing events in oriental studies. ACTA ORIENTALIA is published in
yearly volumes of four numbers by AKADÉMIAI KIADÖ, H -l 117 Budapest. Prielle Kornélia и. 4.
http://www.akkrt.hu
Subscription information: Orders should be addressed to AKADÉMIAI KIADO, H -l519 Budapest,

Subscription price for Volume 52 (1999) in four numbers is USS 156.00 including normal postage,
E d it o r ia l c o r r e s p o n d e n c e : Manuscripts and correspondence with the editors should be sent to

ACTA ORIENTALIA, House o f Professors, Room 303, 11-1146 Budapest, Ajtôsi Dürer sor 19-21.
© Akadémiai Kiadö, Budapest 1999
AOrient 52 (1999) 2
ACTA
ORIENTAIT A
HUNGARICAE
VOL. 52 (1999) NO. 2

CONTENTS
A. Г. С а зы к и н , П ятый (анонимный) монгольский перевод «Алмазной су т р ы » ........... 129
Books and journals received ...................................................................................................................... 223

ПЯТЫЙ (АНОНИМ НЫ Й) МОНГОЛЬСКИЙ ПЕРЕВОД

«АЛМ АЗНОЙ СУТРЫ»
А . Г. С а з ы к и н
(Санкт-Петербург)
М онгольские переводы буддийской канонической литературы, представленные в де-

сятках коллекций монгольских рукописей и ксилографов, казалось бы, достаточно
хорош о известны востоковедам. Однако наряду с общ ераспространенными переводами
наиболее именитых монгольских литераторов, тиражированными как в рукописном,
так и в особенности в ксилографированном виде, в монгольских коллекциях время от
времени удается обнаружить редкие, порой уникальные, письменные образцы неизвест-
ных прежде переводов буддийских сочинений.
К числу таких редкостей принадлежит включенный в настоящ ую публикацию
текст монголького перевода «Алмазной сутры» («Ваджраччхедики»). Этот, появившийся
не позднее XVIII столетия, остающийся пока анонимным, перевод выявлен в настоящее
время только в рукописных собраниях Санкт-Петербурга. Таким образом , подготов-
ленное издание вводит в научный об и х о д еще одну неизвестную ранее монгольскую
версию «А лмазной сутры», одного из распространеннейших буддийских сочинений
в рамках старописьменной монгольской словесности.
В рамках буддийской канонической литературы, бытовавшей у монголь-

ских народов в виде отдельных рукописных и ксилографированных изда-
ний, можно выделить несколько сочинений, преимущественно сутр, поль-
зовавшихся особой популярностью у читателя. К их числу следует от-
нести, например, «Üliger-ün dalai» («Море притч»), «Altan gerel-tü» («Сутра
золотого блеска»), «Tharpa chenpo» («Великий освободитель»).
Однако по степени известности, распространенности они не идут ни
в какое сравнение с небольшой канонической сутрой, в научном обиходе
именуемой обычно как «Vajracchedikä» («Алмазная сутра»). Действитель-
но, трудно найти собрание монгольских рукописей и ксилографов, где бы
не были представлены печатные или рукописные экземпляры этой сутры.
Не составляет исключения и монгольский фонд Санкт-Петербургского
филиала Института востоковедения Российской Академии наук (далее:
СПбФ ИВ РАН), где в настоящее время выявлено 24 ксилографированных
издания и 72 рукописи сочинения.

130 А. Г. С А З Ы К И Н
Многочисленность рукописей сутры при наличии десятков ее кси-

лографированных изданий объясняется в первую очередь положением
буддийской догматики о пользе и спасительной силе именно переписки
священных текстов. Важнейшим среди них стал текст «Алмазной сутры»,
содержащий изложение учения о праджняпарамите, т.е. о пути спасения
и достижения святости, в виде беседы Будды со своим учеником по имени
Субхути.
Следует отметить, что тибетскими и монгольскими ламами немало
было сделано для пропаганды этой сутры. В буддийской повествователь-
ной литературе, например, в качестве средства к избавлению от страданий
и накоплению прижизненных благих деяний чаще всего называлась «Ал-
мазная сутра».1 В Тибете был также составлен сборник 15 рассказов, спе-
циально посвященный разъяснению пользы чтения и переписки сутры.
Переведенный на монгольский язык не позднее XVII в.,2 он долгое
время распространялся у монгольских народов в рукописном виде, пока
во второй половине XIX в. в Агинском3 и Эгитуевском4 дацанах Бурятии
не появились ксилографированные издания двух редакций сборника. По-
пав к монголам, сборник рассказов о пользе «Алмазной сутры» был суще-
ственно переработан и дополнен новыми рассказами.5 Эта новая редакция
сборника, включавшая 22 рассказа, в конце XIX-начале XX в. была также
ксилографирована бурятскими ламами, на этот раз Цугольского дацана.6
В XVIII столетии монгольская версия рассказов была переложена на
заяпандитовское «ясное письмо». Ойратская версия сборника подобно
большинству литературных памятников западных монголов распростра-
нялась только в рукописном виде.7
Из четырех известных прежде монгольских переводов «Алмазной
сутры» наиболее распространен был перевод Банджа (Банча) Тиристы,
выполненный в 1629 г. по повелению Лигдан-хана Чахарского.8 В собра
1 История Чойджид-дагини. Факсимиле рукописи. Транслитерация текста, пере-

вод с монгольского, исследование и комментарий А. Г. Сазыкина. Памятники письмен-
нести Востока. T. ХС (1990) (Bibliotheca Buddhica. XXXVII). М осква, с. 190, 191, 199, 201.
2 Ёндон, Д. (1975), Нэгэн зохиолын тегсгвлийн у г . Хэл зохиол судлал. T. XI.
Улаанбаатар, с. 245-250.
3 Сазыкин, А. Г. (1988), Каталог монгольских рукописей и ксилографов Институ-
та востоковедения Академии наук СССР. T. 1. Москва, № 366.
4 Там же, № 365; Ёндон, Д .-С азы кин А . Г. (1998), Монгольская версия рассказов
о пользе «Алмазной сутры». Mongolica-IV. Санкт-Петербург, с. 3 6 -4 7 .
5 Сазыкин, А. Г. (1986), Монгольские версии рассказов о пользе «Ваджраччхе-
дики». Письменные памятники и проблемы истории культуры народов Востока. T. XX,
ч. 1. Москва, с. 70-74.
6 Сазыкин, см сноску 3, № 364.
7 А. Г. Сазыкин (1997), Ойратская версия рассказов о пользе «Ваджраччхедики».
Петербургское востоковедение. Выпуск 9. Санкт-Петербург, с. 139-160.
8 Heissig, W. 1966, Zur Bestandsaufnahme und Katalogisierung mongolischer Handschrif-
ten und Blockdrucke in Japan. UAJ, Vol. 38. 1966, pp. 69, 76; Poppe, N. (1971), The Diamond
Sutra. Three Mongolian Versions of the Vajracchedika Prajnäpäramitä Texts, Translations, Notes
and Glossaries. (Asiatische Forschungen, Bd. 35.) Wiesbaden, pp. 11-97.
ПЯТЫ Й М О Н ГО Л ЬСКИ Й ПЕРЕВОД «АЛМ А ЗН ОЙ СУТРЫ » 131
ниях СПбФ ИВ РАН этот перевод представлен 8 пекинскими и 12 бу-

рятскими ксилографированными изданиями, а также 51 рукописью.
Пекинские ксилографы сутры имеются теперь во многих зарубеж-
ных коллекциях и потому достаточно хорошо известны востоковедам.
Тем не менее в монгольском фонде СПбФ ИВ РАН удалось обнаружить
еше два издания, не зафиксированные ни в одном из существующих ката-
логов.9
Больший интерес все же представляют бурятские ксилографы, по-
скольку из 12 изданий 10 выявлены пока только в коллекции СПбФ ИВ
РАН. Три из них появились уже в первой половине XIX в., на начальной
стадии бурятского книгопечатания.101Остальные были напечатаны во вто-
рой половине XIX в. в Агинском," Худонском,12 Гусиноозерском,13 Эги-
туевском 14дацанах.15
Рукописи «Алмазной сутры» в переводе Банджа Тиристы, имеющиеся
в собрании СПбФ ИВ РАН, преимущественно бурятского происхождения.
Десять из них написаны «серебряными», «золотыми» или пятицветными
«драгоценными» чернилами на листах глянцевой синей или черной бума-
ги, что по убеждению буддистов значительно увеличивало меру исполнен-
ной добродетели.
Следует еще отметить, что в 27 рукописях, во всех пекинских и 8 бу-
рятских изданиях указанного перевода сутры помещено также изложенное
9 Q 122: Vcir-iyar oytolaji orosiba, 41 л.+1 л. (гравюра), размер листа: 25,5x8, раз-
мер рамки текста: 22x6,7, 21-22 строки на странице; Q 3178/ 3728/ 3736: Vcir-iyar
oytaluyci orosiba, 49 л. (нет конца), размер листа: 30x10, размер рамки текста: 27x8,3, 1 8 -
19 строк на странице.
'‫ ״‬Сазыкин, А. Г. (1987), Раннепечатные бурятские ксилографированные издания
в собраниях монгольского фонда Института востоковедения А Н СССР. Историогра-
фия и источниковедение истории стран Азии и Африки. Выпуск X . Ленинград, № 2, 3,
с. 136-139.
11 Q 1519, экз. 1: Bilig-ün cinadu kijayar-a kiirügsen vcir-iyar oyta=luyci neretü yeke köl-
gen sudur orosiba, 68 л., размер листа: 22,5x9, размер рамки текста: 18,5x6, 1 6 -1 7 строк на
странице.
12 Q 1937: Без заглавия, 37 л., размер листа: 44,5x9, размер рамки текста: 28,5x7,5,
22-23 строки на странице.
13 С 186, экз. 1 : Qutuy-tu vcir-iyar oyta=luyci sudur orosi=ba, 50 л., размер листа: 36x9,
размер рамки текста: 26,5x7, 21 -2 2 строки на странице.
14 С 8, экз. 1: Qutuy-tu vcir-iyar oyta=luyci sudur orosi=ba, 40 л., размер листа:
35,7x11, размер рамки текста: 28,5x7, 22-23 строки на странице.
15 Места появления трех бурятских изданий «Алмазной сутры» в переводе Бан-
дж а Тиристы пока не установлены: a) Q 2516: Qutuy-tu vcar-iyar: oytaluyci bilig-ün cinadu-a
kürügsen yeke kölgen sudur orosibai, 52 л., размер листа: 22x7, размер рамки текста: 17,5x6,
11-18 строк на странице.
б) Q 436, экз. 1: Qutuy-tu vacir-iyar oyta=luyci nertü sudur orosi=ba, 42 л., размер листа
35,5x9, размер рамки текста: 30,5x7, 24 строки на странице;
в) Q 919, экз. 2: Без заглавия, 55 л., размер листа: 35,5x8,8, размер рамки текста:
26,5x7, 1 9 -2 0 строк на странице.

132 А . Г. С А З Ы К И Н
в форме сутры учение о пути спасения, не включенное в буддийский Ка-

нон .16
О другом монгольском переводе «Алмазной сутры», осуществлен-
ном Ширегету-гуши-цорджи в 1612 г., впервые сообщил В. Хайссиг на ос-
новании сведений, содержащихся в колофоне монгольской рукописи из
собраний Дворцовой библиотеки Пекина.17 Этот же перевод был идеи-
тифицирован им в изданном без колофона пекинском ксилографе, обна-
руженном в одной из библиотек Австралии.18 Впоследствии H. Н. Поппе
опубликовал текст ксилографа, прибавив к нему колофон пекинской руко-
писи.19
В фондах СПбФ ИВ РАН находятся полный (К 17) и неполный
(К 25) экземпляры этого пекинского ксилографа «конца XVII или, скорее,
начала XVIII в.»20, являющегося, таким образом, одним из самых ранних
пекинских изданий сутры на монгольском языке.
Перевод Ширегету-гуши-цорджи (тоже без колофона) выявлен здесь
еще в одной бурятской рукописи сутры21 и в монгольской рукописи, вклю-
чающей выборку из сборника тарни и сутр «Сундуй»,22 в состав которого
входит и интерсующее нас сочинение.23
В XVII в. ойратским Зая-пандитой был выполнен еще один перевод
«Алмазной сутры». Наибольшее количество рукописей и ксилографов сут-
ры в записи на «ясном письме» находится в немецких коллекциях24 и в со-
браниях Улан-Батора.25 17 рукописей ойратской версии сутры вошли в со-
став монгольского фонда СПбФ ИВ РАН. Среди них имеется копия ой-
ратского ксилографа 1741 г. издания (В 159), в котором колофон перевод
16 Это небольш ое, написанное в стихах, сочинение во всех случаях имеет устой-
чивое монгольское заглавие: Bilig-ün cinadu kijayar-a küriigsen oytaluyci vein bodi mör-i
toyulaqui neretü sudur. В подражание каноническим сутрам в начале его помещены сан-
скритское и тибетское названия сочинения в монгольской графике.
17 Heissig, W. (1971), Zur Bestandsaufnahme, pp. 71-74.
18 Там же, pp. 70, 75, 76, Abbildung 1.
19 Poppe, см. сноску, pp. 98 -1 5 8 .
20 Кара, Д. (1972), Книги монгольских кочевников. М осква, с. 124.
21 Q 2317: Qutuy-tu bilig-ün cinadu kijayar-a küriigsen veir-iyar ebdegei nertü yeke kölgen
sudur oroäiba, 54 л., размер листа: 22,6x8,5, размер рамки текста: 16x7, 15-16 строк на
странице, русская белая бумага, черные чернила, перо.
22 С 435: Gjodba orosiba (л. 76); Qutuy-tu bilig-ün cinadu kijayar-a küriigsen veir ebdegei
neretü yeke kölgen sudur (л. 8a), листы 7 б -5 7 а , размер листа: 37x12, размер рамки текста:
28x8, 18 строк на странице, плотная белая бумага, красная и черная тушь, калам.
23 Заметим, что в рукописном экземпляре «Сундуя», датированном 1673 г. (К 6),
и в четырех известных нам пекинских ксилографированных изданиях сборника, подго-
товленных в 1707 (К7а), 1718 (Н 338), 1723 (К 7-6) и 1723 (I 100) годах, помещен перевод
Б андж а Тиристы. Имеющиеся в фондах С П бФ ИВ РАН издание 1729 г. (Q 2581) и неда-
тированное издание п од шифром Q 2580/ 2587 - неполные. Листы с текстом «Алмазной
сутры » в них отсутствуют.
24 Heissig, W .-Sagaster, К (1961), Mongolische Handschriften, Blockdrucke, Landkarten.
Wiesbaden, № 240-258.
25 Лувсанбалдан, X. (1975), Tod усэг, m ‫ ץ‬уний дурегалууд. Улаанбаатар, с. 232, №
0241 (72 рукописи).
П Я Т Ы Й М О Н ГО Л ЬС К И Й ПЕРЕВОД «А ЛМ А ЗН О Й СУТРЫ » 133
чика отсутствует, а есть лишь колофон издателей.26 В двух других более

поздних ксилографированных изданиях сочинения на «ясном письме»27 по-
мещен колофон, где указаны имена переводчика и донатора. Сохранился
колофон переводчика и в 12 ойратских (калмыцких) рукописях из фондов
СПбФ ИВ РАН, причем представлен он здесь в двух редакциях: краткой,
состоящей из 8 стихотворных строк, и пространной - в 16 строк.
Наиболее распространены и известны востоковедам ойратские изда-
ния «Алмазной сутры» с кратким колофоном.28 Тогда как пространная ре-
дакция заяпандитовского колофона прежде была зафиксирована только
в рукописях, вошедших в состав немецких коллекций ксилографирован-
ных и рукописных монгольских книг.29 Теперь к ним прибавились три ой-
ратских рукописи из собрания СПбФ ИВ РАН (В 18, В 133, С 267).
Существование разных редакций колофона к одному и тому же
переводу Зая-пандиты в записи на ойратской графике - явление довольно
редкое. Нечто подобное можно наблюдать, например, в ойратской версии
переведенной им «Повести о Чойджид-дагини», известной по двум руко-
писям на «ясном письме», где помещены две редакции колофона, одна из
которых является переложением со старомонгольской письменности,
а вторая - собственно ойратское творчество.30 Естественно, что и в дан-
ном случае возникает предположение, что из двух колофонов ойратской
версии «Алмазной сутры» один сохранил переложение на «ясное письмо»
первоначального монгольского перевода, а второй - уже последующая ой-
ратская переработка. И такое предположение вполне оправдалось, когда в
фондах СПбФ ИВ РАН впервые был обнаружен заяпандитовский перевод
сутры на старомонгольской графике, сохранивший своеобразную терми-
нологию этого ойратского литератора и снабженный пространной редак-
цией колофона переводчика.31
Четвертый монгольский перевод сутры, вышедший из-под пера
Тойн-гуши, был выявлен в двух рукописных экземплярах, находящихся
26 Единственный ксилографированный экземпляр этого издания хранится в фон-

дах краеведческого музея Убсунурского аймака М онголии, см. сноску 25, с. 118, № 8.
27 Экземпляры одного из этих изданий находятся в собраниях Улан-Батора, см.
сноску 25, с. 117-118, № 6; Ринчэн (1966), Ойратские переводы с китайского. RO X X X
(1), рр. 66-67; В торое издание известно по ксилографированному экземпляру, имеющ е-
муся только в коллекции СПбФ ИВ РАН , см. сноску 25, с. 118, № 7; Сазыкин, А . Г.,
М онгольские рукописи и ксилографы, поступившие в Азиатский М узей Российской
Академии наук от Б. Я. Владимирцова. Mongolica. Памяти академика Б. Я. Владимир-
цова. 1884-1931. М осква, 1986, с. 288, № 6 .
28 Poppe, см. сноску, рр. 159-217; Лувсанбалдан, сноска 25, с. 262, № 5; Алмазная
сутра - Д ордж и дж одва - Сутра о Совершенной М удрости, рассекающей [тьму неве-
жества], как удар молнии. На старокалмыцком, калмыцком и русском языках. (П еревод
со старокалмыцкого А . В. Бадмаева, с санскрита В. П. А ндросова). Элиста, 1993.
29 Heissig-Sagaster см. сноску 24, № 244, 246.
3(1 История Чойджид-дагини, см. сноску 1, с. 37 -3 9 .
31 Q 2078: Dorji codba-yin sudur orosiyul-un tegüsbe, 33 л., размер листа: 33,5x10,
размер рамки текста: 28,5x8, 21 -2 3 строки на странице, русская голубая бумага начала
X IX в., красные и черные чернила, перо.
134 А. Г. С А З Ы К И Н
в коллекции монгольских рукописей и ксилографов Будапешта.32 Помимо

самой сутры Тойн-гуши заново перевел и текст упомянутого выше учения
о пути спасения.
Поиски перевода Тойн-гуши в монгольском фонде СПбФ ИВ РАН
успеха не принесли. Однако в процессе атрибуции рукописей и ксилогра-
фов «Алмазной сутры» был получен совершенно неожиданный результат.
Среди экземпляров, в которых отсутствуют колофоны переводчиков либо
утрачены последние листы, а таковых здесь насчитывается 32 единицы хра-
нения, был обнаружен еще один, неизвестный прежде, монгольский пере-
вод сочинения. Причем зафиксирован он сразу в трех изданиях - одном
рукописном и двух печатных.
Рукопись вновь найденного перевода сутры, вероятнее всего, бурят-
ского происхождения, поскольку под шифром Q 2859 она входит в коллек-
цию, составленную преимущественно из бурятских материалов.33 Мон-
гольское название сочинения «Qutuy-tu bilig-ün cinadu kijayar-a kürügsen vcir-
iyar oytolugci kemegdekü yeke kölgen sudur», помещенное в рукописи на листе
26 сразу после санскритского и тибетского заглавий, транскрибированных
буквами монгольского алфавита, отлично от того, что использовано в из-
даниях буддийского Канона «Ганджур» на монгольском языке.34 Испол-
йена рукопись «золотыми» чернилами на 68 листах черной лакированной
бумаги с синими полями35 и заключена в деревянную обложку, обтянутую
зеленой материей. На первом листе помещены две миниатюры с изобра-
жениями буддийских святых. К рукописи приложен отдельный лист с ри-
сованными фигурами четырех «махарадж» - защитников веры и мира. Са-
ма сутра занимает в рукописи 64 листа, на остальных четырех находится
сопутствующее ей стихотворное учение о пути спасения в том же самом
переводе, что и в аналогичных прибавлениях к «Алмазной сутре», пере-
веденной на монгольский язык Банджа Тиристой. К сожалению, рукопись
Q 2859 пострадала от сырости и потому прочтение текста в ней предста-
вляет порой нелегкую задачу.
Одно из ксилографированных изданий анонимного перевода «Ал-
мазной сутры» появилось, по всей видимости, в одном из многочисленных
монгольских монастырей. Этот ксилограф (В 266), поступивший в фонды
Азиатского музея (ныне СПбФ ИВ РАН) в составе коллекции Казанской
Духовной Академии, включает тибетскую и монгольскую версии сутры.
32 Särközi, А. (1972), Toyin Guisi’s Mongol Vajracchedikâ. AOH XXVII (1). pp. 43-102.
33 Сазыкин, A. Г. (1981), К характеристике собрания монгольских рукопи-
сей и ксилографов Ленинградского отделения Института востоковедения А Н СССР
(шифр Q). Письменные памятники и проблемы истории культуры народов Востока.
T. X V , часть 1(1). Москва, с. 5 6 -6 0 .
34 Ligeti L. (1942), Catalogue du Kanjur Mongol imprime. T. 1. Budapest, 1942, № 771;
К аталог петербургского рукописного «Ганджура». Составление, введение, транслите-
рация и указатели Касьяненко 3. К. (1993). - Памятники письменности Востока. Т. СП
(Bibliotheca Buddhica. XXXIX). М осква, № 669.
35 Размер листа: 21x8,3, размер рамки текста: 16x6, 14-20 строк на странице.
П Я Т Ы Й М О Н ГО Л ЬС К И Й ПЕРЕВОД «А ЛМ А ЗН О Й С У ТРЫ » 135
Заглавие на лицевой стороне первого листа отсутствует. Монгольское же

название сочинения, помещенное на листах 26 и 90а, совершенно такое же,
как и в рукописи Q 2859. Напечатан ксилограф на 91 двухслойном листе
малого формата,36 склеенном из тонкой китайской бумаги, бывшей в упо-
треблении в XVII-XVIII в. От времени бумага обветшала и побурела, в ре-
зультате чего данный ксилографированный образец имеет теперь не впол-
не удовлетворительную сохранность. К тому же уровень ксилоглиптики,
свойственный матрицам монгольского издания, как, впрочем, и качество
печатания этого единственного известного нам экземпляра, далеки от со-
вершенства.
Второе ксилографированное издание анонимного перевода «Алмаз-
ной сутры» было подготовлено в Бурятии. В фондах СПбФ ИВ РАН оно
представлено тремя экземплярами, один из которых (Q 1807) использован
при подготовке настоящей публикации. Как и в предыдущем случае, это
издание содержит тибетский и монгольский тексты сочинения. На лице-
вой стороне первого листа помещены соответственно тибетское заглавие
сутры: «’Phags-pa ses-rab kyi pha-rol-tu phyin-pa rdo-rje gcod-pa zes-ba bzugso» и
его монгольский перевод: «Qutuy-tu bilig-ün cinadu kijayar-a kürügsen vcir
oytaluyci kemegdekü orosibai».
Ксилограф Q 1807 выполнен на 68 листах37 белой русской бумаги
с рельефным штемпелем «Говард и Ко. № 7». Рельефные штемпели вошли
в употребление в России во второй половине XIX в. Следовательно, дан-
ный экземпляр мог быть напечатан никак не ранее середины прошлого
столетия, что отличает его от двух других ксилографированных образцов
этого же издания, имеющихся в собрании СПбФ ИВ РАН (Q 1967, Q 1987),
появление которых следует отнести к более раннему времени.
Оба экземпляра были напечатаны на грубоватой белой русской
бумаге с водяными знаками, среди которых встречаются даты «1828»
и «1829». В отличие от позднего ксилографа Q 1807, в каждом из ранних
экземпляров имеется отдельный лист с изображениями четырех «маха-
радж». Этот дополнительный лист, а также листы 16 и 2а, тоже украшен-
ные гравюрами,38 исполнены красной краской.39
Для датировки появления этого бурятского ксилографированного
издания «Алмазной сутры» полезными оказались некоторые его экземпля-
ры, включенные в коллекцию «Polyglota» (№ 46-49) библиотеки Восточ
36 Размер листа: 28,5x8,5, размер рамки текста: 20x6, на странице по 3 строки

тибетского текста с подстрочным монгольским переводом.
37 Размер листа: 34,5x10,3, размер рамки текста: 27,5x7,5, на странице по 3 строки
тибетского текста с подстрочным монгольским переводом, тибетская и монгольская
пагинации.
38 На листе 16, судя по тибетским подписям, изображены: слева - Шакьямуни
(Säkyagrub), справа - Дипанкара (M ar т е mdzad). На листе 2а помещены две гравюры
со священным слогом « о т » на письме «ланча».
39 В экземпляре Q 1967 красной краской напечатаны также и заглавия на лицевой
стороне первого листа.
136 А. Г. С А З Ы К И Н
ного факультета Санкт-Петербургского университета.40 До поступления

туда в 1855 г. все они находились в собрании Казанского университета,
причем два ксилографа были уже отмечены в опубликованном в 1834 г.
каталоге университетской коллекции восточных рукописных и ксилогра-
фированных книг.41
На экземпляре под шифром «Polyglota 49», помещена дарственная
надпись, из которой узнаем, что в библиотеку Казанского университета
он был передан шанцзодбой42 Гусиноозерского дацана Жигмед-Чойдар
Жамбалаевым осенью 1832 г. Эта надпись позволяет не только установить
дату, наиболее приближенную ко времени появления интересующего нас
издания, но и дает возможность назвать место его происхождения - Гуси-
ноозерский дацан, один из трех бурятских дацанов, замеченных в печата-
нии ксилографированных книг на монгольском языке в 20-30 годах XIX
столетия.43
Текст анонимного перевода «Алмазной сутры», как, впрочем, и тек-
сты остальных известных ныне монгольских переводов сутры, не имеет де-
ления на части. Использованная в данной публикации разбивка текста на
32 главы заимствована из упомянутой выше работы А. Шаркёзи, приняв-
шей, в свою очередь, за образец издание санскритской версии «Ваджрач-
чхедики».44
4,1 Сведения об университетских экземплярах обсуж даем ого здесь бурятского из-
дания «Алмазной сутры» лю безно предоставил нам сотрудник С П бФ ИВ РАН , канди-
дат исторических наук В. Л. Успенский.
41 [Ковалевский, О. М ] (1834), Каталог санскритским, монгольским, тибетским,
маньнджурским и китайским книгам и рукописям, в библиотеке Императорского
Казанского университета хранящимся. Казань, № 73, 74.
42 Ш анцзодба - одна из высших должностей в буддийских монастырях.
43 Сазыкин, см. сноску 10, с. 130-147.
44 Conze, Е. (1958), Buddhist Wisdom Books. The Diamond Sutra. The Heart Sutra. Lon-
don, pp. 21-71.

П Я Т Ы Й М О НГО Л ЬСКИ Й ПЕРЕВОД «АЛМ А ЗН ОЙ СУТРЫ » 137
[la] Qutuy-tu bilig-ün cinadu kijayar-a kürüg=sen* v[c]ir oytaluy=ci

kemegdekii orosi=bai:
[lb] Enedkig-iin kelen-dür: Ary-a bajar cce-de-ka bray-a-jna bar-a-mida nä-ma

mahä-ya-na su’=dar-a: Töbed-ün kelen-tür: ‘Pags-ba se-[2a]-rabkiy parôl-du pyagba
rdô-rje gcôd-ba jes byaba tigba cinbô yi mdô: Mongyol-un kelen-diir: Qutuy-tu bilig-
iin cinadu kijayar kiiriig=sen vacir oyta=luyci kemegdekii yeke kölgen sudur: [2b]
qamuy burqan kiged bô=dhi=sadova-nar-a mör=gü=mü:
1. eyin kemen bi ber sono=suysan nigen cay-tur: ilaju tögüs nö’g=cigsen Sar-
a=vasda-dur ilyun iiiled=diigci qan köbe=gü’n-ü ceceg=lig itegel iigegiin=e idegen
öggüg=ci-yin qotala bayasqu=lang-tu kiiriyen-diir: min’g=yan qoyar jayun tabun
ayay=qa tegim=lig-iin buyan-i abuyci-yin yekes quva’=ray-ud kiged: masi ola[n]
bô=dhi=sado=va ma=hâ=sadova-nuyud’-luy-a qamtu [За] nigen sayuy=san bolai:
tende-ece ilaju tegüs nög=cigsen örlö’ge cay-un kir-tiir sa’m=tab kiged пот-tu debil-
iyen ögöde bolji: badar ayay-a-ban bann Sir-a=vasd-yin yeke baly’-a=sun-tur
banvad-un tula=da ajara=bai: tende-ece ilaju tegüs nög=cigsen Sir-a=vasd-yin yeke
baly’-a=sun-tur binvad-un tulada ajira=ju ide=gen-i joyay=layad ide=gen=ii iiile-i
[3b] Uiled=cii: qoyitu idegen-й binvad tebci=ged badar kiged п о т -tu debil-iyen
talbiju: ama=ban jayilaju belediig=sen debis=ker-tiir jabalal jasa=yad bey-e-ben
sidur=yu-da jalaju durad=qui-yi ilete talbi=ju sayu=bai: tende-ece olan ayay-a
tegim=lig ber ilaju tegüs nög=cigsen qamiy-a bükü tegün-tür joraji=yad kürcü: ilaju
tegüs nög=cigsen-ii öl’müi-tür terigü=ber-iyen mörgü=jü yurban-ta erge=ged [4a]
nigen jüg-tür bayibai:
2. basa tegün-ü cay-tur amin qabiy-a-tu Subudi mön kü tere nököd-tür

quralca=ju sayuy=san bolai: te[n]de-ece amin qabiy-a-tu Sub[u]di debis=ker-ecegen
boscu degedü qubc[a]sun-iyan nigen egem-tür-iyen orki=yad barayun ebüdüg-ün
toyin-i yajar-tur qadqu=ju il[a]ju tegüs nög=cigsen qamiy-a bükü tere tus-tur alay’-a-
ban qabsur-un böküyijü: ilaju tegüs nög=cigsen-tür eyin kemen [4b] öci=bei: ilaju
tegüs nög=cigsen tegün=cilen ireg=sen dayin-i daruy=san üne’=ger toyu=luy=san
burqan ber bodhi=sadova maha-sadova-nuyud-tur tedüi bükü dege=dü tusa kürge=
küi ber tusa kürgeg=sen kiged: tegün=cilen ireg=sen ber bodhi=sadova maha=
sadova-nuyud-tur kedüi bükü dege=dü oy’o=yat[a] ög=küi ber oyo=yat[a] öggü=gsen
inu: ilaju tegüs nög=cigsen â [5a] yayiqam=siy bolai: sayibar oduy=san ä yayiqam=
siy bolai: ilaju tegüs nög=cig=sen ä bö=dhi=sadova-yin kölgen-tür üneger oroy=sad
ber yambar metü orosin üile’=dümü: yambar metü butü=gen üile=dümü yambar
metü sedkel-i masida barin üile=dümü: teyin kemen öci=meg=ce: ilaju tegüs nög=
cig=sen ber amin qabiy-a-tu Subudi-dur eyin kemen jarliy bolbai: Subudi ä [5b]
sayin sayin: Subudi ä tere tegüncilen bol[a]i: tere tegün=cilen bülüge: tegün=cilen
ireg=sen ber bö=dhi=sadova mahä-sadova-nuyud-tur dege=dü tusa kürge=küi ber
tusa kürgeg=sen bolai: tegün=cilen ireg=sen ber bö=dhi=sadova ma=hä=sadova-
nuyud-tur dege=dü oyo=yata ögküi ber oyo=yata ög=gügsen Subudi ä teyimü-yin
*Знак = обозначает конец строк.

Acta Orient. Hung. 52, J999
138 А. Г. С А З Ы К И Н
tulada sonosu=yad sayitur masida sedkil-dür-iyen barid=yad: bô=dhi=[6a]=sadova-

yin kölgen-dür iineger oroy=sad ber yambar metü orosin iiiled=kiii kiged: yambar
metii bütü=gen iiiled=kiii kiged: yambar metii sedkel-iyen masida barin iiiled=kiii-yi
biber cima’=dur nomlasu=yai: ilaju tegüs nög=cigsen tere tegün=cilen bolai kemen
öcijü:
3. amin qabiy-a-tu Subudi ilaju tegüs nögcig=sen-ü jarliy metii cingla’=mayca:

ilaju tegüs nög=cigsen ber amin qabiy-a-tu Subudi-dur eyin kemen jarliy [6b] bol=
bai: Subudi ä egün-dür bôdhi=sadova-yin kölgen-dür ün-e=ger oroy=sad ber: eyin
kemen bi ber kedüi cinegen amita[n]-i ami=tan-dur quriy-a=qui ber quri=yaysan:
öndegen-ece törög=sen ba: umai-aca törög=sen ba: dulayan cigiyg-ece törög=sen ba:
qubilju törög=sen ba: öngge’-tü ba: ö[n]gge-tü busu ba: quran mede=küi-tü ba: quran
medeküi ügei quran medekü ügei busu ba: ami=tan-u oron [7a] kedüi cinegen ami=
tan-dur nereyid=küi ber nereyid=dü’gsen tedeger qamuy-i соуса’ üile=bü’ri ügei-yin
yasala[n]g-aca nögcig=sen-ü töb-tür oyo=yata yasala[n]g-aca nögcig=sügei tere metü
cayla=si ügei amitan-i oyoyat[a] yasala[n]g-aca nögciy-e=becü: alin cü ami=tan-i
oyo=yata yasa’=la[n]g-aca nögcig=sen bolqu ügei bolai kemen sedkil egüs=ken
üiled=küi bolai: tere yayun-u tulada kemebesü: Subudi ä ker=be [7b] bö=dhi=sadova
ami=tan-dur quran medeküi oro=basu: tegün-i bö=dhi= sadova ülü kemegde=küi-yin
tula bolai: tere yayun-u tulada kemebesü Subudi ä al’in amita[n]-dur quran medeküi
oro=qui ba: amin-dur quran medeküi oro=qui ba: bod=gali-dur quran medeküi oro=
basu tegüni bö=dhi=sadova ülü kerne’g=deküi-yin tula bolai:
4. basa Subudi ä bö-dhi-sadova boda-tur orosil üge=güy-e ögli=ge öggün

üiled=deküi: yayun-dur cü orosil [8a] üge=güy-e ögli=ge öggün üiled=deküi: öngge-
dür cü orosil üge=güy-e ögli=ge öggün üiled=deküi: tegün=cilen dayun kiged:
kürüg=deküi kiged: nom-dur cü orosil ögküi ögli=ge-yi öggün üiled=deküi: Subudi ä
kerke=becü belgen-dür quran mede’=küi-dür cü orosil üge=güy-e tere metü bödhi=
sadova ber öglige ög=besü: tere yayun-u tulada kemebesü: Subudi ä bö=dhi=sadova
alin-dur cü orosil üg[e]güy-e [8b] ögli=ge ög=gügsen tegün-ü buyan-u coy=ca’ inu:
Subudi ä kemjiy-e bari=quy-a kilbar busu-yin tula bolai: Subudi ä egün-i yayun
kemen sedki=mü: doron-a jüg-ün oytar=yui-yin kemjiy-e bari=quy-a kilbar kemekü
güü: Subudi ber öcir-ün: ilaju tegüs nög=cigsen ä tere anu busu bolai: ilaju tegüs
nög=cigsen ber jarliy bolur-un: Subudi ä tegün=cilen emün-e kiged örön-e kiged:
umar-a kiged: degere kiged: doora-yin jüg kiged jüg [9a] jobkis kiged: arban jüg-ün
oytar=yui-yin kemjiy-e bari=quy-a kilbar kemekü güü: Subudi ber öcir-ün: ilaju
tegüs nög=cigsen ä tere anu busu bolai: ilaju tegüs nög=cigsen ber jarliy bolur-un:
Subudi ä tegün=cilen bö=dhi=sadova alin-dur cü orosil üge=güy-e ögli=ge öggüg=
sen tegün-ü buyan-u coy=ya’ anu: Subudi ä kemjiy-e bari=quy-a kilbar busu-yin tula
bolai:
5. [9b] Subudi ä egün-i yayun kemen sedki=müi: beige qotala tegü=sü’g=sen-

iyer tegün=cilen ireg=sen-dür üjen üiled=dümü kemekü güü: Subudi ber öcir-ün:
ilaju tegüs nög=cigsen ä tere inu busu bolai: beige qotala tegü=sü’gsen-iyer

П ЯТЫ Й М О Н ГО Л ЬС КИ Й ПЕРЕВОД «А ЛМ А ЗН О Й СУТРЫ » 139
tegün=cilen ireg=sen-dür üjen ülü üiled=kü bolai: tere yayun-u tulada kemebesü
tegün=cilen ireg=sen ber belge qotala tegü=sü’sen kemen ali nomlay=san mön kü
[10a] tere: beige qotala tegü=sü’gsen ügei-yin tula bolai: teyin kemen öci=megce:
ilaju tegüs nög=cigsen ber amin qabiy-a-tu Subu=di-dur eyin kemen jarliy bolbai:
Subudi ä kedüi cinegen beige qotala tegü=sü’g=sen tedüi cinegen qudal bolai: kedüi
cinegen beige qotala tegüsüg=sen ügei tedüi cinegen qudal busu bülüge: tere metü
bögesü tegüncilen ireg=sen-dür beige kiged [10b] belge=degsen ügei-dür uq[a]yda=
qui: teyin kemen jarliy bolu=may’ca:
6. ilaju tegüs nögcig=sen=dür amin qabiy-a-tu Subudi ber eyin kemen öci=
bei: ilaju tegüs nög=cigsen ä irege edüi-yin tabun jayun cay-un ecüs-dür: degedü-yin
nom masida teyin ber ebde[r]eküi bolba=su: ali amita[n] jarim-yin ene metü sudur-
un ayimay-un ügesi nomlaqui nuyud-tur üneger quran mede’=küi-yi egüske=küi
bolqui ire=küi [11a] metü bui buyu-’uu: ilaju tegüs nög=cigsen ber jarliy bolur-un:
Subudi ä ali ci eyin kemen irege edüi tabun jayun cay-un ecüs-tür: dege=tü-yin nom
masi teyin ber ebdere=küi bolba=su: ali ami=tan jarim-ud ene metü-yin sudur-un
ayimay-un üges-i nomlay=san nuyud-tur üneger quran medeküi egüske=küi bolqu
ire=küi metü buyu-’uu kemen ci teyin kemen buu ögüled=kün: [11b] Subudi ä irege
edüi-yin tabun jayun cay-un ecüs-tür degedü-yin nom masi teyin ber ebdereküi
bolba=su: bö=dhi=sadova mä=hä=sadova say’sa’=bad’-luy-a tegüsü’gsen: erdem-
lüge tegüsüg=sen: bilig-lüge tegüsüg=sen nuyud ireküi bülüge: Subu=di ä bö=dhi=
sadova mä=hä=sadova tedeger cü: yayca burqan-dur ergül kündü=lel-i üile=dügsen
busu: yayca burqan-dur buyan-i ija=[12a]=yur-i egüskeg= sen busu bui j-a: Subudi ä
olan gbum toy-a-tu burqan-dur ergül kündü=lel-i üiledü=ged: olan gbum toy-a-tu
burqan-dur buyan-u ija=yur-nuyud-i egüs=keg=sen-ü bö= dhi=sadova tede=ger ire=
küi bolai: Subudi ä ali=mad ene metü-yin sudur-un ayimay-un üges-i nomlay=san
nuyud-tur sedkil süsülküi nijiged-ün tedüi olqui bolqu tedeger inu: tegüncilen ireg=
sen ber ayilad=qu [12b] bolai: Subudi ä tede=ger-i inu tegüncilen ireg=sen ber üje=
küi bülü’ge: Subudi ä tedeger qamuy ami=tan inu buyan-u coy=ca cay=lasi ügei-yi
egüske=ged oyo=yata quri=yaqu bolqu bolai:
tere yayun-u tulada kemebesü: Subudi à bö=dhi=sadova mä=hä=sadova tede=
ger inu bi-dur quran mede=küi oroju ülü bolqu böged ami=tan-dur quran mede=küi
busu: amin-dur quran medeküi busu: bod=gali-dur [13a] quran medeküi oroju ülü
bolqu-yin tula bolai: Subu=di ä bö=dhi=sadova mä=hä=sadova tede=ger cü nom-dur
quran mede=küi kiged: nom ügei-dür cü quran medeküi ülü oro=qui bülüge: tedeger
inu quran medeküi ügei-dür cü oro=ju ülü bolqu bolai:
tere yayu-u tulada kemebesü Subudi ä ker=be bö=dhi=sadova mä=hä=sadova
tede=ger inu: nom-dur quran medeküi oro=basu cü mön kü tegün-i tedeger ber bi-
[13b]-dur bariqu bolu=yad: amita[n]-dur bari=qui kiged: amin-dur bari=qui kiged:
bod=gali-dur bariqui bolqui-yin tula bolai: ker=be nom-i bi ügei-dür quran mede=küi
oroba=su cü mön kü tegün-i tedeger ber bi-dur bari=qui bolu=yad: amitan-dur bari=
qui kiged: bod’gali-dur bari=qui bolqu-yin tula bolai: tere yayun-u tulada kemebesü
Subu=di ä bö=dhi=sadova nom-i cü buruyu=da barin ülü üiled=kü bülü’ge: nom
[14a] busu-yi cü ülü bariqu-yin tula bolai: teyimü-yin tulada tegün-dür tay[a]laju

140 А . Г. С А З Ы К И Н
tegün=cilen ireg=sen ber nom-un jüil egün-i sal metü medeg=ci nuyud ber nom-
nuyud-i cü tebcin üiledkü bögesü: nom busu nuyud-i edü’ge yayun ögülekü: kemen
nomlay=san bolai::
7. busu basa ilaju tegüs nô’g=cigsen ber amin qabiy-a-tu Subudi-dur eyin
kemen jarliy bol=bai: Subudi à egün-i yayun kemen sedki=müi tegün=cilen ireg=sen
ber [14b] ali degere ügei üneger toyu=luysan bô=dhi-dur ilete toyul=ju burqan
bolqui-yin tere nom alin cü bui kemebesü: tegün=cilen ireg=sen ber tere nom-i alin-i
cü üjü=gülbei kemekü güü: teyin kemen jarliy bolu=may=ca: ilaju tegüs nôg=cigsen-
dür amin qabiy-a-tu Subudi ber eyin kemen ôci=bei: ilaju tegüs nögcig=sen à biber
ilaju tegüs nôg=cigsen ber nomlay=san-u ud’q-a-yi medekü metü bôge=sü tegün=
eilen ireg=sen ber ali degere [15a] ügei üne’=ger toyu=luysan bô=dhi-dur ile’=te
toyulju burqan bolqu-yin tere nom alin cü ügei bolai: tegün=cilen ireg=sen ber
üjügü=lügsen-ü tere nom alin cü ügei bolai:
tere yayun-u tulada kemebe=sü: tegün=cilen ireg=sen ber tere nom alin cü
ile’=te toyul=ju burqan bolqu ba: üjügü=lügsen tere inu baray=dasi ügei ôgü=lesi
ügei bülü=ge: tere inu nom cü busu: [15b] nom ügei cü busu-yin tulada bolai: tere
yayun-u tulada kemebesü qutuy-tanu bod=gali-nuyud inu quran ese üile=dügsen-iyer
masida ily-a=qu-yin tula bolai:
8. ilaju tegüs nôgcig=sen ber jarliy bolur-un: Subudi à egün-i yayun kemen
sedki=mü: ijayur-tanu köbegün ba ijayu=r’-tanu ôkin ali jarim=ud nigen ber yurban
mingya’n yeke mingya’n yirten=cü-yin oron egün-i doloyan jüil erdeni ber dügür=
gen üiled=cü ögli=ge’ ög=besü: tere ijayur-tan-u köbegün [16a] ba ijayur-tanu ôken
tere sitügen-ece buyan-u coy=ca olan egüske=be kemekü güü: Subu=di ber ôcir-ün:
ilaju tegüs nôg=cigsen â olan bolai: sayibar oduy=san à olan bolai: tere ija’=yur-tanu
köbe-gün ba ijayur-tanu ôkin tere sitügen-ece buyan-u соуса olan-i egüske=kü bolai:
tere yayun-u tulada kemebesü: buyan-u coy=ca mon kü tere buyan-u соуса ügei-yin
tula bülü’ge: teyimü-yin tulada tegün=[16b]=cilen ireg=sen ber buyan-u coyca-yi
buyan-u coy=ca kemen nomlay=san bolai: ilaju tegüs nôg=cigsen ber jarliy bolur-un:
Subudi à ija=yur-tanu köbegün ba ijayur-tanu ôken alin ber yurban mi[n]g=yan-u
yeke mi[n]g=yan yirten=cü-yin ene oron-i doloyan jüil erde’=ni ber masi=da dügür=
gen üiled=cü öglige ôggüg=sen-ece: alin ber nom-un jüil egün-ece adqala=basu
dörben üge=tü-yin silü’g-ün tedüi=ken-i bariju busud [17a] nuyud-tur nomla’=yad
üneger masida üjü=gülbesü tere sitü=gen-ece buyan-u coy=ca neng olan toy-a=lasi
ügei cayla=si ügei-yi egüske=kü bolai: tere yayun-u tulada kemebesü: Subudi à
tegün=cilen ireg=sen dayin-i daruy=san ün-e=ger toyu=luy=san burqan-nuyud-un
degere ügei ün-e=ger toyu=luy=san bô=dhi inu egün-ece boluy=san bülü’ge: ilaju
tegüs nög=cigsen burqan-nuyud cü egün-ece törög=sen-ü tula [17b] bolai: tere
yayun-u tulada kemebesü: Subu=di â burqan-u nom-nuyud-i burqan-’u nom-nuyud
kemeg=deküi inu: burqan-’u tedeger nom-’i ügei-tür tegün=cilen ireg=sen ber
nomlay=san-u tulada bülü’ge: teyimü-yin tulada burqan-’u nom-nuyud kemeg-dekü
bolai:

9. Subudi à egün-i уауип kernen sedki=mü: urus=qal-dur oroy=sad eyin

kernen biber urus=qal-dur oroy=sad-un üre-yi olbai kernen sedki=be kemekü güü:
Subudi ber öcir-ün: [18a] ilaju tegüs nög=cigsen ä tere inu busu bolai: tere yayun-u
tulada kemebesü: ilaju tegüs nög=cigsen à tere inu yayun-dur cü oroy=san ügei-yin
tulada bülü=ge: teyimü-yin tulada urus=qal-dur oroysan kemeg=dekii bolai: öngge-
dür cü ese oro=bai: dayun-dur cü busu: ünür-tür cü busu: amta-tur cü busu: kürüg=
deküi-dür busu: nom-nuyud-tur cü ese oroy=san bülü’ge: teyimü-yin tulada urus=
qal-dur oroysan kemeg=dekü bolai: tere yayun-u tulada kemebe=[ 18b]=sü: ilaju
tegüs nög=cigsen ä kerbe urus=qal-tur oroy=sad eyin kemen bi ber urus=qal-dur
oroy=san-u üre-yi olbai kemen sedki=küi bolba=su: mön kü tere anu tegün-i bi-dur
bari=qu bolqu bolai: ami=tan-dur bari=qui kiged: amin-dur bar=qui kiged: bod=gali-
dur bari=qui bolqu bolai:
ilaju tegüs nög=cigsen ber jarliy bolur-un: Subudi ä egün-i yayun kemen
sedki=mü: nigen üy-e jici ireg’ci eyin kemen bi ber jici [19a] ireg=ci ene üre-yi olbai
kemen sedki=be kemekü güü: Subudi ber öcir-ün: ilaju tegüs nög=cig=sen ä tere inu
busu bolai: tere yayun-u tulada kemebesü: ali nigen üy-e mön kü jici ir[e]g=ci-dür
oro=quyin tere nom alin cü ügei-yin tulada bülü=ge: teyimü-yin tulada nigen üy-e-
dür ireg’ci kemegdekü bolai: ilaju tegüs nög=cigsen ä ker=be nigen üy-e jici iregci
eyin kemen bi ber nigen üy-e ireg=ci-yin üre-yi olbai kemen [19b] sedki=küi bolba=
su: mön kü tere tegün-i bi-dur bari=qui bolqu bolai: ami=tan-dur bari=qui kiged
amin-dur bari=qui kiged: bod=gali-dur bari=qui bolqu bolai:
sedki=mü: jici ülü ireg’ci eyin kemen bi ber jici ülü ireg’ci-yin üre-yi olbai kemen
sedki=be kemekü güü: Subu=di ber öcir-ün: ilaju tegüs nög=cig=sen ä tere inu busu
bolai: [20a] tere yayun-u tulada kemebe=sü: alin-u tula mön kü jici ülü ireküi-dür
oroqui-yin tere nom alin cü ügei-yin tulada bülü=ge: teyimü-yin tula jici ülü iregci
kemeg=dekü bolai:
sedki=mü: dayin-i daruy=sad eyin kemen bi ber mön kü dayin-i daruy=san-i olbai
kemen sedki=be kemekü güü: Subu=di ber öcir-ün: ilaju tegüs nög=cigsen ä tere inu
busu bolai: [20b] tere yayun-u tulada kemebe=sü: ali=mad dayin-i daruy=sad
kemeg=deküi-yin tere nom alin cü ügei-yin tula bolai: ilaju tegüs nög=cigsen ä kerbe
dayin-i daruy=sad eyin kemen bi ber mön kü dayin-i daruy=san-i olbai kemen sedki=
küi bol=ba=su: mön kü tere tegün-i bi-dur bari=qui bolqu bolai: ami=tan-dur bari=
qui kiged: amin-dur bari^qui kiged: bod=gali-dur bari=qui bolqu bolai:
ilaju tegüs nög=cigsen ä namayi inu tegün=cilen ireg=sen dayin-i [21a]
daruy=san ün-e=ger toyu=luy=san burqan ber nisva’=nis ügei ber oro=siyci nuyud-
un degedü-dür üjü=gülü’g=sen bülüge: ilaju tegüs nög=cigsen ä bi taci=yang=yui-
luy-a qayacay=san dayin-’i daruy=san bolba=cu: ilaju tegüs nög=cigsen ä bi eyin
kemen bi inu dayin-i daruy=san bolai kemen ülü sedki=küi bolai: ilaju tegüs nög=
cigsen ä kerbe eyin kemen bi ber mön kü dayin-i daruy=san-i olbai kemen sedki=küi
bolba=su: tegün=cilen [21b] ireg=sen ber nadur ija’=yur-tan-u köbegün Subudi inu
nisva’=nis üge=güy-e oro=siyci nuyud-un degedü mön bülü’ge: yayun-dur cü ülü
oro=siqu-yin tulada nisva’=nis üge=güy-e oro=siyci kemen ali ülü üjü=gülkü bolai:

142 А. Г. С А З Ы К И Н
10. ilaju tegüs nög=cigsen ber jarliy bolur-un: Subudi ä egün-i yayun kemen
sed=kimü tegün=cilen ireg=sen ber-i tegün=cilen ireg=sen day in-’i daruy=san ün-e=
ger toyu=luysan Diy=bang=gara burqan-aca ali abuy=san tere nom alin [22a] cü bui
kemekü giiü: Subudi ber öcir-ün: ilaju tegüs nög=cigsen ä tere inu busu bolai: tegün=
eilen ireg=sen ber tegün=cilen ireg=sen dayin-i daruy=san ün-e=ger toyu= luysan
Diy=ba[n]g=gara burqan-aca ali abuy=san tere nom alin cü oyo=yata tong ügei bolai:
ilaju tegüs nög=cigsen ber jarliy bolur-un: Subudi ä bö=dhi=sadova-nar ali jarim=ud
nigen eyin kemen bi ber oron bayidal nuyud-’i [22b] bütü=gebei kemen ögülel=
debesü tere inu ünen busu ber ögüle’=bei: tere yayun-u tulada kemebesü: Subudi à
oron bayidal nuyud-’i oron bayidal nuyud kemeg=deküi inu: tede=ger bayidal-i ügei-
dür tegün=cilen iregsen ber nomlay=san-u tulada bülü’ge: teyimü-yin tulada oron
bayidal-nuyud kemeg=dekü bolai:
Subudi ä tere metü-yin tulada: bö=dhi=sadova mä=hä=sadova ber ene metü
oro=sil üge=güy-e sedkil egüs=ken üiled=kü bülü’ge yayun-dur orosil üge=güy-e
sedkil-i egüsken [23a] üiled=deküi: öngge-dür cü oro=sil üge=güy-e sedkel-i egüs=
ken üiled=deküi: dayun kiged ünür kiged amta kiged kürü’=cen üiled=küi kiged
nom-durcü orosil üge=güy-e sedkil-i egüsken üiled=deküi:
Subudi ä ene metü bülü’ge üliger=lebesü: nigen törö’l=kite[n] ene metü bey-
e-tü bolju: ene metü bülüge ayulas-yin qayan Sümber-yin tedüi bolba=su: Subudi ä
egün-i yayun kemen sedki=mü: tere bey-e-ni yeke bolai kemekü güü: Subu=di ber
öcir-ün: ilaju tegüs nög=cigsen ä tere [23b] bey-e yeke bolai: sayibar oduy=san à tere
bey-e yeke bolai: tere yayun-u tulada kemebesü: tegün=cilen ireg=sen ber tegün-i
boda ügei-dür nomlay=san-u tulada bülü=ge: teyimü-yin tulada tere bey-e-yi yeke
kemegdekü bolai: tere boda ügei-yin tulada tegüber bögesü bey-e kemeg=dekü bolai:
11. ilaju tegüs nög=cigsen ber jarliy bolur-un: Subudi ä egün-i yayun kemen
sedki=mü: Gangga-a müren-’ü qumaki kedüi bükü Gangga-a müren cü imay=ta
tedüi bükü bolu=yad: tede=ger=ün qumaki ali bükün tede=[24a]=ger-i olan bolai
kemekü güü: Subu=di ber öcir-ün: ilaju tegüs nög=cigsen ä mön kü tede=ger
Gangga-a müren cü olan böget[e]le tede=ger-ün qumaki-yi edü’ge ögüle’=küi cü
yayun kereg: ilaju tegüs nög=cigsen ber jarliy bolur-un Subudi ä ci ber üjen üiled=
deküi: ci ber dotora=ban bayta=yan üiled=deküi: tedeger Gangga-a müre=nü qumaki
kedüi bükü tedüi bükü-yin yirten=cü-yin oron-dur eregdüi ba: qatuy=tai jarim nigen
ber doloyan jüil erdeni [24b] ber masi=da dügür=gen üiled=cü: tegün= eilen ireg=sen
dayin-’i daruy=san ün-e=ger toyu=luysan burqan-nuyud-tur ög=lige ög=besü: Subudi
à egün-i yayun kemen sedki=müü: tere eregtei ba qatuy=tai tere sitügen-ece olan
buyan-i egüske=be kemekü güü: Subu=di ber öcir-ün: ilaju tegüs nög=cigsen à olan
bolai: sayibar oduy=san ä olan bolai: tere eregtei ba qatuy=tai tere sitü=gen-ece olan
buyan-i egüske=[25a]=kü bolai: ilaju tegüs nög=cigsen ber jarliy bolur-un: Subudi ä
alin ber yirtin=cü-yin oron tedüi bükün-i doloyan jüil erdeni ber masida dügür=gen
üiled=cü: tegün=cilen ireg=sen dayin-i daruy=san ün-e=ger toyu= luysan burqan-
nuyud-tur ögli=ge ög=gügsen-ece: alin ber nom-un jüil egün-ece adalqa=basu dörben
üge-tü silü’g-ün tedüi=ken-i bari=ju: busud-nuyud-tur ün-e=ger masi=da nomlaju

П Я ТЫ Й М О Н ГО Л ЬС КИ Й ПЕРЕВОД «А ЛМ А ЗН ОЙ СУТРЫ » 143
[25b] ün-e=ger masi=da üjü=gülbe=sü: mon kü tere tere sitü=gen-ece neng toy-
a=lasi ügei calay=si [=caylasi] ügei buyan-u egüskekü bolai:
12. basa Subudi â ali yajar-un jüg-tür nom-un jüil egün-ece adalqa=basu
dörben üge-tü-yin silü’g-ün tedüi-yi ungsiy=san ba üjegü=lüysen yajar-un jüg tere
anu tngri kiged kiimün kiged asuri-luy-a selte-yin yirten=cü-yin takil-un sitü=gen-diir
bolqu mön bôge=sii ken nigen nom-un jüil egün-i abqui ki=ged: bici=küi kiged:
bari=qui kiged: qady-a=laqui [26a] kiged: surulca=qui kiged: qotola-yi onon üiled=
küi kiged: yosu=cilan sed=kel-dür üile=dügci tere masi yayiqam=siy-luy-a tegüsküi
bolqui inu edüge yayun ögülemü tere ayur-un jüg-tür üjügülüg=ci ba blam-a metü
alin cü bolun dayun-dur orosi=qui bolai:
13. teyin kemen jarliy bolu=mayca: ilaju tegüs nôg=cigsen-dür amin qabiy-a-
tu Subu=di ber eyin kemen ôci=bei: ilaju tegüs nög=cigsen ä nom-un jüil egün-ü
nere yayun bolai: egün-i yambar metü [26b] barin üiled=dümü: teyin kemen ôci=
megce: ilaju tegüs nôg=cigsen ber amin qabiy-a-tu Subudi-dur eyin kemen jarliy
bol=bai: Subudi â nom-un jüil egün-ü nere anu bilig-ün cinadu kijayar kürü’g^sen
kemeg=dekü bülü’ge egün-’i tere metü bari’y=yad:
tere yayun-u tulada kemebesü: Subudi à tegün=cilen ireg=sen ber bilig-ün
cinadu kijayar-a kürüg=sen-i ali nomlay=san mon kü tere cinadu kijayar-a kürkü
ügei-yin tulada bülü’ge: teyimü-yin tulada bilig-ün cinadu kijayar-a kürügsen
kemeg=dekü bolai:
[27a] Subudi â egün-i yayun kemen sedki=mü: tegün=cilen ireg=sen ber
nomlay=san-u tere nom alin cü bui kemekü güü: Subu=di ber ôcir-ün ilaju tegüs
nög=cigsen ä tegün=cilen ireg=sen ber ali nomlay=san tere nom alin cü ügei bolai:
ilaju tegüs nôg=cigsen ber jarliy bolur-un Subudi à egün-i yayun kemen
sedki=mü: yurban mingy’an yeke mingy’an yirtin=cü-yin oron-dur yajar-un toyo=sun
kedüi [27b] bükü bui tegün-i olan bolai kemekü güü: Subu=di ber ôcir-ün: ilaju tegüs
nög=cigsen ä yajar-un toyo=sun tere olan bolai: sayibar oduy=san à olan bolai: tere
yayun-u tulada kemebesü: ilaju tegüs nög=cigsen à yajar-un toyo=sun ali bükün
tegün-’i tayus=un ügei-dür tegün=cilen ireg=sen ber nomlay=san-’u tulada bülü=ge:
teyimü-yin tulada yajar-un toyosun kemegdekü bolai: yirtin=cü-yin oron ali bükün
tede=ger-i oron ügei-dür tegün=cilen ireg=sen ber nomlay=san-u tulada bülü=
[28a]=ge: teyimü-yin tulada yirtin=cü-yin oron kemeg=dekü bolai:
ilaju tegüs nôg=cigsen ber jarliy bolur-un: Subudi â egün-i yayun kemen
sedki=mü: yeke töröl=kite=nü yucin qoyar belge tede=ger-iyer tegün=cilen ireg=sen
dayin-i daruy=san ün-e=ger toyu=luysan burqan-tur üje[n] üile=dümü kemekü güü:
Subu=di ber ôcir-ün: ilaju tegüs nög=cigsen ä tere inu busu bolai: tere yayun-u tulada
kemebe=|28b]=sü: yeke tôrôT=kiten-ü yucin qoyar belge ali=mad-i tegün=cilen
ireg=sen ber nomlay=san tede=ger-i belge ügei-dür tegün=cilen ireg=sen ber
nomlay= san-’u tulada bülü=ge: teyimü-yin tulada yeke tôrôlki=ten-ü yucin qoyar
belge nuyud kemeg=dekü bolai:
ilaju tegüs nôg=cig=sen ber jarliy bolur-un: Subudi ä ereg=tei ba: qatuy=tai ali
jarim-ud nigen ber Gangga-a müren-’ü qumaki-yin tedüi bey-e-yi oyo=yata ôg=

144 А . Г. С А З Ы К И Н
gügsen-[29a]-ece: ali ber nom-un jüil egün-ece adalqa=basu dörben üge=tü-yin

silü’g-ün tedüiken-’i bariju busud-nuyud-tur üneger oroyul=basu: mon kü tere tere
sitügen-ece neng olan toy-a=lasi ügei: cayla=si ügei buyan-i egtiske=kü bolai:
14. tendece amin qabiy-a-tu Subudi anu nom-un öngge ber nilbusun yarya=ju:
tere ber nilbusun-iyan arciju ilaju tegüs nög=cigsen-dür eyin kernen ôci=bei: [29b]
nom-un jüil egün-i tegün=cilen ireg=sen kedüi bükü ber nomlay=san inu: ilaju tegüs
nôg=cigsen â yayiqam=siy bolai: sayibar oduy=san à yayiqam=siy bo=lai: ilaju tegüs
nôg=cig=sen à biber belge bilig törög=sen-ece inaysi bi ber nom-un jüil egün-i urida
kejiy-e cü ese so=no=suy=san bolai: tere yayun-u tulada kemebe=sü: ilaju tegüs
nög=cigsen à sudur-un ayimay-i nomlay=san egün-dür alimad üneger quran
medeküi-yi [30a] egüske=kü bolqu-yin tedeger ami=tan inu masi yayiqam=siy-luy-a
tegüsüg= sen bolqu bolai: tere yayun-u tulada kemebe=sü: ilaju tegüs nôg=cigsen ün-
e=ger quran mede=küi ali bükün mon kü tere quran mede=küi ügei-yin tulada
bülü’ge: teyimü-yin tulada üneger quran medekü kemen tegün=cilen ireg=sen ber
nomlay=san bolai
ilaju tegüs nög=cigsen ä bi inu nom-un jüil egün-i nomlay=san-dur ajiy-
a=layad sanay=san inu: nadur yayiqam=[30b]=siy-tu busu bui j-a: ilaju tegüs nôgcig=
sen â qoyitu ene üy-e qoyitu-yin cay tabun jayun-u ecüs-tür alimad ami-tan nom-un
jüil egün-i abqui kiged: bici=küi kiged: bari=qui kiged: qaday-a=lqui kiged: surul=
caqui kiged: qotala-yin onon üile=dügci tedeger inu masi yayiqam=siy-luy-a tegüs=
kü bolqu bolai:
basa ilaju tegüs nôgcig=sen â tedeger inu bi-tur quran medeküi oro=ju ülü
bolqu böged: ami=tan-dur quran medeküi ki=ged: amin-dur quran [31a] medeküi
kiged: bod=gali-dur quran mede=küi oro=ju ülü bolqu bo=lai: tere yayun-u tulada
kemebesü ilaju tegüs nög=cigsen à bi-dur quran mede=küi kiged: ami=tan-dur quran
mede=küi kiged: amin-tur quran mede=küi kiged: bod=gali-tur quran mede=küi ali
bükün mon kü tere qamuy quran mede=küi-luy-a qayacay=san-’u tula bolai: tere
yayun-u tulada kemebe=sü: burqan ä ilaju tegüs nôg=cigsen nuyud’ inu qamuy quran
mede= küi-[31b]-luy-a qay-a=cay=san-u tula bolai: teyin kemen ôci=megce: ilaju
tegüs nôg=cigsen ber amin qabiy-a-tu Subudi-dur eyin kemen jarliy bol=bai: Subudi
ä tere tegüncilen bolai: tere tegün=cilen bülü’ge: ali=mad ami=tan sudur-un ayimay
egün-i nomlaqui-dur ülü ayu=yad ülü emiyen qotala-dur emiyekü ülü bolqu tede=ger
inu masi yayiqam=siy-luy-a tegüs=küi bolqu bolai: tere yayun-u tulada kemebesü:
Subudi â degedü cinadu kijayar-’a kürü’g^sen egün-i inu [32a] tegün=cilen ireg=sen
ber nomlay=san bülü’ge: tegün=cilen ireg=sen ber nomlay=san ali degedü cinadu
kija=yar-a kürü’g=sen tegün-i cay=la’=si ügei kemjiy-e=lesi ügei ilaju tegüs
nôgcigsen burqan-nuyud ber cü nomlay=san-u tulada bülü=ge: teyimü-yin tulada
degedü cinadu kijayara kürügsen kemeg=dekü bolai:
basa Subudi ä tegün=cilen ireg=sen ber külicing=güi-yin cinadu kija=yar-a
kürüg=sen ali bükü mon kü tegün-i cinadu kijayara kürkü ügei bolai: tere yayun-u
tulada keme’=besü: [32b] Subudi ä alin-u cay-tur Kaling=ga-yin qayan ber minu
gesigün kiged üyes-nuyud-’i oyta=ciqui boluy=san tere cay-tur bi bidur quran mede=
küi ba: ami=tan-tur quran medeküi ba: amin-dur quran medeküi ba: bod=gali-tur

П Я Т Ы Й М О Н Г О Л ЬС К И Й ПЕРЕВОД «А Л М А ЗН О Й СУТРЫ » 145
quran medekiii cü ese bolu=yad: nadur quran medeküi yayun cü ügei boluyad quran
medeküi ügei boluy=san cü busu-yin tula bo=lai: tere yayun-u tulada kemebe=sü:
Subudi ä kerbe tere cay-tur bi bidur quran medeküi bolba=su: tere cay-tur qoor=tu
[33a] sedkil-ün quran medekü cü bolqu böged: ami=tan-tur quran mede=küi kiged:
amin-tur quran medeküi kiged: bod=gali-tur quran medeküi bolbasu: tere cay-tur
qoor=tu sedkil-ün quran medeküi cü bolqui-yin tula bolai: Subudi ä bi ber ilete
mede=jü: nögcig=sen cay-tur bi tabun jayun törö’l üyes-tür Külice[n]g=güi ögülegci
kemeg=deküi-yin isi boluy=san tere cay-tur cü: bi bidur quran mede’=küi esi bolbai:
ami=tan-tur quran medeküi kiged: [33b] amin-tur quran medeküi kiged: bod=gali-
dur quran medeküi ese boluysan bolai:
Subudi ä tere metü-yin tulada: bödhi=sadova mä=hä=sadova ber qamuy quran
mede=küi-yi teyin ber t’o=ci]u degere ügei üneger toyu=luy=san bod’=dhi-dur sed=
kil egüs=ken üiled=deküi: öngge-tür cü orosil üge=güy-e sed=kil egüs=ken üiled=
deküi: dayun kiged ünür kiged amta kiged: kürü’l=cen üiled=küi-dür cü orosil üge=
güy-e sedkil egüsken üiled=deküi: nom-dur cü orosil üge=güy-e sedkil egüs=ken
üiled=deküi: [34a] nom ügei-dür cü oro=sil üge=güy-e sedkil egüs=ken üiled=deküi:
yayun-dur cü orosil üge=güy-e sedkil egüsken üiled=deküi: tere yayun-u tulada
kemebesü: oro=siqu ali bükün mön kü tere ülü orosi=quyin tulada bülü’ge: teyimü-
yin tulada tegün=cilen ireg=sen ber eyin kemen: bö=dhi=sadova alin-dur cü orosil
üge=güy-e ögli=ge öggü’n üiled=deküi: kemen nomlay=san bolai::
basa Subu=di ä bö=dhi=sadova ene metü ken ami=[34b]=tan’u tusa-yin tulada
ögli=ge oyo=yata öggün üiled=deküi: arni=tan-dur quran mede=küi ali bükün mön
kü tere cü quran medekü ügei bülü=ge: tegün=cilen ireg=sen ber qamuy ami=tan
kemen ali=mad nomlay=san-’u mön kü tede=ger amita[n] cü ügei bolai: tere yayun-u
tulada kemebe=sü: Subudi ä tegün=cilen ireg=sen inu ün-e=ger nomlay=ci: ünen
jarliy-tu: mön kü tegün=cilen nomlaqui bülü’ge: tegün=cilen ireg=sen inu buru=yu
[35a] nom=loysan-u busu-yin tula bolai:
basa Subudi ä tegün=cilen ireg=sen ber ali nom cü ilete toyul=Ju burqan
boluy=san ba: tere üjü=gülü’g=sen-dür ünen cü ügei: qudal cü ügei bo=lai:
Subudi ä ene metü bülüge: üliger=lebesü nidün-lü’ge tegüsüg=sen nigen
kümün qara[n]g=yui-tur oro=basu yayun cü ülü üjükü tegün=cilen ali boda-tur
unaqui ber oyo=yat[a] ögli=ge ög=gügsen-ü bö=dhi=sadova-yi [35b] qaran üiled=
de’=küi: basa Subudi ä ene metü bülü’ge üliger=lebesü: ör[i] cayiyad naran uryu=ju
nidün-lü’ge tegüsüg=sen kümün ber eldeb jüil ö[n]gge-nuyud-’i üje=kü tegün=cilen:
ali boda-dur ülü unaqui ber ögli=ge-yi oyo=yata öggü’g=sen-’ü bö=dhi=sadova-yi
qaran üiled=deküi:
basa Subudi ä ijayur-tanu köbegün ba: ijayur-tanu: öken ali=mad nom-un jüil
egün-’i abqui kiged: bari=qui kiged: qaday-a=laqui kiged: qurulca=qui [36a] ki=ged:
qotala-yi onon üiled=kü kiged: busud-nuyud-tur ayui yekede ün-e=ger masida
üjügü=lügci tedeger-i inu tegüncilen ireg=sen ber ai=la’=dumu: tegün=cilen ireg=
sen ber üjekü bülü=ge: tede=ger qamuy ami=tan inu cayla=si ügei buyan-u coy=ca-yi
egü[s]ke=kü bolqu bolai:

146 А . Г. С А З Ы К И Н
15. basa Subudi ä ereg=tei ba: qatuy=tai ali=mad-tan ôrlüge-yi-yin cay-un üy-
e-dür bey-e-yi Gangga-a müren-ü qumaki-yin [36b] tedüi oyo=yata ôggü=ged: üde-
yin cay kiged üdesi-yin cay-un üy-e-dür cü bey-e-yi Gangga-a müren-’ü qumaki-yin
oyoyat[a] ôg’cü: ene metü jüil-iyer bau-a grag grag gbum toy-a-tan olan galab-tur
bey-e-yi oyo=yat[a] ôg=gügsen-ece: ali ber nom-un jüil egün-yi sonos=cu tebci=
besü: mon kü tere tere sitü=gen-ece neng olan toy-a=la’=si ügei cayla’=si ügei
buyan-i egüske=kü bögesü: ali ber üsü’g-tür bicijü ab=qui kiged: [37a] bari=qui
kiged: qadayala’=qui kiged: quralca=qui kiged: qotala-yi onon üiled=küi kiged:
busud-nuyud-tur ayui yeke=de üneger masida üjügü=lügsen-’ü edüge yayun ögülekü:
basa Subudi à nom-un jüil ene inu sedki=si ügei bolu=yad adalid=qasi ügei
bülü’ge: egün-ü teyin bolba=sural inu cü mon kü sedki=si ügei-dür on’an [=onon]
üiled=deküi: nom-un jüil ene anu dege=dü kölgen-dür ün-e=ger oroy=san ami=tan-
[37b]-nuyud-un tusa kiged: ketür’=kei-dür kürü’g=sen kölgen-dür ün-e=ger oroy=
san ami=tan-nuyud-un tusa-yin tula tegün=cilen ireg=sen ber nomlay=san bolai: ali=
mad nom-un jüil egün-i ab=qui kiged: bari=qui kiged: qadayala’=qui kiged: surulca=
qui kiged: qotala-yi onon üiled=küi kiged: busud-nuyud-tur ayui yekede ün-e=ger
masi=da üjü=gülügci tedeger-i inu tegün=cilen ireg=sen ber ayila’=dumu: tegün=
eilen ireg=sen ber üjekü bülü=ge: [38a] tede=ger qamuy ami=tan inu cayla’=si ügei
buyan-’u coy=ca-luy-a tegüs=kü boluyu: buyan-’u coy=ca sed=ki=si ügei kiged:
adalid=qasi ügei kiged: üli=si ügei kiged: kemjiy-e=lesi ügei-lüge tegüs=kü bolqu
bülü’ge: tede=ger qamuy ami=tan minu bô=dhi-yi e[r]kem-tür bari=qui bolqu
bülü’ge tere yayun-u tulada kemebesü: Subudi à door-a-du-yi sanay=ci-nuyud ber
[38b] nom-un jüil egün-i sonos=un ülü cidaqu bülü’ge: bi-tur üji’g=ci-nuyud ber
busu: ami=tan-tur üji’g=ci-nuyud ber busu: emiyn-tür [=amin-dur] üjigci-nuyud ber
busu bo=lu=yad bod=gali-dur üjigci-nuyud ber sonos=qui kiged ab=qui kiged
bari^qui kiged: surul=caqui kiged qotala-yi onon üiled=cü ülü cidaqu bülü’ge: tere
inu oron ügei-yin tula bolai:
basa Subudi à yajar-un jüg alin-tur sudur-un ayimay egün-i üjegü=lügsen-ü
tere yajar-un jüg tngri kiged kümün kiged asuri-[39a]-luy-a selte-yin yirten=cü-yin
takin üiled=küi-tür jokistu bolqu bolai: tere yajar-un jüg mörgün üiled=küi-tür
jokis=tu kiged: ergin üiled=küi-tür jokis=tu bolqu bülüge: tere yajar-un jüg takil-un
sitügen metü bolqu bola’i:
16. basa Subudi ä ijayur-tan-u köbegün ba: ijayur-tanu ôken ali=mad ene
metü-yin sudur-un ayimay-un de=ger[e] üges-’i ab=qui kiged: bari=qui kiged: qaday-
a=laqui kiged: surul=caqui kiged: qotala-yi onon üile’=dügci tede=ger inu nerbe’g=
[39b]=deküi bolqu: masi=da nerbe’g^deküi bolqu bolai: tere yayun-‘u tulada keme-
besü: Subudi ä tede=ger ami=tan inu urida törö’l üyes-ün buyan busu-yin üile mayui
jayayan-tur törö=kü bol=qui-yin ali=mad üile ü[i]le’=dügsen tede=ger inu mon kü
ene nasun-tur nerbeg=digsen-iyer: uridu törö’l üyes-ün buyan busu-yin üile mayui
jayayan-tur törö=küi bolqu ali=mad üile=dügsen tede=ger aril=qui bolqu bülü=ge:
burqan-’u bô=dhi-yi cü olqui bolqu bolai:
Subudi à bi ber ilete mede=jü: nôg=cig=sen cay [40a] toyala=si ügei-ece cü
neng toy-a=lasi ügei galab-tur: tegün=cilen ireg=sen dayin-i daruy=san ün-e=ger

toyu=luy=san Diy=ba[ng]=gara burqan-’u urida-aca cü neng uri=da: bau-a grag

nayan dörben gbum toy-a-tu burqan-nuyud boluy=san-i biber bayasqan üiled=dügsen
biilüge: ese bayas=qan ese üiled=cü sedkil-i inu côkü=gen ese üiled=dügsen bülüge:
Subudi ä ilaju tegüs nôg=cigsen tede=ger burqan-’i bi ber bayasqan üiled=[40b]=cü
sedkil-i inu côkü=gen ese üiled=düg=sen ali bükün-lüge: üdei-yin cay tabun jayun-’u
ecüs bolba=su: sudur-un ayimay egün-i abqui kiged: bici=küi kiged: bari=qui kiged:
qaday-a=laqui kiged: surul=caqui kiged: qot[a]la-yi onon üiled=kü ali bükün-ece:
Subudi â ene buyan-’u coy=ca-tur uri=du buyan-’u coy=ca ber: jayun qubi-tur cü
oyira=dun ülü odaju: mingyan qubi kiged: gbum-un qubi kiged: toy-а kiged kemjiy-e
kiged toy-a=laqui kiged: üliger kiged: [41a] sar-a kiged o’n-tur cü ülü teskü bolai:
Subudi à ker=be tere cay-tur: ijayur-tan-u kôbe=gün ba: ijayur-tan-’u ôken
ali=mad buyan-u coy=ca kedüi bükü masida bari=qui bolqu-yin ija=yur-tanu
köbegün ba ijayur-tanu ôken tede=ger-ün buyan-’u coy=ca-yi ber ôgülebe=sü: ami=
tan-nuyud toyiran soytayu=yiqu bolu=yad sed=kil kimura’=qui bol=qu bolai: basa
Subudi à nom-un jüil ene sedki=si ügei [41b] bülü=ge: egün-i teyin bolba=sural inu
cü mon kü sedki=si ügei-dür uqan [=onon] üiled=düküi:
17. tende-ece ilaju tegüs nôg=cigsen-tür amin qabiy-a-tu Subudi ber eyin
kernen ôci=bei: ilaju tegüs nög=cigsen à bô=dhi=sadova-yin kölgen-tür ün-e=ger
oroy=san-nuyud ber: yambar metü oro=sin üile=demü: yambar metü bütü=gen üile=
dümü: yambar metü sedkel masi=da barin üile=dümü: teyin kemen ôci=megce: ilaju
tegüs nôg=cigsen ber amin qabiy-a-tu Subudi-tur eyin [42a] kemen jarliy bo’l= bai:
Subudi à egün-dür bô=dhi=sadova-yin kölgen-tür ün-e=ger oroy=sad ber eyin kemen
bi ber qamuy ami=tan-i coy=ca-yi olal (?) üge=güy-e yasa=lang-aca nôg= cigsen-ü
töb-tür oyoyat[a] yasalang-aca nôgciy-e=sü’gei: tere metü cayla=si ügei ami=tan-’i
yasa=lang-aca nögciy-e=becü: alin cü ami=tan-i oyoya=ta yasa=lang-aca nôgciy-e=
küi boluy=san ügei bolai [42b] kemen sed=kel-’i egüs=ken üiled=deküi: tere yayun-u
tulada kemebesü: Subudi à ker=be bô=dhi=sadova ami=tan-tur quran mede=küi oro=
basu tegün-i bô=dhi=sadova ülü kemegde=kü böged: bod=gali kürtele quran mede=
küi ôci=besü cü: tegün-’i bô=dhi=sadova ülü kemegde=küi-yin tula bolai: tere
yayun-u tulada kemebesü: Subudi â bô=dhi=sadova-yin kölgen-tür ün-e=ger oroy=
san kemeg=deküi-yin tere nom alin cü ügei-yin tula bolai:
Subudi ä egün-i [43a] yayun kemen sedki=mü: tegün=cilen ireg=sen ber
tegün=cilen ireg=sen Diy=bang=gari-aca ali degere ügei ün-e=ger toyu=luy=san bô=
dhi-tur ilete toyul=ju burqan bolqu-yin tere nom alin cü bui kemekü güü: teyin
kemen jarliy bolumay=ca’: ilaju tegüs nôg=cigsen-tür nasun qabiy-a-tu Subu=di ber
eyin kemen ôci=bei: ilaju tegüs nög=cigsen ä tegün=cilen ireg=sen ber tegün=[43b]
=cilen ireg=sen Diy=’bang=gari-aca ali degere ügei ün-e=ger toyu=luysan bô=dhi=
tur ile=te toyul=ju burqan bolqu-yin tere nom alin cü ügei bolai: teyin kemen ôci=
meg=ce: ilaju tegüs nôg=cigsen ber amin qabiy-a-tu Subu=di-tur eyin kemen jarliy
bol=bai: Subudi â tere tegüncile[n] bolai: tere tegün=cilen bülüge: tegün=cilen ireg=
sen ber tegün=cilen ireg=sen Diy=’bang=gara-aca ali degere ügei ün-e=ger toyu=
luysan bô=[44a]=dhi-dur ilete toyul=ju burqan bolqu-yin tere nom ali cü ügei bolai:
Subudi ä ker=be tegün= eilen ireg=sen ber ali ile=te toyul^ju burqan bolqu-yin tere

148 А. Г. С А З Ы К И Н
nom jarim nigen bui bolba=su: tegün=cilen ireg=sen Diy-bang=gara ber nadur
biraman-’u kôbe=gün ci irege edüi cay-tur: tegün=cilen ireg=sen dayin-i daruy=san
ün-e=ger toyu=luy=san Saky-a-muni kemeg=deküi [44b] burqan bolqu bolai: kemen
esi ülü üjü’gül=kü nigen bögetele: Subudi a ene metü tegün=cilen ireg=sen ber ali
degere ügei ün-e=ger to= yuluy=san bô=dhi-tur ilete toyul=ju burqan bolqu-yin tere
nom alin cü ügei-yin tula teyimü-yin tula tegün=cilen ireg=sen Diy=bang=gara ber
nadur biraman-u köbegün ci irege edüi cay-tur tegün=cilen ireg=sen dayin-i daruy=
san ün-e=ger toyu=luysan Saky-a-muni [45a] burqan kemeg=dekü bolqu bolai
kemen esi üjügü=lügsen bolai:
tere yayun-u tulada kemebesü: Subudi â tegün=cilen ireg=sen kemekü inu: ün-
e=ger mön cinar-un tübden nereyid=dügsen üge mön-ü tula bolai:
basa Subudi à ali=mad jarim-ud nigen eyin kemen tegün=cilen ireg=sen
dayin-i daruy=san üneger toyu=luy=san burqan ber degere ügei ün-e=ger toyu=luy=
san bô=dhi-tur ilete toyul=[45b]=ju burqan boluy=san ôgülel=dübesü: tere buruyu
ôgüle’=kü mon bolai: tere yayun-u tulada kemebesü: Subudi â tegün=cilen ireg=sen
ber ali degere ügei ün-e=ger toyu=luy=san bô=dhi-tur ile=te toyul=ju burqan bolqu-
yin tere nom alin cü ügei-yin tula bolai: Subudi ä tegün=cilen ireg=sen ber ali nom
anu ile=te toyul=ju burqan bolqu ba: üjügü=lügsen tegün-tür ünen cü ügei: qudal cü
ügei bolai: teyimü-yin tulada tegün=cilen ireg=sen [46a] ber qamuy nom-yi burqan
nom kemen nomlay=san bolai: Subudi à qamuy nom kemeg=deküi anu: tede=ger
qamuy nom ügei mon bülü=ge: teyimü-yin tulada qamuy nom-’i burqan-’u nom
kemeg=deküi bolai:
Subudi ä ene metü bülüge üliger=lebesü: nigen töröl=kite=n kümün-ü bey-e-
lüge tegüskü=ged yeke bey-e-tü boluy=san metü bolai: amin qabiy-a-tu Subudi ber
ôcir-ün: ilaju tegüs nôg=cigsen à tegün=cilen ireg=sen ber törö’l=kite[n] [46b]
kümün-ü bey-e-lüge tegüsü=ged yeke bey-e-tü kemen ali nomlay=san tegün-i inu
tegün=cilen ireg=sen ber bey-e ügei-tür nomlay=san bülüge: teyimü-yin tula=da bey-
e-lüge tegüsü=ged yeke bey-e-tü kemeg=dekü bolai:
ilaju tegüs nôg=cigsen ber Jarliy bolur-un: Subudi â tere tegün=cilen bolai:
tere tegün=cilen bülüge: ali bô=dhi=sadova eyin kemen bi ber ami=tan-nuyud-’i
oyo=yata yasa=lang-aca nôgciy-e=sü’gei kemen ôgülel=dübesü: tegün-i bô=dhi=
sadova ülü kemeg=dekü bolai: tere yayun-u tulada [47a] kemebe=sü: Subudi â bô=
dhi=sadova kemegdekü-yin tere nom alin cü bui kemekü bi: Subu=di ber ôcir-ün:
ilaju tegüs nög=cigsen à tere inu busu bolai: ilaju tegüs nôg=cigsen ber jarliy bolur-
un: Subudi â teyimü-yin tulada: tegün=cilen ireg=sen ber qamuy nom-i ami=tan
ügei: aman [=amin] ügei bod=gali ügei bolai: kemen nomlay=san bolai::
basa Subudi à ali nigen bô=dhi=sadova eyin [47b] kemen: bi ber oron-’i
bayidal-nuyud-i bütü=gebei kemen ögülel=dübesü: tegün-i cü tegün=cilen ögülen
üiled=deküi: tere yayun-u tulada kemebesü: Subudi à oron-’i bayidal-nuyud-i oron-’i
bayidal-nuyud kemegdeküi anu: tede=ger-’i bayidal ügei-dür tegün=cilen ireg=sen
ber nomlay=san-u tulada bülü=ge: teyimü-yin tulada oron-i bayidal-nuyud kemegde=
küi bolai:
Subudi à ali nigen bô=dhi=sadova eyin kemen: nom-nuyud inu bi ügei-yi
[48a] nom-nuyud inu bi ügei bolai kemen sanayci tegün-i inu: tegün=cilen ireg=sen-

П Я ТЫ Й М О Н ГО Л ЬС К И Й ПЕРЕВОД «А ЛМ А ЗН О Й СУТРЫ » 149
ten dayin-i daruy=san iin-e=ger toyu=luy=san burqan ber: bô=dhi=sadova mâ=hâ=

sadova kemen ögüle’=kü bolai:
18. Subudi à egün-’i yayun kemen sedkemü: tegiin=cilen ireg=sen-tiir miqan-

’u nidiin bui kemekü güü: Subu=di ber öcir-ün: ilaju tegiis nög=cigsen à tere inu tere
metü bülüge: tegiin=cilen ireg=sen-diir miqan [48b] nidiin bui bolai: ilaju tegiis nög=
cigsen ber jarliy bolur-un: Subudi ä egiin-i yayun kemen sedki=mii: tegiin=cilen
ireg=sen-tiir tngri-yin nidiin bui kemekii giiii: Subu=di ber öcir-ün: ilaju tegüs nög=
cigsen â tere inu tere metü bui bülüge: tegün=cilen ireg=sen-tür tngri-yin nidün bui
bolai: ilaju tegüs nög=cigsen ber jarliy bolur-un: Subudi ä egün-i: yayun kemen
sedki=mü: tegün=cilen ireg=sen-dür bilig-ün nidün bui kemekü güü: Subu=di [49a]
ber öcir-ün: ilaju tegüs nög=cigsen ä tere inu tere metü bui bülüge: tegün=cilen ireg=
sen-tür bilig-ün nidün bui bolai:
19. ilaju tegüs nög=cig=sen ber jarliy bolur-un: Subudi ä egün-i yayun kemen
sedki=mü: tegün=cilen ireg=sen-tür nom-un nidün bui kemekü güü: Subu=di ber
öcir-ün: ilaju tegüs nög=cigsen ä tere inu tere metü bui bülüge: tegün=cilen ireg=sen-
dür nom-un nidün bui bolai: ilaju tegüs nög=cigsen [49b] ber jarliy bolur-un: Subudi
ä egün-i yayun kemen sedki=mü: tegün=cilen ireg=sen-tür burqan-’и nidün bui
kemekü güü: Subu=di ber öcir-ün: ilaju tegüs nög=cigsen ä tere inu tere metü bui
bülüge: tegün=cilen ireg=sen-tür burqan-u nidün bui bolai:
sedki=mü: Gangga-a müre=nü qumaki kedüi bükü Gangga-a müre=nü imayta tedüi
bükü bolu=yad: tede=ger-ün qumaki [50a] kedüi bükün-’ü yirtin=cü-yin oron
bolbasu: tede=ger yirtin=cü-yin oron-’i olan kemekü güü: Subu=di ber öcir-ün: ilaju
tegüs nög=cigsen ä tere inu tere metü bui bülüge: tede=ger yirtin=cü-yin oron inu
olan bolai: ilaju tegüs nög=cigsen ber jarliy bolur-un: Subudi ä tede=ger yirtin=cü-
yin oron-tur ami=tan kedüi bükü bui tede=ger-ün tusburi sanay-a=tan-’u sedkil-ün
ürgül=jilel-i biber [50b] masida medeg=sen bolai: tere yayun-u tulada kemebe=sü:
Subudi ä sedkil-ün ürgülji=lel-i sedkil-ün ürgül=jilel kemegde=küi inu: tegün-i
ürgül=jilel ügei-tür tegün=cilen ireg=sen ber nomlay=san-u tulada bülü=ge: teyimü-
yin tulada sedkel-ün ürgül=jilel kemeg=deküi bolai: tere yayun-u tulada kemebe=sü:
Subudi ä nögcig=sen sedkel cü joriy=daqu ügei: irege edüi sed=kel cü joriy=daqu
ügei: edüge boluy=san-’u sedkel cü joriy=daqu ügei-yin tula bolai:
Subudi ä egün-’i yayun kemen sedki=mü: [51a] alin ber yurban mingyan yeke
ming=yan yirtin=cü-yin oron egün-i dolo=yan jüil erdeni ber masida dügür=gen
üiled=cü öglige ög=besü: ijayur-tanu köbe=gün ba ijayur-tanu öken tere inu tere
sitü=gen-ece olan buyan-i egün gekü güü: Subu=di ber öcir-ün: ilaju tegüs nög=
cigsen ä olan bo=lai: sayibar oduy=san ä olan boloi: ilaju tegüs nög=cigsen ber jarliy
bolur-un: Subudi ä tere inu tegüncilen [51b] bolai tere inu tegünci=len bülü=ge:
ijayur-tanu köbe=gün ba ija=yur-tanu öken tere inu tere sitügen-ece buyan-u coy=ca
olan-i egü[s]ke=kü bolai: basa Subu=di ä ker=be buyan-’u соуса inu buyan-u соуса-
tur bolba=su: buyan-u coy=ca-yi buyan-’u coy=ca kemen tegün=cilen ireg=sen ber
ese nomlay=san bolai:
A da Orient. Hung. 52, 1999

150 А. Г. С А З Ы К И Н
20. Subudi â egün-i yayun kemen sedki=mii: öngge-yin bey-e oyo=yata

bütüg=sen-iyer tegiin=cilen ireg=sen-dür iijen [52a] üiled=dümü kemekü güü: Subu=
di ber öcir-ün: ilaju tegiis nög=cigsen ä tere inu busu bolai: öngge-yin bey-e oyo=
yata bütüg=sen-iyer tegiin=cilen ireg=sen-tiir iijen ülii üiled=dekü bolai: tere yayun-u
tulada kemebe=sü: ilaju tegiis nög=cigsen ä öngge-yin bey-e oyo=yata biitiig=sen
kemeg=deküi inu: tegiin-i oyo=yata bütüg=sen iigei-diir tegiin=cilen ireg=sen ber
nomlay=san-u tulada biilii=ge: teyimii-yin [52b] tulada öngge-yin bey-e oyo=yata
biitiig=sen kemegdekii bolai:
ilaju tegiis nög=cigsen ber jarliy bolur-un: Subudi à egiin-i yayun kemen
sedki=mü: beige qotala tegiisiig=sen-iyer tegiin=cilen ireg=sen-diir iijen üiled=dümü
kemekü güü: Subu=di ber öcir-ün: ilaju tegüs nög=cigsen ä tere inu busu bo=lai:
beige qotala tegü=süg=sen-iyer tegün=cilen ireg=sen-tür üjen ülü üiled=deküi bolai:
tere yayun-u tulada kemebe=sü: tegün=cilen [53a] ireg=sen ber beige qot[a]la tegü=
süg=sen ali nomlay=san tegün-’i: beige qotala tegü=sü’g=sen ügei-tür tegün=cilen
ireg=sen ber nomlay=san-u bülü=ge: teyimü-yin tulada beige qotala tegü=sü’g=sen
kemeg=dekü bolai:
21. ilaju tegüs nög=cigsen ber jarliy bolur-un: Subudi ä egüni yayun kemen
sedki=müü: tegün=cilen ireg=sen ber eyin kemen bi ber nom-i üjü=gü’lbei kemen
tayalay=san bolai kemebesü Subudi [53b] à tegün-i inu tere metü ülü üjekü bülüge:
tegün=cilen ireg=sen ber ali üjü=gülüg=sen tere nom alin cü ügei-yin tula bolai:
Subu=di ä ken nigen eyin kemen tegün=cilen ireg=sen nom-i üjügül=bei kemen
ögülel=dübesü: Subudi ä tere inu ügei kiged buruyu=da bariy=daqui ber namai-yi
mayud=qaysan bolqu bolai: tere yayun-u tulada kemebesü: Subudi ä nom-i üjügü=
lügsen kemeg=deküi inu: ali nom-i üjügü=lügsen inu nom-i üjügü=lügsen kemegde=
küi-tür joriyul=ju bolqu-yin tere nom alin [54a] cü ügei-yin tula bolai:
tende-ece ilaju tegüs nög=cigsen-dür amin qabiy-a-tu Subu=di ber eyin kemen
öci=bei: ilaju tegüs nög=cigsen ä irege edüi cay-tur: ali=mad ami=tan ene metü-yin
nom-i nomlay=san egün-i sonos=cu: nisvan (!) itege=küi bolqu ireg=ci edü=ge bui
buyu-’uu: ilaju tegüs nög=cigsen ber jarliy bolur-un: Subudi ä tede=ger inu ami=tan
cü bu=su: ami=tan [54b] ügei cü busu bolai: tere yayun-u tulada kemebesü: Subudi ä
ami=tan-nuyud kemeg=deküi inu: tegün=cilen ireg=sen ber tede=ger ami=tan-i ügei-
tür nomlay=san-u tulada bülüge: teyimü-yin tula=da ami=tan-nuyud kemeg=deküi
bolai:
22. Subudi ä egün-i yayun kemen sedki=mü: tegün=cilen ireg=sen ber ali
degere ügei ün-e=ger toyu=luy=san bö=dhi-tur ile=te toyul=ju burqan bolqu-yin tere
nom alin cü bui kemekü güü: [55a] amin qabiy-a-tu Subu=di ber öcir-ün: ilaju tegüs
nög=cigsen ä tegün=cilen ireg=sen ber ali degere ügei ün-e=ger toyu=luy=san bö=
dhi-tur ilete toyul=ju burqan bol=qu-yin tere nom alin cü ügei bo=lai: ilaju tegüs
nög=cigsen ber jarliy bolur-un: Subu=di ä tere inu tegünci=len bolai: tere inu tegün=
eilen bülüge: tegün-dür nom ücü=ken cü ülü joriy=dayad ügei bolai: teyimü-yin
tulada degere ügei ün-e=ger [55b] toyu=luy=san bö=dhi kemegde=kü bolai:

П ЯТЫ Й М О НГОЛЬСКИ Й ПЕРЕВОД «АЛМ А ЗН ОЙ СУТРЫ » 151
23. basa Subudi â tere nom inu tegsi büliige: tegün-dür ülü tegsid=kiii kiged
tegsid=küi alya=ju iigei-yin tulada: teyimü-yin tulada degere ügei ün-e=ger toyu=
luysan bô=dhi kemeg=dekü bolai: degere ügei ün-e=ger toyu=luy=san tere bô=dhi
inu bi ügei kiged: ami=tan ügei kiged: ami ügei kiged: bod=gali ügei-tür tegsi
bülüge: buyan-u qamuy nom-iyar ile=te toyul=ju burqan bol=bai: [56a| Subudi à
buyan-u nom-nuyud-’i buyan-u nom-nuyud kemeg=deküi inu: tede=ger-i mon kü
nom ügei-tür tegün=cilen iregsen ber nomlay=san bülüge: teyimü-yin tulada buyan-u
nom-nuyud kemeg=dekü bolai:
24. basa Subudi ä ijayur-tan-u köbe=gün ba: ijayur-tanu ôken ali jarim-ud
nigen ber yurban mingya’n yeke mingya’n yir=tin=cü-yin oron-da ayulas-yin qayan
Sümber nuyud kedüi bükü bui tedüi cinegen dolo=yan jüil erdeni-[56b]-yin coyca-yi
ilete quri=yaju ögli=ge ôg=gügsen-ece: alin ber bilig-ün cinadu kijayar kürügsen
egün-ece adqala=basu dörben üge-tü-yin silü’g-ün tedüi=ken-i bariju: busud-nuyud-
tur ün-e=ger üjü’gül=besü: Subudi ä buyan-u ene соуса-dur buyan-u uridu coy=ca
tere ber yayun qubi-tur cü oyira=dun ülü cida=qui-aca uy-un kürtele cü ülü teskü
bolai:
25. Subudi ä egün-i inada kemen sedki=mü: tegün=cilen ireg=sen ber eyin
kemen [57a] bi ber ami=tan-nuyud-i tonilya’=bai kemen tayalay=san bolai kemebe=
sü: Subudi ä tegün-i inu tere metü ülü üjeg=deküi: tere yayun-u tulada kemebesü:
Subudi à tegün=cilen ireg=sen ber ali tonilya[y]=san-u ami=tan alin cü ügei bolai:
Subudi ä ker=be tegün=cilen ireg=sen ber ali jarim nigen ami=tan-i boluy=san
bögesü: mön kü tere anu tegün=cilen ireg=sen-ü bi-dur bari=qui boluyu: ami=tan-tur
bari=qui [57b] kiged: amin-tur bari=qui kiged: bod=gali-tur bari=qui bolai: Subudi à
bi-tur bari=qui kemeg=deküi inu: bari=qui ügei-tür tegün=cilen ireg=sen ber
nomlay=san bögesü: tegün-i cü ilyan öber-e öber-e töröl=kite[n]-nuyud ber bari=qui
bolai: Subudi à ilyan obere obere töröl=kite[n]-nuyud kemeg=deküi inu: tede=ger
mön kü töröküi ügei-tür tegün=cilen ireg=sen ber nomlay=san bülü’^ge: teyimü-yin
tulada ilyan obere obere törölke=ten-nuyud kemeg=dekü bolai:
26. Subudi a egün-’i [58a] yayun kemen sedki=mii: belge qotala tegü=sü’g=
sen-iyer tegün=cilen ireg=sen-tür üjen üile’=dümü kemekü güü: Subu=di ber ôcir-
ün: ilaju tegüs nög=cigsen ä tere inu busu bolai: belge qotala tegü=sü’g=sen-iyer
tegün=cilen ireg=sen-tür üjen ülü üiled=kü bo=lai: ilaju tegüs nôg=cigsen ber jarliy
bolur-un: Subudi â tere inu tegüncilen bolai: tere inu tegün=ciien bülü’=ge: belge
qotala tegü=sü’g=sen-[58b]-iyer tegün=cilen ireg=sen-tür üjen ülü üiled=kü bülü’ge:
Subudi â ker=be belge qotala tegü=sü’g=sen-iyer tegün=cilen ireg=sen-tür üje’=küi
bolba=su: kürdü ber orci=yuluy=ci qayan cü tegün=cilen ireg=sen bolqu bülü=ge:
teyimü-yin tulada belge qotala tegü=sü’g=sen-iyer tegün=cilen ireg=sen-tür üjen ülü
üiled=kü bolai:
tende-ece ilaju tegüs nôg=cigsen-tür amin qabiy-a-tu Subu=di ber eyin kemen
ôci=bei: ilaju tegüs nög=cigsen à bi [59a] ber yambar metü ilaju tegüs nôg=cigsen
ber nomlay=san-u udqa-yi medeg=sen metü bögesü: belge qotala tegü=sü’g=sen-iyer

152 А. Г. С А З Ы К И Н
tegün=cilen ireg=sen-tür üjen ülü üiled=kü bolai: tendece ilaju tegüs nôg=cigsen ber
tegün-ü cay-tur silü’g-üd-’i jarliy bol=bai
ali=mad namai-yi öngge-tür üjeg’ci:

ali=mad namai-yi dayun-tur medeg’ci:
buru=yu mör-tür oroy=san bülü’=ge:
tede=ger törölki=ten namai-yi ülü üjeyü:
burqan-[59b]-nuyud inu nom-un cinar-i üjemü:
uduri=duyci-nuyud inu nom-un bey-e:
nom-un cinar-i meden üiled=kü busu-yin tula’=da:
tegün-i inu teyin ber meden ülü cidayu:
27. Subudi à egün-’i yayun kemen sedki=mü: belge qotala tegü=sü’g=sen-iyer

tegün=cilen ireg=sen dayin-i daruy=san ün-e=ger toyu=luy=san bô=dhi-tur ile=te
toyul=ju burqan bol=bai kemen bari=basu: Subu=di à ci tere metü dein ülü üiled=kü
bülü’ge: Subudi à belge qotala tegü=sü’g=[60a]=sen-iyer tegün=cilen ireg=sen
dayin-i daruy=san ün-e=ger toyu=luysan bô=dhi-tur ile=te toyul=ju burqan bolqu
ügei bolai: Subudi à ci eyin kemen bô=dhi=sadova-yin kölgen-tür ün-e=ger oroy=
san-nuyud ber: ali jarim nom-’i teyin ber ebdeg=sen ba: tasur’=qai-tur nereyid=
dügsen bolai kemen bari=basu: Subudi à tegün-i inu tere metü üjen ülü üiled=kü
bülü’ge: bô=dhi=[60b]=sadova-yin kölgen-tür ün-e=ger oroy=san-nuyud ber: nom
alin cü teyin ber ebdeg=sen ba: tasur’=qai-tur nereyid=dügsen ügei bolai:
28. basa Subudi à ijayur-tan-u kôbe=gün ba ijayur-tan-u ôken alin ber yirtin=
cü-yin oron Gangg’a müre=nü qumaki-yin toyo=sun-u tedüi nuyud-’i dolo=yan jüil
erdeni ber masida dügürgen üiled=cü ö’glige ôg=gü’gsen-ece: ali bô=dhi=sadova ber
nom-un jüil egün’i bi ügei bolu=yad törökü ügei-tür küliceng=güi-yi [61a] olba=su:
mon kü tere inu tere sitügen-ece neng olan toyala’=si ügei cayla=si ügei buyan-u
coy= ca-yi egüske=kü bolai:
basa Subudi â bô=dhi=sadova ber buyan-’u coy=ca-yi oyo=yat[a] barin ülü
üiled=deküi: amin qabiy-a-tu Subu=di ber ôcir-ün: ilaju tegüs nög=cigsen ä bô=
dhi=sadova ber buyan-u coy=ca-yi oyo=yat[a| barin ülü üiled=kü buyu-’uu: ilaju
tegüs nog=cig=sen ber [61b] jarliy bolur-un: Subudi a oyo=yata bari=qui amui ja-a:
buruyu=da ülü bariqu bülü’ge teyimü-yin tulada oyo=yata bariqu kemeg=dekü bolai:
29. basa Subudi à ali jarim-ud nigen eyin kemen: tegün=cilen ireg=sen-i odqui
ba ucira=qui ba: bos=qui ba: sayu=qui ba: noyir=san üiled=bei kemen ôgülel=
dübesü: tere ber minu nomlay=san udq-a-yi ese medegsen bolai: tere yayun-u tulada
kemebesü: Subudi à tegün=cilen ireg=sen kemeg=deküi inu: alin-tur cü ese oduy=
san alin-aca cü [62a] ese ireg=sen-ü tulada bülü’ge: teyimü-yin tulada tegün=cilen
ireg=sen dayin-’i daruy=san ün-e=ger toyu=luy=san burqan kemeg=dekü bolai:
30. basa Subu=di ä ijayur-tanu köbegün ba ijayur-tanu ôken ali jarim-ud nigen
ber: yurban mi[n]g=yan yeke mi[n]g=yan yirtin=cü-yin oron-tur yajar-un toyo=sun
П Я ТЫ Й М О Н ГО Л ЬС КИ Й ПЕРЕВОД «А ЛМ А ЗН ОЙ СУТРЫ» 153
kediii biikii bui tedeger-’i ene metii biiliige iiliger=lebesii masi narin toyo=sun-u
ciyul=yan metii kemkel=ün üiled=besü: Subudi ä egün-i yayun kemen sedki=mii:
masi [62b] narin toyo=sun-i ciyul=yan tere olan bolai kemekü güü: Subu=di ber öcir-
ün: ilaju tegüs nög=cigsen ä tere inu tere metü bülü=ge: masi narin toyo=sun-’u tere
ciyu=lyan olan bolai: tere yayun-u tulada kemebe=sü: ilaju tegüs nög=cigsen ä kerbe
masi narin toyo=sun‫־‬u nigen ciyul=yan bui bolba=su: ilaju tegüs nög=cigsen ber
masi narin toyo=sun-u ciyul=yan kemen jarliy ülü bolqu-yin tula bolai: tere yayun-u
tulada kemebe=sü: ilaju tegüs nög=cig=sen ber masi narin toyo=sun-u ciyul=[63a]
=yan kemen ali nomlay=san tegün-i ciyul=yan ügei-dür tegün=cilen ireg=sen ber
nomlay=san-u tula’=da bülü=ge: teyimü-yin tulada masi narin toyo=sun-u ciyul=yan
kemeg-de[kü] bolai:
tegün=cilen ireg=sen ber yurban mi[n]g=yan yeke mi[n]g=yan‫־‬u yirtin=cü-yin
oron kemen ali nom=loy=san tegün-’i oron ügei-dür tegün=cilen ireg=sen ber
nomlay=san bülü’ge: teyimü-yin tulada yurban mi[n]g=yan yeke mingyan yirtin=cü-
yin oron kemeg=dekü bolai: [63b] tere yayun-u tulada keme’=besü: ilaju tegüs nög=
cigsen ä kerbe yirtin=cü-yin oron nigen bui bol=ba=su: mön kü tere inu büküli-de
bariy=san bolqu-yin tula bolai: tegün=cilen ireg=sen ber büküli-de burqan ene-dür
nomlay=san tegün-i inu bariqu ügei-tür tegün=cilen ber nomlay=san-u tulada bülü=
ge: teyimü-yin tulada büküli-de bariysan kemeg=dekü bolai: ilaju tegüs nög= cigsen
ber jarliy bolur-un: Subudi ä mön kü büküli-dür bari=qui inu tomiga=[64a]=lal
bülü’=ge: tere nom inu ögüle’=si ügei mön bögetele: tegün-i cü ilyan obere obere
töröl=kite[n] nuyud ber bariqu bolai:
31. Subudi ä ali jarim-ud nigen eyin kemen: tegün=cilen ireg=sen ber bi-dur
üjekü ber nomla’=bai: tegün=cilen ireg=sen ber ami=tan-dur üje=küi kiged: amin-
dur üjeküi kiged: bod=gali-tur üjeküi ber nomlay=san bolai: kemen ögülel=dübesü
tegün-i inu üneger ögüle’=küi ber ögü=legsen mön buyu-’uu: Subu=di ber öcir-ün:
[64b] ilaju tegüs nög=cigsen ä tere inu busu bolai: sayibar ajiray=san ä tere inu busu
bolai: tere yayun-u tulada kemebe=sü: ilaju tegüs nög=cigsen ä tegün=cilen ireg=sen
ber bi-tur üjeküi ber ali nomlay=san tegün-i inu üjeküi ügei-tür tegün=cilen ireg=sen
ber nomlay=san-u tulada bülü=ge: teyimü-yin tulada bi-tur üjekü kemeg=deküi
bolai: ilaju tegüs nög=cigsen ber jarliy bolur-un: Subudi ä egün-tür bö=dhi=sadova-
yin kölgen-dür ün-e=ger oroy=sad ber: ene [65a] metü qamuy nom-i mede üiled:
üjen üiled: sanan üiled=kü bülü’ge: kerkejü nom-tur quran medeküi-tür orosil üge=
güy-e mede üiled=deküi: tere yayun-u tulada kemebesü: Subudi ä nom-dur quran
medeküi-yi nom-tur quran medeküi kemeg=deküi inu: tegün-i quran medeküi ügei-
tür tegün= eilen ireg=sen ber nomlay=san-u tulada bülü’ge: teyimü-yin tulada nom-
tur quran medeküi-yi nom-dur quran medeküi kemegde=kü bolai:
32. basa Subudi ä bö=dhi=sadova mä=hä=sadova alin [65b] ber cayla’=si ügei
toyala=si ügei yirtin=cü-yin oron-nuyud-i dolo=yan jüil erdeni ber masida dügür=gen
iiiled=cü ögli^ge ög=gügsen-ece: ijayur-tanu köbe=gün ba ijayur-tanu öken alin ber
belig-ün cinadu kijayara kürüg=sen egün-ece adqala=basu dörben üge-tü-yin silüg
tedüi=ken-i bieijü bariy=san ba: qaday-a=laqui ba: surul=caqui ba: qotala-yi onon

154 А. Г. С А З Ы К И Н
üiled=küi ba: busud-nuyud-tur ayui yekede ün-e=ger masi=da üjü=gülbesü: mon kü

tere [66a] inu tere sitü=gen-ece neng olan toyalay=si ügei cayla=si ügei buyan-u
coyca-yi egüske=kü bolai: yambar metü ün-e=ger masida üjügü=lümii kemebe=sü:
yambar metü ün-e=ger masi=da ülü ü]ü=gülkü tegün=cilen ün-e=ger masi=da üjü=
gülkü biilü=ge: teyimü-yin tulada iin-e=ger masi=da üjü=gülkü kemeg=dekü bolai:
odun bürügül=jekü jula’ kiged:

yelvi sigüde=ri usun-u côbü=rigüü kiged:
jegüdefn] cakil=yan kiged: egiilen metü:
amitan quran üile=dügsen-i tere [66b] metü üjegde=büi:
ilaju tegüs nôg=cigsen ber teyin kernen nom=la=ju: batud=dugsan oron

Subudi kiged: tedeger ayay-q-a tegim=lig kiged: tede=ger bô=dhi=sadova-nar kiged:
nökör dörbe=güle ayay‫־‬a tegem=lig kiged: eke ayay-a tegem-lig kiged: uba=si kiged:
uba=sanca kiged: tngri kiged: kümiin kiged asu=ri kiged: ünür idesi=ten selte-yin
yirtin=cü’ bayasul=caju: ilaju tegüs nôg=cigsen ber nom=lay=san-[67a]-dur ile—te
mayta’=bai:
qutuy-tu bilig-ün cinadu kija=yar-a kürüg=sen: vcir oyta=luyci kemeg=dekü

yeke kölgen sudur tegüs=bei:
namô Bha-ga-va-di: br-a-jna bara-mida-ya-a: ôm à dad-di-da: ili-si: ili-si: mi-
li-si: mi-li-si: bhan-[67b]-yan-a: bhan-yan‫־‬a: namô Bha-ga-va-di: bradi-dy-a bradi:
i-ri-di: i-ri-di: mi-ri-di: mi-ri-di: su-ri-di: su-ri-di: u-su-ri: u-su-ri: bhu-yu-yi bhu-yu-
yi suvva ha::
vacir oyto=layci-yin ene jirüken-i nigen-te’n urigsan-iyar vacir oyto=lugci-yin
tüme[n] yisü’n mingy’-a u[n]gsiy=san-luy-a saca’=yu bol=qu bolai:
[68a] yi dh[a]-rma-a hidu br-a-bha va-hi dondi-ka-a ta-ta-a-gatô hy-a-va dad:
di-ka-a gcig yô-gi rova-dha-va va-ba‫־‬a-di maha-a sr-a-ma‫־‬na soova-a ha-a::
Алексей Георгиевич Сазыкин

Пискаревский пр. д. 17. кори. 2, кв. 1.
195176, Санкт-Петербург, Россия

ПЯТЫ Й М ОНГОЛЬСКИЙ П ЕРЕВО Д «А Л М А ЗН О Й СУТРЫ »
[la] Qutuy-tu bilig-ün cinadu kijayar-a kürüg=sen v[c]ir oytaluy=ci kemegdekü orosi=bai:
156 А . Г. С А З Ы К И Н

П Я Т Ы Й М О Н ГО Л ЬС К И Й ПЕРЕВОД «А ЛМ А ЗН О Й СУТРЫ» 157

158 А. Г. С А З Ы К И Н
Лcfa Orient. Hung. 52, 1999

ПЯТЫЙ МОНГОЛЬСКИЙ ПЕРЕВОД «АЛМАЗНОЙ СУТРЫ» 159

160 А. Г. С А З Ы К И Н


162 А. Г. САЗЫКИН



П ЯТЫ Й М О Н ГО Л ЬС К И Й ПЕРЕВОД «АЛМ А ЗН ОЙ СУТРЫ» 165



168 А . Г. С А З Ы К И Н

П Я Т Ы Й М О Н ГО Л ЬС К И Й П ЕРЕВОД «А ЛМ А ЗН О Й СУТРЫ» 169

170 А. Г. С А З Ы К И Н
Дега Orient. Hung. 52, 1999

П Я ТЫ Й М О НГО Л ЬС КИ Й ПЕРЕВОД «А ЛМ А ЗН ОЙ СУТРЫ» 171

172 А . Г. С А З Ы К И Н
ОО

П Я ТЫ Й М О Н ГО Л ЬС КИ Й ПЕРЕВОД «А ЛМ А ЗН О Й СУТРЫ » 173

174 А. Г. С А З Ы К И Н

П Я Т Ы Й М О Н ГО Л ЬС К И Й ПЕРЕВОД «А Л М А ЗН О Й СУТРЫ » 175

176 А. Г. С А З Ы К И Н
-О Й
(Ч
СЧ (Ч

П ЯТЫ Й М О Н ГО Л ЬС КИ Й ПЕРЕВОД «А ЛМ А ЗН ОЙ СУТРЫ » 177
23a

178 А. Г. С А З Ы К И Н

П ЯТЫ Й М О Н ГО Л ЬС К И Й ПЕРЕВОД «А ЛМ А ЗН О Й СУТРЫ» 179

180 А . Г. С А З Ы К И Н
26a

П Я Т Ы Й М О Н ГО Л ЬС К И Й ПЕРЕВОД «А Л М А ЗН О Й СУТРЫ» 181
Acfa Orient. Hung. 52, 1999

182 А. Г. С А З Ы К И Н
cd
00
(N
Acta Orient. Hung. 52, J 999

П Я Т Ы Й М О Н Г О Л Ь С К И Й ПЕРЕВОД «А Л М А ЗН О Й СУТРЫ » 183
(Я
CN

184 А . Г. С А З Ы К И Н
30a

П Я Т Ы Й М О Н Г О Л ЬС К И Й П ЕРЕВ О Д «А Л М А ЗН О Й СУТРЫ » 185

186 А. Г. С А З Ы К И Н

а
со
со

188 А . Г. С А З Ы К И Н
Л е т Orient. Hung. 52, 1999

П Я Т Ы Й М О Н Г О Л Ь С К И Й ПЕРЕВОД «А ЛМ А ЗН О Й СУ ТРЫ » 189
35a

190 А . Г. С А З Ы К И Н

П Я Т Ы Й М О НГО Л ЬС КИ Й ПЕРЕВОД «АЛМ АЗНОЙ СУТРЫ » 191

192 А. Г. С А З Ы К И Н

П ЯТЫ Й М О Н ГО Л ЬС К И Й ПЕРЕВОД «А ЛМ А ЗН О Й СУТРЫ » 193
<‫מ‬
‫אס‬
r<‫־‬,
к Acta Orient. Hung. 52, 1999

194 А. Г. С А З Ы К И Н

П ЯТЫ Й М О Н ГО Л ЬСКИ Й ПЕРЕВОД «АЛМ А ЗН ОЙ СУТРЫ » 195
41a
Д67а Orient. Hung. 52, 1999

196 А. Г. С А З Ы К И Н

П ЯТЫ Й М О Н ГО Л ЬС К И Й ПЕРЕВОД «А ЛМ А ЗН О Й СУТРЫ » 197

198 А. Г. С А З Ы К И Н
‫ס‬
44a
(‫ס‬


200 А . Г. С А З Ы К И Н

П Я Т Ы Й М О Н ГО Л ЬС КИ Й ПЕРЕВОД «А ЛМ А ЗН ОЙ СУТРЫ » 201
Дега Orient. Hung. 52, 1999

202 А. Г. С А З Ы К И Н

49a

204 А. Г. С А З Ы К И Н

П Я Т Ы Й М О Н ГО Л ЬС КИ Й ПЕРЕВОД «А ЛМ А ЗН О Й СУТРЫ » 205
51a

206 А. Г. С А З Ы К И Н

П Я Т Ы Й М О Н ГО Л ЬС КИ Й ПЕРЕВОД «А ЛМ А ЗН О Й СУТРЫ » 207
03
СО
‫י‬/‫ר‬
Acte Orient. Hung. 52, 1999

208 А. Г. С А З Ы К И Н

П Я Т Ы Й М О Н Г О Л ЬС К И Й П ЕРЕВОД «А Л М А ЗН О Й СУТРЫ» 209

210 А. Г. С А З Ы К И Н
.‫ם‬
56a
ю
io

П Я Т Ы Й М О Н ГО Л ЬСКИ Й ПЕРЕВОД «А Л М А ЗН О Й СУТРЫ» 211
03
г-
ю
Acta Orient. Hung. 52, /9 9 9

212 А. Г С А З Ы К И Н
Х>
г-
U~)

П Я Т Ы Й М О Н ГО Л ЬС КИ Й ПЕРЕВОД «А ЛМ А ЗН ОЙ СУТРЫ » 213

214 А . Г. С А З Ы К И Н
O n
ю

П Я ТЫ Й М О НГОЛЬСКИ Й ПЕРЕВОД «АЛМ АЗНОЙ СУТРЫ » 215
Лета Orient. Hung. 52, 1999

216 А. Г. С А З Ы К И Н
62a

П ЯТЫ Й М О Н ГО Л ЬС КИ Й ПЕРЕВОД «А ЛМ А ЗН ОЙ СУ ТРЫ » 217
63a
Лета Orient. Hung. 52, 1999

218 А . Г. С А З Ы К И Н
64a

П ЯТЫ Й М О Н ГО Л ЬС КИ Й ПЕРЕВОД «А ЛМ А ЗН О Й СУ ТРЫ » 219
Л
1/‫ר‬
40

2 20 А . Г. С А З Ы К И Н
66a
A cfa Orient. Hung. 52, 1999

П Я ТЫ Й М О Н ГО Л ЬС К И Й ПЕРЕВОД «А ЛМ А ЗН О Й СУТРЫ » 221
-О
67a
40
40

222 А. Г. С А З Ы К И Н
68a

Acta Orientalia Academiae Scientiarum Hung. Volume 52 (2), 223 (1999)
Brown, K. ed. (1999), Sikh Art and Litera- Lau, A. (1998), A Moment of Anguish. Singa-
ture. London-New York, Routledge. GBP pore in Malaysia and the Politics o f Dis-
17.99. engagement. Singapore, Times Academic
Futaky, I. (1991), ‫״‬Seilige Tage im Musensitz Press.
Göttingen‘‘ Stadt und Universität in unga- Ôztopçu, K .-Abuov, Zh.-Kambarov, N .-A ze-
rischen Berichten aus dem 18. und 19. moun, Y. (1999), Dictionary of the Turkic
Jahrhundert. Göttingen, Vandenhoeck und Languages. London-New York, Routledge.
Ruprecht. USD 19.99.
Gülberg, N. (1999), Buddhistische Zeremoniale Schneider, I. (1999), Kinderverkauf und Schuld-
(Koshiki) und ihre Bedeutung für die Lite- knechtschaft. Untersuchungen zur frühen
ratur des japanischen Mittelalters. (Mün- Phase des islamischen Rechts. (Abhandlun-
chener Ostasiatische Studien 76.) gen für die Kunde des Morgenlandes 52:1),
Gungwu, W. (1998), The Nanhai Trade. The Stuttgart, Steiner.
Early History of Chinese Trade in the South Vikar, L.-G. Bereczki, (1999), Tatar Folksongs.
China Sea. Kuala Lumpur, Times Аса- Budapest, Akadémiai Kiadö. USD 120.
demie Press.
Haig, G. (1998), Relative Constructions in Turk- Studies and Essays. History, Archaeology, Ge-
ish. (Turcologica 33), Wiesbaden, Harrasso- ography 19.
witz. The Tibet Journal XXIII: 4.
Akadémiai Kiadö, Budapest

ABBREVIATIONS OF JOURNALS AND SERIES
AEMA Archivum Eurasiae Medii Aevi, Wiesbaden

A OH Acta Orientalia Academiae Scientiarum Hungaricae, Budapest
ArO r Archiv Orientalni, Praha
B ASOR Bulletin of the American Schools of Oriental Research
B GA Bibliotheca geographorum arabicorum
BIFAO Bulletin de l ’Institut Française d'Archéologie Orientale du Caire
BiOr Bibliotheca Orientalis
E l, E l2 Encyclopaedia of Islam (1st and 2nd edition)
FO Folia Orientalia, Krakdw
IC Islam Culture
KCsA Korösi Csoma Archivum, Budapest
KK Keletkutatds, Budapest
MFOB Mélanges de la Faculté Orientale de l ’Université St. Joseph de Beyrouth
M ID E O Mélanges de l ’Institut Dominicain d’Études Orientales du Caire
MRDTB Memoirs of the Research Department of the Toyo Bunko, Tokyo
NAA Narody Azii i Afriki, Moskva
N yK Nyelvtudomänyi Kôzlemények, Budapest
O LZ Orientalische Literaturzeitung, Berlin
OsmAr Osmanli Ara§t1 rmalar1 , Istanbul
REJ Revue des Études Juives
Stil Studien zur Indologie und Iranistik. Reinbek
TD A Y B Türk Dili Araçtirmalari Yilligi Belleten, Ankara
TDKY Türk Dili Kurumu Yayinlan
TP T ’oung Pao, Leiden
TS AB Turkish Studies Association Bulletin, Bloomington
UAJb N F Ural-Altaische Jahrbücher: Neue Folge, Wiesbaden
W ZKM Wiener Zeitschrift für die Kunde des Morgenlandes, Wien
ZD M G Zeitschrift der Deutschen Morgenländischen Gesellschaft, Wiesbaden
tat Bonn, Wiesbaden
ZV O IR A O Zapiski vostocnago otdelenija Imperatorskago russkago arheologiceskago obscestva, Sanktpeterburg
PRINTED IN HUNGARY
INFORMATION FOR CONTRIBUTORS
oriental studies, including Turkic, Mongolian, Manchurian, Chinese, Tibetan, Indian, Iranian and Semitic
philology, literature and history. A copy o f publishing agreement w ill be sent to the authors o f papers accepted
by two reviewers and the board of editors for publication. Manuscripts w ill be printed only after receiving the
Andrds Rona-Tas
Editor-in-Chief
H - l 146 Budapest
A jtôsi Durer sor 19-21
from the authors together with the first proof.
T e x t. The total length o f the text (including illustrations, tables, etc.), if not agreed otherwise, should not
exceed 20 typewritten pages. The first page should contain the title o f the paper, the author’s names and
Abstract. Manuscripts o f articles should be accompanied by an abstract o f about 100 words / 15 lines
printed on a separate sheet o f paper summarising the conceptual content and the new results o f the article.
Key words. The authors are kindly asked for 6 to 8 key words added to their articles.
Please indicate the top o f the figure and the name of the author o f the paper on the back. High quality half-
Footnotes. A ll footnotes must be placed on the same page, as the text and numbered continuously.
References. Full citation o f literature referred to should be given in an alphabetically arranged list at the end
the text. In the references use full names o f journals if they are not to be found on the last page o f A O H . In the
text they should be cited by the names o f the authors and the year o f publication, e.g. Doerfer and Tezcan
(1980), or if pages are referred to: Doerfer and Tezcan (1980, pp. 21 -2 2 ). In case o f publications o f the same
year use letters: Doerfer and Tezcan (1980a).
Reference to books:
Doerfer, G.-Tezcan, S. (1980), Wörterbuch d es Chaladsch. (Dialekt von Charrab). Budapest (Bibliotheca
Orientalis Hungarica X X V I) .
Gervers, M . (1997), Nomads as gatherers and creators o f world culture. The artistic sources o f the great felt
hanging from Pazyryk, Kurgan 5. Summaries o f congress papers. 7th International Congress o f M ongolists
1 2 -1 6 A ugust 1997, Ulaanbaatar. p. 3. Ulaanbaatar.
Berta, Â. (1994), Zum Wandel des auslautenden -G im Kiptschakischen. Journal ofT u rko lo g y 2:2, pp. 163-195.
Transcription. The authors should use the transcription internationally accepted in their respective field o f
studies. Please try to avoid the usage of special character that is not to be found in W inW ord 95 or 98 Symbol
True Type font. Try to avoid Chinese, Arabic, etc. non Latin or C yrillic characters. Use the Pinyin transcrip-
special symbols, please enclose a floppy containing the fonts (in case o f Chinese preferably Big5, G B or
Chinese Star).
Proofs. The author w ill receive the first proof. Please read it carefully and return it promptly, i f necessary by
Reviews. General rules o f scientific publication (see above) should be followed.
Addresses of the authors. We publish the mailing addresses o f the authors as they are given in the publish-
HU ISSN 0001-6446
ACTA
ORIENTALIA
HUNGARICAE
VOL. 52 (1999) NOS 3 -4

ACTA ORIENTALIA
FOUNDED
BY
L. LIGETI
E DITO RIAL B O A R D
Ildikô E csedy
Pal Fodor
Éva J eremias
Andrâs R öna-T as , Editor-in-Chief
Robert S imon
Gyula WojTiLLA
A D V IS O R Y B O A RD : G y. HAZAI, Z s . K a k u k , Gy. K a r a , M . M a r o t h , F. TÖKEI
T E C H N IC A L EDITORS: É va CsÂKl and Imre H a m a r
brief accounts on past and com ing events in oriental studies. ACTA ORIENTALIA is published in
yearly volumes of four numbers by AKADEMIAI KIADO, H-l 117 Budapest, Prielle Kornélia u. 4.
http://www.akkrt.hu
Subscription information: Orders should be addressed to AKADEMIAI KIADO, H-l519 Budapest,

Subscription price for Volume 52 (1999) in four numbers is US$ 156.00 including normal postage,
Editorial correspondence: Manuscripts and correspondence with the editors should be sent to
ACTA ORIENTALIA, House o f Professors. Room 303, H-l 146 Budapest, Ajtösi Dürer sor 19-21.
© Akadémiai Kiadô, Budapest 1999
PRINTED IN HUNGARY
AOrient 52 (1999) 3 -4
ACTA
ORIENTAL!A
HUNGARICAE
VOL. 52 (1999) NOS 3 - 4

CONTENTS
G yula W0JTILLA, The Sanskrit Godhûma apropos of a short excursion in Indo-European and
Indo-Aryan prehistory .................................................................................................................. 223
Röbert S imon, Muhammad and the Jihäd ............................................................................................. 235
A ndrey Korotaev- V ladimir Klimenko- D mitry Proussakov, Origins o f Islam: Political-
anthropological and environmental c o n te x t............................................................................. 243
L äszlö J. N agy , Un gouverneur d’origine hongroise à Alger 1580-1582 .................................... 277
Imre H amar , Buddhism and the Dao in Tang China: The Impact o f Confucianism and
Daoism on the philosophy of C hengguan................................................................................. 283
К л а р а А дягаш и , К вопросу об этногенезе м а р и й ц е в ............................................................. 293
Reviews
Chlodwig Н. Werba, Verba indoarica. Die primären und sekundären Wurzeln der Sanskrit-
Sprache. Pars I.: Radices primariae (G y . W ojtilla) ............................................................. 309
Klaus G. Witz, The Supreme Wisdom of the Upanisads. An Introduction (F. Ruzsa) ............... 310
Läszlö Vikdr-Gâbor Bereczki, Tatar Folksongs (J. S ipos) ................................................................ 311
Klaus Karttunen, India and the Hellenistic World (G y . W ojtilla) ..................................................... 313
Klaus Mylius, Wörterbuch Päli-Deutsch (G y . W ojtilla) ................................................................. 314
Untersuchungen im Totentempel des Merenptah in Theben unter der Leitung von Horst Jaritz
III. Tore und andere wiederverwendete Bauteile Amenophis’ III von Susanne Bickel
mit Beiträgen von Horst Jaritz, - Horst Jaritz-Uwe Minuth-Raphael A. J. Wüst
(L. K äkosy) ................................................................................................................................... 316
Edmund Schütz, Armeno-Turcica. Selected Studies (A. R öna -TAS) .............................................. 317
Books and journals received ..................................................................................................................... 321
Addresses of the authors ............................................................................................................................ 323
Contents o f Volume 52, Nos 1 - 4 .............................................................................................................. 327

Acta Orientalin Academiae Scientiarum Hung. Volume 52 (3-4), 223-234 (1999)
THE SANSKRIT GODHÜMA

APROPOS OF A SHORT EXCURSION
IN INDO-EUROPEAN AND INDO-ARYAN PREHISTORY
G y u l a W o jt il l a
(Szeged)
It has been suggested that research into the history o f agriculture throws light on details o f the pre-
history o f Indo-European speaking peoples. This issue has been the subject o f heavy debates among
archaeologists; however, it has only sparsely been dealt with by students of the history o f Indo-
European languages. The Sanskrit godhiima appears to be one o f the most promising terms that can
be traced down from Anatolia to India. The present paper lays special emphasis on the way the word
got into India, and offers a plausible solution to the problem o f its absence in the Rgveda. Subject-
ing its cognates to closer examination allows us to identify cultural and language borrowings in
a much broader temporal and spatial framework than it has been assumed.
Key-words: Indo-European, Indo-Aryan, Sanskrit, godhiima, agriculture, Vedic India.
Renfrew’s process model offers a hypothesis about the spread of Indo-European

(hereafter: Œ) languages to West and East. Renfrew sees a direct connection be-
tween the spread of IE languages and the spread of agriculture (Renfrew 1989, pp.
126-131; cf. Zvelebil-Zvelebil 1990, p. 243). Renfrew’s hypothesis met strong criti-
cism as well as support from archaeologists and linguists; nevertheless, it gave fresh
impetus to the research of the interrelationship between the language and the culture
of its speakers (cf. Lehmann 1992, p. 6). In defence of Renfrew, Erdosy calls attention
to his greatest merit, namely that he replaces the old equation: language = ethnos or
‘people’ = culture with the formula of language change <-> socio-economic change
<-» change in material culture. Since socio-economic changes likely cause language
changes, their study is “well within archaeologists’ competence” (Erdosy 1995, pp.
22-23). Nevertheless, studies in lexicology from the viewpoint of Wörter und Sachen
are justified (Elizarenkova 1998, p. 71), and so treating words, as bearers of certain
cultural information in IE languages or even in their early reconstructed stages, prove
to be very fruitful (cf. Gamkrelidze-Ivanov 1984, vol. 2).
There are numerous linguistic problems around Renfrew’s hypothesis. Baldi
and Coleman raise doubts about the assumption that the first farmers of Anatolia -
from where Renfrew believes Ш speakers moved - spoke Indo-European (Baldi
1988, p. 446; Coleman 1988, p. 453). Coleman and Norman see an unbridgable gap
between the proposed date of the process around 6000 BC and the appearance of Ш
0001-6446 /99/ $ 5.00 © 1999 Akadémiai Kiadö. Budapest
224 GYULA WOJTILLA
languages in history (Coleman 1988, p. 452; Norman 1988, p. 73). Diakonoff tries to
reconciliate this obvious difficulty. He accepts the main thesis, i.e. the connection of
the spread of agriculture and Œ dispersals (Diakonoff 1988, p. 83); however, he also
has doubts about whether the first agriculturists were indeed IE speakers (Diakonoff
1988, p. 84). He reckons that the farmers of Çatal-Hüyük spoke a Pre-Proto-Indo-Euro-
pean language “which might have been the ancestor of PIE and other languages”
(Diakonoff 1988, p. 86). In a letter to Lehman on 5th April 1992 he calls the people
of Çatal-Hüyük Proto-Proto-Indo-Europeans, plus Proto-Proto-Kartvels, and calls
those, who migrated from there to the Balkans, Proto-Proto-Indo-Europeans (cf. Leh-
mann 1992, p. 42). Dolgopolsky subscribes to Renfrew’s conclusions drawn from ar-
chaeological evidence, insofar as he believes that this verifies his own ideas (Dolgo-
polsky 1990-1993, p. 230). In fact, there is nothing new in his argumentation, since
he simply uses the so-called Indo-Hittite theory set up by Sturtevant in the thirties
(Dolgopolsky 1990-1993, p. 235). Under the influence of criticism Renfrew acknowl-
edges in his latest paper that the spread of agriculture and IE dispersal to the East are
not without difficulties (Renfrew 1996, p. 82).
In a former paper on terms denoting ‘ploughing’ and ‘plough’ we also tried to
expand the spatial and temporal limits by proposing a special Sanskrit term karsati-
‘ploughs’ denoting the procedure which can be identified in both Western and East-
ern IE languages (Wojtilla 1986, pp. 30-33). Of course this assumption also has its
shortcomings. Firstly, the verb is onomatopoeic, and secondly we could not find evi-
dence of an antecendent to the Vedic language - therefore the ‘grey zone’ preceeding
it is shrouded in obscurity - and at best we are left to rely on reconstructions where
the verb has other meanings such as ‘scratches’, ‘draws a line’, etc. In spite of the evi-
dent obstacles, we do not think this idea should be dropped altogether. Let us be re-
minded that verbs denoting such procedures belong to the core vocabulary of many IE
languages, and are not easily liable to change, such as the names of tools, for instance.
In short, this single example encourages us to proceed in Wortforschung fur-
ther. However, we must be highly circumspect, as is Mayrhofer in his short paper on
an alleged Indo-Aryan legal term in a Mitanni letter (Mayrhofer 1996, pp. 161-162).
Our choice fell on wheat, a cereal from the Neolithic Age, commonly pro-
duced throughout Asia. It is called godhüma in Sanskrit. We believe that the study
of the history of this word can considerably enrich our knowledge, and also support
Renfrew’s hypothesis.
It cannot be denied that the word was a disputed problem for specialists for
many years, and as it will be pointed out, there is still no common consent about it.
We regard Harmatta’s paper treating the early history of the word as a landmark in
research. Harmatta genuinely seeks to explain its origin and spread in IE and Semitic
languages. His sweeping conclusion is worth quoting: “the history of the term truly
reflects both the spread of agriculture from the Near East and the cultural relations
between Anatolians, Semites, Indo-Iranians and the Western European tribes in the
course of the developoment of mankind during the last five-six millennia” (Harmatta
1988, p. 248). One should bear in mind that the original Hungarian version of the
paper was published in 1984 - well before the appearance of Renfrew’s book!

THE SANSKRIT GODHÜMA 225
The word escaped the attention of Gamkrelidze and Ivanov. Being an archae-
ologist, who uses linguistic material only sparsely, Renfrew also fails to mention it,
or at least he does not earmark wheat for meticulous study. Mayrhofer, in his new
etymological dictionary, records Harmatta’s view without comments (EWA, pp. 498-
499). In his opinion the Sanskrit word is of foreign origin, and took its present form
by means of popular etymology. Otherwise he seems to basically use the results of
the respective part of a paper by Berger (cf. Berger 1959, pp. 39-43). On the other
hand, it is a pity that Harmatta, who treats the developments in the Indian linguistic
area only briefly, did not read Berger’s writing.
Berger derives the Sanskrit godhüma- and its New Indo-Aryan (hereinafter:
NIA) developments from an Old Burusaski *yendum-lyondum- and connects it with
the Gypsy khurmin-, khurmi- ‘Weizen, Hirse, Griitebrei’. According to him the varia-
tion -a-/-e- in the first syllable forms a kind of morphological opposition. On the ba-
sis of the appearance of the initial aspirated consonant in the Gypsy word he postu-
lates a *ghendüma-/ghandüma- as the antecendent of the NIA forms, and he reckons
that the de-aspiration is due to a sudden shift (Umsprung) that is caused by the influ-
ence of the popular etymology in question (Berger 1959, p. 40). In his view the -й- in
the second syllable cannot represent the earliest stage in the Indian linguistic area
because the -u- in the Tamil kotikuma- kotuma- reflects a more archaic one (Berger
1959, p. 42). Since the word has the form gantuma- in Avesta, a fluctuation of -dh-
and -t- is conceivable, and this must account for a Burusaski origin, owing to the fact
that this kind of interchange is quite normal in that language (Berger 1959, p. 41). By
linking Burusaski with languages of the Mediterranean, he shifts the origin of the
plant denoted by the word also, to a remote place in India (Berger 1959, p. 44).
We think that some of his statements should be viewed with scepticism, and
only regret that neither Harmatta nor Mayrhofer gave them consideration. First of all,
the parentship of Burusaski is a puzzle (cf. Klimov-Edel’man 1970, p. 10 and p. 19).
We do not see any reason why we should accept the fact that Old Burusaski words are
comparable with Vedic ones. For the appearance of the initial aspirated consonant
Harmatta’s reconstruction of an Anatolian *ghondu- may be quoted (Harmatta 1988,
p. 245). The presence of this aspirate consonant can be observed in NLA: Gujarati
ghan-/ghen- (Prasad 1959-1966, p. 17) and in Pamirian dialects: ghidim- (Schrader
1883, p. 358). Ascribing a short -u- to a Dravidian/Tamil origin is not as simple as
Berger believes. This short vowel occurs in a nineteenth-century Sanskrit lexicon, the
Sabdakalpadruma (cf. pw II, p. 180 and Apte 1957, p. 672). In Pali and Jain Prakrit
one can find forms with -ü- (PTS p. 255; Sheth 1963, p. 303). At the same time it ap-
pears in Niya documents as goduma- in the 3rd century AD (CDIAL, p. 4287). Simi-
larly, -u- is common in Iranian: Avestan gantuma- (ibid) and Afghan yanum- (Horn
1893, p. 209). Beside the nasalisation of -u-, there is a development of goü- in Nepali,
owing to the length of the vowel developing a nasal resonance. This nasal was intro-
duced no later than the Prakrit stage (Srivastava 1962, p. 27). A counter-check in
Dravidian material also contradicts Berger’s statement. Burrow and Emenau do not
record godhüma- as a Dravidian borrowing in Sanskrit (cf. DED and DEDS). Gonda
also rules out the possibility of a Dravidian derivation, together with a Finno-Ugric

226 GYULA WOJTILLA
or Semitic one (Gonda 1971, pp. 215-216). The assumption of the presence of Drav-
idian speakers in the area of the genesis of the Rgveda rests on slender foundations.
This theory, constructed on Brahui speakers, has been challenged by Elfenbein who
speaks of a much later migration of these people into that region (Elfenbein 1987, p.
217). Curiously enough, the Dravidian derivation put down deep roots and even seri-
ous scholars such as Southworth (Southworth 1979, p. 204 and 1988, pp. 655-660)
and even Witzel (Witzel 1995, p. 102) repeat it. The ‘hunt’ for Dravidian loan-words
in Sanskrit was (and is, to some extent) a decreasing fashion (cf. Thieme 1994, p.
327). As Das formulates it, “ ...any statement on interaction with Dravidian or Austro-
Asiatics in Rgvedic times is, in the light of our present knowledge, nothing but specu-
lation, which may be justified, but equally well may not” (Das 1995, p. 218). Its ad-
herents do not even bother about McAlpin’s well-founded statement, that it is defi-
nitely clear that the terminology of wheat is lacking in Proto-Dravidian (McAlpin
1979, p. 181). Berger owes us an explanation of the disappearance of -n- in Sanskrit,
Prakrits and some NIA languages. It can be ascribed to the frequently mentioned popu-
lar etymology. This -n- has been retained in Iranian languages: Avestan gantuma-,
Choresmian yndym- (Humbach 1989, p. 193), Pashto (Waned dialect) yandan-
(Skjærvp 1989, p. 403). Berger’s explanation of the variation of -a-/-e-/-o- is too broad.
Curiously, Southworth contents himself with merely raising the issue (Southworth
1979, p. 201). It must be confessed, however, that the problem remains unresolved.
Before discussing the different opinions, we must refer to two isolated re-
marks also unnoticed in EWA. Masica believes that the word is confined to Indo-
Iranian (Masica 1979, p. 76). It is noteworthy that he locates its source in an un-
known pre-IE neolithic language of North Iran/Afghanistan, from where bread wheats
are thought to have originated (Masica 1979, p. 104). What interests us here is the
time span. The relegation of the term to the category of the unknown languages can
be methodically understood, but it hardly brings us closer to a plausible solution.
Nevertheless, Erdosy also appears to adhere to it. He toys with the idea that the
undeciphered Harappa script might have preserved this unknown language (Erdosy
1995, p. 16). Finally, the brief opinion of Kuiper deserves mentioning. He clearly puts
it in the group of foreign words of the Rgveda, adopted before the Aryans reached
India (Kuiper 1991, p. 90).
As it has been mentioned, Harmatta paves the way to further research and
immensely extends the temporal and spatial frames of the prehistory of the Sanskrit
godhuma- when he puts forward the reconstructed go~dhi1- which might have entered
Proto-Indie (Harmatta 1988, p. 245). He also convincingly points out that the Avestan
gantuma- goes back to a later Anatolian *gantu- (Harmatta 1988, p. 242), and doing
so he solves the problem of deriving the Avestan form (cf. Hübschmann 1890, p.
556). Accepting Harmatta’s views, the alleged Dravidian, the explanation of a Bum-
saski origin or a derivation from a lost, unknown language is dismissed. At the same
time, there are still details in the Anatolian and especially in the Indian linguistic area
which require further research.
It is rather surprising that Harmatta does not exploit the linguistic evidence pro-
vided by the historic languages of Anatolia. His work completely lacks the rich mate

rial collected by Hoffner (Hoffner 1974). Hoffner tells us that the Hittite kant- ‘wheat’,
mentioned by Harmatta, figured in horse-related texts as kdnza- and denoted ‘horse
fodder’, may have been offered to gods, or may even have been fit for human con-
sumption. It would probably be translated as Einkorn (Hoffner 1974, p. 73). He sug-
gests that the word should be studied in the proper context, as opposed to other
cereals (Hoffner 1974, p. 69). Harmatta also disregards Tischler’s assumptions based
on a critical survey of the Anatolian material. Tischler compares the word to the
Ugaritic hndrt- ‘alter Weizen(?) als Pferdefutter’ (Tischler 1980, p. 485). Here one
must bear Diebold’s important warning in mind. He tells us that it is very difficult to
assign a given IE cereal name to a particular variety below the genus level (Diebold
1992, p. 348). Before turning to the Indian situation we should like to call attention
to a minor item which was not touched upon by Harmatta. Schrader records a Greek
yavSoy1r!v ‘wheat’ which is confirmed in Hesychius’ lexicon (Schrader 1883, p.
189). Although it is missing from the standard modem lexicons of the Greek lan-
guage, the item is verifiable (cf. Latte 1953-1966).
Thanks to Harmatta’s painstaking research into the antecedents of the Sanskrit
godhiima-, it can be traced down in the vast territory between Anatolia and the Indo-
Iranian borderland; however, there are problems that still give food to thought. The
trouble begins with the non-attestation of the words in the Rgveda. Southworth is
apparently content by establishing the fact that in the earliest period only barley and
pulses are mentioned, while later texts, such as the Atharvaveda, record wheat and
rice also (Southworth 1995, p. 263). It is puzzling, because this earliest extant text in
Sanskrit is supposed to be linked with the earliest Indo-Aryan speakers who entered
India. Moreover, the geographical area of the genesis of the Rgveda is considered
a fertile wheat producing region (cf. Farmers, p. 37; Randhawa 1980, p. 24) which
coincides with the area where Pänini, the grammarian, lived. Pan ini records the word
in his Ganapätha IV, 3 and 136 (cf. Chatterji 1969, p. 63 and p. 65). As a matter of
fact, there is abundant archaeological evidence of wheat remains from the Punjab
(Randhawa 1980, p. 104) and the not-so-distant Kashmir (Singh 1990, p. 20) from
the period before the invasion of Indo-Aryan speakers. Lallanji Gopal, drawing on
M. D. N. Sahi’s results, speaks of six geographical zones in the Neolithic cultures of
India, the first of which is the wheat-barley culture associated with North-Western
India, including Kashmir (Gopal 1984, p. 91).
In many instances migrating peoples who enter an environment to which a plant
produced by them in their previous habitation(s) is not indigenous, simply forget its
name. This explanation could work if we suppose that wheat production was insig-
nificant in Punjab at the time of the genesis of the Rgveda - a hypothesis that seems
to contradict the above statements. Wheat grows everywhere in Punjab except in un-
irrigated areas (Farmers, p. 38). As it is well known, hydraulic changes in the Sara-
svati-Drsadvati system may anyway be linked to the weakening of Indus civilisation
(cf. Chakrabarti 1997, p. 140). With another word natural changes, together with the
neglecting of canalisation, may have resulted in the decline of wheat production.
However, barley (yava) will grow in dry, unirrigated areas. Details of this process, or
even climatic changes, can be assumed in Punjab. There is a good analogy from the

228 GYULA WOJTILLA
systematically excavated Inamgaon and Daimabad of Rajasthan (cf. Dhavalikar

1988, p. 30). At these sites the climate was relatively dry during 1700-1500 BC. Af-
ter 1500 BC the territory became relatively wet again, and the inhabitants of Inam-
gaon were able to divert the floods of the Ghod into irrigation channels (Dhavalikar
1988, p. 33). Only much broader archaeological evidence than what is at our disposal
will ultimately help solving this mystery.
It may be simply due to an accident: the poets of the Rgveda did not specify
each and every cereal name. We can compare the situation in the Avesta where gan-
tuma- is attested only in Nirangistän 28, a very late text, a Pahlavi book that contains
citations from the Jungavesta (cf. Hoffmann-Forssman 1996, p. 33).
The absence may be attributed to the varied geographical scenery Indo-Aryan
speakers passed by. There are barley-dominated areas adjacent to India (Tosi 1982-
1983, p. 37 and Meadow 1996, p. 393), which is why yava- ‘barley’ has been thought
to be associated with Indo-Aryan speakers; however, this assumption can also be chal-
lenged. This is an old Indo-Aryan or rather Indo-Iranian word that entered Finno-
Ugrian (Joki 1973, p. 265) and probably even the language that existed before the
splintering of Finno-Ugrian and Samoyedic (cf. Wojtilla 1977, p. 118). The matter is
not as simple as it looks, though. The word also denotes ‘grain’ or ‘seed’ (Joki 1973,
p. 265) and perhaps ‘cereal’ in general (Gimbutas 1985, p. 186). It is also uncertain
whether barley was a favourite cereal in ancient India prior to the arrival of Indo-
Aryan speakers (cf. Vishnu-Mittre 1969-1970, pp. 1-2). One should not forget that
Sahi and Gopal speak of wheat-barley culture in this connection.
A critical examination of cultural information inherent in Sanskrit texts, in
which the word features, may yield some important data, too. To our best knowledge
the first occurrence can be found in the Paippaläda recension of the Atharvaveda
(AV(P) IX, pp. 11-12), and other places of occurrence are scattered over the whole
Sanskrit literature up to the modern times: MaiS I, p. 28 - VäjaS XVIII, p. 12 -
SatBr V, p. 2, pp. 1-6, etc. - SänkhäSS XIV, p. 40, pp. 6-9 - BrÄUp VI, p. 3, p. 13
- P IV, p. 3, p. 106 - ArthSä II, p. 24, p. 14. - AinaK II, p. 9, p. 18. - BrSam XV,
p. 6 - NâmaMâ, p. 548 - VaijaK III, p. 8, p. 53 - Vrks(S), p. 213, p. 321 - RasäK,
p. 332 - SivaK, p. 271 and p. 336.
According to AV(P) IX, pp. 11-12, it is mentioned together with other grains
and pulses as an ingredient of a remedy (cf. Zysk 1985, p. 285). The loci in MaiS,
VäjaS, SatBr, SänkhGS and BrÀUp speak of it as a material used in sacrifice. Pänini,
ArthSä, BrSam and Vrks(S) define the word as a type of grain distinct from barley
and rice. It is a triticum type, perhaps triticum aestivum. AmaK presents its synonym
sumanas- ‘good-minded’, NämaMä lists sumanas, sumana-. All these are clearly
euphemisms and attest to the uncertainty of the authors about the etymology of
godhüma-. Additionally, NämaMä makes a curious remark: it is a mlecchabhojya
‘a food of barbarians’. The polysemy of the word in Sanskrit is also a matter of
confusion: it means an ‘unidentified medical plant’ (pw II, p. 180 and MW p. 365) as
well as ‘orange’ (MediK p. Ill, v. p. 43). The situation in Prakrits and NLA languages
is much more balanced: Pali godhüma- (PTS p. 255) or the NIA developments of the
word definitely denote ‘wheat’ (CDIAL p. 4287). In short, the occurrence of the word
Acta Orient. Hung. 52,1999

depends on various factors: one is the geographical milieu, another is food habits,
and thirdly, possibly, the Brahmanical tradition which dominates Sanskrit texts. The
latter concerns especially the sacrificial or medical employment of wheat.
Mention must be made of the attestation in AV(P). As it has been said the edi-
tors of AV(P) use the word in a context where it means a remedy. It is conceivable
that the AV(P) preserves such lexical elements that are extinct from Rgveda. This
circumstance can be ascribed to the special geographical aspect of the place of genesis
of the work. One of the basic manuscripts of the text comes from Kashmir where
wheat has been cultivated since the third millennium BC (cf. Singh 1990, p. 20). Un-
fortunately this assumption could no longer be maintained after the discovery of an-
other important manuscript of the text in Orissa in the Sixties, and consequently the
traditional North Indian provenance had to be given up (Gonda 1975, p. 276). The
existence of the above mentioned type of lexical elements cannot be ruled out fully
on account of the fact that - as Witzel puts it - “some quite old material” can be found
in AV, as well as interaction between the Aryan groups represented by the AV around
1000 BC and those in East Iran or Bactria (Witzel 1980, p. 94). According to Witzel
the links between Iranian and Indie were still active in Pänini’s days, for he, “in all
probability, was a citizen of the Persian province of Hindus” (Witzel ibid.).
The uncertainty of the etymology and meaning in the lexicons is truly re-
fleeted in the efforts of Indian grammarians and modem scholars. Traditional dériva-
tions either from gudh- (MW, p. 365) or guh- (cf. Patyal 1979, p. 117) do not offer
real explanations. Bloch aptly calls it an “absurd signification” (Bloch 1975, p. 53).
Even the modem translations of the compound differ slightly. On the one hand it is
translated as ‘cowsmoke’ (Gonda 1971, p. 215), in accordance with Wackemagel’s
‘Rauch der Rinder’ (Aig II/l, p. 268) and Debrunner’s ‘Kuh-rauch’ (AiG ПУ2, p.
499), and on the other hand as ‘earth-smoke’ (MW, p. 365 and Patyal 1979, p. 117).
It is difficult to decide whether the first or second meaning can be established as defi-
nite. At the level of hypothesis ‘earth-smoke’ would appear to be better than ‘cow-
smoke’. This hypothesis draws on an idea raised by Diakonoff. Dwelling on the ety-
mology of IE * dhoHna- ‘grain’, ‘bread’ he proposes a connection with the Semitic
Duha- ‘millet, or strictly speaking a variety of millet ‘grey as smoke’ (Diakonoff
1985, p. 127). But why couldn’t it be ‘white-coloured smoke’? If it were, one might
go on to assume ‘wheat-flour’ as being a product of earth (go). Of course, this as-
sumption is purely conjectural.
The picture that emerges from this discussion shows that certain traits, such as
godhüma-, permit us to follow agricultural production in a large area from Anatolia
to Greece and to India, and to identify certain stations of the route of this word both
in space and time. We are fully aware of the gaps and pitfalls; nevertheless - like
Harmatta and Renfrew - we are optimistic and do not hesitate to present our modest
results as at least a working hypothesis waiting for analysis and judgement.

230 GYULA WOJTILLA
References
AiG see Wackemagel
AmaK = Amarakosa. Sardesai, N. G.-Padhye, D. G. (eds) (1969), Poona.

Apte, V. S. (1957-1959), The Practical Sanskrit English Dictionary. Revised and enlarged edition.
Poona.
ArthSä = Arthasästra. Kangle, R. P. (ed.) (1970), Bombay.
A V (P) = Atharvaveda o f the Paippalädas. Raghu Vira (ed.) (1979), New Delhi. (Kashmiri
recension) cf. Paippalaäda Samhitä o f the Atharvaveda. Bhattacharyya, D. (ed.) (1 9 6 4 -
1970), Calcutta. Hitherto two volumes.
Baldi, P. (1988), Review on Renfrew 1987. Current Anthropology 29:3, pp. 445-447.
Berger, H. (1959), D ie Burusaski-Lehnwörter in der Zigeunersprache. IIJ 3, pp. 1 7 4 3 ‫־‬.
Bloch, J. (1975), Sanskrit and Dravidian. In: Pre-Aryan and Pre-Dravidian by Sylvain Lévi, J.
Przyluski and J. Bloch. Tr. by R. C. Bagchi. Calcutta, pp. 35 -5 9 . (The French original was
published in 1929.)
BrÄUp = Brhadäranyakopanisad. In: The Principal Upanisads. Radhakrishnan, S. (ed.) (1953),

London.
BrSarh = Brhatsamhitä. Bhat, M. R. (ed. ) (1981), Delhi-Varanasi-Patna, etc.
Burrow, T - Emenau, M. B. (1961), A Dravidian Etymological Dictionary. London.
CDIAL see Turner
Chakrabarti, D. C. (1997), The Archaeology o f Ancient Indian Cities. Calcutta-Chennai-Mumbai.
Chatterji, S. K. (1969), Indo-Aryan and Hindi. Calcutta.
Coleman, R. (1988), R eview on Renfrew 1987. Current Anthropology 29:3, pp. 449-453.
Compendium = Compendium linguarum Iranicarum. Hrsg, von Rüdiger Schmitt (1989), Wies-
baden.
Das, R. P. (1995), The Hunt for Foreign Words in the Rgveda. IIJ 38, pp. 207-238.
DED see Burrow-Emenau
DEDS = Supplement to DED. London 1968.
Deshpande, M. М.-H ook , P. E. (eds) (1979), Aryan and non-Aryan in India. Ann Arbor.
Dhavalikar, M. K. (1988), The First Fanners of the Deccan. Pune.
Diakonoff, I. M. (1985), On the original home o f speakers o f Indo-European. JIES 13:1-2, pp.
92-174.
Diakonoff, I. M. (1988), R eview on Renfrew 1987. Annual of Armenian Linguistics 9, pp. 79-87.
Diebold, A. R. (1992), The traditional view o f the Indo-European paleoeconomy: contradictory

evidence from anthropology and linguistics. In: Reconstructing Languages and Cultures.
Polomé, E. С.-W inter, W. (eds), Berlin-New York, pp. 317-368.

Dolgopolsky, A. (1990-1993), More about the Indo-European homeland problem. Mediterranean

Language Review 6 -7 , pp. 230-248.
Elfenbein, J. (1987), A periplus o f the Brahui problem. Studia Iranica 16, pp. 215-235.
Elizarenkova, T. (1998), Concerning future studies in the history o f development of the Indo-Aryan
languages. In: Trends in Indian Studies. Proceedings of the ESIS. Vacek, J.-Dvorak, J.
(eds), Prague, pp.71-78.
Erdosy, G. (1995), Language, Material Culture and Ethnicity: Theoretical Perspectives. In: The
Indo-Aryans, pp. 1-31.
EWA see Mayrhofer
Farmers see Randhawa-Kaw
Gamkrelidze, T. V .-Ivanov, V. V. (1984), Indoevropeyskoy jazik i indoevropeyci. Tbilisi.
Gimbutas, M. (1985), Primary and secondary homeland o f the Indo-Europeans. Comments on

Gamkrelidze-Ivanov articles. LIES 13:1-2, pp. 185-202.
Gonda, J. (1971), Old-Indian. Leiden-Köln. (Handbuch der Orientalistik. Zweite Abteilung. Erster
Band. Erster Abschnitt.)
Gonda, J. (1975), Vedic literature. Wiesbaden. (A History o f Indian Literature, Volume 1. Fascicle 1)
Gopal, L. (1984), Beginnings of agriculture in India. In: Technology in India (Ancient and Medie-
val Periods. Proceedings o f the Seminar Held in February 1983. Ananthacharya Indologi-
cal Research Institute. Bombay.
Harmatta, J. (1984), Egy anatöliai eredettf vdndorszö. Antik Tanulmànyok 21:1, pp. 1-6.
Harmatta, J. (1988), Semitica et Anatolica. In: Occident and Orient. A Tribute to the Memory of
Alexander Scheiber. Dân, R. (ed.), Budapest-Leiden, pp. 241-249.
Hoffmann, K.-Forssman, B. (1996), Awestische Laut- und Flexionslehre. Innsbruck.
Hoffner, H. A. (1974), Alimenta Hethaeorum. Food Production in Hittite Asia Minor. New Haven
1974.
Horn, P. (1893), Grundriss der neupersischen Etymologie. Strassburg.
Hübschmann, H. (1890), Geiger’s Etymologie des Baluci. ZDMG 44, pp. 554-561.
Humbach, H. (1989), Chorasmian. In: Compendium... pp. 193-203.
The Indo-Aryans = The Indo-Aryans of Ancient South Asia. Language, material culture and
ethnicity. Erdosy, G. (ed.) (1995), Berlin-New York. (Indian Philology and South Asian
Studies. Wezler, A.-W itzel, M. (eds), Volume 1)
JIES The Journal o f Indo-European Studies, Washington.
Joki, A. J. (1973), Uralier und Indogermanen. Die älteren Berührungen zwischen den uralischen
und indogermanischen Sprachen. Helsinki.
Klimov, G. A .-E del’man, D. I. (1970), Jazik burushaski. Moskva.

232 GYULA WOJTILLA
Kuiper, F. B. J. (1991), Aryans in the Rigveda. Amsterdam-Atlanta.
Latte, K. (1953-1966), Hesychii Alexandrini Lexicon. Vols 1-2. Copenhagen. We are indebted to
Imre Tegyey M .A. Ph. D. University o f Debrecen for obtaining the data from this work.
Lehmann, W. P. (1992), Die gegenwärtige Richtung der indogermanischen Forschung. Budapest.
MaiS = Maiträyanisamhitä. Ed. by L. von Schroeder (1881-1886), Leipzig.
Masica, С. P. (1979), Aryan and non-Aryan elements in north-Indian agriculture. In: Aryan and
Non-Aryan..., pp. 55 -1 5 1 .
Mayrhofer, M. (1990), Etymologisches Wörterbuch des Altindoarischen. Lieferung 7. Heidelberg.
Mayrhofer, M. (1996), Ein indo-arischer Rechtsterminus im Mitanni-Brief? Zeitschrift für verglei-

chende Sprachforschung 109:2, pp. 161-162.
McAlpin, D. W. (1979), Linguistic prehistory: the Dravidian situation. In: Aryan and Non-Aryan...,
pp. 175-189.
Meadow, R. H. (1996), The origins and spread o f agriculture in northwestern South Asia. In: The
Origins..., pp. 3 90-412.
MediK = Medinikosa. Jagannath Sastri Hoshing (ed.) (1968), Varanasi.
MW = Monier-Williams, M. (1960), A Sanskrit-English Dictionary. Oxford.
NämaMä = Nämamäla o f Bhoja. Kulkarni, E. D.-Gokhale, V. D. (eds) (1955), Poona.
Norman 1988 = Norman, R., Review on Renfrew 1987. Lingua 76, pp. 91-114.
The Origins = The Origins and Spread of Agriculture and Pastoralism in Eurasia. Harris, D. R. (ed.)
(1996), London.
P = Astädhyäyi of Pänini. Hrsg, von O. Böhtlingk (1886), Leipzig.
Patyal, H. C. (1979), Etym ology and Sanskrit dictionary on historical principles. Indian Linguistics
40:2, pp. 115-122.
Prasad, V. (1959-1966), Krsikos. Patna.
PTS = The Pali Text Society’s Pali-English Dictionary. Rhys Davids, T. W .-Stede, W. (eds)
(1966), London.
pw = Böhtlingk, О. von (1883-1886), Sanskrit Wörterbuch in kürzerer Fassung. St Petersburg.
Randhawa, M. S. (1980), A History of Agriculture in India. I. New Delhi.
Randhawa, M. S.-K aw , P. N. (1959), Farmers o f India. I. Punjab, Himachal Pradesh, Jamnu and
Kashmir. New Delhi.
RasäK = Rasärnavakalpa. Roy, M.-Subbarayappa, В. V. (eds) (1976), New Delhi.
Renfrew, R. C. (1989), Archaeology and Language. The Puzzle o f Indo-European Origins. Har-
mondsworth. (Under the preparation o f the present paper this edition was at our disposal.
The first edition reviewed by a large group o f specialists was published in London in 1987.)
Renfrew, R. C. (1996), Language families and the spread o f farming. In: The Origins..., pp. 70-92.

RV = Rgveda. Van Nooten, B. A.-Holland, G. B. (eds) (1994), Cambridge, Mass.
Sânkhâés = Éânkhâyana Érauta Sutra. Hillebrandt, A. (ed.) (1888-1897), Calcutta.
SatBr = Éatapathabrâhmana. Weber, A. (ed.) (1855), Berlin-London.
Schrader, O. (1883), Sprachvergleichung und Urgeschichte. Jena.
Sheth, H. (1963), Päia-sadda-mahannavo. Varanasi.
Singh, R. P. (1990), Agriculture in Prehistoric India. Delhi.
éivaK = Sivakosa of Sivadatta Misra. Harshe, R. G. (ed.) (1952), Poona.
Skjærv0, O. (1989), Pashto. In: Compendium..., pp. 384-410.
Southworth, F. C. (1979), Lexical evidence for early contacts between Indo-Aryan and Dravidian.
In: Aryan and N on-Aryan..., pp. 191-233.
Southworth, F. C. (1988), Ancient economic plants o f South Asia: linguistic archaeology and early
agriculture. In: Languages and Cultures. Studies in Honour of Edgar C. Polomé. Jazayery,
Ali-Winter, W. (eds), Berlin-New York, pp. 649-668.
Southworth, F. C. (1995), Reconstructing social context from language: Indo-Aryan and Dravidian
prehistory. In: The Indo-Aryans..., pp. 258-277.
Srivastava, D. (1962), Nepali Language. Its History and Development. Calcutta.
Thieme, P. (1994), On M. Mayrhofer’s Etymologisches Wörterbuch des Altindoarischen. BSOAS

57, pp. 321-328.
Tischler, J. (1980), Hethitisches etymologisches Glossar. Lieferung 3. Innsbruck.
Tosi, M. (1982-1983), Baluchistan in prehistory: reversing the center-periphery paradigm for a

future generation of studies. Newsletter of Baluchistan Studies 1. Naples, pp. 32-46.
Turner, R. L. (1966), A Comparative Dictionary o f the Indo-Aryan Languages. London.
VaijaK = Vaijayantikosa. Haragovinda Sästri (ed.) (1971), Varanasi.
VâjaS = Vâjasaneyïsamhitâ. Jagadishlal Shastri (ed.) (1971), Delhi-Patna-Varanasi.
Vishnu-Mittre (1969-1970), Studies o f barley in Indian archaeology with remarks on the Aryan
hypothesis. Purätattva 1969-1970:3, pp. 1-2.
Vrksa(S) = Sürapälas Vrkssäyurveda. Kritisch ed. iibers. von R. P. Das (1988), Stuttgart.
Wackemagel, J. (1954), Altindische Grammatik. II/l. Einleitung zur Wortlehre. Nominalkomposi-

tion. Göttingen 1905 and II/2 Die Nominal-Suffixe von A. Debrunner. Göttingen.
Witzei, M. (1980), Eastern Iran and the Atharvaveda. Persica 9, pp. 86-127.
Witzei, M. (1995), Early Indian history: linguistic and textual paramétrés. In: The Indo-Aryans...,
pp. 85-125.
Wojtilla, Gy. (1977), Terminological studies o f selected plant names o f the Krsiparasara. Sri Ven-
kateswara Oriental Journal 20:1-2, pp. 111-119.

234 GYULA WOJTILLA
Wojtilla, Gy. (1986), Notes on Indo-Aryan terms for ‘ploughing’ and the ‘plough’. JIES 14:1-2,
pp. 27-37.
Zvelebil, M .-Zvelebil, К. V. (1990), Agricultural transition, “Indo-European origins” and the

spread of farming. In: When worlds collide: the Bellagio papers. Markey, T. L - Greppin,
J. A. C. (eds), Ann Arbor.
Zysk, K. G. (1985), Religious healing in the Veda. Philadelphia. (Transactions o f the American
Philosophical Society. Volume 75, Part 7.)
Gyula Wojtilla
Department o f Ancient History
University o f Szeged
H-6722 Szeged
Egyetem u. 2
Hungary
e-mail: wojtilla@antiq.u-szeged.hu

Acta Orientalia Academiae Scientiarum Hung. Volume 52 (3-4), 235-242 (1999)
MUHAMMAD AND THE JIHÄD

R ö b e r t S im o n
(Budapest)
In mass media reports as well as in general surveys o f Islam there are a lot o f prejudices about
Jihad which were not treated and analysed properly in the scholarly literature so far. According to
the common opinion the Prophet would have regulated the Jihad as Holy War once and for all in
an unequivocal way. The unbiased analysis does not support this opinion. The Prophet’s stand-
point underwent several phases on the one hand and showed up variegated elements in the same
time, on the other, and, moreover, He was always inclined to reasonable bilateral agreement.
Key-words: Jihad, Holy War, tolerance and intolerance in Islam, Jihad and conversion, Jihad and
Arabs, Jihad and non-Arabs.
Everybody who is giving university courses on Islamic topics may share an uneasy
feeling that the biased knowledge of most of the students about Islam consists of
some basic prejudices which are motivated by the commonplaces of the mass media.1
One of these prejudices refers to the Jihad, which, according to the widespread West-
ern belief would have motivated from the very beginnings of Islam the relationship
of the Muslims towards non-Muslims and it is the Prophet who would have ordained
it in no uncertain terms once and for all. This belief is supported by a number of mili-
tant fundamentalist groups all over the Arab world which called themselves “Jihad”
from the late seventies.‫ ־‬The members of these militant groups interpret Jihad une-
quivocally as “Holy War” and consider it as one of the main pillars of Islam which
1To the mechanism o f the mass media in this respect, see E. Said (1981), Covering Islam.
How the Media and the Experts Determine How to See the Rest of the World. New York, Pantheon
Books; Akbar S. Ahmed (1992), Postmodernism and Islam. Predicament and Promise. London-
New York, Routledge, especially pp. 1-50, 2 22-265. The ignorance amounts sometimes to total
indifference or contempt o f the “otherness”. Guy Monnot reports a story o f a seminar organised by
Aramco in the United States where the participants have been asked about Islam. Somebody an-
swered: “it is a gamble similar to the bridge”, see Islam et religions (1986) Paris, Maisonneuve et
Larose, p. 9.
' See the article “Jihad Organizations” in John L. Esposito (ed.) (1995), The Oxford Ency-
clopaedia of the Modern Islamic World. New York-Oxford, OUP, II, pp. 373-376 (Ascad Abu
Khalil): most o f them are to be found in Palestine and Egypt.
0 0 0 1 1 9 9 9 © 5.00 $ /99/ 6446‫ ־‬Akadémiai Kiadö, Budapest
236 RÖBERT SIMON
has been badly neglected as a foremost duty by the Muslims in the last decades.
Muhammad cAbd as-Saläm Faraj who founded the Egyptian Jihad Group in 1979
wrote his famous treatise, al-Farida al-ghä’iba (“The Missing Duty”) on Jihad as a
sort of Manifesto. He renewed, through Saiyid Qutb,3 4 the leader of al-Ikhwän al-
Muslimùn, the Neohanbalite position on “Holy War”, formulated by Ibn Taiymlya.5
This Neohanbalite standpoint has been shared, before the Jihad group, by most of the
pre-modem and modern revivalist movements: by the wahhäbiya, by cAbd al-Qädir
against the French army and by the Sudanese mahdiya which instead of the pilgrim-
age, regarded the Jihäd as one of the pillars of Islam. All these experiences of the last
two and a half centuries might have abundantly motivated the Western common opin-
ion about Jihäd and its overall importance within Islam. The fact that the right inter-
pretation of Jihäd was a problem even for the early Muslims themselves, is convinc-
ingly shown by the experiences of the first civil war (fitna) which resulted from the
murder of cUthman, the third “right-guided” caliph (35/656) and the legitimacy of his
successor, cAli b. АЫ Tälib. As a consequence of the antagonism not to be solved by
peaceful negotiations, the khawärij seceded from cAll’s camp and probably they were
the first ones to state that the main enemies were not to be found outside but within
the Muslim community. It was mainly Naff b. al-Azraq, the founder of the extremist
al-azäriqa who declared infidel those who remained at home (qa cadat al-muslimin)
instead of waging the war with them against their enemies and regarded as licit kill-
ing the children of these infidels as well as those who fought against them.6That this
opinion was not a commonly shared standpoint even within the khawärij is testified
by the letter of Najda al-Hanaft in which the founder of Najdiya, the moderate branch
of Khawärij passionately criticised the Jihäd conception of Näficb. al-Azraq.7 This
early debate shows in illuminating way that even in early Islam the opinions differed
from each other about how the Prophet regarded Jihäd.
Although a number of studies were devoted to the Jihäd,8 several problems
remained unsolved or contraversial. This uncertainty is amply testified by the rather
3 The text is now available in two, by no means critical, editions: (1) R ifat Saiyid Ahmad
(ed.) (1991), an-Nabi al-mussallah I. ar-Räfidün. London, Riad al Rayyes Books, pp. 127-149; (2)
al-liwâ’ Hasan Sädiq (19932), Judhür al-fitna f i ’l-firaq al-islämiya mundhu cahd ar-rasul hattä igh-
tiyäl as-Sädät. al-Qähira, Matbaca Madbüli, pp. 395-412. J. J. G. Jansen made of it a commented,
though not always reliable, translation: The Neglected Duty. The Creed of Sadat’s Assassins and
Islamic Resurgence in the Middle East (1986). N ew York-London, Macmillan.
4 See mainly his Milestones / МаЪИт f i ’t-tariq / (1398/1978) Beirut-Damascus, The Holy
Koran Publishing House, pp. 73-140. To Saiyid Qutb see Esposito (ed.), op. cit. s.v. Qutb, Saiyid
(111:400-4 - Shahrough Akhavi, with abundant further bibliography).
5 See e.g. in his important work (as-Siyäsa ash-sharTya f l isläh ar-räT wa’r-raTya. (1409/
1988) Bayrüt, Dâr al-kutub al-cilmïya, pp. 105-124). See on him: El2, III. pp. 976-979 (H. Laoust);
Esposito (ed.), op. cit. II, pp. 165 ff (Ronald L. Nettler) and E. Sivan (1985), Radical Islam. Medie-
val Theology and Modern Politics. New Haven-London, Yale University Press, pp. 96-104.
6 See his standpoint in his own words: in: N. Mahmüd Ma'rüf (ed.) (1403/1983), Diwân al-
Khawärij. al-Qähira, Dar al-Masira, pp. 278 ff.
7 Op. cit. pp. 281 ff.
8 See EI, s.v. Djihäd (1:1041 ff. - D. B. Macdonald); EI2, s.v. Djihäd (11:551-3 - E. Tyan);
M. Eliade (ed.) (1995), The Encyclopedia of Religion. New York, Macmillan, s.v. Jihäd (V IIE88-
91 - R. Peters, with well-informed further bibliography). Some major treatments: A. Noth (1964),

MUHAMMAD AND THE JIHÀD 237
diverse interpretations represented by the legal schools and especially by the sects as
well as by the different contemporary “counter-movements”.
According to the lexicographers*910the basic meaning of the third verbal form of
jhd (with Lane’s wording) is: “using, or exerting, one’s utmost power, efforts, en-
deavours, or ability, in contending with an object of disapprobation, and this is of
three kinds, namely, a visible enemy, the devil, and one’s self’, thus its semantic field
would range from actual fighting (mujähada = qitäl) to a morally tainted spiritual or
inner struggle (this latter meaning was used by the süfis).
The main problem with the Prophet’s attitude to Jihäd is that during the Medi-
nan period it underwent several changes and, which aggravates the situation, the reve-
lations referring to it can be dated in a few cases only. Beside the historical-philo-
logical difficulties, the interpretation of Jihäd is connected with the major religious
issue of how Islam had been regarded by the Prophet in the course of his activity.
Was it a universal religion (this would be his likely standpoint in the Meccan period),
or was it a special revelation sent only to the Arabs (one may guess that after the
elaboration of millat Ibrâhîm, in 623^4-, the Prophet seems to have had a preference
for this idea). For the very nature of Islam, as it was thought of by the Prophet, should
have determined or influenced the idea of active mission. Last but not least, there is
the problem, whether “the fighting or fervency on the road of Allah” during its for-
mative period was an individual or a communal duty (fard cayn or fard kifäya)? The
attitude of the Prophet to his adversaries or enemies during the 20 years of his Pro-
phetic activity (if we accept that he began his preaching in 612)'° underwent at least
four major phases. Being in Mecca, he responded to the outrages and injuries of his
adversaries with forbearance, endurance and with his firm belief in the divine reve-
Heiliger Krieg und heiliger Kampf im Islam und Christentum. Beiträge zur Vorgeschichte und Ge-
schichte der Kreuzzüge. Bonn, (Bonner historische Forschungen 28); R. Paret (1970), Toleranz und
Intoleranz im Islam. Saeculum 21, pp. 344-361; B. Scarcia-Amoretti (1974), Tolleranza e guerra
santa nell Islam. Firenze, Sansoni; R. Peters (1977), Jihäd in Medieval and Modern Islam. Leiden,
E. J. Brill; Esposito (ed.), op. cit. s.v. Jihäd (11:369-373 - R. Peters). The material o f the Prophetie
tradition is best to be consulted in: Sahih al-Bukhäri (1378/1959-60) al-Qähira, Matäbi‘ as-shacb,
IV: 17-54.
9 See Ibn Manzür, Lisän al-cärab. (n.d.) al-Qähira, Dar al-Ma'ärif, 1:710 (qâtala; al-mu-
bälagha wa' stifrâgh al-wuscf i ’l-harb a w ’l-lisân aw-mâ atâqa min sha’iy)‫׳‬, ar-Râghib al-Isfahânï,
Mu jam mufradat alfâz al-Qur’ân (n.d.) Karachi, Mir Muh. Kutubkhâna, 99 (= istifrägh al-wuscf î
mudäfai al-caduww, w a’l-jihäd thalâthatu adrub: mujähadat al-caduww az-zähir, wa-mujâhada
ash-shaytàn wa-mujâhada an-nafs) and refers to the following Qur’an verses: 8:72; 9:30,80; E. L.
Lane (1863-1893), An Arab-English Lexicon. London, Williams and Norgate, reprint: (1980),
Beirut, Librairie du Liban, Pt. 2: p. 473 (referring to the before mentioned interpretations, he ex-
plains its basic meaning as “using, or exerting, one’s utmost power, efforts, endeavours or ability,
in contending with an object o f disapprobation”); R. Dozy (1991), Supplément aux Dictionnaires
arabes. Beyrouth, Librairie du Liban, originally published: (1881) Leyde, E. J. Brill, I: p. 227
(among the later interpretations he mentioned the mujähada an-nafs which occurs e.g. in Ibn Khal-
dim’s treatise on süfism: Sifä’-Us-Sä’il Li-tahdzäb-il-Masä’il, éd. par le Père Ignace-Abdo-Khalifé
(1959), Beyrouth, Impr. Catholique, pp. 35 -3 9 , pp. 57-59.
10 It is more than probable that after the beginnings of the nubuwwa there was a two-year
interruption (the so-called fatra) and only about 613 he appeared before the public, see W. M. Watt
(1953), Muhammad at Mecca. Oxford, OUP, pp. 47 -5 2 , pp. 58 ff.
238 RÖBERT SIMON
lation": “The servants of the All-merciful are those who walk in the earth modestly
and who, when the ignorant address them, say, ‘Peace’.”112 In this period the Prophet’s
duty is not to fight but to deliver the message: “And say to those who have been given
the Scripture and to the Gentiles (al-ummiyin): ‘Have you converted to Islam (a-
aslamtumyi If they have converted (fa-in-aslamü), they are rightly guided, but if they
turn their back, the your duty is only to deliver the message.”'3 Or: “So shout that
thou art commanded and turn away from the idolaters (mushrikin). We suffice thee
against those who set up with Allah another god. Certainly they will soon know. We
know indeed thy breast is straitened by the things they say. Proclaim thy Lord’s
praise, and be of those that bow.”14 In this period it is Allah’s duty to punish those
who are going obstinately astray and scoff at the Prophet. This attitude can be grasped
in the so-called stories of punishment (which are probably to be identified with the
seven mathäni5). These stories may be interpreted as a kind of theodicy which,
though cherishing the hopes of the Prophet in the future, are exemplifying the same
pattern (viz. rejection of a messenger and punishment of disbelieving people) in the
past. The Jihad, if it could this time be mentioned at all, is not the duty of the com-
munity or the individual but it is a part of the divine “salvation project” which is to
be realised in the long run.
After the Hijra the position of the Muslim community whose number ranged
at that time from 45 to 15016 underwent far-reaching changes. Among the problems
resulting from these changes, I refer only to two main issues. The Prophet had to solve
in the short run and first within the changing tribal structure the financial difficulties
of the increasing Muslim community. Among the different attempts17 the military ex
11 Qur’an 34:4 ff; 35:25 ff; 2:109 ff; 17:107; 25:63; 23:144; 15:94-98; 3:20; 88:21-24.
12 25:63 (if not indicated otherwise, I used, sometimes modified, A. J. Arberry’s rendering:
The Koran Interpreted (1982), Oxford, OUP.
12 3:20. The translation is mine, Arberry doesn’t give the real meaning o f aslamü which in
this case is probably not “to surrender” but “to embrace Islam”, “to convert” (see the previous
verse: inna’d-dina cinda’llähi al-isläm ) and al-ummiyün is not “common folk” but those who did
not get a Scripture so far, e.g. “the Gentiles”. Among the major translations it was R. Blachère who
interpreted correctly the text (Vous-ètes-vous convertis à lTslam?”). R. Bell interpreted the ummiyûn
as “those belonging to the community” which is evidently a derivation from umma ( The Qur’an.
Translation with a critical re-arrangement of the Surahs. (1960), Edinburgh, T.-T. Clark, 1:46, no. 4.
14 15:94-98.
15 These stories exemplify the admonishing punishment o f those peoples who disbelieved
their prophets and ridiculed them. These peoples have generally been destroyed for their disbelief
by Allah (Noah and his contemporaries; Hud and cAd; Sàlih and Thamüd; Lot and his people being
immersed in indecency and sodomy; Shucayb and Madyan; M oses and Pharao; Abraham and his
people including his idol-worshipping father). To these stories see B ell’s Introduction to the
Qur’än, completely revised and enlarged by W. Montgomery Watt (1970), Edinburgh at the Uni-
versity Press, pp. 127-135; Simon R. (1987), A Koran vildga (The World o f the Qur’an). Budapest,
Helikon, pp. 483-485.
16 See the data and the sources in Watt (1953), pp. 170-179.
17 One of these endeavours was the short experiment o f mu ’äkhah which was very important
from both historical as well as religious points o f view. A few months after the Hijra the Prophet
made a fraternal bond (äkhä ) between the muhâjirûn and ansär in order to foster among them equity
and mutual assistance, see on it my paper in: Meccan Trade and Islam. Problems of Origin and
Structure (1989), Budapest, Akadémiai Kiadd, pp. 115-119, pp. 178-182.
peditions (ghazawät) against Meccan caravans played a major role and the Prophet
himself participated in most of them.1* These expeditions have put on the agenda
without delay the question of the attitude of Islam being in the process of formation
to the basic norms of the tribal society, since the Meccan Muslims (muhäjirün) started
to fight against their former tribal and clan-fellows. Due to these far-reaching changes
the Prophet had to take up an unequivocal position on some religiously motivated
questions which affected the previous coexistence (e.g. on the question of the so-
called “Holy Months”1819) and he had to take a clear stand on the issue of the relation-
ship of the Muslims to the Non-Muslims. This new relationship was intimately con-
nected with another momentous problem leading ultimately to the shaping of the so-
called “Abraham-religion” (d‫־‬m/millat Ibrahima) which meant the clear and definitive
distinction of the peculiarities of Islam marking it off from the other great “book-
religions”, from Judaism and Christianity. In the Meccan period and a short time
after Hijra Islam indeed was seeking after a proper place within the community of
the “book-religions” (ahl alkitâb). Now the state of affairs changed altogether. The
former peaceful attitude underwent a radical transformation assuming different forms
of confrontation and fighting. This time it meant the permission and even the pre-
scription of fighting aiming first at repelling the injustice and “the seduction from the
truth”20: “Leave is given to those who fight because they were wronged {zulimü) —
surely Allah is able to help them - who were expelled from their habitations without
right, except that they say “Our Lord is Allah”.21 The fighting against kinsmen would
have been a startling experience, whose overcoming was formulated in more Qur’an-
verses: “Prescribed for you is fighting, though it be hateful to you. Yet it may happen
that you will hate a thing which is better for you; and it may happen that you will
love a thing which is worse for you; Allah knew, and you know not.”22 Up to the end
of 623 the Muslims respected the custom inherited from the Jähiliya, that during the
Holy Months they did not attack their Qurayshite kinsmen. This religious unwritten
law had been violated by the expedition of Nakhla in January 624/rajab 2. The overt
18 See Ibn Hi sham (1375/1955), as-Sira an-nabawiya. al-Qähira, Must. al-Bâbî al-Halabi,
11:608; Ibn Scad (1904-1940), Kitäb at-tabaqät al-kabïr. Leiden, E. J. Brill, 11/1:1; at-Tabari
(1879-1901), Ta’rîkh ar-rusul w a’l-mulük. Leiden, E. J. Brill, 1:1756 ff. See the list o f expeditions
and their dates: W. Montgomery Watt (1956), Muhanvnad at Medina. Oxford, Oxford University
Press, pp. 339-344.
9 To the “Holy Months” see R. Simon (1989), p. 74, p. 161, nos 10 ff.
20See the verses in: Muh Fu’äd cAbd al-Bäqi (n.d.), al-Mujam al-mufahras li-alfäz al-
Qur’an al-kartm. Beirüt, Dar ihyâ’ at-turâth al-carabï, s.v ./m (p. 511 ff). On the different meanings
o f fitna in Islam see El2, s.v.fitna (11:952 ff - L. Gardet). In the Qur’ân its basic meaning is “temp-
tation or trial o f faith” (see J. Penrice (1970), A Dictionary and Glossary of the Koran. New York,
Praeger Publ., pp. 107 ff).
21 22:39f (instead o f “God” we preferred “Allah”, see our paper: “Allah or God? The Se-
mantic and Religious Meaning of ’Allah on the Eve o f Islam and in the Qur’an”: Acta Classica
Univ. Scient. Debrecen. (1991), XXVII, pp. 129-134). Similar verses: 2:216, 224 ff; 47:4.
” 2:216, comp. 2:246 and especially 4:77: “Hast thou not regarded those to whom it was
said, ‘Restrain your hand, and perform the prayer, and pay the alms’? Then, as soon as fighting is
prescribed for them, there is a party o f them fearing the people as they would fear Allah, or with
a greater fear, and they say, ‘Our Lord, why hast thou prescribed fighting for us? Why not defer us
to a near term?” See further 9, pp. 13-15, 18.
240 RÖBERT SIMON
break with the former norms became unavoidable and the preference of Islam against
everything else had to be clearly formulated: “They will question thee concerning the
holy month, and fighting in it. Say: ’Fighting in it is a heinous thing, but to bar from
Allah’s way, and disbelief in Him, and the Holy Mosque, and to expel its people from
it - that is more heinous in Allah’s sight; and persecution (al-fitna) is more heinous
than slaying.’ They will not cease to fight with you till they turn you from your relig-
ion, if they are able.”23
Most of the revelations relating to the fighting against unbelievers have taken
their origins from concrete events (after the battle of Uhud or during the expedition
of Tabük) but in most of the cases we cannot date the respective verses, which means
that it is almost impossible to say with certainty what was really the Prophet’s stand-
point in his last years. For those who regard the well-known “proclamation” revealed
in February-March 631 (dhü 1-hijja 9.) (9:1-37) and within it the verse 9:524 as the
last and decisive revelation in this respect, they take no notice of the fact that the
whole bard ’a (renunciation” or “acquittal” or “declaration of immunity”) and the so-
called “sword-verse” refer to Arab pagans, who further on (they had for their deci-
sion one and a half month) had to choose between conversion or fighting because in
Mecca and shortly later in the whole Arabian Peninsula there was no infidel Arab
left. Thus, this verse may hardly be accepted as an âya nâsikha which would abro-
gate all the previous verses relating to Jihad depriving them of their validity. The
extension of Jihâd is closely connected with the problem whether the Prophet wanted
to propagate Islam or aimed only at the acknowledgement of his community. This
problem can be formulated more clearly so, that whether the Prophet before his death
did conceive Islam as a universal religion which would not tolerate any other religion
or regarded it a special faith revealed to the Arabs. On the basis of our meagre infor-
mation we are not able to give an unequivocal answer to this question. In the Meccan
period the Prophet’s conception of Islam can be guessed more or less clearly. In the
Medinan period, however, Islam had a complex development and, though it retained
some basic features of the beginnings, “the religion of Abraham” formulated some
ideas which were not devoid of some inherent contradictions. What is it all about? In
the Meccan period the Prophet regarded his preaching as a part of the universal “sal-
vation-plan” which would have continued as well as ended the series of the so-called
“Book-religions”25comprising Judaism, Christianity and in all probability Manicheism.
2:217. See my commentary on it (R. Simon, 1987, pp. 63 ff). According to another dat-
ing the expedition would have happened on 29 Dec. 623. See further 9:32: ‘T h e number of the
months, with Allah, is twelve in the Book of A llah... four o f them are sacred. That is the right
religion. So wrong not each other during them and fight the unbelievers totally even as they fight
you totally.”
24 “Then, when the sacred months are drawn away, slay the idolators wherever you find
them, and take them, and confine them, and lie in wait for them at every place o f ambush...” See to
this verse and its context my commentary (R. Simon, 1987, pp. 198-205).
35 To the “Book-religions” see: S. Lorenz (1950), Entstehung und Wesen der Buchreligio-
nen. ThlZ 75, pp. 709 ff; F. Altheim (1950), Literatur und Gesellschaft im ausgehenden Altertum
II. Halle a.S., Niemayer, pp. 225-243; idem (1957), Der unbesiegte Gott. Hamburg, Rowohlt, pp.
4 7-66; F. Altheim-R. Stiehl (1954), Ein asiatischer Staat. Feudalismus unter den Sasaniden und
From the common origins resulted the idea of the seriality of the Prophets and the
importance of the Scripture (the same revelation has been sent down from the same
God to the Prophets in different languages). For this reason called the Prophet ahl al-
kitäb*26 those peoples who received by their Prophets revelations from the same uni-
versai religion or a part of it which, however, is identical with the whole. This idea
comprised in an implicit way the thought of the mission. In a particular, though un-
derstandable, way, just in this phase, when the universal character of Islam was obvi-
ous and undeniable, the prescription of the compulsory mission never came up. All
the Qur’an verses quoted before are stressing peace and forbearance. The Prophet
hopes that the punishment of unbelievers and the victory of Islam will come from
Allah. We mentioned before that the Prophet, after having broken with the Jews and
Christians, started at the end of 623 to elaborate the din/millat Ibrâhîm. This far-
reaching repristination act in comparison with the universal character of the original
Islam stressed, beyond any doubt, the Arab peculiarities of the remote beginning
which manifested itself in almost all the prescriptions of the shaping cult (e.g. in the
abrogation of the qibla towards Jerusalem in favour of Mecca27; introduction of the
fasting in ramadan 8 and later in taking over the Meccan pilgrimage and its élabora-
tion29). These prescriptions stressed the central role of Mecca. This new attitude can
be grasped e.g. in the emphasis laid upon Arabic as the language of revelation30 which
had greatly influenced the preponderant role of Arabic in Classical Islamic culture
and was instrumental in the dogma of intranslatability of the Qur’än.31
These new features emphasising the Arab peculiarities of Islam may probably
be regarded as tendencies which modified its original background to a certain extent,
bringing it from the universal towards a sort of ethnicity without, however, changing
its basic universal character. Nevertheless, these features are not to be dismissed
lightly and this warning is supported by the first century of Islam, for, in the Omay-
yade period Islam itself as well as the new world-empire conquered in the name of
Islam, have been regarded by the Arab-Muslim conquerors as a pure Arab phenome-
non in which the non-Arabs even if converted to Islam, were treated as second-class
subjects. It is a well-known fact that the universal character of Islam with its organic
culture, law and institutions has been firmly established only in the first century of
cAbbaside rule due to the development of the Islamic orthodoxy.
As to the Jihad, the elaboration of “the Abraham-religion” connected with the
Arabs might have meant that the mission organised peacefully or by sword should
have been directed primarily or exclusively against Arab unbelievers (this standpoint
ihren Nachbarn. Wiesbaden, О. Harrassowitz, pp. 175-189; J. Pedersen (1984), The Arabic Book.
Princeton, Princeton University Press, pp. 12-15; Bell-W att (1970), pp. 141-144.
26 See J. Horovitz (1926), Koranische Untersuchungen. Berlin-Leipzig, W. de Gruyter, pp.
66-69; EI2, s.v. ahl al-kitäb (1:272-4 - G. Vajda).
27 2:144.
2* 2:185.
29 2:196-203; 22:27-32, see further 3:97.
‫ג‬0 12:2 ; 13:17 ; 16:103 ; 20:113 ; 39:28 ; 41:3 ; 42:7 ; 43:3 ; 4 6 ;12.
31 See the theologian implications o f the inimitability o f the Qur’an: EI2, s.v. Udjäz (III:
1044-6 - G. E. v. Grunebaum); R. Simon (1987), 530, no. 151.
242 RÖBERT SIMON
has been worded in the afore-mentioned important verses of the Sura Bam ’a). As far
as the Jews and Christians are concerned there is no question of their conversion con-
nected with Jihad. This lesson can be drawn from the analysis of the hotly debated v.
29 of Sura Bara'a.'1The unequivocal meaning of this verse is that they are not to be
converted which is consonant with the important though hardly interpretable utterance
of 2:256: “No compulsion is there in religion (lä ikräh“fi'd-dln).” On the basis of
such utterances and according to the almost unanimous opinion of the research we,
too, contest the authenticity of those six letters’3, which the Prophet would have sent
a. H.7/May 628 to the king of Abyssinia, to the governor of Busra, to al-Muqawqis,
the governor of Alexandria, to the Ghassänide al-Härith b. АЫ Shamir and to Hawdha
b. CA11, the chief of Banü Hanifa calling them upon to convert to Islam. This alleged
call was not made possible by the actual historical circumstances but the main argu-
ment against the supposition of this step is that the Prophet most likely did not think
of the conversion of the “People of the Scripture”. If the Prophet’s standpoint was
like this, it circumscribed the content of Jihad. According to the arguments referred
to above I am inclined to think that there is no way to characterise the Prophet’s
standpoint about Jihäd in a consistent and unequivocal manner. If things are thus, we
may strongly suspect that both the common opinion emerging soon after the success-
ful conquests as well as the point of view of the contemporary revivalist movements
would be identical with the Prophet’s standpoint which underwent different phases
on the one hand and comprised variegated elements in the same period on the other
and, moreover, He was always inclined to a reasonable bilateral agreement.
Röbert Simon
H-1065 Budapest
Bajcsy-Zsilinszky u. 31
Hungary
32 “Fight those who believe not in Allah and the Last Day and do not forbid what Allah and
His Messenger have forbidden - such men as practise not the religion o f truth, being o f those who
have been given the Book - until they pay the tribute (jizya) out o f hand and have been humbled
(wa-hum säghirüna).” See on it our detailed commentary in R. Simon (1987), pp. 201-203.
33 On these letters see our commentary, op. cit. pp. 88 ff.
ORIGINS OF ISLAM: POLITICAL-ANTHROPOLOGICAL

AND ENVIRONMENTAL CONTEXT
A n d r e y K o r o t a e v , V l a d im ir K l im e n k o , D m it r y P r o u s s a k o v
(Moscow)
The authors suggest to view the origins o f Islam against the background of the 6th century AD
Arabian socio-ecological crisis whose model is specified in the paper through the study o f clima-
tological, seismological, volcanological and epidemiological history o f the period. Most socio-po-
litical systems of the Arabs reacted to the socio-ecological crisis by getting rid of the rigid supra-
tribal political structures (kingdoms and chiefdoms) which started posing a real threat to their very
survival. The decades o f fighting which led to the destruction o f most o f the Arabian kingdoms and
chiefdoms (reflected in Ayyäm al- ‘Arab tradition) led to the elaboration o f some definite “anti-
royal” freedom-loving tribal ethos. At the beginning o f the 7th century tribes which would recog-
nise themselves as subjects o f some terrestrial super-tribal political authority, the “king”, risked to
lose its honour. However, this seems not to be applicable to the authority o f another type, the “ce-
lestial” one. At the meantime the early 7th century evidences the merging of the Arabian tradition
o f prophecy and the Arabian Monotheist “Rahmanist” tradition which produced “the Arabian pro-
phetic movement”. The Monotheist “Rahmanist” prophets appear to have represented a supratribal
authority just of the type many Arab tribes were looking for at this very time, which seems to ex-
plain to a certain extent those prophets’ political success (including the extreme political success o f
Muhammad).
Key-words: Islam, Arabia, ecology, political anthropology, history, climate, evolution
South Arabian puzzle
For many years we were a bit puzzled by a strangely quick collapse of the South Ara-
bian Empire of the “Kings of Saba’ and dhù-Raydân and Hadramawt and Yamanat
and Their Arabs in the Highland and the Coastal Plain” ( ’mlk S‫׳‬B' w-d-RYD" w-
HDRMWT w-YMNT w -”rb-hmw TWDmw-THMT) in the second half of the 6th cen-
tury AD.
Of course, at the beginning of this century South Arabia experienced a series
of rather turbulent events: dhü-Nuwâs’ coup, violent persecutions of the Christians,
Ethiopian invasions and conquest, rebellion (successful) of the Ethiopian soldiers
deployed in Yemen, their leader (Abraha) getting the royal power etc. - see Sabaic
0001-6446 /99/ $ 5.00 © 1999 Akadémiai Kiadö, Budapest

244 ANDREY KOROTAEV-VLADIMIR KLIMENKO-DMITRY PROUSSAKOV
inscriptions C621; Ry507; 508; 510; Ja 1028; as well as: Pirenne and Tesfaye
(1982); Carpenter (1869); Mpberg (1924); Berzina and Kubbel’ (1990, pp. 203-
249); Shahid (1971); Lundin (1961); Kobishchanov (1980, pp. 10-88); Piotrovskij
(1985, pp. 17-23); Smith (1954); Robin et al. (1996) etc. Then, however, under
Abraha’s rule the Empire seemed to have stabilised and achieved reasonable
florescence by the end of the 540s: Abraha managed to organise the successful
repairs of the famous Mârib Dam ( ‘RM" [C 541]), campaigns to Central and Northern
Arabia etc. (Ry 506; Vasil’ev 1907; Kobishchanov 1980, pp. 64—89; Piotrovskij
1985, pp. 23-24)
And then in the second half of the century the Empire (together with the 1500-
year-old South Arabian civilisation) simply collapses without any apparent serious
reason. The study of this collapse is further complicated by the fact that the catastro-
phe appears to have been so profound that the written texts seem to have stopped to
be produced in South Arabia - since the 7th decade of the 6th century (this decade
including) we have no authentic dated South Arabian texts up to the Islamic Age -
which stands in a sharp contrast with the comparatively well documented first
5 decades of the century.1
The collapse seems to have been so profound that when in 570 AD (Shahid
1995, p. 365) Khusraw [I] Parwëz reluctantly sent (as a sort of punishment) a few
hundred convicted criminals to put Yemen into the Persian sphere of influence (con-
sidering this such an adventure that it would be wiser not to risk with the proper
troops), they (the convicted criminals) did manage to overthrow the dynasty of
Abraha, though, of course, not without the help of the Yemenites opposed to the dy-
nasty - see e.g. al-Tabari (1964, pp. 950-956).
North Arabian puzzle
Of course, it is evident that what happened in the 6th-century Yemen was not an iso-
lated event. Already if we look at Arabia as a whole, we shall get a bit different
perspective.
To begin with, in the Soviet Islamology up to the 1980s the dominant theory
of the origins of Islam connected it with the crisis and degeneration of the clan-tribal
system in the 6th-early 7th century Arabia, the process of the state and class forma-
tion (Tolstov 1932; Smirnov 1954, pp. 180 ff; Beljaev 1965; Petrushevskij 1966, pp.
5-11; Mavljutov 1974; Zhukov 1974, p. 29; Fil’shtinskij 1977, p. 22, p. 107; Negija
1981 etc.; a preliminary critique of this point see e.g. Bol’shakov 1989, p. 40).
A somewhat strange theory, we must say, as the very well-known facts show quite
clearly that the actual processes were simply contrary to the ones described above.
The clan-tribal systems in pre-Islamic Arabia were strengthening and consolidating,
1The last dated Sabaic text (C 325 - see Müller 1991) is (see line 5) o f the year 669 o f the
“Himyarite” Era ~ AD 554/555, or much more likely AD 559/560, depending on the solution of
the problem of the beginning o f this era - for the current state o f this question see de Blois (1990);
Shahid (1994); Kitchen (1994, pp. 1-9); and especially Robin et al. (1996).

ORIGINS OF ISLAM 245
whereas these were precisely the state structures which degenerated and disintegrated
in the first century before al-Hijrah. Indeed at the beginning of the 6th century we
see a few kingdoms controlling most of the Arabian territory: the already mentioned
huge Kingdom of the tabäbi'ah in Yemen (dominant not only over the whole Ara-
bian South but also considerable parts of Central Arabia), the second Kindite King-
dom (the vassal of the first one) in Central Arabia, the Lakhmid Kingdom (depend-
ent on the Sassanid Empire) in the Arabian North-East (controlling also considerable
parts of Northern and Central Arabia), and the Ghassanid Kingdom (dependent on
the Byzantine Empire) in the North-West - see e.g. Nöldeke (1879; 1889); Rothstein
(1899); Olinder (1927); Pigulevskaja (1964) etc.
What is more, even in the territories outside the direct control of the above-
mentioned kingdoms we normally find what should be more correctly described as
chiefdoms rather than true tribes. Their heads often explicitly call themselves amläk
(sg. malik) “kings” - see e.g. Negrja (1981, pp. 103-104).
The situation at the beginning of the next century (say, at the time of the be-
ginning of Muhammad’s Prophecy) differs dramatically. All the above-mentioned
great Arabian kingdoms had disappeared together with most smaller ones. There were
almost no “kings” left in Arabia; and where there were chiefdoms a century before,
now we see true free tribes.'
Some neglected causes of the crisis
It appears that the 6th century AD evidenced a simultaneous global climate détériora-
tion and the peak of the tectonic and volcanic activity in the whole world (including
the Mediterranean region [see Appendix C for detail]). Of course, on the face of it, it
is not quite self-evident what this has to do with the 6th century AD Arabian crisis.
Naturally, the earthquakes affected in some way the evolution of the 6th-early 7th
century AD Arabian societies, leaving even some trace in al-Qur’än - cf. e.g. the be-
ginning of the famous Earthquake sürah ([XCIX:] {1.} idhä zulzilat1 ’l-arct zilzäla-
hä {2.} wa-akhrajat‘ ‘l-arçt athqäla-hä {3.} wa-qäla ’l-insän“ mä la-hä “When
the earth is shaken with an earthquake, and the earth lifts its loads, and the man asks:
‘What has happened to it?”’ etc.). Stookey (1978, p. 22) and Grjaznevich (1994, p.
34) have already proposed to connect the final decline of the pre-Islamic South Ara-
bian civilisation with the seismic activity - indeed it may well have produced the
final deadly blow to the most ancient civilisation centers of the edges of the internal
Yemeni desert, which were already on the brink of final collapse by the 6th century
AD and which depended heavily on relatively large-scale irrigation structures that
2 Even for the 6th-century Mecca there seem to be some grounds to suspect the transforma-
tion of a quasi-chiefdom polity into a tribal confederation (e.g. Dostal 1991, pp. 193-199; al-Tabari
1964, pp. 1083-1100). At the age of Muhammad local kings are still attested in al-Yamämah (see
e.g. ibn Hischäm 1858-1860, II, p. 971), but even there at this age we seem to observe a clear trend
towards the replacement o f the “royal” authority with a political authority o f quite a different type
(see below in the section on “Origins o f Islam: socio-political context”).
could be significantly affected by the earthquakes. But this does not seem to be the
case with the kingdoms and chiefdoms of the Arabian North which could not be ap-
parently affected by the earthquakes to a critical extent. Thus, the most significant
outcome of the seismic activity seems to be volcano eruptions rather than earth-
quakes. Again, it is not self-evident how, say, the volcano eruptions on the New Brit-
ain Island near New Guinea could affect the evolution of the Arabian communities.
Again, what is significant here is not the direct effect though some of the South Ara-
bian sites were destroyed just in this way (though not necessarily in the 6th century -
see e.g. Müller and Wissmann [1976]). What is really important are volcanic gases
and tephra which are thrown to the atmosphere in great quantities during such erup-
fions. And this could affect significantly really huge areas. E.g. sulphuric aerosols
would halt partially solar radiation, causing the cooling of the Earth surface and,
hence, droughts, or otherwise floods, and various disbalances in the ecological sys-
terns, which could result in the outbursts of the numbers of the epidemic disease bear-
ing animals, plague fleas etc., and the causal link between the tectonic and volcanic
activity and the epidemics was noticed long ago.
However, the most significant factor seems to be the droughts - and there are
documented cases when, say, changing solar activity or massive volcano eruption,
resulting in a global climatic shift, caused severe droughts in various parts of the
world (naturally, North Arabia could have been affected in such cases too [see Ap-
pendix C for detail]).
Hence, global climate deterioration and the peak of the tectonic activity pro-
duced such an array of primary, secondary, and tertiary factors (earthquakes and
volcano eruptions themselves, epidemics, droughts, barbarian invasions caused by
the socio-ecological crises on the barbarian peripheries) which could pose a deadly
threat for the survival of most of affected civilisations of the time. We ourselves
came to terms with the sudden death of the 1500-year-old pre-Islamic South Arabian
civilisation when we realised that this happened simultaneously with the severe crisis
in the Byzantine Empire which put it on the brink of an almost complete collapse
(the early 7th to early 6th century comparison would produce for Byzantine results
rather similar to the ones obtained above for the Arabian North and South in any
case). And what was an almost deadly blow for strong Byzantine appeared to have
been just a deadly blow for the weaker South Arabian civilisation as well as for most
Arabian kingdoms.3
This is not a mere speculation, especially for the Arabian North. Indeed, as
was mentioned above the second half of the 6th-century history of South Arabia is
documented very poorly (especially, in comparison with the earlier periods). But this
is not as true for the Arabian North. It is not simply that by the early 7th-early 6th-
century comparison we can deduce that most North and Central Arabian kingdoms
disintegrated, without knowing what happened in between. No, it is possible not only
3 The weakening o f the state structures o f the Byzantine, Sassanid and Yemeni empires
(caused to a considerable extent by the same socio-ecological factors), of course, led to the further
decline o f the Arabian kingdoms and chiefdoms most of which were to a considerable degree
rather dependent on the support o f those Near Eastern great powers.
to deduce this disintegration, but also to get to know in some detail how this disinte-
gration proceeded. Indeed, we have at our disposal e.g. the wonderful pre-Islamic
Arab historical tradition, the so called Ayyâm al- ‘Arab (“The Days of the Arabs”).
And one of the typical “Days” can be rendered as follows: there was some Arabian
strongman (a head of a kingdom, or a chiefdom) who behaved sometimes in a bad
and arrogant manner. Such a behaviour could consist of, say, shooting an arrow at
a she-camel of some woman,4but, very noticeably, it could be manifested in attempts
to collect taxes in a “lean” year (usually caused by a draught) - Ibn Habib (1942,
p. 249); Ibn al-Athir (1867, pp. 368-369); see also e.g. Kister (1986, p. 46); note also
e.g. the description by the Day o f Hujr tradition of the beginning of the Вапй Asad
uprising against Hujr (which finally [although by no means immediately] led to the
killing of Hujr and the destruction of the respective kingdom):
Inna Hujr" kânafî Banl Asad wa-kanat la-hu ‘alay-him itäwat“"fi kult
sanat1"... thumma ba'atha ilay-him jäbiya-hu ’lladhiкапа yajbi-him, fa-
mana'u dhälik wa-Hujr yawma-idh!" bi-Tihamah - wa-darabii rusula-
hu wa-daraju-hum “Hujr was [the king] of Вапй Asad, and the taxes
from them were due to him every year... Once he sent tax-collectors to
them [Вапй Asad] and they [Вапй Asad] refused [to pay taxes] (Hujr
was that time in Tihämah), beat the messengers and terribly wounded
them” (“the Day of Hujr” - al-Isfahânï [1955-1964, IX, p. 81]; see also
Ibn al-Athir [1867, p. 376]; al-Mawlaü-bik et al. [1942, p. 113]).
The beginning of the “Day of al-Nafrawat” is also not without interest here. It can be
rendered as follows (for the full quotation see Appendix В): when on a “lean" year
the Hawäzin had to bring their taxes to the head of the Hawäzin chiefdom, the chief
did not like the quantity and quality of the tax payment in kind brought by a certain
old woman and rudely pushed her (al-Isfahäni 1955-1964, XI, pp. 77-78; Ibn al-
Athir 1867, p. 413; Ibn ‘Abdi-Rabbi-hi 1949-1965, V, pp. 135-137; al-Mawlâ-bik et
al. 1942, pp. 235-236).
A typical reaction to “royal” misbehaviour would be that some tough bedou-
ins would go to such a chief and just kill him, which would provoke the revenge
attempts on the part of the murdered chiefs relatives, thus producing one more Arab
“Day” which could last for years filled with series of violent actions on both sides
(e.g. “the Day of Hujr” [al-Isfahäni 1955-1964, IX, pp. 81-103; Ibn al-Athir 1867,
pp. 373-382; al-Mawlâ-bik et al. 1942, pp. 112-123]; “the Day of al-Nafrawât” [al-
Isfahan! 1955-1964, XI, pp. 77-83; Ibn al-Athir 1867, pp. 411^114; Ibn ‘Abdi-
Rabbi-hi 1949-1965, V, pp. 135-137; al-Mawlâ-bik et al. 1942, pp. 235-241]; “the
Day of Khazäz” in Yaqüt’s edition [1410/1990, II, pp. 418^419] ; or the biography of
the famous pre-Islamic poet ‘Amr b. al-Kalthüm [e.g. al-Isfahäni 1955-1964, XI, pp.
4 Incidentally, according to Ayyäm , such an action once precipitated 40 year long violent
hostilities, known by the name of the above-mentioned woman as “the Day (or War) of al-Basüs”
(al-Isfahânï 1955-1964, V, pp. 29-55; Ibn al-Athir 1867, pp. 384-397; Ibn ‘Abdi-Rabbi-hi 19 4 9 -
1965, V, pp. 213-250; Yäqüt 1410/1990, IV, pp. 418^*19; al-Mawlâ-bik et al. 1942, pp. 142-169
etc.), c. 490-530 AD.
38-54] who himself took an active part in his tribe’s struggle against the Lakhmid
Kingdom which seems to have contributed significantly to the weakening and final
dissolution of this political entity; on the struggle of Arab tribes against the Ghas-
sanid kings see e.g. Negrja [1981, pp. 36-37]; see also e.g. al-Bayâtî [1407/1987, pp.
232 ff]). In any case, at the end we normally find original chiefdoms or kingdoms
disintegrated with free true tribes in their places.
The Arab adaptation to the 6th century AD socio-ecological crisis
Actually, what was described at the end of the previous section may well be considered
as an important component of this rather effective adaptation. This was simply that
most socio-political systems of the Arabs (or, for the extreme methodological indi-
vidualists, the Arabs themselves, anyhow, it could be well described in both ways)
reacted rather adequately to the socio-ecological crisis by getting rid of the rigid
supratribal political structures (i.e. all those kings, chiefs and their retainers) which
started posing a real threat to their very survival. Indeed, it is rather difficult to imag-
ine anything more nasty than the royal messengers coming to you in a “lean year”
(which may well have been preceded by one or two similar years) and demanding
from you to pay royal taxes when you yourself have nothing to eat and to feed your
children.
However, the Arabs did not only destroy most of those rigid political supra-
communal structures which were alienating the tribal sovereignty; they also devel-
oped their alternatives - soft structures not posing any threat to the sovereignty of
tribes. Most noticeable of them seems to be the development of the system of sacred
enclaves, regular pilgrimages to them and the regular pilgrim fairs (mawäsim) which
accompanied those pilgrimages.
The result was the development of rather effective intersocietal networks, of
which the best known is the Western Arabian religious-political area (the functioning
and evolution of which, incidentally, left a noticeable trace on the history of the World
System as a whole). It seems to have been formed as a result of the expansion of the
zones of influence of the respective sanctuaries, their interweaving into one more or
less integrated religious-political area.
This of course was primarily a religious area, yet it had evident political di-
mensions too. It was in the pilgrimage-fairs (mawäsim) at the above-mentioned sane-
tuaries “that traditional tribal society established its manifold contacts, the exchange
of the religious and cultural ideas, as well as the barter of products with only use-
value. Furthermore, the various legal problems (armistice, debts, benefits, payment
of blood-money, bailing out of prisoners, finding of clients, looking for disappeared
persons, questions of heritage, etc.) of the participants were also settled there. This
exchange of ideas and goods, as well as the spreading of legal customs and cults com-
mon to several tribes, that is, regular social contact in general, played no negligible
role in the extension of particular tribal consciousness” (Simon 1989, p. 90; also see
especially Wellhausen [1897/1961, pp. 88-91]).

As a result we can observe the formation of a certain political area more or

less correlating with the religious one, an area where certain norms of not only relig-
ious, but also the political culture were shared, where the people would avoid killing
travellers in ashhur hurum, the holy months (and would consider the same parts of
the year as the holy months), where the representatives of various tribes would go to
the same places to settle their conflicts, and would observe the same rules of political
mediation etc. The most remarkable fact is the almost complete absence of signifi-
cant intertribal warfare in “the Area of the Four Sanctuaries” (Majannah, dhü-’l-
Majäz, ‘Ukäz and Mecca5) between the time of its final formation (i.e. Harb al-Fijär
in the last decade of the 6th century AD) and the start of the clashes with the Mus-
lims. Actually at this time we can observe in the “Area of Four Sanctuaries”6 cul-
tural-political entity, which in the absence of any significant political centralisation
secured the existence of a huge cultural network within which a very intensive (and
very productive) exchange of information, energy and matter took place7.
Incidentally, these types of cultural-political entities seem to be ignored
(without any reasonable justification) by practically all the “classical”
theories of social evolution (e.g. Claessen and Skalnik 1978; Claessen
et al. 1985; Fried 1967; Hallpike 1986; Lenski 1987; Parsons 1977;
Sanderson 1990; 1995; Service 1971 [1962]; Spier 1996) and do not
seem to fit in all these essentially unilineal evolutionist schemes, espe-
daily in their most popular band - tribe - chiefdom - state version
(with all its modifications). Indeed, all the Western Arabian polities of
the early 7th century appear to have had rather “primitive” socio-politi
5 Of course, ‘Arafah and Mina could be also added to this list, separately from Mecca; how-
ever, they could be also regarded as parts o f the Meccan Haram.
6 In the early 7th century AD it covered not only Western Arabia, but also considerable
parts of the other Arabian regions.
7 Being polycentric (it is rather doubtful that Mecca could be considered as its only impor-
tant centre [see e.g. Crone 1987]) the Western Arabian area seems to have had a considerably het-
erogeneous structure including a few interweaving subsystems centred on the respective sanctu-
aries. The best known is the hums amphyctiony centred on the Meccan sanctuary (e.g. Ibn Hischäm
1 8 5 8 -1 9 6 0 ,1, pp. 126-129; al-Azraqi 1858, pp. 118-125, p. 130; al-Ja‘qübï 1 8 8 3 ,1, pp. 297-298;
Ibn Habib 1942, pp. 178-181). Kister observed that the tribes belonging to this amphyctiony were
mainly spread along the trade routes of Quraysh (Kister 1965, p. 134), from which the conclusion
was made that this was a kind of trading alliance cemented by the shared religious norms (Simon
1989, pp. 63-64). This might appear implausible taking into consideration the fact that the
affiliation to hums was normally transmitted through the matriline (e.g. Ibn Hischäm 1 8 5 8 -1 9 6 0 ,1,
p. 127; al-Azraqi 1858, p. 122), which is apparently not quite convenient for the spread o f any
religious norms through alliances. However, it should be taken into consideration that the Quraysh
before Islam practised extensively exogamie marriages, whereas most o f these marriages were not
fortuitous at all and aimed at establishing alliances with important tribes and clans (Dostal 1991). It
could be also taken into consideration that “the Quraysh gave not their daughters in marriage
unless on the condition that the children would become hums" (e.g. Kister 1965, p. 136, citing an
unpublished Ms by al-Jâhiz; see also al-Azraql 1858, p. 115). It is known of some Arab tribes and
clans that they became hums just in this way (e.g. al-Azraqi 1858, p. 123). Hence, taking into
consideration the fact that the main interests o f the Quraysh were focused along their trading
routes, it is not so surprising to find the hums tribes just in those areas.
cal structures (which seems to be valid even with respect to the Meccan
community [see e.g. Dostal 1991]) and, according, to such schemes
could be only classified as “autonomous communities”, “tribes”, at most
as “chiefdoms” (though most Arabian “chiefdoms” seem to have
disintegrated in the second half of the 6th century AD). However, they
were parts of a much wider cultural-political entity whose overall level
of social complexity may well be compared with the one of an average
“early state”; though lacking the political centralisation this entity fails
to find its place in the above-mentioned schemes (this appears to be
true with respect to any processes of socio-cultural growth which are
not accompanied by the growing political centralisation, or especially
going in hand with the political decentralisation).
In general, the Arabs appear to have developed a rather effective adaptation to the
6th century socio-ecological crisis. The soft intersocietal networks they created even
permitted them to assume a role of the guardians of the important World System
links in the World System Southern area, a role which the Great Powers of the late
6th-early 7th century were already unable to perform.
In the late 1980s two monographs specifically dedicated to the subject
of the Meccan trade were published (Crone 1987; Simon 1989). Crone
questioned the usual point that the Meccan trade was extremely profit-
able and important, and that Mecca itself was the capital of a huge mer-
chant empire. Indeed, the plausibility of the caravan trade route between
Southern Arabia and the Fertile Crescent which the Meccans were sup-
posed to serve does not appear very high after the beginning of the CE;
as is well known, in the 1st century B C -lst century AD the main
trading routes between South Arabia and the North were transferred to
the sea (J. Ryckmans 1951, p. 331; Bowen 1958, p. 35; Irvine 1973, p.
301; von Wissmann 1981, p. 66; Robin 1982a, I, p. 98; 1982b, p. 17;
1984, p. 212; Crone 1987, pp. 23-36; Audouin-Breton-Robin 1988, p.
74; Bauer-Lundin 1994, pp. 105-106). After that the old Transarabian
caravan “incense” route definitely lost all its importance unable to com-
pete with the much more effective sea way. Hence, it is rather difficult
to imagine after that anything like a “Meccan commercial empire”
thriving on just the Transarabian caravan trade. Meccan Trade and
Islam by Robert Simon was published two years after Patricia Crone’s
Meccan Trade and the Rise o f Islam. Actually, Simon’s monograph is
the English translation of his A tnekkai kereskedelem kialakuldsa e's
jellege originally published by Akadémiai Kiadö in Budapest in 1975.
However, the impression is that Simon’s monograph was actually writ-
ten after Crone’s book, as he managed to find the “golden middle” be-
tween the uncritical traditional accounts of the ancient and huge “Mec-
can commercial empire” (e.g. Lammens 1910; 1924; Watt 1953, p. 3;
1964, p. 1; Donner 1977) and the hypercritical position of Crone. Simon

has shown that the reasonably (though by no means excessively) profit-

able Meccan Transarabian trading network existed, but it was a rather
late phenomenon, having arisen in a more or less full-fledged form in
the 90s of the 6th and the very beginning of the 7th centuries AD - e.g.
the Meccan direct trade with Iraq only developed at the beginning of
the 7th century; Simon e.g. cites (p. 70) the following impressive words
pronounced by a rather famous contemporary of Muhammad, АЬй Suf-
yän b. Harb (the father of the first Omayyad Caliph, Mu‘âwiyah), dur-
ing what seems to have been the Meccan trading expedition which only
at the beginning of the 7th century established for the first time the
direct commercial links between Mecca and Mesopotamia: in-nâ min
masïri-nâ la-’alâ khatar" mä qudümu-nä ‘alä malik‘njabbär" lam ya’~
dhan la-nä f l 'l-qudum ‘alay-hi wa-laysat bilâdu-hu la-nä bi-matjar"
(“We are indeed in danger on our present journey till we reach a power-
ful king who did not give us permission to visit him and his country is
no trading place for us” [al-Isfahäni 1955-1964, XIII, p. 207]). Simon
directly suggests that the formation of the full-fledged Meccan trading
network became possible just because of the destruction of the main
Arabian kingdoms which earlier controlled the Transarabian trade (e.g.
any Meccan control over the Arabian trade with Iraq was highly prob-
lematic before the destruction of the Lakhmid Kingdom which actually
controlled this trade). Thus the formation of the Meccan commercial
network could well be seen as a part of the Arab adaptation to the 6th-
century socio-ecological crisis.8
8 However, Simon seems to have overlooked one important factor which appears to have
enhanced significantly the viability o f the Transarabian trade, seriously contributing to the success
of the Meccan trading network at the beginning o f the 7th century. This was the final subjugation
o f Yemen by the Persians at the very end o f the 6th century. The sea trade with Yemen seems to
have been extremely risky for the Greek merchants already in the last quarter o f the 6th century
taking into consideration the very unstable situation in Yemen at that time (the first Persian
expedition, internal strife etc.). And, o f course, one can hardly imagine any reasonable sea trade by
the Greek merchants with Yemen after the final Persian occupation taking into consideration the
immense Greek-Persian hostilities just at that time (Persian-Byzantine war o f 602-627 AD). These
circumstances appear to have been precisely those which created a niche for the reasonably
(though in no way extremely) profitable Meccan overland Transarabian trade (including the trade
between Yemen and Syria). Demonstrating the absurdity o f the traditional uncritical accounts of
the Meccan (and in general, Arabian) trade Crone presents them in the following way: “They were
a curious people in that they sailed to Africa and India, but transported their goods by caravan on
reaching their native shores” (Crone 1987, p. 9, with reference to Kennedy 1898, pp. 248 ff; Rath-
jens 1962, p. 115; D oe 1971, p. 50). This refers to the pre-Meccan epoch o f the Arabian trade, as
nobody seems to insist on the probability o f the extensive Meccan sea-borne trade. Yet this appears
to be valid for the 6th century AD as well - anyhow, it would be strange to imagine that any mer-
chandises could be first delivered by sea to Yemen and then transported to the North by land, rather
than the Red Sea. However, after the final Persian occupation o f Yemen and the beginning o f the
25-year Persian-Byzantine war this really strange arrangement, perhaps for the first time after the
beginning o f the CE, became reasonable. Peters, the author o f the most recent monograph on
“Muhammad and the Origins o f Islam” (Peters 1994), who occupies an intermediate place between
Origins of Islam: socio-political context
Though some pre-Islamic Arabian tribes managed to find a rather effective adapta-
tion to the crisis along the lines described in the previous section, this adaptation does
not appear to have been quite perfect everywhere. After the destruction of the politi
Crone and Simon (whose monograph, although, does not appear to be known to him), maintains
that “the Arab caravan cities like Petra... still bear eloquent testimony to the prosperity o f its mer-
chants in the form o f capital investment in municipal buildings and monuments. Muhammad’s
Mecca, on the other hand, boasted one unroofed stone building, the Ka'bah, amidst its mud-brick
dwellings” (p. 72). Surprisingly Peters fails to recollect at this point (though he manages to do this
a considerable bit later [pp. 102, 138-141]) that there was at least one major “capital investment”
o f this kind in pre-Islamic Mecca. This “investment” is rather well known and that is the “re-con-
struction” of the Ka’bah in c. 603-605 AD, a “reconstruction” which, incidentally, involved the
pulling down of all the sanctuary walls and the construction o f the new ones, twice as high (c. 9 m)
as the old ones, as well as the roofing o f the building (apparently for the first time in its history);
additionally the walls were plastered and artistically painted from inside (al-Azraqi 1858, pp. 104-
118; Ibn Hischäm: 1 8 5 8 -1 8 6 0 ,1, pp. 122-126). Actually, the old unroofed sanctuary building was
pulled down and a new, tw ice as high, roofed one was built. After that “Muhammad’s Mecca”
(contra Peters) could boast at least one decent roofed “municipal” building. Note that this was a
capital investment in the most strict sense o f this word (o f course, if one at least partly believes the
traditional Muslim accounts o f the fact that the Meccans derived a considerable part o f their wealth
from the servicing o f the pilgrims coming to the Ka’bah, and that their commercial links enjoyed
special protection due to their special association with this apparently rather venerated Arabian
sanctuary [e.g. Kister 1965)). O f course this only major pre-Islamic “capital investment” does not
fit well in the traditional portrayal o f the fabulously rich Meccan commercial empire flourishing all
the way through the 6th century AD since the time o f Hashirn, but it fits entirely well Simon’s
reconstruction, according to which the effective Transarabian Meccan trading network was a rela-
tively recent phenomenon with the Meccan community achieving a reasonable (but by no means
fabulous) level o f well-being just by the time o f the “Reconstruction of the Ka’bah”. Anyhow, the
general impression is that the Meccan Transarabian commercial network formed in its full-fledged
form just at the end o f the 6th century and the beginning of the 7th century. Just during the first dec-
ades o f Muhammad’s life the Meccans appear to have managed to finalise the formation o f their
commercial network and to make it work rather effectively (Muhammad him self took his part in
this, e.g. participating in the 590s in the ‘Ukäz Battle [e.g. Ibn Hischäm 1858-1960, I, pp. 118-
119; Ibn Saad 1905, pp. 181-182; Ibn Habib 1942, p. 210, p. 211, p. 218] o f the War o f Fijär “by
which Mecca eliminated the last internal and external middlemen and obtained supremacy over the
transit trade [in] the Arabian peninsula” (Simon 1989, p. 159; see, however, a different interpréta-
tion of these events in Crone 1987, pp. 145-148). Just by the beginning o f Muhammad’s Prophecy
the Meccans seem to have finally and completely succeeded in finding their rather effective way
out o f the 6th-century Arabian crisis, having sorted out all the major economic and political prob-
lems and achieved a rather reasonable (but by no means fabulous) level o f well-being. Note that
the Qur'an itself portrays a rather affluent society where it could be rather difficult to understand
why some other people have to kill their new-born babies when there are lots o f food all around
(VI, ayahs 140-141; ayah 151; XVII, ayah 31; CVIII). Hence, there is no surprise that the Meccan
community (which had sorted out all the major problems and did not really need any radical struc-
tural changes, any new type o f political authority) was not at all the best place where Muhammad’s
Prophecy could be completely accepted, though it seems to have provided a good environment for
the initial development o f the prophecy (with its lack o f the political centralisation and a reason-
able level of well-being, when a considerable proportion o f Meccans could afford to spend lots of
time on thinking about the things other than their daily bread, without being too much afraid of
some state-sponsored persecutions).
cal structures of the Arabian kingdoms and chiefdoms not all the Arabian communi-
ties entirely succeeded in working out effective substitutes for them. They seem to
have succeeded e.g. in the Meccan area (and that is why there was no sufficient
space for Muhammad’s Prophetic activity there), but they do not seem to have been
so successful already on the periphery of this area, say, in Yathrib, where a few
tribes could not sort out their relations in the absence of any effective supertribal
authority.
Such a problem was not new in Arabia. And at the end of the 5th, or the begin-
ning of the 6th century the answer was quite clear - to send messengers to some
great Arabian King and to ask him to appoint a king over those tribes (see Appendix
A; or e.g. Ibn Habib [1942, p. 249]).
However, such a practice appears to have become unacceptable by the 7th
century. The decades of fighting which led to the destruction of most of the Arabian
kingdoms and chiefdoms seem to have also led to the elaboration of some definite
“anti-royal” freedom-loving tribal ethos codified in the tribal historical traditions and
poetry - see e.g. al-ти ‘allaqah of ‘Amr b. Kulthüm [e.g. Ibn Kulthüm 1983, pp. 263-
269; or al-Zawzani 1972, pp. 163-187], or such lines as: wa-laysa ‘alay-nä qatlu-
hum [al-muluk] bi-muharram" (“It is not forbidden to us to kill the kings!” - al-Dabbi
1964, p. 211, note 42, the second half of bayt 20; a line with a similar sense see e.g.
in the “Warning to Imru’-’l-Qays” by ‘Abid b. al-Abras [1983, p. 164/ bayt 20]) etc.
The reflections of this ethos seem to be present even in al-Qur’än - see XXVII/34:
al-mulûk“ idhà dakhalü qaryat‫ "״‬afsadü-hä wa-ja’alü a'izzaf ahli-hä adhillaf” wa-
ka-dhàlika yaf ‘alün (“The kings, when they enter a town, they corrupt it; they make
the most glorious of its folk the most base, they do it this way”). The second Caliph,
‘Umar, would even say: inna-hu la-yaqbuhu bi- 7- ‘arab an yamlik° ba ‘du-hum ba ‘cf"
(“It is disgusting for the Arabs that some of them reigned over others” - al-Tabari
1964, p. 2012) - a striking contrast with the situation a century before when most
Arabs were subject (in one, or another way) to the kings.
Anyway, at the beginning of the 7th century a tribe which would recognise
itself as subject of some terrestrial supertribal political authority, a “king”, risked to
lose its honour. However, this seems not to be applicable to the authority of another
type, the “celestial” one.
Note, e.g. the words of a famous Arab poet al-Hutay’ah said at the time of the
Arab tribes’ revolts after the death of the Prophet during the reign of the first Caliph,
Abù-Bakr:
ata ‘nä rasüf ’lläh‘ idh käna bayna-nä
fa-yä la- 'ibäct ’lläh' mä li-Abï Bakr
a-yürithu-hä Bakr " idhä mäta ba‘da-hu...
“We obeyed the God’s messenger, when he was among us.
We are the servants of the God, not the servants of Abü Bakr!
Will he leave us to Bakr ’ as inheritance?” (al-Isfahäni 1955-1964, II, p. 130).9
9 Abü Bakr literally means “the father o f Bakr”.

Hence, the impression is that whereas for many Arab tribes becoming subjects of
some terrestrial king was entirely unacceptable, was tantamount to an enormous loss
of honour, the recognition of some “celestial” authority (naturally through its terres-
trial representative) was more or less acceptable.
Another group of facts should be also taken into consideration here. The pre-
Islamic Arabia knew rather well the figure of “prophet” (kähin - see e.g. al-Mas‘ùdï
1965, pp. 151-176; Ibn Khaldün 1415/1995, pp. 96-98; Fahd 1966, pp. 91-104).
Anyhow, an average Arab seems to have known quite well how a “prophet” looked
like, what the prophetic trans was, etc. However, all the pre-7th-century Arab proph-
ets (kahanah) were the ones of the pagan deities. Hence, their authority was not the
best possible one, as the recognition of their authority would mean the recognition
of the authority of the respective pagan deity, whereas all the cults of such deities
would be normally connected with a specific tribe, whose protector this deity was -
hence, such a recognition would imply the recognition of the authority of the respec-
tive tribe as well (as is amply evidenced e.g. by the South Arabian epigraphy - see
e.g. Beeston 1984a).
Hence, the best possible figure here would be rather some Monotheist prophet.
However, the prophets of the established Monotheist Faiths would not be entirely
suitable as well, as the recognition of their authority would imply the dependence on
some extra-Arabian powers, or in the case of Judaism would put the Arab Jewish
tribes in an advantageous position.
At the meantime there seems to have been a more or less independent Mono-
theist Arabian (“Rahmanist”Уal-Hanïfiyyah) tradition.10 However, its North Arabian
adherents (hunafä’) do not appear to have given any prophets before the 7th century
(see e.g. Piotrovskij [1984, p. 20]). Yet, in the early 7th century both traditions (the
Arabian tradition of prophecy and the Arabian Monotheist “Rahmanist” tradition)
seem to have merged, producing what Piotrovskij calls “the Arabian prophetic move-
ment” (Piotrovskij 1984). It should be taken into consideration that in addition to
Muhammad there were at least 5 other Monotheist prophets (pseudo-prophets, of
course, from the Moslem point of view) in Arabia at the time of Muhammad. Beside
one Judaic prophet in Yathrib (Ibn Sayyäd - see e.g. Halperin [1976]) and a para-
Christian prophetess, Sajäh (e.g. al-Tabari 1964, pp. 1911-1916), 3 others (al-Musay-
limah, al-Aswad and [rather hypothetically] Tulayhah b. Khuwaylid) seem to have
belonged to the Arabian “Rahmanist” tradition (e.g. Bartol’d 1925; Piotrovskij
1984). Note (e.g. al-Tabari 1964, p. 1933, p. 1937; Balädhüri 1866, pp. 105-107)
that both al-Musaylimah and al-Aswad called the God al-Rahmän (just as was done
by Muhammad - see especially al-Qur'än XVII, p. 110: qui ud‘ü ’{A}lläha aw ud‘û
’l-Rahmän ayyan-mä tad'ü fa-la-hu ’l-asmä’“ ’l-husnä “Say - invoke the God, or
invoke Rahman, however you would call, He has the best names” - it has been even
suggested that bi-smi ’{Ajllähi ’l-rahmäni ,l-rahim may well be translated as “In the
10This hypothesis is still under attack (see e.g. Rippin 1991), however we do not think it
has been either finally proved, or rejected, and can be still regarded as a working hypothesis (see
especially Shahid 1989, pp. 154-156, pp. 162-172, pp. 332-338; and Beeston 1984a; 1984b; 1984c;
Rubin 1990; Peters 1994, pp. 117-128; Korotayev 1996b etc.)
name of the merciful God Rahman”;" note that after с. 460-470 AD the God was
also called Rahman(an) by the authors of pre-Islamic Monotheist inscriptions of South
Arabia (incidentally, most of them could not be identified for sure either as Jewish,
or as Christian [e.g. Robin 1980; 1991, pp. 146-147; Beeston 1984a; 1984b; 1984c]).
By the moment our impression is that there were actually two relatively
independent autochthonous monotheist traditions in pre-Islamic Arabia:
millat/din Ibrahim (= al-Hanifiyyahl) of North-West and West Arabia,
and “Rahmanism” of Yemen and al-Yamâmah. For the most recent (and
best, to our knowledge) collection of data on pre-Islamic Western Ara-
bian millat Ibrahim see Peters 1994, pp. 118-128; see also e.g. Rubin
1990, pp. 85-112, as well as Ibn Hischäm 1858-1860, I, pp. 143-149,
p. 293, pp. 311-312 etc.; Watt 1975, pp. 168-170; Beeston 1984c, p. 151
etc. The Rahmanist tradition which in the 6th century produced at least
two autochthonous Monotheist prophets (or, naturally, pseudo-prophets
from the Muslim point of view), al-Aswad in Yemen and al-Musay-
lirnah in al-Yamâmah (cf. e.g. al-Tabari 1964, p. 1933, p. 1937; Ba-
lâdhürï 1866, pp. 105-107), seems to have originated somehow later1'
than the Northern Abrahamic tradition, apparently among the superfi-
daily judaicised population of these areas (e.g. Korotayev 1996b); the
fact that the North-Western Abrahamic monotheist tradition was con-
siderably more ancient than the Southern Rahmanist one seems to be
shown by the data supplied by Sozomen and Theodorite (see e.g.
Shahid 1989, pp. 154-156, pp. 167-172, pp. 332-338; cf. Robin 1991,
p. 146). The point that unlike millat Ibrahim the Rahmanist tradition
seems to have been virtually unknown in pre-Islamic Mecca appears to
be evidenced by the well-known fact: when Muhammad during his
second Meccan period started using al-Rahmän as the name of the God,
it turned out that this name (unlike the notion of Allah) was unknown to
almost all the Meccans, as is shown e.g. by the following Qur’anic
äyah: idhä qila la-hum ’sjudü li-’l-Rahmän qälü wa-mä ’l-Rahmän“...
“If it is said to them: ‘Prostrate yourselves before al-Rahmän’, they will
say: ‘And what is al-Rahmän?’...” (XXV/60; see also XIII/30; XXI/36).
Indeed, Rahman as the name of the God is attested in the Northern
Jewish texts (Sokoloff 1990, p. 522), including the epigraphic ones, e.g.
l-smh d-Rhmnh “in the name of Rahmän” (MPAT-A22/10; see also
MPAT-A 39/2 ff). However, the pervasive use of RHMN" as the name
of the God (in the Jewish, Christian and indefinite Monotheist texts),
seems to be a peculiar phenomenon of the late 5th-early 6th-century
South Arabia; at this time here the God was named Rahman“" in
" “Dans l ’invocation bi-{i}sm i {A}llah ar-rahmän ar-rahïm, il est clair que ar-Rahmän
était à l ’origine un nom propre et que les sens premier était : ‘au nom du dieu ar-Rahmân le mise-
ricordieux’” (Robin 1991, p. 146).
12 In any case, apparently not earlier than 4 6 0 -4 7 0 AD when RHMN" appeared as the main
name of the God in the South Arabian Monotheist inscriptions (e.g. Robin 1980; 1991, pp. 146-147).
2 56 ANDREY KOROTAEV-VLADIMIR KLIMENKO-DMITRY PROUSSAKOV
virtually all the texts mentioning the God’s name (c. 2 dozen cases)
(Robin 1980, pp. 85-96), which stands in quite a sharp contrast with
e.g. the Jewish Palestinian Aramaic epigraphic texts of the Byzantine
period where the God is named Rhmnh only twice (Sokoloff 1990, p.
522), though the total number of these texts exceeds the number of the
late 5th-early 6th-century South Arabian texts. Within such a context
any Monotheists “deviating” from the established Monotheist faiths
were most likely to call the God just Rahmänän/al-Rahmän. Inciden-
tally, these developments could not be without repercussions in Central
Arabia (including al-Yamämah) where the Yemenites exerted consider-
able influence just at this time through their Kindite vassals (e.g. Gajda
1996; Robin 1996), hence, the presence in al-Yamämah of the inde-
pendent Rahmanist tradition which at the age of Muhammad even pro-
duced a Rahmanist prophet (or, naturally, pseudo-prophet from the Mus-
lim point of view) appears here of no surprise - for more detail see e.g.
Korotayev (1996b). Irrespective of what have been mentioned above, in
this paper we denote as “Rahmanist” all the prophets belonging to the
Arabian autochthonous monotheist traditions for the present-day lack of
any other more adequate term.
The Monotheist “Rahmanist” prophets appear to have represented a supertribal au-
thority just of the type many Arab tribes were looking for at this very time. Note, that
all the “Rahmanist” prophets achieved considerable political success in their areas
(al-Musaylimah in al-Yamämah [e.g. al-Tabari 1964, pp. 1929 ff], Tulayhah in Cen-
tral Arabia [e.g. al-Tabari 1964, pp. 1885 ff], al-Aswad in Yemen [e.g. al-Tabari
1964, pp. 1851 ff], though the political success of Sajäh in the Arabian extreme
North-East [e.g. al-Tabari 1964, pp. 1908 ff] also appears relevant in this respect) -
their success could not be compared with the one of Muhammad, but their political
success was considerable, however; and they seem to show that in the early 7th-cen-
tury Arabia such success could be achieved by a “Rahmanist” prophet rather than by
a king.
In general, our impression is that the origins of Islam could be well considered
as a rather logical outcome of the Arabian processes of adaptation to the 6th-century
socio-ecological crisis in the context of the developing autochthonous Arabian pro-
phetic and monotheist traditions, i.e. this adaptation did not actually cause the forma-
tion of Islam, but rather created in many parts of the Peninsula a socio-political
environment most suitable for the development of an autochthonous Arabian mono-
theist religion (spiritual preconditions of which already existed by the age of the crisis).
Some world system consequences: pilgrimage structures
From what has been mentioned in the previous section it must be rather clear that the
Arab adaptation to the 6th century AD crisis influenced the world system develop-
ment mainly through one of its more or less logical outcomes, the formation of Islam.
Indeed, it seems possible to show that the Islamic civilisation incorporated

many of the important patterns, structures, values and attitudes elaborated by the
Arabs during this adaptation.
One of the most obvious points here is the pilgrimage system of a typically
Arabian type. Of course, the Arabian pilgrimage practices did not arise in the 6th
century AD. They are much older being attested already in the first Arabian written
documents of the beginning of the 1st millennium BC (e.g. in the earliest Sabaean
epigraphy [YM 375 = YM 1064 = CIAS {1977} 95.4l/r4]). However, it was in the
6th century AD when the most effective intersocietal communication network based
largely on the enhanced pilgrimage practices was elaborated in Western Arabia, the
network which served as a rather potent substitute for the rigid supertribal political
structures principally destroyed by the Arabs during their adaptation to the 6th-
century crisis.
Islam was embraced and spread by the people who grew up within the setting
of the Western Arabian intersocietal network were the pilgrimage practices played an
essential “structure-constituting” role. Of course, the pilgrimage (al-hajj) was pre-
scribed by al-Qur’än, but not all such prescriptions and prohibitions were applied by
the Arabs to the same extent'3. The pilgrimage prescription was applied quite consis-
tently and effectively to a considerable extent because the necessity of the pilgrimage
was self-evident for the Arabs.
For a specialist in pre-Islamic Arabia the proliferation of the Islamic civilisa-
tion appears (to a considerable degree) as a spread (in a rather modified form) of
many important ancient Arabian structures following some patterns which could be
traced in Arabia for about 1500 years before Islam. Perhaps, a bit surprisingly some
of these patterns look more like South (rather than North) Arabian.
The South Arabian religious-political areas were created firstly by the political
expansion of respective South Arabian states which was accompanied by the expan-
sion of the conquerors’ religion within the borders of the respective political entity.
The formation of the religious-political area was relatively independent from the fate
of its kingdom-creator (on evolution and functioning of an ancient South Arabian
cultural-political area see e.g. Korotayev 1995; 1996a).
The Western Arabian religious-political area appears to have been mainly ere-
ated by the proliferation of the religious authority of the respective sanctuaries (which
could of course have some connection with the political activities of their guardians,
however, this political activity was entirely different from the mainly violent expan-
sion of the South Arabian states).
The expansion of the religious authority of the sanctuaries led to the spread of
the correlating political culture. Hence, in the process of the formation of the South
Arabian religious-political areas “the politics went in front of religion”, whereas in
Western Arabia religion appears to have “gone in front of the politics”.
15 Note e.g. the prohibition o f wine (a rather popular drink in pre-Islamic Arabia [e.g. Ma-
raqten 1993]) which was not applied quite fully as evidenced by the huge corpus of the Arab “wine
poetry” ( khamriyyät).
Surprisingly, we can observe the South Arabian pattern with respect to the
evolution of the religious-political area with the centre in Western Arabia in the Is-
lamic Age. This religious-political area is just what is usually called the “Islamic
Civilisation”. Indeed, its evolution presents just the familiar South Arabian scheme
of the development of a religious-political area: the political (mainly military) expan-
sion creates an “empire” - the religion of conquerors spreads within this empire -
after the disintegration of the empire religious-political area remains and expands
beyond its borders; the population of the area shares common religious norms and
common norms of political culture; a very important role in the integration of the re-
ligious-political area is played by the pilgrimage to the central sanctuary.
Anyhow, the establishment of the Islamic Pilgrimage system had important
consequences for the world system evolution.
It should be taken into consideration that the pre-Islamic West Arabian pil-
grimage system (on the basis of which the Islamic one was formed) was very well
adapted to serve as an integrating mechanism for an intersocietal communication
network lacking the political unity. It might not be a mere coincidence that the
Islamic one turned out to have rather similar properties. Of course, for the first 150
years of Islam the Moslem pilgrimage area was more or less identical with the
territory controlled by the united Islamic polity. However, after the disintegration of
the latter, this system turned out to work precisely as its pre-Islamic Arabian counter-
parts, serving as an important integrating mechanism for an intersocietal communica-
tion network not united politically.
Hence, one might suppose that one of the world system consequences of the
Arab adaptation to the 6th-century crisis was the formation of an important mecha-
nism securing the integration of a huge intersocietal network covering some most
important central areas of the world system (and many peripheral areas as well),
a mechanism which secured the unity of some significant patterns, values and prac-
tices throughout all this territory, guaranteeing the annual meeting of the représenta-
fives of all the societies covered by the respective network in one place, the exchange
of information between them, the constant reintegration of the network, etc.
Some world system consequences: tribal structures
As has been already mentioned above Arabs elaborated a rather effective adaptation
to the 6th-century crisis to a considerable extent through the massive transforma-
tion of their state and chiefdom structures into the tribal ones. This could hardly be
regarded as a “degeneration” because the newly elaborated tribal structures turned
out to be able to serve the functional needs of rather complex stratified societies.
With the Islamic conquests these tribal structures and tribal ethos (al-qa-
byalah) appear to have proliferated through almost all of the territory of the new
Islamic state.
Of course, it should be stressed that there is not so much of al-qabyalah in
Islam itself. Yet it seems necessary to take into account the following points.

To start with, within the Russian Islamic Studies the Islamic civilisation was
traditionally designated as the “Arab-Moslem” one (which naturally often met strong
objections on the part of our Moslem colleagues from the former Soviet Central Asia
[e.g. Ahmadjonzoda 1988]).
However, we would stress that this designation is rather helpful in some re-
spects. The fact is that this civilisation (especially within the territory of the first
Islamic Empire) seems to contain important Arab non-Islamic elements (and cannot
be understood without taking them into account). And al-qabyalah appears to be one
of them.
It is essential to mention that the Arabs were the dominant ethnos within the
Islamic Empire at least till the Abbasid revolution in the middle of the 8th century
AD; and the Arab culture as a whole (including its non-Islamic components, like al-
qabyalah) acquired a rather high prestige and proliferated within the borders of the
Empire.
The proliferation of the tribal structures and tribal ethos seems to have had
both positive and negative consequences.
On the one hand, in the areas where most of the population acquired the tribal
organisation it often permitted the existence of complex systems of non-oppressed
agriculturists (which is very difficult to find otherwise in the pre-industrial world).
One of the most evident cases is the North-East Yemen Highlands of this millen-
nium, where the tribal organisation for most time effectively prevented the exploita-
tion of most agriculturists (who were honourable armed tribesmen), at the meantime
securing the existence of an intense network of markets, towns, centres of learning
etc. [see e.g. Dresch 1989].
Notwithstanding all the attractiveness of such systems, some negative conse-
quences of their proliferation should neither be overlooked. Looking rather attractive
from inside they often looked entirely unattractive for their non-tribal neighbours,
who often had to deal with rather destructive side-effects of their functioning. In
general, the proliferation of the tribal structures seem to have played a rather impor-
tant role in the inducing of the cyclical “Khaldunian” processes (for a model of such
processes in addition to Ibn Khaldün’s al-Muqaddimah [e.g. Ibn Khaldûn 1958] it-
self see e.g. Gellner [1981, pp. 1-85]) which contributed significantly to the Middle
Eastern “involution” in the 11th-18th centuries, and through this to the loss of its for-
mer central part in the world system.

Appendix A
The “Inviting of the kings” in “pre-crisis” Arabia
1 )
Lcimmä tasäfahat Bakr ibn Wä’il wa-ghalaba-hum sufahä’u-hä ... irta’ä ru’sä’и-
hum fa-qälü inna sufahä’a-nä qad ghalabü ‘alä amri-na fa-akala ‘l-qawiyyu ‘l-da‘if,
wa-lä nastati‘“taghyir dhälik, fa-narä an numallik“ ‘alay-nä malika" nu'ü-hi ‘l-shäf
w a-’l-ba'ïr fa-ya’khudh“ li-’l-d a 'ïf mina ‘l-qawiyy, wa-yarudct ‘alä ‘l-mazlüm
mina ‘l-zälim, wa-lä yumkin an yakün min ba'dt qabä’ili-nä fa-ya'bä-hu ‘l-äkharün
... wa-läkin-nä na’tï Tubba‘“‫ ״‬fa-numalliku-hu ‘alay-nä. Fa-ataw-hu fa-dhakarü la-
hu amra-hum, fa-mallaka ‘alay-him al-Härith“ 'bn ‘Amr Äkil al-Murär al-Kindi
“When [the tribal confederation] Bakr b. Wä’il got corrupted and dominated by
shameless people among it..., the leaders of them [i.e. of the confederation tribes]
came to decision and said: ‘We are unable to change this; our opinion is to make
king over us to whom we shall give (as a tax) sheep and camels, and he will protect
the weak from the strong, and the oppressed from the oppressor; it is impossible that
he will be from one of our tribes, as the other (tribes) will not accept him... Rather
we shall come to the tubba‘ [the Himyarite king of Yemen], and make him a king
over us’; so, they came to him and told him about their problem and he made al-
Härith b. ‘Amr Äkil al-Murär al-Kindi king over them” (Ibn ‘Abd Rabbi-hi 1949-
1965, V, p. 222; see also e.g. al-Isfahäni [1955-1964, XII, p. 207]; Ibn al-Athir
[1867, pp. 374-375]).
2)
Lammä tafäsadat‘ ‘l-qabä ’i t min Nizär atä-hu ashräfu-hum fa-qalü in-nä f i dini-ka
wa-nahnu nakhäf an natafänä fi-mä yahduth“ bayna-nä, fa-wajjih ma ‘a-nä bani-ka
yanzilüna fi-nä fa-yakuffüna ba‘da-nä ‘an ba'd fa-farraqa wulda-hu f i qabä’il‘ 7-
‘Arabfa-mallaka ‘bna-hu Hujr" ‘alä BaniAsad wa-Ghatafän, wa-’bna-hu Shurihbil
... ‘alä Bakr b. Wä’il ... wa-mallaka ‘bna-hu Ma'adkarib ... ‘alä Bani Taghlib ...
wa-mallaka ‘bna-hu ‘Abct ‘lläh ‘alä ‘Abd! ‘I-Qays; wa-mallaka ‘bna-hu Salamah
‘alä Qays
“When the tribes of the Nizär [confederation] got corrupted, their nobles came to him
[i.e. to al-Härith b. ‘Amr, reigning in al-Hirah at that moment] and said: ‘We are in
your power (judgement) we are afraid that we shall annihilate each other because of
what is happening between us; send with us your sons so that they will rule us - they
will secure the order among us (lit.: hold some of us back from the others). So he
distributed his offspring among the tribes of the Arabs: he made his son Hujr king
over Banu Asad and Ghatafän, his son Shurihbil ... - over Bakr b. Wä’il ..., his son

Ma'adikarib ... - over Banu Taghlib ...; his son ‘Abd Allah - over ‘Abdu-’l-Qays;
his son Salamah - over Qays” (the beginning of “the Day of Hujr” - al-Isfahäni
1955-1964, IX, pp. 80-81; Ibn al-Athir 1867, p. 376; al-Mawlâ-bik et al. 1942, p. 112;
see also the beginning of “the First Day of al-Kuläb” - al-Isfahäni 1955-1964, XII,
p. 207; Ibn al-Athir 1867, p. 406; Ibn ‘Abd Rabbi-hi 1949-1965, V, p. 223; al-Maw-
lâ-bik et al. 1942, p. 46; see also the beginning of “the Day of Khazäz” in the edition
of Yäqüt [1410/1990, II, pp. 418-119]; al-Mawlâ-bik et al. 1942, pp. 109-110).
Appendix В
End of Arabian chiefdoms
1)
The end of the previous story (“the killing of the kings” taking place in the 530-40s,
i.e. after the beginning of the Arabian socio-ecological crisis):
Mallakat Barn 'Àmir Sharâhîl ibn Ч-Härith wa-mallakat B am Tamim ... Mu-
harriq" ibn Ч-Härith wa-mallakat Wä’il Shurihbïl ibn Ч-Härith ..., käna malik“Ba-
ni Taghlib ... Salamat“ ‘bn Ч-Härith, wa-mallakat baqiyyat“ Qays ... Ma‘adîkarib
ibn Ч-Härith ... [!!!] fa-qatalat Banü Asad Hujr" ... wa-nahadat Banü Àmir ‘alâ
Sharâhîl fa-qatalü-hu ... wa-qatalat Banü Tamïm Muharriq " ... wa-qatalat Wä’il
Shurihbïl... [!!!]
“Banü ‘Àmir made Sharâhîl the son of al-Härith king [over them]; Banü Tamïm ...
made Muharriq the son of al-Hârith king [over them]; Wä’il made Shurihbïl the son
of al-Hârith king [over them] ...; Salamah the son of al-Hârith became king over
Banä Taghlib ...; and the rest of Qays ... made Ma'adikarib the son of al-Hârith king
[over them]... Then Banü Asad killed Hujr ...; Banü ‘Àmir rose up against Sharâhîl
and killed him...; Banü Tamïm killed Muharriq ...; and Wä’il killed Shurihbïl..."
(the end of the introduction to “the Day of Khazäz” in Yaqüt’s edition [1410/1990,
П, pp. 418-419]).
2)
End of an Arabian chiefdom
[Känat] Hawäzin lä tard Zuhayr b. Jazïmah illä rabbm wa-Hawäzin yawma- ’ idh1" lâ
khayr fï-hâ... wa-inna-mä hum ri‘ä ’u Ч-shä’1f l Ч-jibäl... wa-käna Zuhayr ya'shuru-
hum... fa-ta’tï-hi Hawäzin bi-’l-itäwat' ‘llatï känat la-hu f i a'nâqi-him fa-ya’tüna-hu
bi-’l-samn wa-’l-aqit‘ wa-’l-ghanam... atat-hu ‘ajüz ” min Hawäzin bi-samn" f i nahy,
wa-’4adharat ilay-hi wa-shakat' Ч-sinïna ‘llatï tatäiba 41a ‘alä Ч-näs fa-dhâqa-hu fa-
lam yarda ta ‘та-hu fa-da ‘‘а-hä bi-qaus " f i yadi-hi ‘utul"' f i sadri-hä ja- ’stalqat li-

haläwai Ч-qafä... fa-ghadabat min dhälika Hawäzin wa-haqadat ‘alay-hi ilä met
känafi sadri-hä mina ‘l-ghayz wa-’l-diman... wa-tadhämarat ‘Ä mir b. Sa‘sa ‘ah wa-
cilà Khâlid b. Ja fa r fa-qâl wa-'llah1la-aj‘alanna dhirä ‘ayya ward’a ‘unuqi-hi hattà
uqtaf aw yu q ta f...
wa-qâla Khâlid b. J a fa r yamunn ‘alä Hawäzin bi-qatli-hi Zuhayr"...:
bal kayfa takfuru-ni Hawäzin ba ‘da-mâ
a ‘taqtu-hum fa-tawäladü ahrär"
wa-qataltu rabba-hum Zuhayr" ba ‘da-mâ
jada'a ‘l-unüf wa-akthara ‘l-awzärä
wa-ja ‘altu hazin bilàdi-him wa-jibali-him
ara fada sahlat wa- ishar
“The Hawäzin [confederation] considered Zuhayr b. Jazïmah [their] lord. That time
they lived in poverty,... they grazed [their] goats in the mountains... Zuhayr was
taxing them with ‘ushr [a 10% tax]... The Hawäzin people brought to him the tribute
which he imposed on them, they brought to him melted butter, cheese, sheep... Once
an old Hawäzin women brought him a skinful of melted butter, apologised to him
[for the modesty o f her tribute], and complaint o f the [ “lean”] years which struck
[her] people (The emphasis is ours - A. К., V. K., D. P.). He tried the butter, did not
like its taste and pushed her with a stringless bow which he held in his hand - she
fell flat on her back... The Hawäzin got angry because of that, bore rancour against
him; this added to spite and rancour which they nursed against him. And [the tribe]
‘Àmir b. Sa‘sa‘ah [one of the Hawäzin tribes] began to call for war, and Khâlid b.
Ja‘far swore: I swear in the name of the God that my hands will reach his neck -
either I shall die, or he will...” [Finally [not without lots of trouble] Khâlid and his
fellow tribesmen manage to kill Zuhayr and destroy his chiefdom.] “.. .And Khâlid b.
Ja‘far said, boasting in front of the Hawäzin of killing Zuhayr...
Well, how may the Hawäzin not recognize me now after
I brought free life for them - they have been reborn as free people!
I have killed their tyrant Zuhayr after the time when
He cut off [their] noses and increased [their] burdens,
I have transformed their land which was cramped and rugged
Into an open plain full of camels!...”
(al-Isfahänl 1955-1964, XI, pp. 77-83; see also Ibn al-Athir 1867, pp. 411-414; Ibn
‘Abd Rabbi-hi 1949-1965, V, pp. 135-137; al-Mawlä-bik et al. 1942, pp. 235-241
etc.).

Appendix C
Some environmental backgrounds of the origins of Islam
Lately historical research has been coming into more and more close contact with
natural sciences. Systematic socio-natural study of civilisations is being carried on,
often allowing us to considerably develop our ideas about the historical process. We
don’t think that research of this kind is always quite necessary, but we are sure that
when we deal with such problems as the genesis of civilisations, the socio-natural
approach is the most objective of all the feasible ones. To sufficiently understand the
factors and the course of the social evolution we need to be aware as fully as possible
of the environmental conditions of the social development.
The latest Russian interdisciplinary socio-environmental research in general
history of civilisations (Klimenko 1998), as well as of separate cultures, e.g. China
(Kul’pin 1990) and Egypt (Proussakov 1996), has shown that crucial historical ages
are in close connection with natural processes and sometimes also with catastrophic
natural impacts. All this is utterly relevant for the genesis of the Islamic civilisation,
as its birth was preceded by a series of disastrous environmental phenomena both in
the Mediterranean region and in the Near East, as well as in the rest of the world.
Here we are going to discuss those phenomena, and though this will be rather of geo-
physical than historical pattern, it bears a direct relation to the origin and early his-
tory of Islam, because the Arabian society could not fail to feel consequences of the
socio-ecological crisis of the 6th century AD.
For those who deal with socio-natural problems, the 6th century AD is quite
an outstanding epoch. First of all, the middle of the first millennium AD is character-
ised by a considerable global cooling which culminated probably in the second half
of the 6th century (Klimenko 1997). At that time mean global temperatures were
about 0.7-0.9 °C lower than now, i.e. close to the temperature level corresponding to
the so-called “Little Ice Age” (between the late 16th and late 19th centuries). Evi-
dence of this cooling can be found in various parts of the world: due to glacier
advances in the Alps, Himalaya/Karakorum, Scandinavia, North and South Ameri-
cas, New Zealand (Röthlisberger 1986; Grove 1988); dendrochronological observa-
tions of tree-ring widths and densities (Briffa et al. 1990; Scuderi 1993), palynologi-
cal reconstructions of climate based on fossil pollen residuals (Klimenko et al. 1996);
marine microfauna data, received through the analysis of deep-sea cores (Juillet-
Leclerc and Schrader 1987); and historical documentary sources (Hennig 1904;
Easton 1928).
Among those listed, there are two methods of past climates’ reconstruction
which have the best temporal resolution - close to one year - namely, dendrochro-
nological and historical. Let us have a look what these two methods tell us about the
climate of the middle of the first millennium. Dendrochronological evidence shows
that the years 535, 536, 541 and 685 were the coolest, and the time span 542-561
was the coolest bidecadal period of the past 2000 years (Scuderi 1993). It should be

noted, however, that dendrochronological data provide reliable information concern-

ing only a warm part of the year, covering typically April through September. For
the information concerning a cold part of the year one should consult the results
provided by historical climatology (see e.g. Easton 1928; Weikinn 1958; Chu Ko-
chen 1973). Historical evidence shows that there were several unusually cold winters
in Europe and Byzantine in the 6th and early 7th centuries - in 508, 509 AD
(possibly these two dates refer to the same event), 524, 545, 548, 554, 559, 565, 566,
568 (possibly these three dates refer to the same event), 609, 620 and 642 (Easton
1928; Weikinn 1958; Borisenkov and Pasetskij 1988; Camuffo 1987); Chinese re-
cords also confirm that the 6th century was one of the coldest and driest in the first
millennium AD (Chu Ko-chen 1973, p. 252; Zhang Deer 1984, p. 826; Fang and Liu
1992, p. 162). Later from around the middle of the 7th century the number of reports
about cold winters and droughts sharply decreases. Therefore, there is a lot of direct
evidence that the 6th and early 7th centuries have witnessed a considerable global
cooling which affected undoubtedly all parts of the globe including Middle East and
adjacent regions.
To one of the most important regional manifestations of global climate change
one can attribute the resumption of the intensive monsoon circulation. As was shown
by Fairbridge (1986), between 1400 BC and 500 AD the catastrophic collapse of
regular monsoonal circulation above the North-Western part of the Indian Ocean
took place which was followed by a general desiccation of the savannas of North-
Western India and semi-deserts of South Arabia. After 500 AD the Indian Ocean
monsoon recovered, although since that time it has never reached the intensity
recorded in the early and middle Holocene. Nevertheless, one can assume that in the
course of the 6th-7th centuries AD there was a sudden increase of summer précipita-
tion in South Arabia. It may well be that it contributed to the final destruction of the
famous Märib Dam through a catastrophic flood. There is also another piece of indi-
rect evidence pointing to the reappearance of intensive monsoon rains - the increase
of the Nile discharge up to the levels which were by 30^10% higher than the modem
ones (Fairbridge 1976, p. 535).
One can assume that the Northern part of the Arabian Desert also became
wetter in the middle of the first millennium AD. Palaeoreconstruction data show that,
starting from about the late 6th century, the Dead Sea level, which was extremely
low in the preceding two millennia (or more), experienced a rapid rise (Varushenko
et al. 1987, p. 149).
On the other hand, based on palaeoreconstructions and contemporary instru-
mental data, one can assume that in the middle of the first millennium AD climate of
the major part of the Arabian Peninsula (excluding its northwestern and southern
portions discussed above) was drier than at present (for details see Klimenko [1998]
and references herein).
Apart from massive climatic changes, the 6th and 7th centuries AD were also
a period of large worldwide tectonic and volcanic activities. In the history of the
Mediterranean region, we may compare it, perhaps, only with the tectonic catastro-
phe of the middle of the 2nd millennium BC, which was crowned with the greatest

Santorini eruption in the Aegean Sea (Ninkovich and Heezen 1965; Watkins et al.
1978) that seems to have become fatal for the Minoan Civilisation (Marinatos 1939;
Galanopoulos 1960; Duginov and Strekalov 1997). To understand properly the na-
ture of these phenomena it is important to note that both ecological disasters in ques-
tion, which had the most substantial climatic effect among all the explosive eruptions
of the second half of the Holocene (Borzenkova 1992, p. 203; Stothers and Rampino
1983a, p. 412, tab. 1), correlate closely with minimums of the Northern Hemisphere
temperature curve for the past 4000 years (Klimenko 1997, p. 400, fig. 1). This latest
curve was recently obtained in the Moscow Energy Institute by taking into account
the superposition of three major long-term climate cycles: 2400-yr, 1100-yr and 850-
yr connected probably to solar and planetary periodicities (Klimenko 1997). This
means that the ecological crisis of the 6th century AD was not accidental, and it may
be actually explained from the standpoint of global natural laws, which mould the
appearance of the Earth.
The problem of a relationship between volcanism and solar cycles has been
quite recently disputable (Stothers 1989, p. 17371) because of the lack of data (New-
hall and Self 1982, p. 1235). But after extensive catalogues of the world volcanoes
(Gushchenko 1979; Simkin et al. 1981) as well as of solar activity (Schove 1983)
were published in the late 70s and the early 80s, it became possible to establish a cor-
relation between peaks of volcanism and extremums of solar activity. Moreover, it
turned out that eruptions tend to concentrate first of all near the times of solar mini-
mums, though a few large eruptions may also occur close to the years of solar maxi-
mum (Stothers 1989, pp. 17378-17379). The latest research by Russian scientists has
shown that the later part of the 6th century AD was characterised by decreasing solar
activity (Mikushina et al. 1997, p. 321, fig. 5) after a strong peak around 532 AD. It
may be that this peak which overlaps with great minimum of the superposed climatic
cycles has played its part in the development of the global tectonic and volcanic
disaster of the 6th century.
The mechanism of solar-volcanic relationships may be described briefly as
follows. According to one of the hypotheses, in the scale of thousand-year solar
cycles, increase in the solar activity is accompanied by melting of the polar ice caps
and continental glaciers and, correspondingly, by the eustatic rise of the World
Ocean level; as a result, the angular velocity of the Earth slows down; the centrifugal
forces decrease; the globe becomes less elliptical. On the contrary, lowering of the
solar radiation and growing of the ice caps result in increase of the angular velocity,
and the Earth becomes more “flattened” along the terrestrial axis. Such a pulsation of
the Globe along its axis triggers in the lithosphere the processes towards restoration
of the gravitational equilibrium: the planet figure adjusts to new rotational condi-
tions. During such “adjustments” numerous earthquakes and volcanic eruptions take
place on the Earth (Reznichenko-Reznichenko 1964, pp. 173-175). In addition to this,
minor solar variations cause changes in atmospheric circulation, and these changes
induce additional crustal stresses (Rampino et al. 1979, p. 828; Stothers 1989,
p. 17378). This is the geophysical model that might explain the tectonic phenomenon
of the 6th century.

Nowadays the most reliable method of reconstruction of the Holocene volcan-

ism is through the detection in ancient glaciers of volcanic, first of all sulphuric, acid
which is stored in the ice layers as a result of deposition of acid aerosols shortly
(from 2 [Zielinski et al. 1994, p. 948] to 3-4 or even 5-6 years) after eruptions (Asa-
turov et al. 1986, p. 181). Rich sources of this information for the Northern Hemi-
sphere are the ice cores drilled in the Greenland ice sheet; these cores are distinct,
well dated acidity records. Nowadays several cores are at researchers’ disposal.
Thus, the core from Camp Century, North-West Greenland, 1390 m long (Johnsen et
al. 1970; Johnsen et al. 1972), and the core from Crête, central Greenland, 404 m
long (Hammer 1977, pp. 483-486; Hammer et al. 1980) both cover 10,000 years,
giving us a clear picture of the Holocene volcanic history. A peak of acidity in the
second core corresponds to 553 AD, but unfortunately there is a 500-year gap in the
data of the two cores including the first half of the 6th century: 40 to 540 AD (Ham-
mer et al. 1980, p. 230, fig. 1, p. 231, tab. 1; Stothers-Rampino 1983a, pp. 411-412).
One of the latest, the so-called GISP2 Greenland ice core, providing a record of
volcanism since 7000 BC, due to drilling problems tells us almost nothing about the
6th century (Mayewski et al. 1993a; Zielinski et al. 1994, p. 949, tab. 1). Antarctic
cores are silent also, because they are not deep enough (Legrand and Delmas 1987;
Delmas et al. 1992). The lack of information is supplemented either by special mod-
em volcanological research (Kelly-Sear 1984) or by original sources (direct or indi-
rect atmospheric observations) (Stothers-Rampino 1983a; 1983b), which corroborate
the statement of high world volcanic activity in this period. For example, the year
536 AD was marked by Vesuvio explosion (Simkin et al. 1981, p. 112). Besides this,
the famous historical event of the “dry fog” and other atmospheric phenomena in
Europe and Middle East in 536-537 AD (Stothers 1984) were caused by an unknown
eruption twice as large as that of Tambora in 1815 AD; some scientists believe it to
be the explosion of Rabaul volcano, on the New Britain Island near New Guinea
(Palais-Sigurdsson 1989, pp. 41-42), which took place 1450±60 radiocarbon years
BP (Heming 1974, p. 1258); according to another theory, it might have been a vol-
cano eruption at the upper reaches of the Yukon river, Alaska (Macdonald 1975,
p. 141), preliminary dated to 540±10 AD, of estimated global acid fall-out of 70
million tons (Hammer et al. 1980, p. 235); besides, it is not unlikely that both these
explosions took place almost simultaneously. The information about large explosive
volcanic activities between 533 and 537 AD is confirmed by data from the Dye 3
Greenland deep ice core (Herron 1982, p. 3055). According to recent Russian re-
search (Brajtseva et al. 1997) another giant eruption took place in 606 AD from the
Opala volcano in Kamchatka - this one was the most violent Kamchatka eruption
during the past 1500 years and could match the most powerful history in modem
Tambora eruption (1815 AD) which had long-lasting atmospheric and climatic con-
sequences.
Volcanic explosions inject great volumes of silicate dust and aerosol-forming
sulphur gases into the troposphere and stratosphere; while the fine ash gravitationally
settles down within a few weeks or months, aerosols are spread globally with strato-
spheric winds and suspend in the stratosphere for several years altering the planet

heat balance and, therefore, climate (Hansen et al. 1978; Kelly-Sear 1984, p. 740;
Rampino-Self 1982; 1984). The effect of this phenomenon on the climate is twofold.
During the first few months after an eruption, when aerosol and ash particles are
rather large, a warming of the Earth’s surface takes place due to the absorption of re-
fleeted terrestrial radiation by these particles (Pollack et al. 1976). But “large” grains
soon decrease or are washed down by rains, and the net effect of the aerosol injection
into the stratosphere cools the Earth’s surface because of high reflectivity of sul-
phuric acid to solar radiation (Baldwin et al. 1976; Hansen et al. 1978, p. 1066). It is
known that the Krakatau explosion in 1883 AD caused a 20% reduction of the at-
mospheric transmissivity in the Northern Hemisphere (Hammer 1977, p. 482). The
aerosol distribution after large eruptions is of global or hemispheric scale (Hammer
1977, p. 482; Stothers-Rampino 1983a, p. 411; Zielinski et al. 1994, p. 948) and, in
addition, rather rapid (Hansen et al. 1978, p. 1065): thus, for example, if the volcanic
explosion of 536 AD mentioned above is that of Rabaul near New Guinea, it would
take not more than 2 or 3 weeks for the stratospheric aerosols to reach the Mediterra-
nean region; as for the territorial scale, consequences of this eruption were described
by contemporary authors from Rome, Constantinople and Mesopotamia; for exam-
pie, the latter wrote: “The Sun was dark and its darkness lasted for eighteen months;
each day it shone for about four hours, and still this light was only a feeble shadow
... the fruit did not ripen and the wine tasted like sour grapes” (John of Ephesus,
quoted from: Stothers 1984; Stothers-Rampino 1983b); the winter of 536-537 AD
in Europe and the Middle East was extraordinarily severe, with heavy snowfalls in
Mesopotamia (Stothers 1984; Stothers-Rampino 1983b, pp. 6362-6363). Decrease
of zonal mean temperature is observed from 1-3 to 5-7 years after eruptions (Spirina
1975). Thus, the Tambora event in 1815 AD was followed by the Northern Hemi-
sphere temperature decrease within 0.5-1.0 °C for 2-3 years; 1816 AD is famous as
the “year without summer” in the North-East of the USA and in the North-West of
Europe (Stommel-Stommel 1979). Even more dangerous from this standpoint are
the so-called clustered eruptions that may cause cooling from 10 to 100 years. Influ-
ence of volcanoes on climate is so considerable that the information about volcanic
eruptions was even used as a basis for estimating possible climatic effect of a nuclear
war in various nuclear winter models (Rampino-Stothers 1985). The fact is estab-
lished that massive volcanic eruptions triggered initially the great Pleistocene ice
sheets (Bray 1974; 1976; Rampino-Self 1992; Zielinski et al. 1996). A prolong term
of volcanism is considered to be an actual cause of the so-called Younger Dryas -
the period of near-glacial conditions which began 12-13 thousand years BP and
lasted for 1300±70 years (Alley et al. 1993; Mayewski et al. 1993b). As for the so-
cial history, it is of great importance that cooling of the Earth’s surface in all the
tropical and some of the subtropical latitudes (including the Arabian Peninsula) re-
suits in droughts (Borzenkova 1992, pp. 162-174); droughts lasting for several years
can sometimes lead to significant social transformations. Besides, warming and cool-
ing after explosive eruptions, being abrupt climatic shifts, may disbalance the bio-
sphere and trigger endemic hotbeds of plague or other diseases (Bogolepov 1912,

p. 21); for example, cooling in tropical latitudes creates favourable living conditions
for fleas - carriers of plague (Cavanaugh 1971).
The ancient historical sources do not definitely mention volcanic eruptions in
Arabia proper, but according to modern natural studies the influence of volcanism
upon it is beyond doubt: it is conditioned both by geological structure and geographi-
cal location of the Arabian Peninsula. On the one hand, the Red Sea and the Ocean
coasts of Arabia is a part of the seismic East-African Rift System (Belousov et al.
1974); thus, Western and Southern Arabia is situated in a volcanic zone (Belousov et
al. 1974, I, p. 41; Koronovskij 1971; Macdonald 1975, p. 327, fig. 14-1). On the
other hand, Arabia could feel the force of tectonism and volcanism of the Mediterra-
nean region and the rest of the world, e.g. the consequence of earthquakes and injec-
tions of volcanic aerosols into the atmosphere. It is also important to note that there
is natural hotbed of plague in Arabia (Kozlov-Sultanov 1993, pp. 201-203) and that
earthquakes and plague epidemics often coincide (Biraben 1976, pp. 16-17).
In the 6th century AD the Mediterranean region has been going through one
of the heaviest tectonic disasters in its recent history. According to historical data
(Gezer 1867), it began in 512 AD with the Vesuvio eruption (see also Simkin et al.
[1981, p. 112]). In 526 AD there was a great earthquake in Syria, where 250 thou-
sand people perished only in Antioch; it was followed by heavy earthquakes in Asia
Minor and Europe. In Egypt in summer 547 the highest Nile flooded the Delta and
prevented the sowing; as a result, the whole country suffered from famine. In 551
a new wave of earthquakes rolled from Constantinople to Alexandria. Earthquakes
and volcanism were accompanied by inundations, droughts and widest spread of
plague - the so-called Justinianos pandemia, 531-580 AD, during which about sev-
eral million people perished (Ostrovskij 1978, p. 258). It is important that volcanic
activity at the same time was characteristic also for the Arabian Peninsula proper: in
the 6th-7th centuries AD there were eruptions of at least 5 volcanoes in Arabia
according to the recent volcanological research (Simkin et al. 1981, p. 112; Aprodov
1982, pp. 269-270; Gushchenko 1979, pp. 12-13).
Taking the mechanism of interaction of processes in the atmosphere, hydro-
sphere and lithosphere, as well as after-effects of seismic disasters (droughts, plague
pandemics etc.) into consideration, we have certain grounds to consider the 6th-7th
centuries AD as one of the ecologically most crucial periods in the history of the
Middle East including the Arabian Peninsula. The scale of the ecological catastrophe
under consideration forces us to bear it in mind while studying the early history of Is-
lamic Arabia. We think that a socio-ecological rather than a purely social crisis in the
6th century AD caused the disintegration of most Arabian kingdoms and chiefdoms
and, consequently, the development of some alternative cultural-political structures
which provided the optimum environment for the development of the autochthonous
Arabian monotheist religion.
As volcanologists say, “empirical studies concerning volcanic effects on cli-
mate tend to raise as many, if not more, questions than they answer” (Kelly-Sear
1984, p. 742). But we think the more questions we have, the more interesting our
further work will be. In this paper we tried to raise an important problem of socio-

natural evolution in Arabia at the age of its transition to the Islamic civilisation. We
hope that this approach will demonstrate its value in future research.
Acknowledgements
The research has been supported by the Universities of Russia.

Our thanks also go to Professor Patricia Crone of the Institute for Advanced
Study in Princeton for her very useful comments on the draft of this paper and to our
colleague Leonid Kogan of the Russian State University for the Humanities, Mos-
cow, for his invaluable advice concerning the interpretation of the Aramaic and He-
brew terms.
References
Ahmadjonzoda, A. (1988), O problème tak nazyvaemoj musul’manskoj kul’tury. Vzaimodejstvie i

vzaimovlijanie tsivilizatsij i kul’tur na Vostoke. Moskva, pp. 3-4.
Alley, R .-M eese, D.-Shuman, C .-G ow , A.-Taylor, K.-Grootes, Р.-W hite, J.-Ram, M.-Wadding-
ton, E.-M ayewski, P.-Zielinski, G. (1993), Abrupt Increase in Greenland Snow Accumula-
tion at the End of the Younger Dryas Event. Nature 362, pp. 527-529.
Aprodov, V. A. (1982), Vulkany. Moskva.
Asaturov, M. L.-Budyko, M. I.-Vinnikov, K. Ja.-Grojsman, P. Ja.-Kabanov, A. S.-K arol’, I. L .-
Kolomeev, M. P.-Pivovarova, Z. I.-Rozanov, E. V.-Hm elevtsov, S. S. (1986), Vulkany,
stratosfernyj aérozoV i klimat Zemli. Leningrad.
Audouin, R.-Breton, J.-F.-Robin, C. (1988), Towns and Temples - the Emergence o f South Ara-
bian Civilization. In: Yemen: 3000 Years of Art and Civilization in Arabia Felix. Inns-
bruck-Frankfurt/Main, pp. 63-77.
al-Azraqi, Muhammad b. ‘Abdallah (1858), Kitâb akhbâr Makkah. Leipzig.
Balädhüri, Ahmad b. Yahyä (1866), Futüh al-buldän. Lugduni Batavorum [Leiden],
Baldwin, B.-Pollack, J.-Summers, A .-T oon, O.-Sagan, С.-V an Camp, W. (1976), Stratospheric
Aerosols and Climate Change. Nature 263, pp. 551-555.
Bartol’d, V. V. (1925), Musejlima. Izvestija Rossijskoj Akademii nauk. Serija V I 19, pp. 483-512.
Bauer, G. M.-Lundin, A. G. (1994), Juzhnaja Aravija v X veke do n.é.-V I veke n.é.: osnovnye
étapy istorii. Krasnomorskie zametki. Vol. 1. Moskva, pp. 86-117.
al-Bayâtï, ‘Sdil Jàsim (1407/1987), Al-malâhim al-'arabiyyah: Muqâranât f i daw ’ al-asâtir wa-’l-
malàhim al-kawniyyah ai-sumiriyyah wa-’l-Âmiyyah. Bayrüt.
Beeston, A. F. L. (1984a), The Religions o f pre-Islamic Yemen. In: L ’Arabie du Sud: histoire et
civilisation. 1. Le peuple yéménite et ses racines (Islam d ’hier et d’aujourd’hui, 21), Paris,
pp. 259-269.
Beeston, A. F. L. (1984b), Judaism and Christianity in pre-Islamic Yemen. In: L ’Arabie du Sud:
histoire et civilization. 1. Le peuple yéménite et ses racines (Islam d’hier et d’aujourd’hui,
21), Paris, pp. 271-278.
Beeston, A. F. L. (1984c), Himyarite Monotheism. In: Studies in the History o f Arabia. IL Pre-
Islamic Arabia. Riyad, pp. 149-154.
Beljaev, E. A. (1965), Araby, islam i arabskij halifat v rannee srednevekov’e. Moskva.

Belousov, V. V.-Gerasimovskij, V. I.-Gorjachev, A. V .-D obrovol’skij, V. V.-Kapitsa, A. P.-Lo-

gachev, N. A .-M ilanovskij, E. E.-Poljakov, A. I.-Rykunov, L. N .-Sedov, V. V. (1974),
Vostochno-Afrikanskaja riftovaja sistema. T. 1-3. Moskva.
Berzina, S. Ja.-Kubbel’, L. E. (eds) (1990), Istorija Afriki v drevnih i srednevekovyh istochnikah.
Hrestomatija. M oskva.
Biraben, J.-N. (1976), Les hommes et la peste en France et dans les pays européens et méditerra-
néens. T. 2. Les hommes face à la peste. Paris-La Haye.
de Blois, F. (1990), The Date o f the “Martyrs” o f Nagrim. Arabian Archaeology and Epigraphy 1,
pp. 110-128.
Bogolepov, M. A. (1912), Kolebanija klimata i istoricheskaja zhizn’ (golod i vojna). Moskva.
B o l’shakov, O. G. (1989), Istorija Halifata. I. Islam vAravii (570-633). Moskva.
Borisenkov, E. P.-Pasetskij, V. M. (1988), Tysjacheletnjaja letopis’ neobichajnyh javlenij prirody.
Moskva.
Borzenkova, 1.1. (1992), Izmenenie klimata v kajnozoe. S-Peterburg.
Bowen, R. L. (1958), Ancient Trade Routes in South Arabia. In: Archaeological Discoveries in
South Arabia. (Publications o f the American Foundation for the Study o f Man, II), Balti-
more, pp. 3 5 4 2 ‫־‬.
Brajtseva, O. A.-Sulerzhitskij, L. D.-Ponomareva, V. V .-M elekestsev, I. V. (1997), Geohronolo-
gija krupnejshih éksplozivnyh izverzhenij Kamchatka v golotsene i ih otrazhenie v gren-
landskom lednikovom shchite. Doklady Akademii nauk 352:4, pp. 516-518.
Bray, I. (1974), Volcanism and Glaciation during the Past 40 Millennia. Nature 252, pp. 679-680.
Bray, J. (1976), Volcanic Triggering o f Glaciation. Nature 260, pp. 414-415.
Briffa, K. R.-Bartholin, T. S.-Eckstein, D.-Jones, P. D.-Karlén, W.-Schweingruber, F. H.-Zetter-
berg, P. (1990), A 1,400-year Tree-Ring Record o f Summer Temperatures in Fennoscan-
dia. Nature 346, pp. 434-439.
Camuffo, D. (1987), Freezing o f the Venetian Lagoon since the 9th Century AD in Comparison to
the Climate o f Western Europe and England. Climatic Change 10:1, pp. 43-66.
[Carpenter, E., ed.] (1869), Martirium sancti Arethae et Sociorum in civatate Negran. Acta Sancto-
rum 10, pp. 7 2 1 -7 5 9 .
Cavanaugh, D. (1971), Specific Effect o f Temperature upon Transmission of the Plague Bacillus
by the Oriental Rat Flea X. Cheopis. American Journal o f Tropical Medicine and Hygiene
20, pp. 2 6 4 2 7 4 ‫ ־‬.
Chu Ko-chen (1973), A Preliminary Study on the Climatic Fluctuations during the Last 5,000
Years in China. Scientia Sinica 16:2, pp. 226-256.
CIAS (1977), Corpus des inscriptions et antiquités Sud Arabes. T .l. Louvain.
Claessen, H. J. M .-Skalnik, P. (eds) (1978), The Early State. The Hague-Paris-New York.
Claessen, H. J. М.-v a n de Velde, Р.-Sm ith, M. E. (eds) (1985), Development and Decline: The
Evolution of Sociopolitical Organization. South Hadley, MA.
Crone, P. (1987), Meccan Trade and the Rise of Islam. Oxford.
al-Dabbi, Mufaddal ubn Muhammad. (1964), Al-Mufaddaliyyät. (Diwan al-‘Arab. Majmü'ät min
‘иуйп al-shi‘r, 1), al-Qähirah.
Delmas, R. J.-Kirchner, S.-Palais, J. М.-Petit, J. R. (1992), 1000 Years o f Explosive Volcanism
Recorded at the South Pole. Tellus (Series В) 44:4, pp. 335-350.
D oe, D. В. (1971), Southern Arabia. London.
Donner, F. M. (1977), M ecca’s Food Supplies and Muhammad’s Boycott. Journal of the Economic
and Social History o f the Orient 20, pp. 249-266.
Dostal, W. (1991), M ecca before the Time o f the Prophet - Attempt o f an Anthropological Inter-
pretation. Der Islam 68, pp. 193-231.

Dresch, P. (1989), Tribes, Government, and History in Yemen. Oxford.

Duginov, B. A.-Strekalov, S. S. (1997), Katastroficheskoe tsunami na poberezh’e ostrova Krit
i v Egipte v X IV -X V vekah do nashej éry. BUT о pravah cheloveka i prirody. Moskva, pp.
142-151.
Easton, C. (1928), Les Hivers dans L ’Europe Occidentale. Leiden.
Fahd, T. (1966), La divination arabe. Leiden.
Fairbridge, R. W. (1976), Effects o f Holocene Climatic Change on Some Tropical Geomorphic
Processes. Quaternary Research 6:5, pp. 529-556.
Fairbridge, R. W. (1986), Monsoons and Paleomonsoons. Episodes 9:3, pp. 143-149.
Fang, Jin-Qi-Liu, Guo (1992), Relationship between Climatic Change and Nomadic Southward
Migrations in Eastern Asia during Historical Times. Climatic Change 22:2, pp. 151-169.
F il’shtinskij, I. M. (1977), Arabskaja literatura v srednie veka. Moskva.
Fitzmyer, J. A.-Harrington, D. J. (1978), A Manual o f Palestinian Aramaic Texts. Rome.
Fried, M. H. (1967), The Evolution o f Political Society: An Essay in Political Anthropology. New
York.
Gajda, I. (1996), Hujr b. ‘Amr roi de Kinda et l’etablissement de la domination himyarite en Ara-
bie centrale. Proceedings of the Seminar for Arabian Studies 26, pp. 65-73, pi. I.
Galanopoulos, A. (1960), Tsunami Observed on the Coasts o f Greece from Antiquity to Present
Time. Annali Geofisici 13.
Gellner, E. (1981), Muslim Society. Cambridge-New York, NY-Melbourne.
Gezer, G. (1867), Istorija poval’nyh boleznej. S-Peterburg.
Grjaznevich, P. A. (1994), Istoriko-arheologicheskie pamjatniki drevnego i srednevekovogo Jeme-
na. Polevye issledovanija 1970-1971 gg. (Juzhnaja Aravija. Pamjatniki drevnej istorii
i kul’tury, 2/1), S‫־‬Peterburg.
Grove, J. M. (1988), The Little Ice Age. London-New York.
Gushchenko, 1.1. (1979), Izverzhenija vulkanov mira (katalog). Moskva.
Hallpike, C. R. (1986), The Principles o f Social Evolution. Oxford.
Halperin, D. J. (1976), The Ibn Sayyäd Tradition and the Legend o f al-Dajjal. Journal o f the
American Oriental Society 96, pp. 129-141.
Hammer, C. (1977), Past Volcanism Revealed by Greenland Ice Sheet Impurities. Nature 270, pp.
482-486.
Hammer, C.-Clausen, H.-Dansgaard, W. (1980), Greenland Ice Sheet Evidence o f Post-Glacial
Volcanism and its Climatic Impact. Nature 288, pp. 230-235.
Hansen, J.-Wang, W .-C.-Lacis, A. (1978), Mount Agung Eruption Provides Tests of a Global Cli-
matic Perturbation. Science 199, pp. 1065-1068.
Hennig, R. (1904), Katalog bemerkenswerter Witterungsereignisse von den ältesten Zeiten bis zum
Jahre 1800. (Abhandlungen des Königlichen Preußischen Meteorologischen Instituts, 2/4).
Berlin.
Heming, R. (1974), Geology and Petrology o f Rabaul Caldera, Papua New Guinea. Geological
Society o f America Bulletin 85, pp. 1253-1264.
Herron, M. (1982), Impurity Sources o f F (-), C l(-), N 0 3 (-) and S 0 4 (2 -) in Greenland and
Antarctic Precipitation. Journal of Geophysical Research 87, C4, pp. 3052-3060.
Hoftijzer, J.-Jongeling, K. (1995), Dictionary of the North-West Semitic Inscriptions. Leiden-New
York-Köln.
Ibn ‘Abd Rabbi-hi, Ahmad Muhammad al-Andalusi (1949-1965), Kitäb a l-‘iqd al-farid. 7 vols.
Al-Qähirah.
Ibn al-Abras, ‘Abid (1983), Indhar ilâ Imru’-’l-Qays. Al-rawä’i'min al-adab al-'Arabiyy. 1. A l-‘asr
al-Jahiliyy. Al-Qähirah, pp. 162-165.

Ibn al-Athir, Abü-T-Husayn ‘All (1867), Kitäb al-kämil f i ‘l-ta’rikh. Al-juz’ al-awwal. Lugduni
Batavorum [Leiden],
Ibn Habib, Muhammad (1942), Kitäb al-muhabbar. Hyderabad al-Dakkan.
Ibn Hischäm, ‘Abd al-Malik (1858-1860), Das Leben Muhammeds nach Muhammed Ibn Ishäk.
2 vols. Göttingen.
Ibn Khaldün, ‘Abd al-Rahmän ibn Muhammad (1415/1995), Muqaddimah. Saydà-Baymt.
Ibn Khaldün (1958), The Muqaddimah. London.
Ibn Kulthüm, ‘Ami1983) ‫)־‬, A l‫־‬Mu‘allaqah. In: Al-rawä’i' min al-adab a l-‘Arabiyy. 1. A l-‘asr al-
Jähiliyy. Al-Qähirah, pp. 162-165.
Ibn Saad (1905), Biographien Muhammedes. Bd. 1/1. Leiden.
Irvine, A. K. (1973), The Arabs and the Ethiopians. Peoples of the Old Testament Times. Oxford,
pp. 287-311.
al-Isfahânî, Abû-T-Faraj (1955-1964), Kitäb al-Aghânï. Vols I-XX V. Bayrüt.
Al-Ja‘qubi (al-Ya'qübi), Ahmad b. Abî Y a‘qüb (1883), Historia. 2 vols, Leiden.
Johnsen, S.-Dansgaard, W .-Clausen, H.-Langway C., Jr. (1970), Climatic Oscillations 1200-2000
AD. Nature 227, pp. 482-483.
Johnsen, S.-Dansgaard, W .-Clausen, H.-Langway, C., Jr. (1972), Oxygen Isotope Profiles through
the Antarctic and Greenland Ice Sheets. Nature 235, pp. 429-434.
Juillet-Leclerc, A.-Schrader, H. (1987), Variations o f Upwelling Intensity Recorded in Varved Sedi-
ment from the G ulf de California during the Past 3,000 Years. Nature 329, pp. 146-148.
Kelly, Р.-Sear, C. (1984), Climatic Impact o f Explosive Volcanic Eruptions. Nature 311, pp. 7 4 0 -
743.
Kennedy, J. (1898), The Early Commerce o f Babylon with India. Journal o f the Royal Asiatic So-
ciety, pp. 241-273.
Kister, M. J. (1965), M ecca and Tamim (Aspects o f Their Relations). Journal o f the Economic and
Social History o f the Orient 8, pp. 117-163.
Kister, M. J. (1986), M ecca and the Tribes o f Arabia. Studies in Islamic History and Civilization in
Honour o f David Ayalon. Jerusalem-Leiden, pp. 33-57.
Kitchen K. A. (1994), Documentation for Ancient Arabia. Pt. 1. Chronological Framework and
Historical Sources. Liverpool.
Klimenko, V. V. (1997), О glavnyh klimaticheskih ritmah golotsena. Doklady Akademii nauk
357:3, pp. 399^102.
Klimenko, V. V. (1998), Klimat i istorija v épohu pervyh vysokih kul’tur (3500-500 do h.é.).
Vostok no. 4, pp. 5 -2 4 .
Klimenko, V. V.-Klim anov, V. A.-Fedorov, M. V. (1996), Istorija srednej temperatury Sevemogo
polusharija za poslednie 11000 let. Doklady Akademii nauk 348:1, pp. 111-114.
Klimenko, V. V.-Klim enko, A. V.-Andrejchenko, T. N.-Dovgaljuk, V. V.-Mikushina, O. V .-T e-
reshin, A. G .-Fedorov, M. V. (1997), Energija, priroda i klimat. Moskva.
Kobishchanov, Ju. M. (1980), Severo-Vostochnaja Afrika v rannesrednevekovom mire (VI-seredi-
na VII vv.). Moskva.
Koronovskij, N. V. (1971), Molodoj vulkanizm Éfiopii i Jemena. Vestnik Moskovskogo université-
ta, se г. IV (geologija) по. 3, pp. 52-53.
Korotayev, A. V. (1995), Ancient Yemen. Oxford.
Korotayev, A. V. (1996a), Pre-Islamic Yemen. Wiesbaden.
Korotayev, A. V. (1996b), “Aramaeans” in a Late Sabaic Inscription. ARAM 8, pp. 293-298.
Kozlov, M. P.-Sultanov, G. V. (1993), Epidemicheskie projavlenija chumy v proshlom i nastoja-
shchem. Mahachkala.
Kul’pin, É. S. (1990), Chelovek i priroda v Kitae. Moskva.

Lammens, H. (1910), La république marchande de la Mecque vers l ’an 600 de notre ère. Bulletin
de l ’Institut Égyptien. 5th series 4, pp. 23-54.
Lammens, H. (1924), La Mecque à la veille de l ’hégire. Beirut.
Legrand, M .-Delm as, R. (1987), A 220-Year Continuous Record o f Volcanic H2S 0 4 in the Antarc-
tic Ice Sheet. Nature 327, pp. 671-676.
Lenski, G. (1987), Human Societies: An Introduction to Macrosociology. Fifth edition. N ew York.
Lundin, A. G. (1961), Juzhnaja Aravija v VI veke. Palestinskij Sbornik 8, pp. 1-159.
Macdonald, G. (1975), Vulkany. Moskva.
Maraqten, M. (1993), Wine Drinking and Wine Prohibition in Arabia before Islam. Proceedings of
the Seminar for Arabian Studies 23, pp. 95-115.
Marinatos, S. (1939), The Volcanic Destruction o f Minoan Crete. Antiquity 13, pp. 4 25-439.
Al-M as‘üdi, Abü-T-Hasan ‘A li ibn al-Husayn ibn ‘All (1965), Muriij al-dhahab wa-ma‘ddin al-
jawhar. Vol. 2. Bayrüt.
Mavljutov, R. R. (1974), К voprosu о proishozhdenii islama. Ateizm i religija 1, pp. 202-209.
al-Mawla-bik, Muhammad Ahmad Jâd-al-Bijâwï, ‘Ali Muhammad-Ibrâhîm, Muhammad Abä-’l-
Fadl (1942), Ayyäm a l-‘A rabfi ‘l-Jähiliyyah. Al-Qähirah.
Mayewski, Р.-M eeker, L.-Morrison, M.-Twickler, М.-W hitlow , S.-Ferland, K.-M eese, D .-L e-
grand, M .-Steffensen, J. (1993a), Greenland Ice Core “Signal” Characteristics: An Ex-
panded View of Climate Change. Journal o f Geophysical Research 98:D7, pp. 12839-
12847.
Mayewski, Р.-M eeker, L.-W hitlow, S.-Twickler, M .-Morrison, М.-A lley, R.-Bloom field, P .-
Taylor, K. (1993b), The Atmosphere during the Younger Dryas. Science 261, pp. 195-197.
Mikushina, O. V.-Klimenko, V. V.-Dovgalyuk, V. V. (1997), History and Forecast o f Solar Ac-
tivity .Astronomical and Astrophysical Transactions 12:4, pp. 315-326.
Moberg, A. (ed.) (1924), The Book ofHimyarites. Lund.
MPAT - Fitzmyer-Harrington 1978.
Müller, W. W. (1991), CIH 325: Die jüngste datierte sabäische Inschrift. In: Recueil offen à
Jacques Ryckmans (Publications de l ’Institut Orientaliste de Louvain 39), Louvain-la-
Neuve, pp. 117-131.
Müller,W. W .-von Wissmann, H. (1976), Über die von einem Lavastrom bedrohten Tempel der
Stadt Damhän, des heutigen al-Huqqa, im antiken Gau M a’din (Jemen). Anzeiger der phil.-
hist. Klasse der Österreichischen Akademie der Wissenschaften 113, pp. 112-146.
Negrja, L. V. (1981), Obshchestvennyj stroj Severnoj i Tsentral'noj Aravii v V-VII vv. Moskva.
Newhall, С.-S elf, S. (1982), The Volcanic Explosivity Index (VEI): An Estimate o f Explosive
Magnitude for Historical Volcanism. Journal of Geophysical Research 87:C2, pp. 1231-
1238.
Ninkovich, D.-Heezen, B. (1965), Santorini Tephra. In: Submarine Geology and Geophysics. Lon-
don, pp. 413—452.
Nöldeke, Th. (1879), Geschichte der Perser und Araber zur Zeit der Säsäniden, aus der arabi-
sehen Chronik des Tabari. Leiden.
Nöldeke, Th. (1889), Die Ghassänischen Fürsten aus dem Hause Gafna’s. Abhandlungen der
K. Akademie der Wissenschaften zu Berlin. Phil.-hist. Kl., pp. 3-60.
Olinder, G. (1927), The Kings o f Kinda of the Family o f Äkil al-Murär. Lunds Universitets Ars-
skrift N.F. Avd. 1:23:6, pp. 1-118.
Ostrovskij, G. D. (1978), Chuma. Bol’shaja Sovetskaja Éntsiklopedija. T. 29. Moskva, p. 258.
Palais, J.-Sigurdsson, H. (1989), Petrologie Evidence o f Volatile Emissions from Major Historic
and pre-Historic Volcanic Eruptions. American Geophysical Union Geophysical Monograph
52, pp. 31-53.

Parsons, T. (1977), The Evolution of Societies. Englewood Cliffs, NJ.

Peters, F. E. (1994), Muhammad and the Origins o f Islam. Albany.
Petrushevskij, I. P. (1966), Islam v Irane v VII-XV vekah. Leningrad.
Pigulevskaja, N. V. (1964), Araby и granits Vizantii i Irana v IV-VI vv. Moskva-Leningrad.
Piotrovskij, M. B. (1984), Prorocheskoe dvizhenie v Aravii VII v. In: Islam. Religija, obshchestvo,
gosudarstvo. M oskva, pp. 19-27.
Piotrovskij, M. B. (1985), Juzhnaja Aravija v rannee srednevekov’e. Stanovlenie srednevekovogo
obshchestva. Moskva.
Pirenne, J.-Tesfaye, G. (1982), Les deux inscriptions du negus Kaleb en Arabie du sud. Journal of
Ethiopian Studies 15, pp. 105-122.
Pollack, J.-Toon, O .-Sagan, С.-Summers, A .-Baldw in, В.-V an Camp, W. (1976), Volcanic Ex-
plosions and Climatic Change: A Theoretical Assessment. Journal of Geophysical Research
81, pp. 1071-1083.
Proussakov, D. B. (1996), Vzaimootnoshenie cheloveka i prirody v drevnem Egipte. Moskva.
Rampino, М.-S elf, S. (1982), Historic Eruption o f Tambora (1815) and Agung (1963). Their
Stratospheric A erosols and Climatic Impact. Quaternary Research 18, pp. 127-143.
Rampino, М.-S elf, S. (1984), Sulphur-Rich Volcanic Eruptions and Stratospheric Aerosols. Nature
310, pp. 677-679.
Rampino, М.-S elf, S. (1992), Volcanic Winter and Accelerated Glaciation Following the Toba
Super-Eruption. Nature 359, pp. 50 -5 2 .
Rampino, М.-S elf, S.-Fairbridge, R. (1979), Can Rapid Climatic Change Cause Volcanic Erup-
tions? Science 206, pp. 826-829.
Rampino, M.-Stothers, R. (1985), Climatic Effects o f Volcanic Eruptions. Nature 313, p. 272.
Rathjens, C. (1962), D ie alten Welthandelstrassen und die Offenbarungsreligionen. Oriens 15, pp.
115-129.
Reznichenko, T. D.-Reznichenko, S. D. (1964), O nekotoryh zakonomernostjah v razvitii Zemli.
In: Zemlja vo Vselennoj. Moskva, pp. 172-229.
Rippin, A. (1991), RHM NN and the Hanifs. In: Islamic Studies Presented to Charles J. Adams.
Leiden, pp. 153-168.
Robin, Ch. J. (1980), Judaisme et Christianisme en Arabie du Sud d’après les sources épigra-
phiques et archéologiques. Proceedings o f the Seminar fo r Arabian Studies 10, pp. 85-96.
Robin, Ch. J. (1982a), Les hautes terres du Nord-Yémen avant l ’Islam. Istanbul, 2 vols.
Robin, Ch. J. (1982b), Esquisse d’une histoire de l ’organisation tribale en Arabie du Sud antique.
In: La péninsule Arabique d ’aujourd’hui. T. II. Etudes par pays. Paris, pp. 17-30.
Robin, Ch. J. (1984), La civilisation de l ’Arabie méridionale avant l’Islam. In: L ’Arabie du Sud:
histoire et civilisation. 1. Le peuple yéménite et ses racines (Islam d’hier et d’aujourd’hui,
21), Paris, pp. 195-223.
Robin, Ch. J. (1991), Du paganisme au monothéisme. In: L ’Arabie antique de Karib’il à Mahomet.
Novelles données sur l ’histoire des Arabes grâce aux inscriptions (Revue du Monde Mu-
sulman et de la Méditerranée 61), Aix-en-Provence, pp. 139-155.
Robin, Ch. J. (1996), Le royaume hujride, dit ‘royaume de Kinda', entre Himyar et Byzance. Аса-
démie des inscriptions et belles-lettres. Compte rendus..., pp. 665-714.
Robin, Ch. J.-Beaucamp, J.-Briquel-Chatonnet, F. (1996), La persécution des Chrétiens de Najrân
et la chronologie himyarite. ARAM 10.
Röthlisberger, F. (1986), 10000 Jahre Gletschergeschichte der Erde. Aarau.
Rothstein, G. (1899), Die Dynastie der Lahmiden in al-Hira. Berlin.
Rubin, U. (1990), Hanifiyyah and Ka’ba. An Inquiry into the Arabian pre-Islamic Background of
the Din Ibrahim. Jerusalem Studies in Arabic and Islam 13, pp. 85-112.

Ryckmans, J. (1951), L ’institution monarchique en Arabie Méridionale avant L ’Islam (Ma’in et

Saba ’). (Bibliothèque du Muséon 28), Louvain.
Sanderson, S. K. (1990), Social Evolutionism. A Critical History. Cambridge, MA-Oxford.
Sanderson, S. K. (1995), Social Transformations. A General Theory of Historical Development.
Oxford.
Schove, D. J. (ed.) (1983), Sunspot Cycles. Stroudsburg, PN.
Scuderi, L. A. (1993), A 2000-year Tree-Ring Record o f Annual Temperatures in the Sierra Ne-
vada Mountains. Science 259, pp. 1433-1436.
Service, E. R. (1971), Primitive Social Organization. An Evolutionary Perspective. 2nd ed., New
York.
Shahid, T. (1971), The Martyrs o f Najrän. New Documents (Subsidia Hagiographica 49), Bru-
xelles.
Shahid, T. (1989), Byzantium and the Arabs in the Fifth Century. Washington, D.C.
Shahid, T. (1994), On the Chronology o f the South Arabian Martyrdoms. Arabian Archaeology
and Epigraphy 5, pp. 66 -6 9 .
Shahid, I. (1995), Byzantium and the Arabs in the Sixth Century. Washington, D.C.
Simkin, T.-Siebert, L.-McClelland, L.-Bridge, D.-Newhall, С.-Latter, J. (1981), Volcanoes o f the
World. Stroudsburg, PN.
Simon, R. (1989), Meccan Trade and Islam. Problems of Origin and Structure. Budapest.
Smirnov, N. A. (1954), Ocherki izuchenija istorii islama v SSSR. Moskva.
Smith, S. (1954), Events in Arabia in the 6th Century A.D. BSOS 14, pp. 425-468.
Sokoloff, M. (1990), A Dictionary o f Jewish Palestinian Aramaic of the Byzantine Period. Jerusalem.
Spier, F. (1996), The Structure o f the Big History: From the Big Bang until Today. Amsterdam.
Spirina, L. P. (1975), О dlitel’nosti i intensivnosti vozmushchajushchego vozdejstvija vulkaniche-
skih izverzhenij na temperaturu severnogo polusharija. Trudy Glavnoj geoftzicheskoj ob-
servatorii 330, pp. 126-130.
Stommel, H.-Stommel, E. (1979), The Year Without a Summer. Scientific American 240, pp. 176-
186.
Stookey, R. W. (1978), Yemen. The Politics o f the Yemen Arab Republic. Boulder, CO.
Stothers, R. (1984), Mystery Cloud o f AD 536. Nature 307, pp. 344-345.
Stothers, R. (1989), Volcanic Eruptions and Solar Activity. Journal of Geophysical Research
94:B12, pp. 17371-17381.
Stothers, R.-Rampino, M. (1983a), Historical Volcanism, European Dry Fogs, and Greenland Acid
Precipitation, 1500 BC to A D 1500. Science 222, pp. 411-413.
Stothers, R.-Rampino, M. (1983b), Volcanic Eruptions in the Mediterranean before AD 630 from
Written and Archaeological Sources. Journal of Geophysical Research 88:B8, pp. 6 3 5 7 -
6371.
al-Tabari (at-Tabari), Muhammad b. Jarir (1879-1901), Annales quos scripsit Abu D jafar Mo-
hammed ibn Djarir at-Tabari. Series I-III. Lugduni Batavorum [Leiden],
al-Tabari (at-Tabari), Muhammad b. Jarir (1964), Annales quos scripsit Abu D ja fa r Mohammed
ibn Djarir at-Tabari. Series I. Lugduni Batavorum [Leiden].
Tolstov, S. P. (1932), Ocherki pervonachalnogo islama. Sovetskaja étnografija no. 2, pp. 24—82.
Varushenko, S. I.-Varushenko, A. N .-K lige, R. K. (1987), Izmenenie rezhima Kaspijskogo morja
i besstochnyh vodoemov v paleovremeni. Moskva.
V asil’ev, A. (ed.) (1907), Zhitie sv. Grigentija, episkopa Omiritskogo. Vizantijskij vremennik 14.
Watkins, N.-Sparks, R.-Sigurdsson, H.-Huang, T.-Federman, A.-Carey, S.-Ninkovich, D. (1978),
Volume and Extent of the Minoan Tephra from Santorini Volcano: New Evidence from
Deep-Sea Cores. Nature 271, pp. 122-126.

Watt, W. M. (1953), Muhammad at Mecca. Oxford.

Watt, W. M. (1964), Muhammad, Prophet and Statesman. Oxford.
Watt, W. M. (1975), Hanïf. Encyclopédie de l ’Islam. T. III. Leyde, pp. 165-166.
Weikinn, С. (1958), Quellentexte zur Witterungschichte Europas von der Zeitwende bis zum Jahre
1850. Hydrographie. Teil 1. Berlin.
Wellhausen, J. (1897/1961), Reste Arabischen Heidentums. Berlin.
von Wissmann, H. (1981), Aravija. Materialy po istorii otkrytija. Moskva.
Yäqüt, Shihäb al-Din al-Hamawi al-Rümi al-Baghdädi (1410/1990), Mu'jam al-buldân. 7 vols.
Bayrüt.
al-Zawzanï (1972), Sharh al-mu'allaqät al-sab'. Bayrüt.
Zhang Deer (1984), Synoptic-Climatic Studies of Dust Fall in China since Historic Times. Scientia
Sinica (Series B) 27:8, pp. 825-836.
Zhukov, E. M. (1974), R ol’ religii v mirovoj istorii. Ateizm i religija 1, pp. 25-38.
Zielinski, G.-Mayewski, Р.-M eeker, L.-W hitlow, S.-Twickler, M.-Morrison, M .-M eese, D -
Gow, A .-A lley, R. (1994), Record o f Volcanism Since 7000 BC from the GISP2 Green-
land Ice Core and Implications for the Volcano-Climate System. Science 264, pp. 948-952.
Zielinski, G.-Mayewski, Р.-M eeker, L.-W hitlow, S.-Twickler, M. (1996), Potential Atmospheric
Impact of the Toba Mega-Eruption (71000 Years Ago). Geophysical Research Letters 23,
pp. 837-840.
Andrey V. Korotaev
Institute o f Cultural Anthropology
Russian State University for Humanities
6 Miusskaja Ploscad’
Moscow 125267, Russia
e-mail: andrei@rsuh.ru
Vladimir V. Klimenko
Moscow Institute o f Energy
14 Krasnokazarmennaja ulica
e-mail: gepl@deans.mpei.ac.ru
Dmitry B. Proussakov
Oriental Institute o f the Russian Academy o f Sciences
12 Rozdestvenka
e-mai: prusakov@prusakov.msk.ru

UN GOUVERNEUR D ’ORIGINE HONGROISE À ALGER

1580-1582
L ä s z l ö J. N a g y
(Szeged)
Alger, grâce aux faits militaires des frères Barberousse, Khayreddîne et Arudj, devient province
ottomane. La ville est en même temps le haut lieu de la course où les ra’îs, capitaines de navire,
réunis en une corporation, le tâ’ifa, dominé par les convertis, organisent d’une manière autonome
la vie intérieure de la province.
Pour garder le contrôle de la province la Porte y envoie un gouverneur dans tous les trois
ans. En 1580 Djefar pacha, d’origine hongroise (fait prisonnier dans son enfance au sud de la Hon-
grie) est nommé gouverneur de la province. L ’auteur présente ses activités à Alger - jusque-là non
découvertes - d’après la chronique de D iego de Haedo, en captivité à Alger dans les années 1578-
1581.
Mots-clé: Alger, Djefar, corsaire, course, D iego de Haedo, Khyreddîne, tâ’ifa, ra’îs
Après l’insurrection des Morisques de 1499 à 1502 la Monarchie espagnole - en pre-

mier lieu à l’instigation de Ximenes, l’archevêque de Tolède - décide de prolonger la
reconquête sur la terre africaine. Les places côtières (Oran, Tripoli etc.) tombent au
pouvoir des Espagnols, prises par force ou bien elles se rendent. Pour éviter le mas-
sacre Alger opte en 1510 pour la deuxième solution. Cette année-là les campagnes
espagnoles s’arrêtent, le roi, Ferdinand le Catholique ne les soutient plus. Ses préoc-
cupations vont aux problèmes de l’Italie et d’outre-Pyrénées, considérant l’Afrique
du Nord comme « un théâtre d’opérations secondaires ».'
A la suite du retrait espagnol sont apparus aux côtes nord-africaines les frères
Barberousse, Arudj et Khayreddîne, originaires des îles grecques, qui n’étaient pas
du tout inconnus grâce à la guerre de course ce qu’ils menaient contre les Chrétiens
en Méditerranée occidentale. Lors de la mort de Ferdinand le Catholique, en 1516 les
Algérois font appel à eux pour qu’ils viennent les débarasser des Espagnols. Les frè-
res Barberousse prennent la ville d’où ils comptent diriger des campagnes pour la
conquête des places du Maghreb occidental. Mais ils n’auront pas beaucoup de suc-
cès. Dans la bataille pour Tlemcen Arudj est tué. Les défaites subies, la révolte des
tribus berbères et la préparation de la guerre des Espagnols poussent Khyreddîne à se
1 Braudel (1996, p. 52).

0001-6446/99/ $ 5.00 © 1999 Akadémiai Kiadô, Budapest
278 LÄSZLÖ J. NAGY
placer sous la suzeraineté ottomane. Selîm, victorieux de la campagne d’Egypte Гас-

cepte espérant surtout qu’il pourrait élargir son pouvoir dans la Méditerranée occi-
dentale aussi. En 1519 il envoie 2000 janissaires, un contingent de 4000 volontaires,
de l’artillerie aux Algérois et un firman par lequel il déclare « leur accorder sa pro-
tection ».2
Khayreddîne stabilise son pouvoir à Alger et conquiert de nouvelles places en
Afrique du Nord. Il va à l’aide des Morisques en Espagne qui - dans la lettre adres-
sée au sultan - décrivent Khayreddîne comme « l’auteur de la délivrence de nom-
breux musulmans prisonniers aux mains des infidèles, il les a transportés en terre
d’Islam ... représente la gloire de la religion et la garantie de notre existence » et ils
présentent Alger comme « le rempart protecteur des Musulmans ».3
Soliman le Magnifique convoque à Constantinople et en 1534 le nomme ka-
pudan pacha et beylerbey. Désormais Khyreddîne passera très peu de temps à Alger.
Il mène des campagnes et des guerres de course au service des Ottomans et devient
« le véritable fondateur de la puissance navale ottomane en Méditerranée ».4 C’est lui
qui dirige la flotte ottomane victorieuse de la bataille de Preveza en 1538. Cet acte
militaire est considéré comme la fondation de la flotte ottomane.5 Il charge Hassan,
son adjoint, de gouverner Alger ce qu’il défendra avec succès contre l’armada de
Charles Quint en 1541. Une campagne de telle envergure dirigée par les puissances
européennes n’aura pas lieu jusqu’à 1830. D’après les sources l’armada dirigée perso-
nellement par l’empereur comprenait plusieurs centains de navires et 30 milles soldats.6
A la suite de cette victoire la position d’Alger comme bastion de l’Empire
ottoman dans la Méditerranée occidentale se stabilise. Sa structure administrative est
semblable à celle des autres provinces ottomanes. A sa tête se trouve le gouverneur,
le beylerbey, qui détient - en principe - tous les pouvoirs. Mais en réalité il doit tenir
compte de l’avis de l’odjak, garnison des janissaires, seule force organisée de la pro-
vince. Sans son consentement aucune décision ne peut être prise. L’originalité de la
province d’Alger (ou la Régence d’Alger comme on l’a appelée à l’époque) compa-
rée avec le Proche-Orient est le rôle considérable des ra’îs, capitaines de navire, orga-
nisés en une corporation, le tâ’ifa où les Turcs sont moins nombreux que les conver-
fis. Le tâ’ifa a une influence extrêmement importante dans la vie économique et poli-
tique parce que c’est la course qui est à la base de la pospérité et de la richesse de la
Provence.
La course ce qu’on distingue du banditisme maritim (la piraterie) est un phé-
nomène normal, une activité acceptée et pratiquée et par les Musulmans et par les
Chrétiens, considérée comme « une forme des échanges forcés dans tout l’espace
méditerranéen ».7
2 Chevallier (1988, p. 38).

3 Temimi (1975, pp. 105-106).
4 Mantran (1989, p. 145) Rubigally Pâl, noble hongrois faisant parti d ’une mission diplo-
matique, arrive à Constantinople en 1540 et devant la ville « admire la flotte de Barberousse corn-
posée de 200 vaisseaux comblés de soldats ». Rubigally (1977, p. 200).
5 Belhamissi (1986, p. 118).
6 Chevallier (1988, p. 103); Kdrpâty (1941, pp. 27-31).
7 Braudel (1990, p. 639).
UN GOUVERNOR D’ORIGINE HONGROISE À ALGER 279
Aux 16-17e siècles presque tous les ra’îs sont d’origine européenne. Convertis
à l’islam par une décision volontaire ou bien à la suite de leur capture par les corsai-
res ils se mettent au service de la « République des corsaires » en espérant s’enrichir.
Ils sont nommés « rénégats ». Pierre Boyer a réussi à identifier le lieu d’origine des
22 ra’îs de la deuxième moitié du 16e siècle. Il y avait d’entre eux 11 Italiens, 3 Grecs,
2 Albanais, 2 Espagnols et un Corse, un Français, un juif algérois et un Hongrois.8
Le partage du butin se fait par des règles strictes. Après avoir fait l’inventaire,
l’Etat (beylerbey) reçoit le capitaine, le pilote, le charpentier (métier le plus pré-
cieux !) et les canons. Et encore 12% des prix des captifs et des articles vendus aux
enchères, 1% pour l’entretien du port et 1% pour celui de la mosquée. Une moitié de
la part restante revient aux armateurs (« les actionnaires »), l’autre moitié est parta-
gée par l’équipage. Le ra’îs perçoit les 40% de la deuxième moitié.9
Les captifs constituent la part la plus précieuse du butin. Le trafic d’homme
est l’entreprise la plus lucrative aux 16-17e siècles dans la Méditerranée. (Le béné-
fice de plusieurs centains pour cent n’est pas une chose rare !) Son centre se trouve
à Livourne où nombreuses maisons de commerce s’y spécialisent. Leurs agents se
retrouvent dans toutes les villes importantes du bassin méditerranéen.10 Il y a des or-
dres religieux aussi ayant pour objectif principal le rachat, « la rédemption » des cap-
tifs. Le Vatican lui-même finance des transactions de ce genre. Dans un document de
fin du 16e siècle, écrit au nom du Pape, on peut retrouver les données (nom, lieu et
date de naissance) des 90 Hongrois.11
Jusqu’à ce que le beylerbey soit élu parmi les ra’îs - ou leurs candidats - la paye
de la solde des janissaires est assurée. Mais même pendant cette période-là l’aga, chef
de l’odjak, provoque souvent des émeutes ou des révoltes pour le retardement de la
paye ou de sa dévalorisation. A partir de 1568 le tâ’ifa admet les janissaires dans
l’équipage qui participent ainsi directement aux bénéfices de la course. Ces liens d’in-
térêt entre l’odjak et le tâ’ifa qui renforce - justement à cause de ça - l’autonomie de
la provence vis-à-vis de Constantinople.
La possibilité de l’indépendence de la provence d’Alger devient de plus en plus
réelle dès les années 1570. Après la victoire chrétienne à Lépante (1571) et celle des
Ottomans à Tunis (1574) le Habsbourg et les Ottomans « ces deux monstres politi-
ques de la Méditerranée » renoncent à la lutte et mettent un terme à la guerre menée
pour la suprématie en Méditerranée.12
Les corsaires algérois s’habituent à la période « pacifique » par se livrer entiè-
rement à la course. (Quelques agas envisagent l’occupation de Fès, ville importante
marocaine, mais le sultan l’interdit.) La Porte craignant - à juste titre d’ailleurs - que
le contrôle de la provence lui échappe, supprime le titre beylerbey en 1572 et essaie
d’y introduire le gouvernement direct par l’envoie d’un gouverneur dans tous les
8 Boyer (1985, p. 97).

9 Chevallier (1988, pp. 50-51); Moulaü (1973, p. 94).
10 Mathieux (1954, pp. 157-164).
11 Bono (1998, p. 26).
12 Braudel (1990, vol. 3, p. 299).
2 80 LÄSZLÖ J. NAGY
trois ans (« les pachas triénnaux »).13 II est chargé du maintien du pouvoir ottoman et
de garantir - ou au cas échéant d’établir - la paix intérieure perturbée très souvent
par les luttes intestines. Il est très rare que le pacha-gouverneur accomplit sa mission
de trois ans. D’une part il n’arrive pas à mettre fin aux luttes intestines et d’autre part
accumulant une fortune considérable provenant des impôts levés sur les tributs ber-
bères et de la course, il a plutôt de l’intérêt de partir que de rester dans le pays carac-
térisé par l’insécurité. Cependant il arrive très souvent que les intrigues de Constan-
tinople le forcent de rentrer.
Le sultan Mourad nomme le gouverneur d’Alger Djefar Pacha « Renégat Hon-
grois, eunuque qui l’avait servi et porté sur ses bras dans son enfance, et qui gouver-
nait une provence en Hongrie avec une renommée méritée de justice ... [il] fut pris,
étant enfant, en même temps que sa mère, un frère déjà grand et une soeur, dans une
incursion que les Turcs firent en Hongrie ».14 La tâche de Djefar est d’examiner la
plainte des janissaires contre son prédécesseur Hassan, nommé le Vénitien d’après
son origine, et de mettre fin aux luttes intestines. Arrivé à Alger le 24 août 1580, il ne
touche pas à Hassan qui a un protecteur très influent à Constantinople, Ochali, chef
de l’armée « maître absolu pour tout ce qui concernait la guerre ». Et l’autre raison
d’être si indulgent pour Hassan est bien « compréhensible »: lors de son départ pour
Alger Ochali lui donna « vingt mille écus pour les frais de son voyage, afin de l’en-
gager à la douceur ».15
Par contre il emprisonne pour peu de temps quelques notables. Libérés, ils
montent une conspiration contre Djefar. Ils sont rejoints par l’aga contre qui les janis-
saires portent plainte à Djefar parce qu’il n’a pas payé leur solde. L’aga promettant
de l’argent tente de les pousser à la révolte. Les janissaires - au lieu de le soutenir -
l’arrêtent et le mènent devant Djefar qui le fait exécuter. Derrière ce mouvement se
trouve le riche commerçant maure, financier de la conspiration, Caxes qui - malgré
son comportement néfaste - « obtint le pardon de son crime, en donnant au Roi Dje-
far une grosse somme, qui, selon ce qu’on m’a affirmé, se montait à trente mille du-
cats ».16
En 1581 Ochali lui-même arrive à Alger avec l’intention de diriger une cam-
pagne contre Fès. Il essaie de gagner à son projet les janissaires aussi. Mais eux, ils
le refusent en arguant qu’ils n’y participent qu’aux ordres formels du sultan. En
secret ils envoient un marabout à Constantinople en vue de demander au sultan
l’interdiction de la campagne contre Fès parce qu’elle renforcerait tellement Ochali
13 Boyer (1970, pp. 99—101).

14Haedo (1881, pp. 14, 17) Diego de Haedo, bénédictin sicilien d ’origine espagnole, pas-
sait les années 1578-1581 en captivité à Alger. En 1612 à Valladolid il a publié Topografia e Histo-
ria General de Argel. La partie Epitome de los Reyes de Argel était publiée en français dans la
Revue africaine en 1880-1881. Le chapitre XXII s’intitule « Djefar Pacha, vingt-troisième Roi ».
Ibrahim Pecevi, chroniqueur turc, a rencontré Djefar Pacha en 1595. Il a écrit que Djefar était né
dans le sandjak Gyula, sud de la Hongrie, où il était fait prisonnier dans son enfance par les Turcs.
Il était gouverneur des divers sandjaks en Hongrie et beylerbey à Bagdad, Damas et Tebrize.
Pecevi ne parle pas de la mission d ’Alger. In: Szekfü (1916, p. 85). Voir encore Thury (1892, pp.
399-403.)
15 Haedo (1881, pp. 17,24).
16 Haedo (1881, p. 23).
UN GOUVERNOR D ’ORIGINE HONGROISE À ALGER 281
qu’il « pourrait facilement se soulever et se rendre Seigneur de toute la Barbarie ».17

La mission du marabout a du succès, le sultan ordonne à Ochali d’abandonner le
projet de la campagne contre Fès et qu’il rentre à Constantinople. Cependant dans la
cour impériale il réussit par des intrigues habiles à ce que le sultan convoque Djefar
et à nommer Hassan le Vénitien le gouverneur d’Alger. « Djefar Pacha avait régné
vingt mois environ, du mois d’août 1580 à mai 1582. »18
L’Empire ottoman en cours d’affaiblissement n’est pas capable d’empêcher le
renforcement de l’autonomie de la provence d’Alger de même que de tout le Ma-
ghreb. Dès la deuxième moitié du 17e siècle l’odjak - d’abord seul, plus tard allié
avec le tâ’ifa - reprend le pouvoir réel et mène la provence à un statut quasi-indépen-
dant. Le détenteur réel du pouvoir élu par les deux corps portera le titre du dey à par-
tir de 1671, utilisé déjà dans le pays voisin de Tunis. Le pacha ne joue que le rôle du
paravent de la légalité jusqu’à 1710 quand le dey le renverra à Constantinople. Par
cet acte la période des pachas triénnaux est vécue formellement aussi.19 Le rattache-
ment à l’Empire ottoman n’est plus marqué que plutôt par des signes extérieurs (im-
pots symboliques), la domination ottomane se réduit à une suzeraineté qui n’a pas de
conséquences pour les détenteurs locaux du pouvoir. Par contre le sentiment reli-
gieux et culturel d’appartenance à l’Empire ottoman s’avère très fort au 19e siècle
lors de la résistance contre l’occupation française.
Références
Belhamissi, M. (1986), Histoire de la marine algérienne (1516-1830). Alger.

Braudel, F. (1990), La Méditerranée et le monde méditerranéen à l ’époque de Philippe IL (Livre
de poche). Paris.
Braudel, F. (1996), Les Espagnols et l'Afrique du Nord de 1492 à 1577. In: Autour de la Méditer-
ranée. Paris, pp. 31-91.
Bono, S. (1998), Europe et Islam en Méditerranée. In: La Méditerranée et l ’Europe. Histoire et po-
litique. Kukovecz, György-J. Nagy, Läszlö (eds), Szeged, pp. 23-33.
Boyer, P. (1966), Introduction à une histoire intérieure de la Régence d’Alger. Revue historique,
pp. 297-316.
Boyer, P. (1970), Des Pachas Triennaux à la révolution d’Ali Khodja Dey (1571-1817). Revue
historique, pp. 99-124.
Boyer, P. (1985), Les rénégats et la marine de la Régence d’Alger. Revue de l ’Occident musulman
et de la Méditerranée 39, pp. 93-106.
Chevallier, C. (1988), Les trente premières années de l ’Etat d ’Alger 1510-1541. Alger.
Haedo, D. de. (1881), Djefar Pacha, vingt-troisième Roi. Revue africaine 25, pp. 19-26.
KiSrpdty, E. (1941), V. Kdroly keleti politikâja. Kolozsvâr.
Mantran, R. (1989), Histoire de l ’Empire ottoman. Paris.
Mathieux, J. (1954), Trafic et prix de l ’homme en Méditerranée aux XVIIe et XVIIIe siècles. Anna-
lesE .S. C., pp. 157-164.
Moulaü, J. (1973), Les Etats barbaresques. Paris.
17 Haedo (1881, p. 24).

18 Haedo (1881, p. 26).
19 Boyer (1966, pp. 307-308).
282 LÄSZLÖ J. NAGY
Rubigally, P. (1977), Utazâsom tôrténete Konstantinâpolyban 1540-ben. In: Rabok, követek, kal
mdrok az Oszmdn Birodalomban. Tardy Lajos (ed.), Budapest.
Szekffl, Gy. (1916), Török törte'netirök 1566-1659. III. Budapest.
Temimi, A. (1975), Une lettre des Morisques de Grenade au sultan Suleian Al-Kanuni en 1541
Revue d'histoire maghrébine 3, pp. 105-106.
Thury, J. (1892), Dzsdfer pasa, 1552-1600. Hadtôrténeti Kôzlemények, pp. 399-403.
Ldszlö J. Nagy
Department o f M odem World History
University o f Szeged
H-6722 Szeged
Egyetem u. 2
Hungary
e-mail: jnagy@hist.u-szeged.hu

BUDDHISM AND THE DAO IN TANG CHINA:

THE IMPACT OF CONFUCIANISM AND DAOISM
ON THE PHILOSOPHY OF CHENGGUAN*
Im re H am ar
(Budapest)
Chengguan S Щ (738-839), the fourth patriarch o f the Huayan school o f Chinese Buddhism, de-
dared the primacy of Buddhism over Confucianism and Daoism and criticised these philosophies
from a Buddhist stance. In his subcommentary to the Avatamsaka Sütra, he defines ten differences
between Buddhism and indigenous philosophies, which are discussed in this paper. However, he
also often quoted from Chinese Classics to clarify the meaning of a Buddhist tenet. On these occa-
sions he sometimes adds that he only borrows the words but not their meaning. We investigate how
he places these words into a new, Buddhist context.
Key-words: Buddhism, Daoist philosophy, Huayan, Chengguan
Throughout its history in China it has remained a dominant issue how Buddhism fits
into Chinese thinking, culture and society. As Arthur Wright argues, the success of
the Buddhist conquest can be attributed to the weakness of Confucianism as a result
of the collapse of the Han with which it was entwined.1 In the beginning translators
used terms associated with traditional Chinese philosophy to convey the meaning of
Buddhist expressions. This method was called matching the meaning (geyi Wt Ш)■
With the appearance of the prajnäpäramitä sütras in the third and fourth centuries the
literati engendered great interest in these works since they found similarities with the
concepts of Dark Learning (xuanxue ‫־‬£ , Ф). Erik Zürcher designates this era as gen-
try Buddhism, referring to those men of letters at that time engaged in Buddhist-
Daoist metaphysical speculations.2 These metaphysical speculations, even though not
correct from the aspect of the original, Indian Buddhism, promoted the spread of Bud-
dhism in high society to a great extent. With the advent of Daoan and Kumarajiva,
a new era began. Chinese Buddhist adepts attempted to understand the Buddhist
scriptures in their original, correct sense. However, due to the Chinese way of think-
ing, the Chinese understanding of Buddhism could not be the same as that of the In-
dian adepts.
* This study was completed with the support o f OTKA (F 029471).

1 See Wright (1959, pp. 4 -2 0 ).
2 See Zürcher (1959, pp. 4 -6 ).
0001 -6446 /99/ $ 5.00 © 1999 Akadémiai Kiadd. Budapest
284 IMRE HAMAR
The Sui and Tang is a very creative epoch of the history of Chinese Bud-
dhism, which is called “new Buddhism” by Yùki Reimon.3 The new Chinese schools
of Buddhism such as Huayan, Tiantai, Chan, Jingtu based their doctrines on their
own understanding and realisation of Buddhist teaching. Even if these schools repre-
sent the genuine Chinese version of Buddhism, different from the Indian one in many
respects, they had to create their own Buddhist identity to avoid being swallowed by
the indigenous Chinese religions. To reach it, they had to prove their differences from
them, their superiority over them. In the following analysis of the impact of tradi-
tional Chinese philosophy on Chengguan, two aspects will be considered. On the one
hand, how he criticised the indigenous religions in order to prove the superiority of
Buddhism, and on the other hand how he adapted Chinese philosophy to accommo-
date the Buddhist teaching to his compatriots.
The Huayan masters’ attitude toward indigenous religions
The classical Huayan philosophy was established by the third patriarch, Fazang Йс Ш
(643-712), who, being on good terms with Empress Wu Ä , often preached the
dharma in the palace, explaining the abstruse doctrines of the Huayan philosophy.
Under the auspices of the court, he was not obliged to take the trouble to dispute the
tenets of Confucianism or Daoism. As Buddhism provided the ideology that Empress
Wu used to legitimise her power against the Confucian officials who severely criti-
cised her usurpation, Fazang was not compelled to prove the superiority of Bud-
dhism among Chinese religions. After Fazang’s death this favourable condition for
Buddhism ended.
Xuanzong Ж !‫׳‬.к (713-756) brought Daoism into prominence. Fazang’s main
disciple, Huiyuan Ш Ш (673-743) had to deal with the teachings of rival religions.
He argued that in regard to the creation of Heaven, Earth and the rest of creation
three Chinese books deserved attention: the Yijing, the Daodejing and the Zhuangzi.
He related the doctrine of Buddhism to that of Daoism. According to Daoism, the
nonbeing originated the myriad things, and this was in accord with the Buddhist
teaching of the dependent origination of the tathägatagarbha, which explained the
evolution of the phenomenal world from the absolute pure Buddhahood.4
Chengguan lived in a Buddhist monastery from his childhood, therefore he
first became versed in Buddhist literature. The following thought led him to realise the
importance of the secular works. “On the fifth level the sage learns worldly methods,
realises the Tathatä and his mind rests on the realm of Buddha. He receives wisdom
after his enlightenment, and the thought to benefit the world arises. I am on the level
of learning, how could I forget about it?” This passage sheds light on Chengguan’s
motivation for studying secular literature. He intended to draw on the Chinese non-
Buddhist works to assist the understanding of Buddhist teaching. In his lifetime the
adepts of Buddhism and Daoism often debated fiercely with each other, for this
3 See Gregory (1991, p. 3).

4 XZJ 5 .2 4 b l4 -2 5 a l7 .

BUDDHISM AND THE DAO IN TANG CHINA 285
reason he was not inclined to formulate a synthesis of the three teachings, but rather
he attempted to claim the primacy of Buddhism over Confucianism and Daoism. He
stated that even the most superficial Buddhist tenet overshadows the most profound
non-Buddhist one.5
The question of origination in Chinese philosophy
Chengguan, like his predecessor, Huiyuan, examined the concepts of Chinese phi-
losophies through the Yijing, Daodejing and Zhuangzi. He called these three works
“the three mystical books (sanxuan H. X ): the first of them represented the mystery
of truth (zhenxuan Jrf i 0 , the second explained the mystery of nothingness (xuxuan
Ш Ж), and the third presented a discourse upon mystery (tanxuan Ж iO ”.6 In con-
trast to Huiyuan, he did not correlate the Daoist concept of nonbeing with the Bud-
dhist notion of tathägatagarbha, supposedly, in order to avoid the impression of con-
fusing the identity of the two rival religions. He regarded the five constant virtues
(wuchang E i.e. benevolence, righteousness, propriety, knowledge and sincerity),
spontaneity (ziran [=1 Ш) and causality (yinyuan tS Ш.) as the cardinal principle of
Confucianism, Daoism and Buddhism, respectively.7
Investigating the causality propounded by the Daodejing, he cited the chapter
41: “Tao produced the One. The One produced the two. The two produced the three.
And the three produced the ten thousand things.”8 He argued that Laozi established
a pseudo causality (siyou yinyuan Ш Щ И Ш), and not a correct causality (feizheng
yinyuan ||: IF. й Ш), for the Dao is voidness and spontaneity (xuwu ziran Ш Ш
@ Ш).
Chengguan in his introduction to his commentary and subcommentary to the
Avatamsaka Sutra discusses nonbuddhist teachings. First, he quotes from the Yijing,
Daodejing and Zhuangzi and the commentaries written to them, then he interprets
them from a Buddhist standpoint. He concludes that the Daodejing, which claims that
the Dao originates the One, proposes an erroneous causality (xieyin H ), while the
Zhuangzi, stating that all things are engendered spontaneously, commits the fault of
acausality (wuyin Iff- Ш). If Zhuangzi's views were right then anything could be white
like a crane, or black like a crow. According to the Yijing the transformation of yin
and yang creates the myriad things, which is the fallacy of erroneous causality. The
commentaries explicate One as nonbeing, and it is the error of acausality. Chengguan
denounces the acausality morally and because on a soteriological level it leads to a
consequence which is unfavourable for Buddhist, and any other religious, praxis.
Acausality implies that the bodhi of living beings, the enlightened mind is engen-
dered spontaneously and not by assiduous cultivation.
5 Song gaoseng zhuan ^ ilb Ш T 50.737a20-23.

6 T 36.103c4-5.
7 T 36.513al7-18.
8 See Chan (1963, p. 160).
286 IMRE HAMAR
Ten critiques against Confucianism and Daoism
Chengguan levels ten critiques against Confucianism and Daoism, in which the in-
fluence of the Ten Differences and Nine Fallacies (Shiyi jiumi lun I‫ ־‬J% / L ÜÈ Ira)
written by Li Zhongqing ^ f r ÜP can be detected.
1. Beginning or lack of beginning. According to Buddhism the process of birth
and death and the causality have no beginning. In contrast, the nonbuddhist teachings
suppose that there is a beginning antecedent to the existence of the myriad things
which is called taichu Jk. %J] or taishi fè.
2. The recognition or non-recognition of the vital force. The Daoists believe
that the spirit (shen È|1) is engendered through the transformation of the vital force
(qi Jft) without action, spontaneously. We might recall the Zhuangzi on transforma-
tion: “Little by little he’ll borrow my left arm to transform it into a cock, and it will
be why I am listening to a cock-crow at dawn. Little by little he’ll borrow my right
arm to transform it into a crossbow, and it will be why I am waiting for a roasted owl
for my dinner. Little by little he’ll borrow and transform my buttocks into wheels,
my daemon into a horse, and they’ll be there for me to ride, I’ll never have to harness
a team again.”9 The natural consequence of the teaching of spontaneous transforma-
tion is to “abandon sageliness and discard wisdom”10 (Laozi 19). Buddhism, how-
ever, teaches that every dharma is rooted in the mind, all activities depend on causes,
therefore living beings are responsible for their deeds. He stresses that the law of cau-
sality is the prerequisite for religious cultivation.
3. The existence or nonexistence of the three worlds. Buddhism regards the
mind (xinling M) as uninterrupted, dependent on conditions it evolves in the three
worlds (desire, forms and formless). The Confucianists and Daoists believe that the
concentration of vital force is birth and the dispersal of it is death. After death the
vital force returns to Heaven and Earth, therefore they do not acknowledge the con-
tinuation of the causality.
4. Deeds either have influence or have not. By using the word xi Щ Cheng-
guan here refers to xunxi |R Щ (väsanä), a central concept of the Yogäcära school of
Buddhist philosophy, which means that the impressions of the external manifestations,
deeds constantly “perfume” or influence the eighth consciousness, the storehouse
consciousness (älayavijnäna). During this process of “perfuming” the karmic seeds
become planted in the storehouse consciousness, waiting for the appropriate condi-
tion to be activated. It depends on the karma accumulated whether one becomes wise
or dull, being so, everybody has the opportunity for the religious cultivation through
the kalpas to alter his or her characteristics. The Confucianists and Daoists think that
both good and evil nature originate from Heaven, therefore it gives (fen j f ) wisdom
or stupidity. If somebody receives pure and harmonical (chunhe Ш $1) things from
Heaven, then he or she will be endowed with the most spiritual nature and sagely
wisdom; but if what is received is turbid and impure (hunzhuo рр $!}), he or she will
be stupid.
9 See Chuang-tsu (1981, p. 88).

10 See Chan p. 149.
5. The acceptance of causality or the vital force. According to Buddhism all

phenomena depend on conditions, while the Confucianists and Daoists believe that
prosperity, happiness and unhappiness are determined by the vital force and the rules
of Heaven (qi ruing M 1^3‫)־‬. Consequently, we cannot change the vital force obtained
from Heaven, but we can change the conditions, and in doing so praxis can be car-
ried out.
6. Origination from inside or from outside. Chengguan from the Yogäcära
standpoint argues that the outer world, the Heaven, the Earth and the myriad things
originate through the transformation of the inner consciousness (neishi biansheng A
!01 A
, i), the ancient Chinese religions, on the contrary, regard man as being origi-
nated through the transformation of Heaven and Earth. It depends on this transforma-
tion whether a man is pure or impure.
7. Phenomena do or do not depend on conditions. In Buddhist epistemology, the
four features of phenomena (origin, existence, change, and cessation) depend on condi-
tions and they are not engendered spontaneously. In Confucianism and Daoism things
are such by their own effort and their being or nonbeing is independent of the mind.
8. Fortune does or does not originate from Heaven. The Confucianists and
Daoists divide fortune and misfortune into two parts: those which the Earth brings
about can be avoided, and those which are provoked by Heaven cannot be escaped.
In the Buddhist view, the deeds either increase suffering or abolish it. Those who act
in order to put an end to suffering get rid of the misconception of nihilism and eter-
nity, surmount the false discrimination of being and not being, and overcome the re-
occurrence of birth and death, the samsara. Those who increase their suffering evolve
karma from their mind, wander in the six realms of living being (liuqu А Ш), and
remain fettered in the three worlds (sanyou EL •f] ). Chengguan remarks that the dif-
ferences between Buddhist and nonbuddhist teachings are very obvious.
9. Activity is defiled or not defiled. Laozi refuses all kinds of activities, and
deems any cultivation superfluous. Only in the epoch of upheaval are ceremonies (/г
Ш) needed. Therefore if we discard ceremonies, confusion terminates by itself with-
out any intervention. In addition, the principle (li Й ) originates from the Dao, so if
the Dao is present, the principle becomes manifest, and the sagely instructions be-
come redundant. If you do not seek, you will find; if you do not act, you can do any-
thing. However, if you act, you fail; if you seek, you find nothing. According to
Buddhism, meritorious deeds bring advantageous birth as man or god, while evil
deeds result in disadvantageous birth. Desire is the root of the samsara, hence with
the removal of it, birth and death come to an end. In accordance with the traditional
Chinese search for longevity, Chengguan states that a man of benevolent deeds will
enjoy a long life. In contrast to Daoism, he emphasises the role of activity and culti-
vation to reach the soteriological aim of Buddhism. He concludes that Buddhism
takes the opposite course to Daoism. They are as contradictory as cold and heat.
10. Returning or not returning. Buddhism teaches that all suffering of samsara
is due to false concepts (wangxiang Igf Ш)• This ignorance gives rise to covetous-
ness, desire and defilement. Therefore, with the removal of the prime cause, igno-
ranсe, the whole process of suffering can be stopped. By being aware of the confu

288 IMRE HAMAR
sion of essence (tiwang m $ ) a Buddhist adept can annihilate the false concepts and
achieve nirvana‫׳‬, and by understanding the deficient nature (xingjia f t fix) he or she
can discard the deficiency and return to the quiescence (gui jimie Й|1 ‘M M). It is the
way to cross the sea of suffering of birth and death, to coexist with the emptiness
peacefully and happily forever.
Laozi, however, deems that birth and death are ordered by the Dao, that
Heaven decides whether a man becomes wise or stupid, and the will of the Heaven
cannot be ignored. Man can do nothing but accept his or her fate passively, and thus
live and die in peace. At the same time, the Daoist adept should strive to keep his or
her nature intact (shouquan xingqing fît). If his or her nature remains un-
changed, Heaven is not spoiled; if he or she follows the course of birth and death, the
Dao is not lost. Consequently, if the Dao is not lost, then the happy and unhappy
states of mind cease; and if Heaven is not spoiled, then the feelings of pleasure and
anger end. The Daoist adept leaves this dusty world behind, and wanders in the realm
of Dao and De. In the state of nonactivity vicious forces (xieqi IjC) cannot exert
any influence upon him or her, therefore he or she can reach longevity. We see that
Buddhism inspires its followers to carry out active cultivation, since someone’s own
deeds are to change one’s future, to bring about future happiness or unhappiness.
Daoism, however, stresses the importance of the maintenance of the original, ideal
state, the nonintervention into the state of affairs.
A few observations should be made here concerning the ten points discussed above.
Chengguan sees that the doctrines of Buddhism differ from original Chinese relig-
ious concepts in two aspects: ontological and soteriological. On the ontological level,
Buddhism does not set a beginning in time and regards the mind as the source of all
phenomena, while Confucianism and Daoism suppose that there is a beginning in
time (1,6).
Their tenets on the rules which influence or determine the future of beings
originated are divergent as well (7). Buddhism teaches that the law of causality regu-
lates the life of living beings while other religions argue that they change spontane-
ously, and that their fate is decided by Heaven.
As a consequence of the different views on ontological questions, disparities
can be found on the soteriological level as well. Confucianism and Daoism assert
that the existence of human beings depend on an external power, i.e. Heaven (4, 8)
or qi (2, 3, 5). In the last two points Chengguan compares religious cultivation, the
praxis propounded by Buddhism and Daoism. He contrasts the Daoist nonactivity
and free wandering with the Buddhist moral behaviour and assiduous cultivation. Of
course, he does not take Confucianism into consideration in this aspect, since Confu-
cianism could not offer a religious, not-worldly practice in order to liberate humans
from the chains and sufferings of this world in the hope of a transcendental existence.
In elaborating the differences between Buddhist and nonbuddhist religions,
Chengguan points out that these differences are essential, and thereby the three teach-
ings cannot be regarded as one system of doctrines. Elsewhere, in connection with
Cheng Xuanying’s application of Buddhist, especially Madhyamaka, terminology to

explain Daoist philosophy in his commentaries to Laozi, and Zhuangzi, Chengguan

warns against mixing the ideas of the three teachings, since it leads to birth in hell
and it is the source of ignorance which blocks the mystical way of wisdom.11
It is noteworthy that Chengguan lumped Confucianism and Daoism together
in his account of differences, though they held different views on several questions.
Chengguan’s main disciple, the fifth patriarch of the Huayan school, Zongmi took
over his master’s critiques against the indigenous religions and elaborated further in
his Inquiry into the Origin o f Humanity (Yuanren lun Ш À àra)■12 He also discussed
Confucianism and Daoism together under the rubric of two teachings.13 Peter N.
Gregory argues that the reason behind it was that Zongmi presumed familiarity with
Confucianism and Daoism, as the audience was the literati.14 Chengguan had a very
good relationship with the literati, hence even if he preached the Avatamsaka Sutra
to the monks, there must have been literati in the audience, from which arose the
need for references to Confucianism and Daoism.
Adaptation of classical Chinese philosophy
Above it was shown that Chengguan was very critical with Daoism and Confucian-
ism, and did not intend to harmonise the three teachings, but rather asserted the pri-
macy of the Buddhist viewpoint. However, to a certain extent he was indebted to clas-
sical philosophy. Its thought and way of expression infiltrated into his writings.
Chengguan’s erudition on Chinese classical works is proven by the ample
quotations from them in his subcommentary to the Avatamsaka Sutra. He cites the
following books: Book o f Odes, Book of Documents, Book o f Change, Zuozhuan,
Erya, Rites of Zhou, Book of Filial Piety, Daodejing, Zhuangzi, Liezi, Annalects,
Guanzi, Records o f the Historian, Chronicle o f the Han Dynasty, Guangya, Shuowen
Jiezi to illustrate the meaning of some passages of the Avatamsaka Sutra. While
quoting these books he keeps his distance by saying: “I only borrow the words but
do not adopt their meaning.”15
The next passage, which reveals the impact of the Book o f Change and Wang
Bi’s commentary to it,16 shows the way how Chengguan endowed the terms of in-
digenous philosophies with new meaning.
“Going and returning are limitless, motion and quiescence [originate] from the
same source. It contains all subtleties (zhongmiao M>), yet has surplus; it is
beyond speech and thought. It is called dhanna-dhâtu.”17
11 T 36.107al 1-13.
12 For an annotated translation of this text along with the Chinese text, see Gregory (1995).
13 See Gregory (1991, pp. 255-294); Jan (1980).
14 See Gregory (1995, p. 80).
15 T 36.2b9 or 3b 13.
16 See Kojima (1998).
17 T. 35.503a7-8. One o f Chengguan’s most important teachings is the four dharma-dhâtus.
On his theory see my article.
290 IMRE HAMAR
To clarify the meaning of “all subtleties” he cites the Daodejing (1):

The Tao (Way) that can be told of is not the eternal Tao;
The name that can be named is not the eternal name.
The Nameless is the origin of Heaven and Earth;
The Named is the mother of all things.
Therefore let there always be non-being so we may see their subtlety,
And let there always be being so we may see their outcome.
The two are the same,
But after they are produced, they have different names.
Deeper and more profound,
The door of all subtleties.18
Chengguan defines one real dharma-dhätu (yizhen fajie —■ÿ Й I?•) as the
essence of the subtleties (xuanmiao ti А й ‫ ׳‬Ш) and all subtleties as the feature of
this essence.19 In the Daodejing the term, “all subtleties” is a direct reference to the
ontological basis of existence, the Dao. Chengguan, in contrast, attributes a lower
rank to it by identifying it with a feature which can be conceived through the senses.
Thereby he upgrades the central concept of Huayan philosophy.
Sometimes the expressions of Chinese philosophies are organically embedded
in the Buddhist context, and the origin of them is not mentioned in the text. The sec-
ond chapter of Zhuangzi, The Equality o f Things reads: “heaven and earth were bom
together with me, and the myriad things and I are one.”20
Chengguan rewrote Sengzhao’s allusion to this passage:21 “The mystical Dao
lies in the wonderful enlightenment. The wonderful enlightenment lies in establish-
ing the truth (jizhen BP jfc). Establishing the truth is the discernment of sariisära and
nirvana as identical (shengmie qiguan ,А1.Ш Щ M). The discernment of samsara and
nirvana as identical means that they are not two. Therefore the Tathatä and I sprang
from the same root, and the dharma-nature (faxing if: 1É) and I form one body.”22
Chengguan substituted the terms being and nonbeing for samsara and nirvana,
Heaven and Earth for Tathatä, and myriad things for dharma-nature. This modifica-
tion can be attributed to his Yogäcära and Huayan standpoint which prompted him to
illustrate the all inclusive nature of the absolute, or Tathatä. In addition, this passage
sheds light on the difference between Sengzhao’s and Chengguan’s application of
Daoist works. First of all, Sengzhao avails himself of Daoist allusions for substantial-
ing the arguments of Madhyamaka philosophy introduced to China by Kumarajiva,
unlike in Chengguan’s case where the Huayan standpoint is obvious. Moreover, in
Zhaolun, which is a masterpiece of the synthesis of Daoist and Buddhist philosophy,
18 Chan (1963, p. 139).

19 T 36.2b 10-11.
20 Chuang-tsu (1981, p. 56).
21 T 45.159b 26-29. Liebenthal rendition o f this passage: “The Inscrutable is found in an
intuitive experience which opens insight into the Middle Path. The Middle Path is insight into the
sameness o f existence and non-existence, (to one who has this insight) subject and object are not
two. It follows that heaven-earth and he (the Sage) sprang from the same root, that all single things
and he form one body.” See Chao Lun (1968, p. 117).
22 T 35.884012-14. For the explanation o f this in the Subcommentary, see T 36.634a5-12.

the Daoist environment in which the Buddhist doctrines are embedded is far more ac-
centuated than in Chengguan’s works. With these remarks we can agree with Kamata
Shigeo who aligns Chengguan with Sengzhao.23
The influence of Daoism on the following definition of dharma-dhätu is un-
doubted.
“The dharma-dhätu is not dhâtu and not non-dhätu, not dharma and not non-
dharma. It is unnameable so it was given forcibly the name ‘non-obstructed dharma-
dhätu'. It is quiescent, void and vast. It reaches deepness and includes all forms of
existence (wanyou Ш 'ft)■ It is the essence of one-mind (yixin ti —‫) §| ־‬, surpasses
the features (xiang fl |) of existence and non-existence. It is not bom and it does not
perish. If the beginning and the end of it cannot be found, how could the middle and
the side of it be found?”24
The sentence “If the beginning and the end of it cannot be found, how could
the middle and the side of it be found” is parallel with “Meet it and you will not see
its head. Follow it and you will not see its back.”25 (Daodejing 14)
*
In our analysis of Chengguan’s attitude toward indigenous religions, it was shown

that his standpoint was characterised by ambivalence. During a time when a keen
contest for primacy among the various religions was the part of the religious and cul-
tural life of China he, as a prominent leader of Buddhism, had to proclaim the supe-
riority of Buddhism. Therefore he condemned those who attempted to mix the three
teachings by correlating their doctrines. He even attacked his own predecessor Hui-
yuan to whose teaching in fact he was indebted in many aspects, since he deviated
from Fazang’s classification of doctrines (panjiao Ш) by taking up the outer
teachings in his own system. It was these arguments that gave a reason for later
generations to exclude Huiyuan from the patriarchal lineage of the Huayan school.
Chengguan based his arguments against Confucianism and Daoism on moral grounds
which had implication for soteriology.
The Buddhist acknowledgement of the law of karma, the retribution in future
lives emphasised the role of the individual responsibility for deeds, an idea quite dif-
ferent from social-oriented Confucianism or cosmology-centred Daoism. As a con-
sequence, Buddhism presented the opportunity for an individual intervention into the
current state of affairs through one’s own religious cultivation to attain liberation from
the fetters of society and cosmos which was impossible in the view of the indigenous
religions.
On the other hand, Chengguan was aware of the importance of the acknowl-
edgement and appreciation of Buddhist teaching to the literati strata, as the spread of
Buddhism in China was always connected to the generous support of the emperor,
the aristocrat and literati. In the case of Buddhist schools with highly sophisticated
23 See Kamata (1959, p. 90). For Sengzhao’s influence on Chengguan, see Kamata (1965,
pp. 338-357).
24 XZJ 7:498311-14.
25 Chan (1963, p. 146).
292 IMRE HAMAR
system of doctrines, such as Tiantai or Huayan, it was even more compelling since
their abstruse teachings were addressed to the elite of Buddhist and nonbuddhist cir-
cles and thereby incomprehensible to the masses. Actually they even could play a role
in the ideology of those rulers, who wished to declare the uniqueness of their power
by adherence to a school different from the one supported by the previous ruler.26 To
attract the attention of the literati toward Buddhist philosophy Chengguan had to make
some concession to indigenous philosophies and religions, and sometimes had to
express the Buddhist ideas in terms with which the literati was familiar. Nonetheless,
the Buddhist meaning that his words conveyed remained dominant.
References
Chan, W. (1963), A Source Book in Chinese Philosophy. Princeton University Press, Princeton.
Chao Lun - The Treatises o f Seng-chao (1968), translated by Walter Liebenthal, Hong Kong.
Chuang-tsu (1981), The Inner Chapters, translated by A. C. Graham. London.
Gregory, P. N. (1991), Tsung-mi and the Sinification of Buddhism. Princeton University Press,
Princeton.
----- (1995), Inquiry into the Origin of Humanity: An Annotated Translation o f Tsung-mi’s
Yuan jen lun with a Modern Commentary. University of Hawai’i Press, Honolulu.
Hamar, I. (1998), Chengguan’s Theory o f the Four Dharma-dhatus. AOH 51:1-2, pp. 1-19.
Jan Yün-hua, Tsung-mi’s questions regarding the Confucian absolute Philosophy. East and West
30:4, pp. 495-504.
Kamata Shigeo 8fflÆ i (1959), Chökan no kegon to Röshöshisö Ш Ш <0 Шt 45 JÎES
S . Komazawa Daigaku Bukkyô Gakubu Kenkyii Kiyd 19.
----- (1965), Chûgoku kegon no shisöshi no kenkyii Ф Ш Щ Ш ,S. S JÈ ® (iff Tokyo dai-
gaku shuppankai, Tokyo.
Kojima Taizan /Jn JJ f§ 1L| (1998), Chökan ni okeru röekigon itchi no kegon shisö to shihökai ffi
Ш и If £ 3 5 13 Ш & <0 Ш Ш Ж ffi t R4 i f ft-■ Indogaku Bukkyôgaku Кепкуй
46:2, pp. 578-582.
Weinstein, S. (1973), Imperial Patronage in the Formation of T’ang Buddhism. In: Perspectives on
the T’ang. Arthur Wright-Denis Twitchett (eds). Yale University Press, New Haven.
Wright, A. (1959), Buddhism in Chinese History. Stanford.
Zürcher, E. (1959), The Buddhist Conquest o f China: The Spread and Adaptation of Buddhism in
Early Medieval China. 2 vols. E. J. Brill, Leiden.
Imre Hamar
Department o f East Asian Studies
University o f Budapest
H-1088 Budapest
Muzeum krt. 4
Hungary
26 See Weinstein (1973).

К ВОПРОСУ ОБ ЭТНОГЕНЕЗЕ М АРИЙЦЕВ1

Клара А дягаш и
(Дебрецен)
Klära Agyagâsi: On the reconstruction o f the Cheremiss ethnogenesis
The author summarises the previous viewpoints on Cheremiss ethnogenesis. According to some,
the Cheremiss people are autochthonic inhabitants in their present-day homeland. According to
others, they escaped to this area from the Tatar invasion during the middle o f the 13th century.
The present paper solves this contradiction with the analysis o f three types o f sources: (1)
historical and linguistic sources on the ethnonym Cheremiss; (2) archaeological foundings from
Viatka, Vetluga and the Volga Region and also from the northern part o f the Rostov-Suzdal Prin-
cipate; (3) the stock of loan-words in present-day Cheremiss dialects.
Results: the Cheremiss people, mentioned by Russian and Khazar sources in the 11th
century, and geographically placed between the Russians and the V olga Bulghars were assimilated
by the Russian population. From the middle of the 12th century, the name Cheremiss denoted any
people living between the Russians and the Volga Bulghars. From the middle of the 12th century
the peoples o f the Ananino and Diakovo-Gorodec cultures inhabited this area. The early history o f
the present-day Mari people and language is independent from the historical Cheremiss’. The Mari
people lived next to the Proto-Permian groups till the 13th century and they settled in the Viatka-
Vetluga-Volga Region fleeing from the Tatars. After their settlement they assimilated the local
population and the topos Cheremiss began to be used for them.
Key-words: Cheremiss ethnogenesis, Archaeology o f the V olga Region, Ethnonym Cheremiss, Cha-
sarian sources, Cheremiss people, Mari language
Вопрос о локализации марийского этноса до татарского нашествия в на-

учных исследованиях до сих пор не имеет единого решения. В литературе
последних трех десятков лет отражено два основных, друг друга исклю-
чающих мнения. Соответственно первому, современные марийцы являются
аборигенами вятского и поветлужского междуречия, по другому же мне-
нию - они являются пришельцами (вследствие татарского нашествия).
1 Данная работа является результатом исследований при поддержке Alexander

von Humboldt-Stiftung.

2 94 КЛАРА А Д ЯГА Ш И
Первого мнения придерживаются в основном археологи (см. Бадер

1951; Смирнов-Трубникова 1965; Архипов 1967; Халиков-Архипов 1969;
Халиков 1969, 1970; Архипов 1976, 1976а; Халиков 1990, 1991 - вм есте
с об о б щ аю щ ей м о н о гр а ф и ей и с то р и к а К о зл о во й 1978), считая, что
предками марийцев являются представители ананьинской, азелинской
и дьяково-городецкой археологических культур, и те лингвисты, которые
принимают теорию о былом существовании в Поволжье волжско-финского
праязыка (см. Ж ираи 1937; Тойвонен 1952; Дечи 1965; Серебренников
1967; Казанцев 1985; Серебренников 1985).
По теории марийской автохтонности, в вятском и поветлужском
междуречье марийский этнос сформировался в первой половине I тысяче-
летая нашей эры, и во второй половине I тысячелетия на левом берегу
Волги он уже распался на две части. Это значит, что дифференциация об-
щемарийского единства на горную и луговую ветвь произошла до конца
I тысячелетия.
Второе мнение высказал проф. Г. Берецки на основании изучения
словарного запаса финно-угорского происхождения марийских диалектов
(Берецки 1992, 1994), утверждая, что марийская лексика финно-угорского
происхождения не доказывает распада древнемарийского языкового един-
ства до середины XIII в.
Что касается археологических исследований, нужно сказать, что для
этнического отождествления населения ананьинской и дьяково-городецкой
культур с предками марийцев нет достоверных критериев; оно основыва-
ется на том единственном факте, что место обитания населения названных
археологических культур и современного марийского народа совпадает.
Те лингвисты, которые поддерживают мнение о марийской автохтонности
на территории между реками Волгой и Ветлугой, ссылаются на письмен-
ные источники и языковые данные мордовского и марийского языков.
Среди письменных источников обыкновенно цитируются ими русские ле-
тописи, в которых черемисы (= марийцы) упоминаются задолго до татар-
ского нашествия (об этом см. еще ниже), а в качестве языковых «доказа-
тельств» приводится никогда не доказанная, но долгое время принимав-
шаяся за аксиому гипотеза о существовании волжско-финского праязыка.
(По мнению этих лингвистов, из этого праязыка дифференцировались поз-
же мордовский и марийский языки. На появление такой гипотезы, види-
мо, тоже повлиял тот факт, что в русских летописях «Мордва» и «Чере-
миса» одновременно упоминаются на восточных границах Киевской Руси,
и языки этих же народов бытуют сегодня близко друг к другу в поволж-
ском регионе.)
Последняя публикация по теме этногенеза марийцев - это моногра-
фия венгерского археолога Ласло Климы (1996), вышедшая одновременно
со статьей Т. Никитиной (1996). В своей книге Клима реконструирует эта-
пы этногенеза поволжско-финских народов, основываясь на трех группах
источников: (1) данные русских летописей, (2) памятники ананьинской,

К ВОП РОСУ О Б Э Т Н О Г Е Н Е ЗЕ М А РИ Й Ц Е В 295
дьяковской и Городецкой археологических культур, (3) лексика поволж-

ско-финских языков финно-угорского происхождения. Клима приходит
к выводу, что сведения русских летописей единичные, они не подтвержда-
ются другими письменными источниками, поэтому могут быть неточными
или ошибочными. Возможность отождествления марийцев с народом ме-
ря (а были такие попытки в прежней литературе, см. Горюнова 1967) он от-
клоняет из-за отсутствия исторической обоснованности и устанавливает,
что марийский и мордовский языки не восходят к общему праязыку. (К та-
кому выводу пришли уже раньше и другие лингвисты-финно-угроведы,
см. Берецки 1974; Гено 1981; Керестеш 1986, 1987; Берецки 1992, 1994). Дан-
ные археологии, по мнению Климы, указывают на то, что в Среднем По-
волжье с эпохи неолита образовалась промежуточная зона разных куль-
тур. Восточные и западные культурные влияния регулярно перемешива-
лись на вятско-поветлужском междуречье, и этногенез марийцев произо-
шел под влиянием этих процессов. В этногенезе горных марийцев играла
определяющую роль смешанная дьяково-городецко-ананьинская куль-
тура, в этногенезе же луговых кроме них участвовала и азелинская. В ко-
нечном счете можно сказать, что синтеза Климы входит в круг работ,
утверждающих теорию автохтонности марийцев на вятско-поветлужском
междуречье, она является более уточненным вариантом этой теории. Ни-
китина (1996) в своих выводах приводит положение, сходное с мнением
Климы, она подчеркивает только, что городецко-ананьинская культура бе-
рет свое начало в волго-окском междуречий, а азелинская восходит к при-
камским началам.
Однако к реконструкции этногенеза марийцев Клима не использо-
вал все имеющиеся источники. Вовлечение в исследование более широкого
круга источников дает возможность прийти к новым результатам, преодо-
левая взаимоисключаемость двух прежних теорий.
Источники с этнонимом «черемис»
Тот народ, который известен европейским языкам вслед за русским под

названием черемисский, называет себя этнонимом мари. Этноним мари
в письменных исторических источниках не встречается. А этноним черемис
встречается в двух, близких по хронологии, но в то же время независимых
друг от друга памятниках: в русских летописях и в письме хазарского
кагана Йосифа. Этот же этноним сохранился в современном русском, чу-
вашском и казанско-татарском языках в разных фонетических вариантах.
Сведения «Повести временных лет» относительно ранней истории
восточных славян признаются историками достоверными, независимо
от того, что этот источник сохранился в поздних списках. Шахматов еще
в начале века (1908) провел подробный филологический анализ этого
памятника и реконструировал порядок редакции его самых ранних частей.

296 КЛА РА А Д Я ГА Ш И
По мнению Ш ахматова, вводная, не датированная часть начального сво-

да, в которой встречается этноним черемис, была составлена в середине
XI в. Таким образом, нижеследующий отрывок текста отражает реальные
исторические условия середины XI в.
«На БЪлЪозЪрЪ сЬдять Весь а на Ростовскомъ 03epk МЪря а на
КлещинЪ езерЪ сЬдять МЪря же. А по O u t pkirk где втечеть
в Волгу М урома языкъ свой и Черемис свой языкъ Мордъва
свой языкъ.» (ПСРЛ, т. I, с. 5)
Другой источник, упоминающий черемисов - это письмо хазарского
кагана Йосифа, которое сохранилось в двух редакциях. Одна простран-
ная, другая сокращенная, и обе они восходят к третьему оригиналу. В этом
письме хазарский каган отвечает на письменные вопросы Хасдая Ибна
Шапрута, министра Кордованского халифата, заданные по поручению ха-
лифа. Так как кордованский халиф жил с 912 до 961 г., их переписка отно-
сится к периоду ранее 961 г.
Некоторые исследователи в первой половине века ставили под сом-
нение достоверность ответов хазарского кагана. Но считать данный источ-
ник фальсификатом можно в крайнем случае только относительно к X в.,
поскольку с конца XI в. он уже был известен. Это означает, что письмо ха-
зарского кагана Йосифа могло передать достоверную информацию XI в.
П р о ст р ан н ая и сокращенная редакция письма были опубликованы
в критическом издании П. К. Коковцовым (1932). Из критического изда-
ния цитируется отрывок в русском переводе (с. 98):
«Я тебе сообщаю, что живу у реки по имени Итиль а)2 в конце
реки Г-р-гана б) подле реки примыкающей к морю Г-р-гана.
Начало этой реки обращено к востоку на протяжении 4 меся-
цев пути. У этой реки расположены многочисленные народы
в селах и городах, некоторые в открытых местностях, а другие
в укрепленных (стенами) городах. Вот их имена: Бур-т-с, Бул-
г-р, С-вар, Арису, Ц-р-мис, В-н-н-тит, С-в-р, С-л-виюн».
Этнонимы (как и весь текст) написаны еврейским (квадратным) пись-
мом. Это письмо, как известно, фиксирует согласные звуки, но в нем есть
и три буквы (алеф, йод и вав), которые могут быть использованы для пере-
дачи гласных3.
Чтение этнонимов, встречающихся в письме кагана, не представляет
трудности, потому что они хорошо известны из других исторических ис-
точников. Первые три названия читаются как burtas, bulgar и suvar - это
2 П од пунктами а) и б) цитируется разница в двух редакциях письма. Название

Г-р-ган встречается и у И бна Фадлана, это арабское название города Ургенча, торго-
вого центра на Ш елковом пути, возле А ральского моря.
‫ ג‬Благодарю Л асло Хуняди за помощ ь, оказанную при разборе цитированного
еврейского текста.
К ВОП РОСУ О Б Э ТН О ГЕ Н Е ЗЕ М А РИ Й Ц Е В 297
в X и XI вв. народности волжско-булгарской империи. Bulgar и suvar в то

же время означали и названия городов. Форма arisu отождествляется с са-
моназванием мордовцев erzäd. Написание c-r-mis имеет две возможности
чтения: cermis и ceremis. (В еврейском письме не имеется буквы для
согласного с, поэтому с может передать и звук с чужого происхождения.)
После этого слова следующее читается как venentit, в чем обнаруживается
испорченное написание названия восточнославянского племени vçticb,
(позже известное как вятич). За ним следует sever, что соответствует на-
званию другого славянского племени severjan, только без славянского суф-
фикса -jan. Наконец форму s-1-vijun можем читать как slavjan. Правиль-
ность чтений еврейских написаний разных этнонимов отчасти подтвер-
ждается отрывком из Никоновской летописи:
«... А Казари имаху дань на Полянехъ, и на СЬверянехъ, и на
Вятичахъ ...» (ПСРЛ, т. 9, с. 8.)
Этноним черемис сохранился кроме русского в современном чуваш-
ском языке как sarmäs и в казанско-татарском как cirmës. В научной лите-
ратуре известны разные попытки определения этимологии слова. Булгаро-
тюркским считал происхождение слова М. Жираи (1937, репринт 1994).
Жираи связывал форму ceremis с древнебулгарским словом cerig ‘войско’
и предполагал, что ceremis имеет значение ‘военный народ’. Но он сам
признает, что объяснить компонент -mis не может. То, что Жираи не смог
объяснить происхождение компонента -mis на (булгаро-)тюркском осно-
вании, не случайно. Принципиально этот элемент воспринимается или как
отыменный суффикс, или как самостоятельное слово. Но ни в современ-
ном чувашском языке, ни в письменных памятниках древнетюркского нет
следа существования словообразовательного суффикса -mis. Компонент
-mës/-mâs, по данным Обратного словаря чувашского языка (см. Сергеев
и соавт. 1985), встречается в чувашском языке очень редко, в заимствован-
ных словах (cp. namäs ‘стыд’ <— тат. <- перс, närnüs ‘слава’, см. Шернер
1977, с. 43, 69; где второй слог, как видно, восходит к -mus, а не к -mis. Дру-
гой пример чув. kâmâs ‘кумыс’ <- тат. qumïs ‘то же’, где слог -mi's относится
к корню) и в одном слове неизвестного происхождения (ср. simës
‘зеленый’, см. Федотов 1996, II, с. 50-51). В обратном индексе древнетюрк-
ского словаря Клоусона (Клоусон 1981), а также в монографии Эрдаля
о древнетюркском словообразовании нет ни одного случая фигурирова-
ния словообразовательного суффикса -mis тюркского или иного происхо-
ждения. А чтобы принять форму ceremis за сложное слово, нужно было бы
иметь самостоятельное тюркское слово mis с приблизительным значением
‘народ, группа людей’, что вообще невозможно, так как оригинальные
тюркские слова с согласного т не начинаются.
Следующее мнение о происхождении слова ceremis - это мнение
Ф. И. Гордеева (1964). Гордеев выражает мысль о том, что слово *черемис
является сложным образованием, первый компонент которого (cere) вое-

298 КЛА РА А Д ЯГА Ш И
ходит к иранскому корню, имеющему параллели, как cärun- ‘жить’ в осе-

тинском, саг- ‘бродить, кочевать’ в древнеиранском, car- ‘id.’ в древнеин-
дийском. Второй компонент «вероятно означало понятие ‘человек’, и гене-
тически оно связано с иранским заимствованием, встречающимся во всех
финно-угорских языках, ср. фин. mies ‘муж, человек’, эстон. mees самона-
звание манси, санскрит, m ams ‘человек’», (ук. соч., с. 212-213). Таким об-
разом, ceremis означает, по Гордееву, ‘кочевой человек’. Гордеев осно-
вывает свою идею на той гипотезе, что предки марийцев всегда обитали
на территории современной Марийской Республики, и относит появление
этнонима ceremis к II тысячелетию до н. э., когда по памятникам археоло-
гии в этом регионе по соседству с тогдашними «марийцами» жили ирано-
язычные племена. К сожалению, идея Гордеева строится на не доказанных
и в языковом, и в историческом отношении гипотезах и не может являться
решением вопроса.
В 1978 году была опубликована книга П. Хайду и П. Домокоша,
в которой (с. 46) обновляется гипотеза булгаро-тюркского происхождения
этнонима ceremis, но она опять же остается без доказательства.
Другое объяснение происхождения слова ceremis предлагает И. Г. Ива-
нов (1978). Он предполагает, что слово ceremis возникло у финноязычных на-
родов (меря и мурома) и состоит из двух частей: cere с гипотетическим
значением ‘человек’ и -mis (- прибалтийско-финск. mies, mees ‘человек’). Но
поскольку никаких следов этих языков не осталось, отождествление со-
ставных элементов современного этнонима с конкретными архетипами
языков-источников не имеет оснований.
Явный методологический недостаток гипотезы Иванова - отсутст-
вие реально существующих данных - был замечен С. Я. Черных в его ста-
тье, опубликованной в 1980 г. Черных, желая доказать правильность эти-
мологического толкования Иванова, ищет доказательства существования
компонента cere в марийской топонимике. Он трижды находит в ревизских
сказках 1722-1727 гг. написание «Черенур черея» (ук. соч., с. 38), на основа-
нии которого cere считает родовым названием или именем родоначальни-
ка. Далее, он приходит к выводу, что форма cere подверглась многосто-
ронней фонетической трансформации, и все варианты марийской топони-
мики с компонентом сога, сига, сога, сига восходят к форме cere.
К сожалению, работа Черных опять же не выдерживает критики. Он
сам не замечает, что его аргументация построена на том, что cere является
словом марийского происхождения, а не элементом из другого финно-
угорского языка (как это предполагал Иванов). В этом случае в прото-
марийском языке должен был бы иметься архетип слова cere ‘человек’
и должны были бы существовать его современные формы, распространен-
ные во всех марийских диалектах, чего нет (ср. Берецки 1992). Конечно,
никто не сомневается в том, что в XVIII в. существовал родоначальник
с именем Чере, но возвести это слово в неизменном виде хотя бы к XI в. и
перепутать с татарским заимствованием сога, сига, сога, сига - грубая

К ВОП РОСУ О Б Э ТН О ГЕН ЕЗЕ М А РИ Й Ц Е В 299
ошибка, не говоря о том, что компонент -mis со значением ‘человек’ так

и не известен марийскому языку.
Наконец, иную этимологию предложил Д. Е. Казанцев в своей книге
о формировании марийских диалектов (1985). Казанцев (ук. соч., с. 9-10),
ссылаясь на высказывание Перетятковича (1877), утверждает, что слово
ceremis имеет мордовское происхождение. Оно расчленяется на корень се-,
связанный с мордовским чи ‘солнце’, и суффикс -ре-, который образует в
современных мордовских языках имя прилагательное. A mis, по его мне-
нию, имеет связь с прибалтийско-финскими формами mies, mees, и его су-
шествование в древнемордовском и древнемарийском языках подтвер-
ждает детерминатив -мае (-маш) в местных названиях. Таким образом,
форма ceremis имеет значение ‘солнечный человек’. Толкование Казанцева
не учитывает факты исторической морфологии мордовских языков, а так-
же не дает убедительных доказательств былого существования слова mis
‘человек’ в языках-предках мордовских.
В заключение этимологического экскурса можно прийти к выводу,
что, по имеющимся данным, этимология этнонима ceremis неизвестна.
Можно знать только, что он не являлся самоназванием марийцев. Несмо-
тря на то, что слово не имеет достоверной этимологии, соответствия этно-
ниму в других языках хорошо объясняют друг друга. Современный чуваш-
ский sarmâs восходит к древнебулгарскому cärmis. Форма cärmis существо-
вала до X в.: в X в. фонема с, по свидетельству булгарских заимствований
в протопермском языке (ср. Вихман 1903; Редеи-Рона-Таш 1972; Насибул-
лин 1992), изменилась в s. Татарская форма cirmês тоже восходит к древне-
булгарскому cärmis, только она была заимствована из другого диалекта,
в котором фонема с не спирантизировалась в s. (О наличии такого диа-
лекта мы знаем по волжско-булгарским эпитафиям, ср. Рона-Таш 1976.)
В этом случае первый слог древнебулгарской формы в татарском законо-
мерно развился в cir-, а на произношение второго слога -mis повлиял очень
употребительный татарский суффикс причастия прошедшего времени
-mis, таким образом из -mis развился -mes. Все это значит, что для данного
этнонима мы имеем два структурных варианта: первый - двухсложный,
который представлен в поволжско-тюркских языках, а второй - трехслож-
ный, он известен из древнерусского памятника. Поскольку формы cärmis ~
ceremis не этимологизируются ни на основании тюркских, ни на основа-
нии славянских языков, обосновано предположение, что данный этноним
попал и в тюркские, и в древнерусские диалекты из третьего источника.
Если бы слово в языке-источнике было бы трехсложным, то и древнебул-
гарские диалекты заимствовали бы его как трехсложное. (В оригинальных
трехсложных словах тюркского происхождения гласный е ~ ä второго ело-
га не выпадает, ср. ДБ täkänä ‘корыто’ > чув. takana.) А если в языке-источ-
нике этноним имел двухсложную структуру, то все данные могут объяс-
няться без проблем. В древнерусский язык в X-XI вв. могла попасть двух-
сложная форма cärmis, но эта форма не соответствовала древнерусским

структурным законам. В это время (до падения редуцированных) в древ-

нерусском языке еще существовала тенденция к открытому слогу. Так как
форма cärmis состоит из двух закрытых слогов, было необходимо ее пре-
образование в слово с открытыми слогами. Преобразование первого за-
крытого слога в открытый осуществилось путем вставления второго глас-
ного е по аналогии с явлением восточнославянского полногласия4. Второй
закрытый слог стал открытым в результате прибавления к корню редуци-
рованного гласного заднего ряда. Это значит, что из первичной формы
cärmis возникла вторичная форма ceremisb соответственно структурным
правилам древнерусского языка.
Суммируя сведения источников с этнонимом «черемис», можно еде-
лать следующие выводы. В X-XI вв. существовал исторический народ, ко-
торый имел контакты с разными булгаро-тюркскими (древнечувашским,
волжско-булгарским и хазарским) народами. Он известен булгаро-тюркам
под названием cärmis. Этноним cärmis попал в древнерусский язык до
XI в., возможно, из булгаро-тюркских диалектов. Составитель вводной
части «Повести временных лет» перечисляет народы, жившие на окраинах
Киевской Руси в середине XI в. В перечне народы следуют друг за другом
по направлению с юго-запада к северо-востоку, начиная с тех, которые
ближе к Киеву. Согласно этому перечню, черемисы обитают в волго-ок-
ском междуречье между народами «Мурома» и «Мордва», не соприка-
саясь непосредственно со славянскими племенами. Хазарский источник
тоже перечисляет народы, живущие на окраинах хазарской империи; в этом
случае исходным пунктом является столица Хазарии Итиль в устье Волги,
и перечень ведется с юго-восточного направления к северо-западному.
Этот источник созвучен с древнерусским в том, что эрзя-мордовцы и чере-
мисы являются соседями. Хотя в хазарском источнике точное место оби-
тания двух народов не указывается, дается их релятивная локализация: эрзя
и черемис находятся между восточнославянскими племенами и народно-
стями волжско-булгарской империи. Их ближайшие соседи - вятичи.
Археологические источники
Перед тем, как использовать археологические источники при реконструк-

ции этногенеза любого народа, следует напомнить о том, что по своему
характеру этот тип памятников дает сведения только о материальной и ду-
ховной культуре какого-либо этноса, о его бывшем наличии или отсутствии
4 Восточнославянское полногласие появилось в восточнославянском диалекте

праславянского языка тож е в результате стремления к открытому слогу: праславянская
структура СегС (где С - это лю бой твердый согласный) изменилась в СегеС, например,
прасл. *bergb > beregb. Обыкновенно изменение сочетания ег > ere осущ ествилось между
твердыми согласными, но известен пример, в котором оно произош ло меж ду мягкими:
прасл. * сег.ппа > др. русск. сегеЗьпа, ср. Трубачев 1977.

К ВОП РОСУ О Б Э ТН О Г Е Н Е ЗЕ М А РИ Й Ц Е В 301
в определенной географической среде и в определенный исторический пе-

риод. Но сам по себе он только в редких случаях пригоден для прямого
отождествления представителей археологических культур с современными
этническими группами.
Что касается археологического описания исторических культур, об-
наруженных на современной территории марийского этноса, можно со-
гласиться с Архиповым (1995, с. 31), что оно в основных чертах решено
(см. важнейшую литературу выше), но недостаточно изученной осталась
проблема контактных зон. Основное значение имеют здесь временные
пределы самостоятельного существования культуры, отнесенной к народу
меря, а также границы географического распространения древнерусского
населения до X II1 в.
Об истории и местонахождении мери новая синтеза на основании
археологических памятников опубликована Леонтьевым (1996). Конечный
итог этой книги, важный с точки зрения нашей проблемы, можно цитиро-
вать одним предложением: «Дальнейшее освоение финских земель за пре-
делами центральных исторических районов Волго-Клязьминского между-
речья происходило на протяжении второй половины XI-начала XIII в. и
было связано с расширением границ Ростово-Суздальского княжества»
(ук. соч., с. 288). К подобному выводу пришла Бернштам (1973) в результа-
те анализа летописных данных. А в новейшей монографии по вопросу
колонизации северных окраин Древней Руси в XI-XIII вв. Макаров (1997)
указывает на то, что период X I-X III вв. был временем интенсивного рас-
селения прибалтийско-финских, поволжско-финских и славянских групп
на Севере, что можно понимать как процесс «жестких барьеров, закрепля-
вших водные пути и крупные массивы прилегающих к ним земель за нов-
городцами или ростовцами» (ук. соч., с. 167).
Заимствованная лексика марийских диалектов
При вовлечении заимствованной лексики в круг источников, указываю-

щих на локализацию носителей прамарийского языка, мы исходим из того
принципа, что в лексическом составе любого языка имеются элементы, за-
имствованные в разные эпохи из соседних языков в результате их непо-
средственного контактирования. Применяя историко-сопоставительный
метод языкознания, мы можем назвать непосредственных соседей данного
языка в разные периоды его существования, а значит и определить реля-
тивную локализацию его обитания.
Исследование лексики марийских диалектов имеет более чем сто-
тридцатилетнюю историю. Эта история начинается с деятельности Регули,
включает исследования таких ученых, как Буденц (1864), Вихман, Беке
(1997-1998), Рясянен (1920), Итконен (1969), Берецки (1992, 1994). Опреде-
ляющее значение работ перечисленных ученых заключается в том, что бы

ли осуществлены сбор и издание марийской диалектной лексики и фольк-

лорных памятников, а также лингвистическое описание диалектов. Все же
в разграничении разных лексических пластов марийской лексики по про-
исхождению долго не имелось реальных результатов, потому что не был
выделен марийский лексический фонд финно-угорского происхождения,
история которого служила бы основанием для определения внутренних
тенденций развития языка до дифференциации прамарийского единства.
(Очень ценны работы Рясянена и Итконена по постановке проблемы о ма-
рийско-булгаро-тюркских языковых связях, но поскольку Рясянен исходил
из предположения родства урало-алтайских языков, а Итконен основал
свое исследование на гипотезе волжско-финского праязыка, ошибочный
теоретический подход не позволил им установить внутренние закономер-
ности истории марийского языка.) В недавнем прошлом выделение марий-
ского лексического фонда финно-угорского происхождения и описание фо-
нетических процессов, происшедших после отделения прамарийцев от фин-
но‫־‬волжского единства, до дифференциации прамарийского единства было
проведено Берецки (1992, 1994). Благодаря его исследованиям науке стали
известны те фонетические изменения, которые происходили во время пра-
марийского единства, и стало возможно определить те изменения, кото-
рые произошли после его распада.
Благодаря исследованиям Берецки теперь мы уже знаем, что в ма-
рийских диалектах имеется 496 слов финно-угорского происхождения.
Только незначительная часть их совпадает со словами финно-угорского
происхождения мордовских языков. (Лексический фонд финно-угорского
происхождения мордовских языков был выделен в работах Керестеша
1985, 1986).
Предыдущие исследования доказали и то (см. Берецки 1977, 1992),
что до распада прамарийского единства прамарийский язык позаимство-
вал значительное количество прапермских элементов. (Под прапермским
понимается исторический период развития пермских языков, когда языки-
предки удмуртского и коми образовали единство, просуществовавшее
до X в. Определение абсолютной хронологии начала распада протоперм-
ского единства стало возможным на основании анализа булгаро-тюркских
заимствований пермских языков, см. Редеи-Рона-Таш 1972; Насибуллин
1992.)
В марийских диалектах очень много заимствований, которые были
определены прежними исследованиями (Рясянен 1920; Федотов 1968; Ит-
конен 1969; Федотов 1990) как пласт булгаро-тюркских заимствований.
Но уже Рясянен отметил (1920), что внутри этой группы много таких слов,
которые, хотя и имеют соответствия в чувашском языке, по фонетическим
и структурным причинам не могут быть словами тюркского происхожде-
ния. Они являются такими элементами, которые в очень похожей форме
присутствуют и в марийских, и в чувашских диалектах, но в марийском
языке они не имеют финно-угорского происхождения, а в чувашском не

могут быть тюркского происхождения. Количество булгаро-тюркских за-

имствований в марийских диалектах и без того значительное, но судя по их
фонетическим особенностям, они попали туда не раньше конца XIII в.5
Важное место занимает среди заимствованной лексики марийских
диалектов слой русских заимствований. Недавние исследования по уста-
новлению хронологии самых ранних русско-марийских языковых контак-
тов (см. Адягаши 1994) доказали, что ранний элемент древнерусского языка
с носовым гласным (*kçzel’a ) перешел в марийскую лексику через булга-
ро-тюркское посредство в XIII в. Это значит, что с тех исторических вре-
мен, когда русские летописи упоминают черемисов вблизи границ Киевской
Руси, и с более ранних в марийском языке не осталось следа контактиро-
вания с древнерусскими диалектами. Саваткова (1969), изучая русские за-
имствования марийского языка, считает, что русско-марийские языковые
контакты начались с XVI в.
Перечисляя все источники6, необходимые для локализации марийцев до та-

тарского нашествия, попытаюсь при решении проблемы интерпретиро-
вать имеющиеся сведения.
Так как два независимых друг от друга источника утверждают, что
в X -X I вв. существовал исторический народ под названием «черемис»,
сомневаться в этом нет причины. Этот народ обитал, по всей вероятности,
на территории между Киевской Русью и страной волжских булгар, извест-
ной как «Magna Bulgaria», вместе с другими финно-угорскими народами
по нижнему течению р. Оки. Но черемисский народ так же ассимилировал
с русским после XI в., как и чудь, меря и мурома, или участвовал в рас-
селении поволжско-финских народов на Севере.
В вятском и поветлужском междуречье жил другой народ, который
сформировался из разных этнических компонентов, о котором историче-
ские памятники не упоминают, но о нем дают знать ананьинская, дьяково-
городецкая и азелинская археологические культуры. Нет причины сомне-
ваться в выводах археологов, соответственно которым во второй полови-
не I тысячелетия единство этих народов прекращает свое существование,
и дальше они развиваются по двум ветвям.
С середины XII в. Киевская Русь тоже распадается на княжества, не
имеющие между собой тесных связей. Новгородское и Ростово-Суздаль-
ское княжества распространяют свои границы на север и северо-восток,
и с этого же времени между русскими и волжскими булгарами фактически
5 Для определения их хронологии дает возможность группа среднемонгольских

заимствований в марийских диалектах, которая попала туда через булгаро-тюркское
посредство (см. Рона-Таш 1982).
6 Среди источников я не использовала топонимический материал, потом у что,
на мой взгляд, вовлечение разнообразны х топонимов для доказания локализации лю-
бого этноса в Поволжье может иметь место только после создания базы данны х всего
региона и после проведения структурного и этимологического анализа территориаль-
ных типов на основании всего материала.
304 КЛА РА А Д Я Г А Ш И
оказывается народ ананьинской, дьяково-городецкой и азелинской куль-

тур. Так как волжские булгары издавна торговали с русскими (об этом
пишет уже и Ибн Ф адлан в 922 г.), и русские, и волжские булгары, а также
хазары по традиции хранили знания о том, что между ними живет народ
под названием «черемис», этноним «черемис» был перенесен для обозна-
чения народа ананьинской, дьяково-городецкой и азелинской культур.
Название «черемис» стало топосом. С этого времени оно обозначало не
конкретный народ со своим определенным языком, а жителей любого
происхождения с любым языком, обитавших до татарского нашествия
между русскими и волжско-булгарскими территориями.
Ранняя история того народа, который называет себя этнонимом ма-
ри, независима от исторических и вторичных черемисов. Прамарийцы,
после того, как они (без предков мордовцев) вышли из финно-волжского
единства, прошли долгий путь своего самостоятельного развития. На
позднем этапе прамарийского периода, до X в. прамарийское и праперм-
ское население опять оказалось близко друг к другу, в непосредственном
соседстве. Точная географическая локализация - задача будущих археоло-
гических исследований.7 В качестве гипотезы можно предположить, что
прамарийцы могли жить к западу от прапермского народа. После распада
прапермского единства прамарийско-удмуртские контакты не прекрати-
лись. В 1236-м году, во время татарского нашествия, самый тяжелый удар
монголов был сконцентрирован на волго-камском междуречье. От глав-
ного нападения монголов прамарийцы ушли в юго-западном направлении
и поселились на территории проживания народов ананьинской, дьяково-
городецкой и азелинской культур, в вятско-волжско-поветлужском между-
речье, защищенном лесами и болотами. С того момента, как прамарийцы
появились в регионе между русскими и волжско-булгарскими террито-
риями, соседи тоже стали их называть топосом «черемис».
Группа волжско-булгарского населения (будущие чуваши) тоже ушла
от монголов из волго-камского междуречья и, появившись на правом
берегу Волги, вместе с прамарийцами ассимилировала местное население.8
Таким образом становятся понятными и сведения русских письменных ис-
точников (см. Димитриев 1964), которые с XVI в. называют «черемисами»
то жителей чувашеязычных территорий, то настоящих горных марийцев
тех времен.
Из всего вышесказанного следует, что в этногенезе марийцев участ-
вовал и местный компонент поветлужско-вятского междуречья, сформи-
ровавшийся в первой половине I тысячелетия и распавшийся на две части
7 Связям меж ду волжскими булгарами и финно-уграми посвящена статья Каза-

нова (1997). И сследования Казакова проведены на основании археологического мате-
риала и сконцентрированы на следах угров в Великой Булгарин, но в его статье содер-
жатся сведения о локализации разных археологических культур, отнесенных к другим
финно-угорским народностям X I-X I1 вв.
8 Об обитании на правом берегу Волги других групп волжских булгар свидетель-
ствую т археологические памятники.

уже во второй половине этого же столетия, а также пришлый элемент,

ушедший со своей родины вследствие татарского нашествия в середине
XIII века. Языковая принадлежность местного компонента неизвестна,
языковым памятником является субстратная лексика в марийском и чу-
вашском языках. Прамарийцы во время их пришествия в поветлужско-
вятское междуречье в языковом отношении еще образовывали единство,
но так как они поселились на территории проживания народа, уже имев-
шего два разных, территориально отдаленных друг от друга диалекта, то
при ассимилировании двух диалектных групп их язык тоже стал распа-
даться на диалекты.
Литература
(Адягаши 1994) Agyagäsi, К., Überlegungen zur Differenzierung der tscheremissischen Mundar-
ten anhand von *kqzel’a ‘Hanfhocke’. UAJb NF 13, pp. 29-36.
(Архипов 1967) Архипов, Г. А., П роисхож дение марийского народа по археологическим
данным (с I. тыс. н. э.). В кн. Происхождение марийского народа. Отв. ред. Х леб-
ников, А . В. Йошкар-Ола.
(Архипов 1976) А рхипов, Г. А ., Марийский край в памятниках археологии. Йош кар-Ола.
(Архипов 1976а) Архипов, Г. А ., К вопросу о городецко-азелинских контактах в связи
с проблемой происхождения марийцев. В кн. Древние и современные этнокулъ-
турные процессы в Марийском крае. Отв. ред. Хлебников, А . В.Йошкар-Ола.
(Архипов 1995) Архипов, Г., Этногенез финно-угорских народов Поволжья: современ-
ное состояние и задачи изучения. Финно-угроведение 1, с. 30-41.
(Бадер 1951) Бадер, О. Н ., Древнее Поветлужье в связи с вопросами этногенеза мари и
ранней истории Поволжья. Советская этнография 2, с. 3-32.
(Беке 1997-1998) Веке, Ö., Mari nyelvjârdsi szôtâr I-II-III. (Bibliotheka ceremissica IV, IV2 IV3)
Unter Mitarbeit von Zs. Velenyâk und tJözsef Erdôdi. Neu redigiert G. Bereczki. Bearbei-
tet von M. Kuznecova. Herausgegeben von Jänos Pusztay. Savariae.
(Берецки 1974) Берецки, Г., Сущ ествовала ли праволжская общ ность финно-угров?
A L H 24, с. 81-85.
(Берецки 1977) Bereczki, G., Permi-cseremisz lexikâlis kôlcsônzések. NyK 79, pp. 57-77.
(Берецки 1992) Bereczki, G., Grundzüge der tscheremissischen Sprachgeschichte II. (SUA 34)
Szeged.
(Берецки 1994) Bereczki, G., Grundzüge der tscheremissischen Sprachgeschichte I. (SUA 35)
Szeged.
(Бернштам 1973) Бернштам, Т. А ., Роль верхневолжской колонизации в освоении
русского севера (IX -X V вв.). В кн.: Фольклор и этнография русского севера. Отв.
ред. П утилов, В. Н., Чистов, К. В. Ленинград, с. 5-29.
(Буденц 1864) Budenz, J., Cseremisz tanulmânyok. NyK 3, pp. 397-470.
(Вихман 1903) Y. Wichmann, Die tschuwaschischen Lehnwörter in den permischen Sprachen.
MSFOu 21.
(Гено 1981) Gheno, D., Megjegyzések a mordvin és cseremisz közötti grammatikai egyezésekrol.
NyK 83, pp. 114-121.
(Гордеев 1964) Гордеев, Ф. И ., К истории этнонима черемис. Труды МарНИИ, вып. 18,
с. 207-213.

Г орю нова 1967) Г орю нова, Е. И ., Меря и мари. Происхождение марийского народа.
Йошкар-Ола, с. 7 0 -7 8 .
(Д ечи 1965) Décsy, Gy., Einführung in die finnisch-ugrische Sprachwissenschaft. Wiesbaden.
(Димитриев 1964) Димитриев, В. Д ., О значении этнонима «черемисы» в русских и за-
падноевропейских источниках XVI начала XVIII веков. Ученые записки ЧНИИ,
вып. 27, с. 118-132.
(Ж ираи 1994) Zsirai, М., Finnugor rokonsâgunk. A z 1937-es kiadäs reprintje. A z uräli nyelv-
rokonainkkal kapcsolatos legüjabb ismeret- és forrâsanyag rövid ôsszegzésével közreadja
Zaicz Gäbor. Budapest.
(И ванов 1978) Иванов, И . Г., Еще раз об этнониме «черемис». В кн. Вопросы марийской
ономастики. Н аучны й редактор Галкин, И . С. Йош кар-Ола.
(И тконен 1969) Itkonen, Е., Thesen und Antithesen in der finnisch-ugrischen Vokalforschung.
UAJb 41, 212-251.
(К азаков 1997) Казаков, E. П ., Волжская Булгария и финно-угорский мир. Finno-Ugrica
1, с. 33-53.
(К азанцев 1985) Казанцев, Д . Е., Формирование диалектов марийского языка. Йошкар
Ола.
(Керестеш 1986) Keresztes, L., Geschichte des mordwinischen Konsonantismus II. Etymologi-
sches Belegmaterial. (SU A 26) Szeged.
(Керестеш 1987) Geschichte des mordwinischen Konsonantismus I. (SUA 27) Szeged.
(Клима 1996) Klima, L., The Linguistic Affinity o f the Volgaic Finno-Ugrians and their Ethno-
genesis. (Early 4th Millennium B.C. - Late 1st Millennium A.D.). Studia historica Fenno-
Ugrica I. Oulu.
(К лоусон 1981) Clauson, sir G., An etymological Dictionary of Pre-Thirteenth Century Turkish.
Indexes Vol. II. (SU A 15) Szeged.
(К озлова 1978) К озлова, К. И ., Очерки этнической истории марийского народа. Москва.
(К оковцов 1932) Коковцов, П. К., Еврейско-хазарская переписка в X веке. Ленинград.
(Леонтьев 1996) Леонтьев, А . Е., Археология Мери. К праистории Северо-Восточной
Руси. Москва.
(М акаров 1997) М акаров, Н . А ., Колонизация северных окраин Древней Руси в X I-X II
веках. По материалам археологических памятников на волоках Белозерья и Пооне-
жья. Москва.
(Н асибуллин 1992) Н асибуллин, Р. Ш ., Булгаризмы и их отнош ение к вопросу о вре-
мени распада общ епермской языковой общ ности. Вордскем Кыл, вып. 2, с. 81-95.
(Никитина 1996) Никитина Т., К вопросу о б истоках марийского этноса. Финно-угро-
ведение 2, с. 50-66.
(Перетяткович 1877) Перетяткович, Г., Поволжье в X V -X V I веках. (Очерки из истории
края и его колонизации). Москва.
ПСРЛ: Полное собрание русских летописей, том I. М осква-Ленинград, 1962.
(Редеи-Рона-Т аш 1972) Rédei, K.-R6na-Tas, А., A permi nyelvek öspermi kori bolgär-török jö-
vevényszavai. NyK 74, pp. 281-298.
(Редеи-Рона-Таш 1982) Rédei, K.-Röna-Tas, A., Proto-Permian and Votyak Loan-Words in
Chuvash. In: Röna-Tas, A. (ed.), Studies in Chuvash Etymology I. (SUA 17) Szeged, pp.
158-179.
(Рона-Таш 1976) Röna-Tas, A., Some Volga Bulgarian Words in the Volga Kipchak Languages.
In: Kdldy-Nagy, Gy. (ed.), Hungaro-Turcica. Budapest, pp. 169-175
(Рона-Таш 1982) Röna-Tas, A., Loan-words o f Ultimate Middle Mongolian Origin in Chuvash.
In: Röna-Tas, A. (ed.), Studies in Chuvash Etymology I. (SUA 17) Szeged, pp. 66-134.

К ВОП РОСУ О Б Э ТН О ГЕН ЕЗЕ М А РИ Й Ц Е В 307
(Рясянен 1920) М. Räsänen, Die tschuwassischen Lehnwörter im Tscheremissischen. MSFOu 48.

(Сааринен 1997) Сааринен, С., Структура и развитие волжско-финских языков. Финне-
угроведение 3, с. 49-52.
(Саваткова 1969) Саваткова, А . А., Русские заимствования в марийском языке. Йош кар-
Ола.
(Сергеев и соавт. 1985) Сергеев, Л. П .-Т и хонов, А. H .-С ергеева, Т. Н .-К унгуров, Р. К.,
Обратный словарь чувашского языка. Чебоксары.
(Серебренников 1967) Серебренников, Б. А ., П роисхождение марийского народа по дан-
ным языка. В кн. Происхождение марийского народа. Отв. ред. Хлебников, А . В.
Йошкар-Ола, с. 165-180.
(Серебренников 1985) Серебренников, Б. А., Существовала ли финно-волжская языко-
вая общность? Вопросы языкознания 2, с 10-21.
(Смирнов-Трубникова 1965) Смирнов, А. П .-Трубникова, Н. В., Городецкая культура.
Москва.
(Тойвонен 1952) Toivonen, Y. Н., Zur Frage der finnisch-ugrischen Urheimat. JSFOu 56.
(Трубачев 1977) Трубачев, О. H ., Этимологический словарь славянских языков, вып. 4.
Москва.
(Ф едотов 1968) Ф едотов, М . Р., Исторические связи чувашского языка с волжскими и
пермскими финно-угорскими языками. Чебоксары.
(Ф едотов 1990) Федотов, М. Р., Чувашско-марийские языковые взаимосвязи. Саранск.
(Ф едотов 1996) Этимологический словарь чувашского языка, I—II. Чебоксары.
(Х айду-Д ом окош 1978) Hajdü, P.-D om okos, P., Urâli ne'pek es nyelvek. Budapest.
(Х аликов-Архипов 1959) Халиков, А. Х .-А рхи пов, Г. А ., М арийская археологическая
экспедиция 1957 г. Труды Мар НИИ, вып. 14.
(Халиков 1969) Халиков, А . X., Древняя история Среднего Поволжья. Москва.
(Халиков 1970) Халиков, А . X ., Этническая принадлежность племен ананьинской
общности. Вопросы финно-угроведения, вып. 5.
(Халиков 1990) Халиков, А . X ., Об азелинско-ананьинской подоснове этногенеза ма-
рийского народа. В кн.: Средневековые древности Волго-Камья. Отв. ред. Архи-
пов, Г. А. (Археология и этнография марийского края, вып. 21) с. 7-18.
(Халиков 1991) Халиков, А . X ., Основы этногенеза народов Среднего Поволжья и При-
уралья. Казань.
(Черных 1980) Черных, С. Я ., М арийские антропонимы, связанные с этнонимами. В кн.:
Вопросы марийской ономастики. Научный редактор Галкин, И. С. Йошкар-Ола.
(Ш ахматов 1908) Ш ахматов, А. А., Разыскания о древнейших летописных сводах. Санкт-
Петербург.
(Ш ернер 1977) Schemer, В., Arabische und neupersische Lehnwörter im Tschuwaschischen.
Wiesbaden.
(Эрдаль 1991) Erdal, M., The Old Turkic Word-Formation. Turcologica 7 (hrsg. von Lars Johan-
son). Wiesbaden.
Klära Agyagdsi
Department o f Slavic Philology
University of Debrecen
H-4010 Debrecen Pf. 53
Hungary

REVIEWS
C h l o d w i g H . W e r b a , Verba indoarica. Die fer the Bibliotheca edition of Kuttanimata2 to the

primären und sekundären Wurzeln der San- Kävyamälä one; I would suggest to use C. R.
skrit-Sprache. Pars I.: Radices primariae. Wien Devadhar’s edition o f Amarusataka3, similarly
1997. Österreichische Akademie der W issen- I do not find R. P. Dwivedi’s edition o f Kali-
schäften. VII, 590 pp. dâsa the best choice. Nolens volens at places he
also turns to the Petersburger Wörterbuch for
This bulky volume is the Habilitationschrift, help (Ritusarhhära). As a result Werba presents
targeted to substitute for Whitney’s Roots.' He full documentation o f 178 roots. In case o f 56
enumerates the shortcomings of the latter: Whit- roots the inventory o f forms is missing. Here
ney’s opus is incomprehensive; only the Vedic he relies on the information o f the secondary
items are assessed; examples are taken from literature. Finally there are roots where the loci
secondary literature, mainly from Petersburger o f attestations are given only in selection. The
Wörterbuch‫׳‬, the selection o f items is made at list of secondary works is very impressive. All
random. Accordingly Werba’s first task was the at once I could mention only one significant
stock-taking of the research made since Whitney. gap, a substantial paper by J. Vekerdi.4
It comes from his report that W. Neisser, H. A crucial point is the definition o f primary
Oertel, K. Hoffmann and Toshifumi Goto carried roots where views o f scholars sharply differ.
out more or less work. Werba could get access Werba names a root primary if it surely or at
to unpublished collections o f material and then least with a great likelihood retains an Old-Aryan
did not spare any effort to process the enor- (urarisch) morpheme which can be recon-
mously vast amount o f possible data. He reas- structed in the same function and which by no
sessed 460 texts referred to by his predecessors means can be explained for a secondary root.
and presented variant readings or necessary cor- A root is called secondary if it has the being
récrions in parenthesis. He himself collected data (Vorhandsein) o f another root or rather the root
from 35 texts hitherto not taken into account.
Concerning the editions of these texts I would 2 Kuttanimatam Kävyam by Dämodara Gup-
occasionally raise few objections: I would pre- ta. Madhusudan Kaul (ed.) (1944), Calcutta
(Bibliotheca Indica 266).
3 Amarusatakam with Érngâradipikâ of Ve-
1 Whitney, D. W. (1885), The Roots, Verb- mabhüpàla. Critically ed. by Chintaman Ram-
forms, and Primary Derivatives of the Sanskrit chandra Devadhar (1959), Poona.
Language. A Supplement to his Sanskrit Gram- 4 Vekerdi, J. (1955), On polymorphic pres-
mar. Leipzig. ents in the Rgveda. AOH 12, pp. 249-287.
Akadémiai Kiadô, Budapest
3 10 REVIEWS
o f a stem belonging to it or also one o f a definite than it can be taken in hand as a tool o f learn-
noun conditio sine qua non for his existence ing in everyday usage. In this sense it does not
(Existenz) (p. 142). This definition is sound as replace old Whitney!
far as it is given from a special point o f view. Gyula Wojdlla
Nevertheless I must say that there are other ap-
proaches too. Palsule, w hose paper was con-
suited by Werba, says that “there is nothing The Supreme Wisdom of the
K l a u s G . W it z ,
sacrosanct about” setting up roots.5 Thieme Upanisads. An Introduction. Motilal Banarsi-
would rather cling to Pan ini and thinks that the dass, Delhi 1998. XXV + 558 pp.
naming Sanskrit roots have “nothing to do with
linguistic history, they are items of a purely syn- The voluminous book o f Klaus G. Witz is not
cronic description.6 really a general introduction to the Upanishads;
The structure o f this exquisitely written rather it attempts to familiarise the reader, in
book is quite clear-cut. It has the touch of the considerable depth, with some o f the central
meticulous care o f its author who is a grammar- teachings of the five oldest Upanishads. Its meth-
ian o f first rank and an excellent editor. The od is quoting and analysing focal teachings,
volume begins with a fascinating list of abbrevia- traditionally known as vidyäs, often comparing
tions and secondary literature. This part is fol- them with parallel texts from other Upanishads
lowed by a short explanation o f theoretical issues or other literature (including even Vedic hymns,
on which the whole work is based. The next and quite often sections o f the Brähmanas and
chapter is the inventory o f forms which is in Aranyakas). The selection is generally good and
fact the backbone of the volume. It consists of the wealth of information and insights is im-
three main chapters: primary roots with guna/ pressive; but it seems very difficult to explain
vrddhi Ablaut; primary roots with samprasarana the omission of the Sadvidyä, the rightly world-
Ablaut; primary roots without Ablaut. Each of famous ‘That art thou’ passage (6th chapter of
them is subdivided into three parts including the Chändogya upanisad.)
anit, set and vet roots respectively. The author concentrates on the evolution
At the end three very useful indices are ap- o f the fundamental vedantic/mystic insight, the
pended: Dhâtukramanï (Index o f roots), Artha- presence o f the absolute within everybody’s
änukiamani (Index o f meanings) and Udahara- self, and presents other Upanishadic doctrines
nànukramanï (Index o f examples). The volume (e.g. about breath as the highest vital function)
is closed with a list o f corrected loci o f attesta- as approaches or precursors to the central teach-
tion and quotations. ing. His analysis is generally informative and
This book is doubtlessly the greatest achieve- reliable although there are some blunders. When
ment in Sanskrit studies in recent years. It is on pp. 460-466 he analyses the substitution o f
a superb handbook from which generations of dahara gagana (subtle firmament, Mahä-närä-
linguists will certainly benefit a great deal. yarn upanisad XII.16) for dahara äkäsa (small
Strange to say, but it surpassed the original space, Chändogya upanisad VIII. 1-6) in the
aims of the author. It is more sophisticated and heart, he seems to be unaware o f the fact that
as to the inventory of forms less comprehensive though “in early Vedic literature diere was an
appreciable difference in meaning [...] Not only
kha, vyoman or äkäsa, but also other words like
5 Palsule, G. B. (1973), On setting up roots
gagana , ambara, viyat, nabhas, etc. are freely
for Sanskrit. In: Ghatage, A. M.-Dandekar, R.
N.-Mehendale, M. A. (eds), Studies in Histori- used in later literature as synonyms.” (Chakra-
cal Sanskrit Lexicography. Poona, pp. 95-100. barti 1996, p. 109).
6 Thieme, P. (1994), On M. Mayrhofer’s Perhaps a more serious objection (from
Etymologisches Wörterbuch des Altindoarischen. a scholarly point of view ) to the whole enter-
BSOAS 57, pp. 321-328. prise is its chosen approach to the Upanishads

REVIEWS 311
as “Supreme Wisdom literature”. It means that them. It became evident that it would not be
Witz presupposes that the Upanishadic sages enough to study the folk music o f a single na-
speak the Truth, so in order to understand them tion, but to collect folk music o f all the people
the best way is to listen to other texts and per- living there. Step by step the research, which
sons (including modern saints such as Satya Sai first aimed to find Cheremiss melody parallels
Baba) talking about the same absolute Truth. to Hungarian fifth-shifting melodies, turned to
This is hermeneutically unsound: even if there a complex survey o f the folk music o f a larger
were such a transcendental Truth, and even if it multinational area.
were unmistakably knowable to some gurus, Lâszlô Vikâr, the musician, and Gâbor Be-
and even if it were expressible in any human reczki, the linguist, carried out extended re-
language: how could a scholar tell the divine search work between 1957 and 1979 in the
inspiration from a clever fraud or a pathetic territory bounded by the rivers Volga, Kama
failure to find the higher reality? Of course, and Belaya. The fruits o f the research were in-
Witz knows and extensively uses the results of numerable publications, among which the more
modern scholarship; still I feel that his work complex are the musical monographs about
would be o f greatest use to those people who Cheremiss, Chuvash and Votyak folksongs.
wish to teach others the Supreme Wisdom, as The fourth and - according to the authors - last
seen in the classical Upanishads. volume o f these valuable series came to light
recently: the Tatar Folksongs.
They collected and recorded 580 tunes in
Reference the Tatar administrative area and 634 among
the Bashkirs, totalling 1214. The book presents
Chakrabarti, S. C. (1996), Àkâsa. In: Vatsyayan, less than half o f these tunes.
K. (gen. ed.), Kalâtattvakosa. A Lexicon of In the first chapter we read the short history
Fundamental Concepts o f the Indian Arts. o f the Tatar people and the Tatar language
Vol. Ill: Bäumer, В. (ed.), Primal Elements written by Gâbor Bereczki. This is followed by
- Mahàbhûta. Delhi, pp. 103-142. an overview o f Tatar folksong collections be-
tween 1816-1979. This study was not intended
Ferenc Ruzsa to be complete. As Lâszlô Vikâr wrote: “the
great bulk of old Tatar collections are probably
only available in Russia, or can be hopefully
Tatar Folk-
L â s z l ô V ik â r - G â b o r B e r e c z k i, unearthed there”. However, here one can find
songs. Akadémiai Kiadö, Budapest 1999,516 pp. the most important publication o f this field.
A particular stress was laid on Mahmud Nig-
The story began in 1906, when the wife of Yürö medzyanov, who organised Vikâr’s journeys
Wichman, the Finnish linguist, gave Béla Bar- and published more than 800 Tatar folksongs.
tök some Cheremiss folksongs collected by her. One of the most important parts o f the book
There was a startling resemblance between these is the classification o f the Tatar melody-forms
melodies and the Hungarian “fifth-shifting” and melody-types. In general the goal o f the
melodies. Bartök made up his mind to travel to classification is to place side by side the tunes
the land o f the Cheremiss, but because of World o f the same style, and within a style, those o f
War I, this plan was postponed forever. identical tune type. This task was done by
Some half a century later Zoltân Kodâly Lâszlô Vikâr at the usual high level. But, as he
sent one o f his students, Lâszlô Vikâr, to the himself had put it: “it is almost impossible to
Volga-Kama region to study Cheremiss melo- impose a strict system upon the living and con-
dies. It soon became clear that these melodies stantly changing musical material”. And really
live only along the Cheremiss-Chuvash border one might wonder if it is necessary to separate
and both Cheremiss and Chuvash people sing during classification the two-sectioned melo

312 REVIEWS
dies with AB form from two-sectioned ones identification code in the HAS, and in the notes
with ABB form, if the musical content o f A and we find other useful additional data. The texts
В is the same. were translated by Gdbor Bereczki, the eminent
An important finding is that there is a big Finno-Ugrian linguist. Ärpdd Berta helped him
difference between the folkmusic o f Moslem to understand the unusual meaning o f some
Tatars and that of the “kreshchonni” Tatars, who Tatar texts. The text under the notes were writ-
were converted to Russian Christianity. What is ten down in a simplified version o f the Setälä
more they do not sing the songs o f each other. transcription used in Finno-Ugrian linguistics
According to Vikdr the “kreshchonni” Tatars and not in a transcription used in Turkic schol-
preserve the oldest and very simple melodies. arship. The reason for this was to preserve the
This observation may strengthen the presenti- unity o f the four volumes. Anyhow the Tatar
ment that the large compass pentatonic melo- folklore collection o f H. Paasonen was also pub-
dies o f the Cheremiss, Chuvash, Tatar, Bashkir lished in this transcription.
might be of relatively later origin. At the same W e can read the original texts o f the Tatar
time we should remember that the large com- songs under the notes in handwriting as well.
pass and the pentatonisiem is very typical o f The Hungarian translation was made by Gdbor
Mongolian folksongs. Vikdr himself says: “The Bereczki with affection and care. He prepared
inventive profusion o f motives and easy pliable the beautiful translation o f folk texts in all the
ornamentation o f Tatar pentatonic is some- four volumes. The English prose translations of
where in the middle between the two ends o f the Tatar song-texts are correct.
Euro-Asia, even if all organic connections are In the Indexes we find a detailed list o f
disregarded”. Tatar folksongs in the Institute for M usicology
While in Hungarian folk music the role o f HAS (1968-1977), the Hungarian and English
the four-sectioned melodies is outstanding, this transcription o f the Tatar place-names and the
is not the case in the folk music o f many other list of songs grouped by localities. For musi-
peoples, e.g. the Tatars. Here half of the tunes cians the classification o f tunes by syllable count
is two- or three-sectioned and the other half is of the first stanza, by tone-set and the compass
four-sectioned. The most common scales are of tunes contains valuable information. Those
the so-, la- and do-pentatonic without semi- who are interested in more details concerning
tones. And an important observation: a quarter Tatar and Bashkir folk music find a basic bib-
o f the collected Tatar folksongs contains lower liography. At the end o f the volume we also
fourth or fifth (not quite complete) answer. find some photographs.
In Tatar lyrical songs four-line strophes pre- To sum it up, this Tatar volume is a great
dominate, with symbolical image of nature or asset for both ethnomusicologists and linguists.
some other content in the first two lines which As a result o f primary experiences and field
the second two lines concretise. This is very works it gives an authentic picture o f the living
characteristic o f the lyrical songs o f the Turkic, linguistic and musical processes, and combines
Hungarian and some other people. Among the the precise publication o f data with the seien-
genres we find lots o f lyrical songs, and a few tific classification. The value o f this work is
love songs, epic songs, soldier’s song, tunes con- multiplied by the fact that this is the fourth vol-
nected to folk costumes and calendar feasts and ume o f a series o f the folksongs and folk text of
religious songs. Surprisingly there is not a single the music o f the Turkic and Finno-Ugrian peo-
lullaby, children’s song or a mourning song pies living in the Volga-Kama region. We can
found. only hope that this excellent work w ill be con-
What is very important is that we are given tinued.
all kinds of data: the date and place o f the re-
cording, the name and age o f the singer, their Jänos Sipos

REVIEWS 313
Klaus Ka rttunen, India and the Hellenistic Macedonian) people participated, but not alone”
World. Helsinki 1997. (Studia Orientalia 83) (p. 1). In this sense one can also speak o f Hel-
IX, 439 pp. lenism in North-Western India.
Chapter two is devoted to Alexander’s ac-
Having read Karttunen’s India in Early Greek tivity in India, his image and the so-called Al-
Literature published in 1989 I took the new exander tradition abounding in legends. Karttu-
opus in hand with lively interest. nen departs from the conventional Alexander
The present work is the second o f a series picture and tries to achieve a balanced view
consisting of three parts; the third volume of that does not keep silent about the brutalities
which is under preparation. The work under re- which happened. At the same time he justly criti-
view is divided into seven main chapters, which cises those historians who consider Alexander
are further subdivided. It covers the period from a great rogue o f history. The anti-imperialist
the beginnings o f Alexander’s activity up to the feelings o f the older generation of Indian schol-
decline of local Greek dynasties in Bactria and ars is anyhow understandable and pardonable.
India. As to contents it deals with a wide range At any rate Alexander’s military achievements
o f topics running from politics to commerce, must be properly recognised. I find particularly
science to literature, religion to cultural contacts. fascinating the pages where Karttunen draws
Karttunen writes with a critical acumen and the route map o f the Macedonians in India. It
a very broad knowledge of facts. He has the ca- is really a masterpiece where he confronts ac-
pacity to carry out meticulous investigation in counts o f written sources and the sparse ar-
philological issues, to make sober assessments chaeological material (pp. 30-54). Not less
o f theories or hypotheses and simultaneously to brilliant is the treatment o f the wonderful sto-
offer an overall view and open new vistas for ries about Alexander’s meeting with naked as-
hover further research. He is one o f the few cetics, philosophers and the like.
scholars o f our days who have a superb com- Chapter three deals with Megasthenes and
mand of both classical philology and Indian Greek diplomacy. As usual it is written with a
studies, a circumstance that enables him to do maximum o f source material. Karttunen’s inter-
his job on a high level. He also did not spare pretations here are also sound: we know very
time and energy to find and evaluate even the little about Megasthenes’ origins: he might have
smallest items o f secondary literature written in had several audiences in the court o f Mauryas
a half dozen European languages. At the same in Pätaliputra; he was on the spot and wrote
time he is always conscious o f the limits o f his down what he saw; his evidence can be used
competency. It is very appealing that he shows with much care and criticism; Daimachus was
restraint in such cases where he confesses his only a “shadowy successor” of him.
language barriers (Arameic, Chinese, Tamil) or Chapter four called The natural sciences
his lack of sufficient proficiency in archaeol- presents a rich panoramic view of the geogra-
ogy or numismatics. He is also able to revise phy, fabulous and real, o f the wonder-land.
even his former result for the sake o f sound I read with great pleasure the estimations o f Al-
knowledge, for instance he is ready to modify exander’s scientific staff and the views of Greek
his standpoint concerning the origin and devel- and Roman scholars about India in general.
opment of scripts in India. Chapter five contains various topics ranging
Chapter one offers a very lucid review of from the world o f plants and animals to precious
the extant source material as well as an oppor- stones, things that made India very much sought
tunity for the author to say some words about after during the last millennia. The subchapters
Hellenism. Together with Momigliano and on elephant-lore or the famous precious stones
O. Murray he regards Hellenism as “the mixed of India or Indian physicians are commanding.
culture of the post-Alexander era, in which Chapter six treats the political role of
Greek civilisation and the Greek (as well as Greeks in the East. I fully agree with Karttunen’s

314 REVIEWS
judgement o f Candragupta’s importance in east- the meaningful earlier hypotheses: non-accep-

ern history in post-Alexandrian history. To- tance o f foreign coins or the melting o f them
gether with him I also lend credence to the be- for local issues he highlights the implications
lief that Candragupta perhaps met Alexander at o f geography. The example o f Gujarat where
the beginning o f his career. He proceeds with rich hoards of coins have been unearthed and
utmost care when he goes into the problem o f the testimony o f the Periplus properly advocate
the origins o f the Mauryas. He does not go into for that. Karttunen convincingly dismisses the
the debates about their caste and is satisfied probability of the existence o f a templum Augusti
with asserting that Candragupta was o f humble in Kerala indicated by the Roman map known
origins. Contrary to many noted specialists he as Tabula Peutingeriana (p. 334).
takes both Piyadassi and Asoka as personal To sum up, Karttunen’s monograph is a mag-
names (p. 269). I must praise Karttunen’s out- nificent book. It is richly documented through-
spokenness when he judges the extreme views out and provided with an impressive bibliogra-
represented by Tarn and Narain whom he re- phy where one can hardly find lacks. It is ex-
gards otherwise as “classics”. Tarn simpiy does emplary how the author treats primary sources
not understand oriental history while Narain and theories. His own ideas are imaginative
has a bias towards “Indianism”. Here a new ap- and surely give strong impetus to further re-
proach is needed that takes into account the search in the field. I am very much convinced
role o f the Iranian population and the fresh ar- that this volume together with the other two will
chaeological as well as the numismatical evi- be an indispensable tool for rewriting numer-
dence especially in the territory o f ancient ous chapters o f the ancient history o f India.
Bactria (p. 271). I find it reasonable that he es-
capes haphazard ventures in the intricate later
political history o f Indo-Greek kings (p. 275). References
A healthy scepticism prevails in his estimation
o f the extent of Greek rule in India: only the Karttunen, K. (1994), Yonas, Yavanas and re-
Panjab seems to have been certainly ruled by lated matter in Indian epigraphy. In: Parpola,
them (p. 275). Karttunen does not forget to add A.-Koskikallio, P. (eds), South Asian Ar-
that it is a pity that Panjab sites are mostly un- chaeology 1993. Helsinki, pp. 329-336.
excavated (p. 283). He briefly mentions a pos- (Annales Academiae Scientiarum Fennicae
sible Greek raid on Pâtaliputra and in this con- sec. В tom. 271)
nection says a few sentences about the Yugapu- Narain, A. K. (1957), The Indo-Greeks. Oxford.
râna. It is one o f the very few places in Karttu- Appendix IV.
nen’s book where I feel some lack. Here he is Sircar, D. C. (1963), An account o f the Yava-
even more laconic than in one of his previous nas in the Yugapurâna. JRAS pp. 7-20.
articles (cf. Karttunen 1994, p. 331). I would be
very thankful for a critical assessment of the in- Gyula Wojtilla
terpretations by Narain (1957) or Sircar (1963).
Chapter seven called India and the Greek
West starts with a critical review of the contem- Klaus M y l iu s , Wörterbuch Päli-Deutsch.
porary historians o f Alexander, many o f them Wichtrach 1997. Institut für Indologie. 438 pp.
almost unknown. Then he turns to the history
o f commerce and the conditions for it. He leaves Pali lexicography has experienced many re-
the question of the name Hippalus undecided: verses in this century. Even PTSD was origi-
it can be either the name of a sailor or a par- nally published in eight parts and although the
ticular wind (p. 331). The vexing problem o f idea o f its revised edition was ripe already in
the non-occurrence o f Roman coins in North 1972 it still keeps us waiting (cf. Norman
India arrested also his interest. Beside listing 1985, p. 145 and Hinüber 1998, p. 66). CPD

REVIEWS 315
has its own fate and surely will serve as a prac- aries o f proper names, because the user of
tical dictionary in the foreseeable future. The PTSD looks for these words in vain (Malala-
situation in the German speaking countries is sekera 1937 and Akanuma 1937). A great ad-
still worse. In spite o f the long tradition of vantage of M ylius’ dictionary is the Sanskrit
Buddhist scholarship as well as the existence index appended pp. 378^138).
of Buddhist communities in Germany one can To make the dictionary practical and useful
only deal with the modest list o f words by Mylius rightly includes certain special verb forms
Nyänatiloka alias A. F. W. Gueth (Nyänatiloka in independent word-heading: dissati Pass, ge-
1928) and that of Mayrhofer (1951). In short, sehen werden, erscheinen (p. 189) äro-cäpeti
I took in hand the present volume with great in- Kaus. mitteilen, zur Kenntnis geben, bekannt-
terest. machen (p. 80). Otherwise he is very economic
Prof. Mylius has been working on Sanskrit with recording averbos. Perhaps he might have
and Pâli lexicography for a long time. Success- mentioned Fachs’ list o f verb forms as a useful
fully using his teaching experience and his own tool for students (cf. Fachs 1985, pp. 2 3 0 -3 1 8 )
translatory activity he published a Sanskrit- in the introduction to the dictionary.
German and a German-Sanskrit dictionary (My- The number o f compounds is obviously lim-
lius 1975 and 1988) positively received in аса- ited due to the limits o f space. The ones included
demie circles. are striking, for example the very important
As he puts it he is perfectly aware o f the word mana (p. 280. cf. PTSD pp. 521-522).
tremendous difficulties o f preparing such a die- Redactory work is excellent, the book is
tionary (cf. p. 6) and draws peculiar attention nicely printed. Merit, for this is due to Dr Peter
to the semantic difficulties expressed by seri- Thomi, director o f the publishing house o f the
ous specialists o f Pali. The selection o f sources Institut für Indologie in Wichtrach. This hand-
is also a troublesome task. Mylius basically some dictionary must be warmly congratulated
takes the vocabulary o f the Pali Canon without by German speaking students o f Buddhism and
the commentaries and a great deal o f words from also the learned community of Indologists and
non-canonical texts such as the Milindapanhä, South Asian studies.
Dipavamsa, Mahävamsa and Visuddhimagga.
Samples taken at random from these texts show
me a positive picture. Any editor o f a die- References
tionary has to face the problem o f how far he
is allowed to use terms taken from different Akanuma, Ch. (1937), A Dictionary o f Bud-
branches o f science while translating the origi- dhist Proper Names. Tokyo.
nal ones. M ylius’s interpretations are reason- CPD: A Critical Pâli Dictionary, begun (1924)
able and in accordance with the requirements by V. Trenckner. Copenhagen.
o f modern German usage. Very rarely do cases Fachs, A. (1985), Grammatik des Pâli. Leipzig.
occur where I would not give the meaning Hinüber, О. v. (1998), The Critical Pâli Die-
accepted by him. As to uparajja n. (p. 100) tionary. History and prospects. In: Oguibé-
I regard Vizekönigtum sound, however, I would nine, B. (ed.), Lexicography in the Indian
not take the sense Satrapie because the term is and Buddhist cultural field. Proceedings of
closely associated with Iran and not with Bud- the conference at the University of Stras-
dhist India. On the other hand I find modern bourg 25 to 27 April 1996. München, pp.
interpretations as samma Oh mein Lieber 65-73. (Studia Tibetica Bd. IV.)
(p. 354) appealing. Malalasekera, G. P. (1937), Dictionary o f Pâli
For the sake o f saving space and fulfil prac- Proper Names. London.
tical purposes references to loci have been omit- Mayrhofer, M. (1951), Handbuch des Pâli. II.
ted. It is laudable that Mylius utilises the re- Teil: Texte und Glossar. Heidelberg, pp.
suits o f Malalasekera’s and Akanuma’s diction- 20-76.

316 REVIEWS
M ylius, K. (1975), Wörterbuch Sanskrit- the two others were smaller constructions. The
Deutsch. Leipzig. authors o f the present book aimed at clarifying
M ylius, K. (1988), Wörterbuch Deutsch-San- the architectural context o f the blocks. As the
skrit. Leipzig. result o f the reconstruction imposing elements
Nyänatiloka (1928), Päli-Anthologie und Wör- o f the Amenhotep temple, hitherto unknown,
terbuch. A: Päli Wörterbuch. München. emerge before the eyes o f the reader.
Norman, K. R. (1985), A report on Pâli die- Specialists o f the history o f the Egyptian
tionaries. Buddhist studies (Bukkyö Ken- N ew Kingdom will be first o f all interested in
куй) 15, pp. 145-152. the analysis of the representations on the blocks
PTSD: Rhys Davids, T. W .-Stede, W. (1 9 2 1 - (by S. Bickel) which reflect the religious ideol-
1925), The Pali Text Society’s Päli-Eng- ogy o f three periods. 1) The reign o f Amen-
lish Dictionary. London. hotep III; 2) the crisis of the Amarna Period; 3)
Gyula Wojtilla the restoration of traditional religion in two
phases (Tutankhamun and Sethos I).
Since the monotheistic extremism o f the
Untersuchungen im Totentempel des Meren- Amarna Period did not tolerate Amun and
ptah in Theben unter der Leitung von H o r s t other gods, their names and representations
J a r i t z III. Tore und andere wiederverwendete were cut out in the temples all over the country,
Bauteile Amenophis’ III von S u s a n n e B i c k e l and even the temple o f Akhenaton’s father was
mit Beiträgen von H O R S T JARITZ - H O RST not an exception. The blocks display the de-
J a r it z - U w e M in u t h - R a p h a e l A . J. W ü s t . structive zeal o f the religious revolution. The
Franz Steiner Verlag Stuttgart 1997. 175 pp., image o f Amun was everywhere destroyed and,
61 Abb., 95 Tf. strangely enough, reshaped into the figure o f
Amenhotep III, who seems to have received
The restoration of the mortuary temple o f divine cult. Therefore, instead o f Amun, the
Pharaoh Merenptah situated on the Theban King venerated now on the blocks which origi-
W est Bank represents one o f the ambitious un- nally represented him before Amun, his own
dertakings of the Swiss Institute for Egyptian divine entity, the God Nebmaatre. Similar rep-
Architecture in Cairo. Researches in this area resentations are known from the temple in Soleb,
began at the end o f the 19th century when Wil- although those in Soleb are original composi-
liam Flinders Petrie made excavations at the tions o f the apotheosis o f Amenhotep III, and
site. A new phase set in when Gerhard Haeny, not products of the persecution o f Amun.
as director of the Swiss Institute, began to work What is even more curious, Amun could be
on the territory of Merenptah’s temple in 1971. replaced by the image o f Ptah. On a remarkable
This work had to be interrupted soon because block Amun came to be transformed into the
o f other obligations o f the Institute. The sys- falcon-headed god Ptah-Sokaris-Osiris (pp.
tematic continuous research, resumed in 1988, 109-110). As far as evidence goes, this was
is in progress up to the present. a unique practice not attested in other temples
During the excavations a lot o f limestone during the Amarna Period. Elsewhere the de-
blocks came to light which were originally stroyed images were never superseded by other
parts o f the neighbouring mortuary temple o f figures. In view of this remarkable phenome-
Amenhotep III which was almost completely non it is legitimate to suppose that when these
destroyed in the Antiquity. It was found out by alterations took place in the temple, the mono-
the team of the Institute, during meticulous ex- theistic doctrine of Atenism did not yet receive
aminations, that a group o f these stones be- its final shape and Akhenaton ordered to carry
longed to three gates o f the Amenhotep temple, out the suppression of polytheism in the monu-
and came to be reused in the temple o f Meren- ment o f his father in a tactful way. On the oth-
ptah. One of the gates was o f considerable size, er hand, the reign of Haremhab and the 19th

R E V IE W S 317
Dynasty was marked by wide-ranging efforts to into three groups. Part I contains studies on Ar-
restore the defaced walls in the temples by cov- meno-Kipchak, Part II on other linguistic studies
ering up traces o f Atenist iconoclasm. This re- and Part III studies on history. The grouping is
storing activity is also clearly recognisable on necessarily arbitrary, because one o f the great
the blocks under discussion. merits o f E. Schütz is that he always tries to
S. Bickel’s pioneering research is important investigate linguistic problems against their
to be emphasised. Up to now the destructive historical background and to solve historical
rage o f the Amama Period was mostly charac- riddles with the help o f linguistic data. If some-
terised in general terms, and it is the present body will write a new, comprehensive mono-
book in which one first finds an in-depth analy- graph on the history, culture and languages o f
sis o f this question on the basis o f the evidence the Armenian diaspora in East Europe, the
o f an important building. works o f Schütz will offer a sound and solid ba-
While the original location o f the huge gate sis. Schütz concentrates on how and when did
poses some problems the authors suggest that it a group o f Armenians change their language
may have belonged to the northern enclosure and began to speak, pray and write in a Kipchak
wall o f the temple. One o f the two smaller idiom of Turkic. This process may not only in-
gates again contain remarkable details, al- terest those who investigate Armenian or Turkic
though its significance cannot be compared but would also interest sociolinguists. The grad-
with the former one. Only one block remained ual bottom-up change in the language, while
preserved from another gate. retaining religion and script is not unprece-
O f the numerous topics treated in the book dented in this region. Similar changes occurred
the petrographic analysis o f the limestone by to the Karaim, another Kipchak Turkic group
Raphael J. A. Wüst should be mentioned, that proselytised to the Jewish faith. The key
which demonstrates the growing importance of historical questions (in both cases) are: when
the exact differentiation in the various kinds of and where did the change o f language occur?
limestone. The question o f the quarry and most The scene where both events happened or at
o f the architectural problems are dealt with by least began was the Crimean peninsula. Less
Horst Jaritz. clear is the chronological framework. Schütz
In view o f the reviewer the importance of argues convincingly that the migration o f the
the luxuriously produced book lies very largely masses of Armenians to the Crimea began only
in the fact that every small detail o f the stones after the Mongolian invasion of East Europe,
was subjected to a micro-analysis by the authors. that is in the 13th century. He goes even further
This led on one hand to the reconstruction o f and demonstrates that the Armenians came in
a stately architectural monument o f the New the 13th-14th centuries from the Ilkhanid Persia
Kingdom and to a better understanding o f the and from about 1600 from Cilicia, that is from
destruction in the temples in the Amama Period East Anatolia in connection with the anti-
on the other. Armenian atrocities by the rebellious Turkish
Lâszlô Kdkosy Jelali movement. However, the Ottoman Empire
conquered Crimea gradually after 1475 which
resulted in the emigration to Podolia and Mol-
Edm und S c h ü t z , Armeno-Turcica. Selected davia. Those newcomers who fled in the 17th
Studies. Bloomington, Indiana 1998 (Indiana century from Anatolia found only a very thin
University Uralic and Altaic Series, 164). Armenian layer in the Crimea. Schütz stresses
that the whole migration process was gradual
This volume contains reprints o f 16 selected and warns against oversimplification.
papers, written by E. Schütz, distinguished An extremely interesting and important
scholar on the relationship between the Arme- paper is on Re-Armenisation and Lexikon. From
nians and the Turks. The papers are divided Armeno-Kipchak Back to Armenian {AOH 13,

318 REVIEWS
1966, pp. 123-130). The language o f the Podol- Armenian, the fact, however, that it is a fre-
ian Armenians underwent fundamental changes quent name makes it plausible that they knew
in the second half o f the 17th century: spoken it. Since the publication o f this paper, M. Erdal
Armeno-Kipchak gave place partly to Polish, has dealt with the word and the suffix +sOk
and partly to Armenian. The reasons for the which seems to have been unclear to Schütz.
shift to Polish are more or less clear. The change Erdal also mentions that in the Berliner Turfan-
o f the world trade, which after the discovery of texte XIII 46, 35 the form Tarjgusok appears as
the Americas and the circumnavigation o f the a proper name ( Old Turkic Word Formation,
African continent, reshaped the network o f the Wiesbaden 1991, p. 157).
commercial connections, drastically narrowed One of the highly informative articles sum-
the importance of trade through East Europe. marises the history o f the ethnonym Tat. The
The Armenian archbishop o f Lvov accom- name was used for Iranian ethnic elements in
plished the union with the Catholic Church and the neighbourhood o f Turks, later it denoted
thus the Polish influence grew. The Ottoman also the neighbours (of whatever origin), and
Turks took Podolia with the centre Kamenec. certain subdued people. Schütz analyses the
Shelter against the invasion could have been history o f the Tat on the Crimean peninsula and
found at the Polish landlords. Without going in Dobrudja (The Tat people in the Crimea,
into details the triple linguistic change Arme- AOH 21, 1977, pp. 77-106). Schütz, joining
nian —> Kipchak —> Polish by the same ethnic odiers, claimed that the language of the Goths
group is well documented and Schütz prepared in the Crimea died out early and that the notes
the field for a thorough sociolinguistic investi- on Crimean Gothic by Ogier Ghislain de
gation. Busbecq in 1560-1562 “was highly mislead-
Even more interesting is the question o f re- ing”. However, Macdonald Stearn in a book
Armenisation. As Schütz states “As a matter of published in 1978 ( Crimean Gothic. Analysis
fact ‘re-Armenisation’ sounds paradox since and Etymology of the Corpus, Saratoga, Cali-
the official language o f the Church o f the col- fomia) convincingly demonstrated that Crimean
ony had always been Armenian”. However, the Gothic lived until the 16th century, which is
local idiom of certain regions was Kipchak and important for the special relationship for the
became step-by-step Armenian. Schütz dis- languages of the Crimea. Schütz is quoting
cusses this in connection with the dialect o f Mahmüd al-Khâshgharï as one o f the early
Kuty, a town founded in the 17th century. Ar- sources on the ethnonym Tat. In fact apud
menians who moved to the region from differ- Khäshghari Tat has the meaning ‘Persian’
ent areas o f the Diaspora triggered the change. (farisi) and ‘Uighur infidels’ (kafara Uighur),
Some came from Moldavia, others from Käme- that is the Buddhists or Manicheans. Khäsh-
пес, a third group from Transylvania. Having ghari cites two proverbs to illustrate their posi-
analysed the language o f Kuty Schütz is sum- tion. The first is tätig közrä tikänig tiipre “The
ming up. In the Armenian dialect of Kuty ap- Persian on the eye, the thorn on the root” and
proximately 30 Kipchak loanwords can be de- gives to it the explanation: “the proverb also
tected. These loanwords were acquired from originally refers to them, because they lack
the language of those Armenians who settled in loyalty; just as the thorn should be cut at its
the region earlier and spoke an Armeno-Kip- root, so the Uighuri should be struck on the
chak language. This means that Kipchak words eye”. More interesting is the following prov-
were transmitted by an Armeno-Kipchak dia- erb: tatsiz türk bolmas, bassiz börk bolmas
lect to the Armenian dialect o f Kuty. “There is no Türk without a Tat, as there is no
In one o f his papers Schütz follows the cap without a head”. In the Arabic “Persian”
history of the Turkic word taqsoq (Tangsux in and “Türk” are reversed (see Dankoff, R. -
Armenian, AOH 17, 1964, pp. 105-112). Though Kelly, J., Mahmud al-Khasyari, Compendium
the word itself could not have been detected in o f the Turkic Dialects. 1984, II, p. 103). Which

REVIEWS 319
ever of the two is correct, the proverb refers to kirdly ‘König’ in kiptschakischen Sprachen],
the close coexistence of the two people or to Proceedings o f the IXth Meeting of the Per-
the interdependence o f the nomads and the manent International Altaistic Conference ...,
settled people. Naples 1970, pp. 259-267) Schütz differenti-
To the history o f the ethnonym Tat we can ated the data which refer to the king o f Poland
add two cases. Among the people o f Caucasia, (korol, korel, körel, etc.) from those denoting
there exists a group that is called Tat. In fact the Hungarian king (kiräl etc.). It may be de-
they are Jewish by religion but ethnically Irani- ceiving to suppose that the Turkic name o f the
ans. They were first mentioned by Berezin in turkey körel and that of the lizard kerel or the
1853. (See A. L. Grjunberg, Tatskij jazyk, in: Hungarian word kirdly played a role in the
Jazyki narodov SSSR, t. pervyj, Indoevropej- emergence o f the front vocalic forms. Foreign
skie jazyki, Red. V. V. Vinogradov, Moskva words with /к/ + back vowels show in many
1966, pp. 281-301). N. Poppe (Reminiscences, cases two reflexes. In Kalmuck we find both
Western Washington 1983, p. 166) who met korül and körül for ‘king’ in card-play, a clear
them in 1942 in Nalchik, writes a few words borrowing from Russian korol’. Interesting is
about them. The Germans, who occupied the the Noghay and Karachay-Balkar qïral ‘land,
territory, were in doubt whether these Jews had country; strana’ quoted by Schütz. I would,
to be annihilated, and Poppe testified that they however, not follow Hasan Eren and Pritsak in
were of Iranian origin, so they were rescued as connecting this word with qara el, neither would
most of the Karaims were in Poland, but not I think of a Hungarian origin. I would con-
the Krimchaks, the Jews of Crimea. nect it with qïr ‘steppe’, the Kipchak-Noghay
In Hungarian the name of the Slavs was equivalent o f the Persian dest ‘desert, country’.
Tdt, it denoted all Slavs and has been narrowed The whole region was called, as we know,
to the designation of the Slovaks only in the early Dest-i-Kipchak, ‘the steppe of the Kipchak’.
19th century. This ethnonym is usually con- The steppe country was the qïr eli, from which
nected with the German name Teut (> deutsch, qïral developed. In the Noghay-Russian die-
cf. Teuton etc.). However, the Old High Ger- tionary of Baskakov (Nogaisko-Russkij slovar’,
man form is diot, the Gothic is Piunda. It is at M oscow 1963, p. 201) we find qïr sesekeyleri
least not impossible that the Hungarians called ‘steppe flowers’.
the Slavs *Tat and the change a > о occurred in All together I think we all are in debt to
Hungarian. Denis Sinor, who published this interesting
In a paper ( Könige und Eidechsen [Berner- volume.
kungen zum Fortleben des ungarischen Wortes Andrds Rôna-Tas

Bagautdinov, R. S.-B ogacev, A. V -Z u b ov, S. Golden, P. B. (1998), Nomads and Sedentary

E. (eds) (1998), Prabolgary na Srednei Vol- Societies in Medieval Eurasia (Essays on
ge (и istokov istorii tatar Volgo-Kam’ja). Global and Comparative History). Wash-
Samara. ington D.C., American Historical Associa-
Bigger, A. (1999), Balaräma im Mahäbhärata. tion.
(Beiträge zur Indologie 30) Wiesbaden, Hasbatar, N. (1999), Mongolische Heftgeschich-
Harrassowitz, DM 118. ten und chinesische Einflüsse auf die
Bogacev, A. V. (1992), Procedurno-metodice- mongolische Literatur. (Asiatische For-
skie aspekty arheologiceskogo datirovanija schungen 135) Wiesbaden, Harrassowitz,
(na materialah pojasnyh narodov IV-VIII DM 98.
vv. Srednego Povolz’j a). Samara. Houtman, G. (1999), Mental Culture in Bur-
Bulut, Ch. (1997), Evliya Çelebis Reise von mese Crisis Politics. Aung San Suu Kyi
Bitlis nach Van. Ein Auszug aus dem Seya- and the National League for Democracy.
hatname. Interpretierende Transliteration, (ILCAA Study o f Languages and Cultures
kommentierte Übersetzung und sprachwis- o f Asia and Africa Monograph Series 33)
senschaftliche Bemerkungen. (Turcologica Tokyo, Tokyo University of Foreign Stud-
35) Wiesbaden, Harrassowitz, DM 98. ies.
Doerfer, G.-Hesche, W. (1998), Türkische Muraoka, T. (1998), Modern Hebrew fo r Bibli-
Folklore-Texte aus Chorasan. (Turcologica cal Scholars. An annotated chrestomathy
38) Wiesbaden, Harrassowitz, DM 182. with an outline grammar and a glossary.
Friedrich, M. (1998), Gabdulla Tuqaj (1886- 2nd revised and enlarged ed. Wiesbaden,
1913). Ein hochgelobter Poet im Dienst von Harrassowitz.
tatarischer Nation und sowjetischer Sozia- Oyunbilig, B. (1999), Zur Überlieferungs-
lismus. (Turcologica 36) Wiesbaden, Har- geschickte des Berichts über den person-
rassowitz. liehen Feldzug des Kangxi-Kaisers gegen
Georg, S.-V olodin, A. P. (1999), Die itelmeni- Galdan (1696-1697). (Tunguso Sibirica 6)
sehe Sprache. Grammatik und Texte. (Tun- Wiesbaden, Harrassowitz.
guso Sibirica 5) Wiesbaden, Harrassowitz, Pletneva, S. A. (1999), Ocerki hazarskoj ar-
DM 108. heologii. Moskva.
Golb, N.-Pricak, O. (1997), Hazarsko-jevrej- Sändor, K. (ed.) (1998), Nyelvi vdltozd - nyelvi
skie dokumenty X veka. Moskva-Jerusalim. vdltozds. Szeged.

322 BOOKS AND JOURNALS RECEIVED
äuäarin, V. P. (1997), Rannij étap etniceskoj J o u r n a ls r e c e iv e d

istorii vengrov. Problemy étniceskogo sa-
Journal o f Asian and African Studies 57,
mosoznanija. Moskva.
March 1999.
Zimmer, Th. (1999), Baihua: zum Problem der
Mediterranean Language Review 10, 1998.
Verschriftung gesprochener Sprache im
Shaman. Journal of the International Society
Chinesischen. Dargestellt anhand morpho-
fo r Shamanistic Research 6:1.
logischer Merkmale in den bianwen aus
Ural-Altaische Jahrbücher. Neue Folge 15.
Dunhuang. (Monumenta Serica Monograph
Series XL) Sankt Augustin, Institut Monu-
menta Serica, DM 70.
Zsoldos, A. (1997), Az Arpâdok és alattvaldik.
Magyarorszdg torténete 1301-ig. (Torténel-
mi kézikônyvtâr) Debrecen.

ADDRESSES OF THE AUTHORS
Klâra Agyagâsi Lâszlô J. Nagy

Department of Slavic Philology Department o f Modern World History
University of Debrecen University o f Szeged
H-4010 Debrecen, Pf. 53 H-6722 Szeged
Hungary Egyetem u. 2
Hungary
Imre Hamar e-mail: jnagy@hist.u-szeged.hu
Department of East Asian Studies
University of Budapest Dmitry B. Proussakov
H-1088 Budapest Oriental Institute o f the Russian Academy
Mûzeum krt. 4 of Sciences
Hungary 12 Rozdestvenka
Vladimir V. Klimenko e-mail: prusakov@prusakov.msk.ru
M oscow Institute o f Energy
14 Krasnokazarmennaja ulica Rôbert Simon
M oscow 111250 H-1065 Budapest
Russia Bajcsy-Zsilinszky u. 31
e-mail: gepl@deans.mpei.ac.ru Hungary
Andrey V. Korotaev Gyula Wojtilla

Institute of Cultural Anthropology Department of Ancient History
Russian State University for Humanities University o f Szeged
6 Miusskaja Ploscad' H-6722 Szeged
M oscow 125267 Egyetem u. 2
Russia Hungary
e-mail: andrei@rsuh.ru e-mail: wojtilla@antiq.u-szeged.hu
ACTA
ORIENTALIA
HUNGARICAE
VOL. 52(1999)

CONTENTS OF VOLUME 52, Nos 1-4
К лара А д я гаш и , К вопросу о б этногенезе м а р и й ц ев ............................................................. 293

Â rpâd B erta, On the Turkic background o f the Hungarian verb ov 'to defend, protect, keep,
guard’ .............................................................................................................................................. 45
Imre H amar , Buddhism and the Dao in Tang China: The Impact o f Confucianism and
Daoism on the philosophy o f C hengguan................................................................................ 283
Y 0SH1Z0 Itabashi, On cognateship o f Old Japanese i/si with Old Korean i/s i .............................. 17
A ndrey Korotaev-V ladimir Klimenko-D mitry Proussakov , Origins o f Islam: Political-
anthropological and environmental con text............................................................................. 243
L âszlô J. N agy , Un gouverneur d’origine hongroise à Alger 1580-1582 ................................... 277
A ndrä S R6 na -T a s , Chuvash and historical morphology .................................................................. 1
А . Г. С азы к и н , П ятый (анонимный) монгольский перевод «А лм азной су т р ы » ......... 129
Claus S chönig, The internal division o f Modern Turkic and its historical implications ........... 63
Röbert Simon, Muhammad and the Jihad ............................................................................................ 235
G yula W ojtilla, The Sanskrit Godhiima apropos o f a short excursion in Indo-European and
Indo-Aryan prehistory .................................................................................................................. 223
Miscellanea
Béla Barték’s Anatolian folk music collection (J. S ipos) .................................................................... 97

The eighth seminar o f the IATS in Bloomington (T. P0RC1Ö)........................................................... 99
Review article
H. Franke-H-L. Chan, Studies on the Jurchens and the Chin Dynasty (H. F ranke) ................... 101
Reviews
Gyula Décsy, The Turkic Protolanguage: a Computational Reconstruction (A. R6 na -Tas ) ..... 107
Seishi Karashima, A Glossary of Dharmaraksa’s Translation o f the Lotus Sutra (I, Hamar) .... 118
Klaus Karttunen, India and the Hellenistic World (G y . W ojtilla) ..................................................... 313
Donald D. Leslie, Jews and Judaism in Traditional China (P. VAmos ) .......................................... 119
Klaus Mylius, Wörterbuch Pâli-Deutsch (G y . W ojtilla) ................................................................ 314
Présence arabe dans le Croissant Fertile avant l’Hégire (I. Ormos) .................................................... 110
André Raymond, Le Caire; Robert Mantran, Histoire d ’Istanbul (I. Ormos) ................................. 116
W olfgang Reuschel, Aspekt und Tempus in der Sprache des Korans (I. Ormos ) ........................ 113
Hugh Richardson, High Peaks, Pure Earth (A. Röna -T as ) ............................................................... 105
Lambert Schmithausen, Maitri and Magic: Aspects o f the Buddhist Attitude toward the Dan-
gerous in Nature ( G y . W o j t i l l a ) ............................................................................................... 120
Edmund Schütz, Armeno-Turcica. Selected Studies (A. R öna -Tas) .............................................. 317
D enis Sinor, Studies in M edieval Inner Asia (A. R öna -T as ) ............................................................ 106
Gotthard Strohmaier, Von Demokrit bis Dante: die Bewahrung antiken Erbes in der arabi-
sehen Kultur (I. O rmos ) ............................................................................................................... 111
Untersuchungen im Totentempel des Merenptah in Theben unter der Leitung von Horst Jaritz
III. Tore und andere wiederverwendete Bauteile Amenophis’ III von Susanne Bickel
mit Beiträgen von Horst Jaritz - Horst Jaritz-Uwe Minuth-Raphael A. J. Wüst
(L. Käkosy) .................................................................................................................................... 316
Läszlö Vikär-Gäbor Bereczki, Tatar Folksongs (J. S ipos) ................................................................ 311
Ewald Wagner, Islamische Handschriften aus Äthiopien (I. Ormos) .............................................. 117
Chlodwig H. Werba, Verba indoarica. Die primären und sekundären Wurzeln der Sanskrit-
Sprache. Pars L: Radices primariae (G y . W ojtilla) ............................................................. 309
Klaus G. Witz, The Supreme Wisdom of the Upanisads. An Introduction (F. R u z s a ) ............... 310
Books and journals received ................................................................................................... 123, 223, 321

ABBREVIATIONS OF JOURNALS AND SERIES
AEMA Archivum Eurasiae Medii Aevi, Wiesbaden

AOH Acta Orientalia Academiae Scientiarum Hungaricae, Budapest
ArOr Archiv Orientalni, Praha
BASOR Bulletin of the American Schools of Oriental Research
BGA Bibliotheca geographorum arabicorum
BIFAO Bulletin de l'Institut Française d’Archéologie Orientale du Caire
BiOr Bibliotheca Orientalis
El, E l2 Encyclopaedia of Islam ( 1st and 2nd edition)
FO Folia Orientalia, Krakdw
IC Islam Culture
KCsA Körösi Csoma Archivum, Budapest
KK Keletkutatds, Budapest
MFOB Mélanges de la Faculté Orientale de l’Université St. Joseph de Beyrouth
M ID E O Mélanges de l’Institut Dominicain d’Études Orientales du Caire
M R D TB Memoirs of the Research Department of the Toyo Bunko, Tokyo
NAA Narody A zii i Afriki, Moskva
N yK Nyelvtudomânyi Kôzlemények, Budapest
OLZ Orientalische Literaturzeitung, Berlin
OsmAr Osmanli Ara$t1rmalan, Istanbul
REJ Revue des Études Juives
Stil Studien zur Indologie und Iranistik, Reinbek
TDAYB Türk Dili Araçtirmalari Ydltgi Belleten, Ankara
TDKY Türk Dili Kurumu Yayinlari
TP T'oung Pao, Leiden
TSAB Turkish Studies Association Bulletin, Bloomington
UAJb NF Ural-Altaische Jahrbücher: Neue Folge, Wiesbaden
WZKM Wiener Zeitschrift für die Kunde des Morgenlandes, Wien
ZDMG Zeitschrift der Deutschen Morgenländischen Gesellschaft, Wiesbaden
tat Bonn, Wiesbaden
ZVOIRAO Zapiski vostocnago otdelenija Imperatorskago russkago arheologiceskago obäcestva, Sanktpeterburg
INFORMATION FOR CONTRIBUTORS
oriental studies, including Turkic, Mongolian, Manchurian, Chinese, Tibetan, Indian, Iranian and Sem itic
philology, literature and history. A copy of publishing agreement w ill be sent to the authors of papers accepted
by two reviewers and the board of editors for publication. Manuscripts w ill be printed only after receiving the
Andres Rdna-Tas
Editor-in-Chief
H -l 146 Budapest
Ajtösi Durer sor 19-21
from the authors together with the first proof.
Text. The total length o f the text (including illustrations, tables, etc.), i f not agreed otherwise, should not
exceed 20 typewritten pages. The first page should contain the title of the paper, the author’s names and
Abstract. Manuscripts o f articles should be accompanied by an abstract o f about 100 words / 1 5 lines
printed on a separate sheet o f paper summarising the conceptual content and the new results o f the article.
Key-words. The authors are kindly asked for 6 to 8 key words added to their articles.
Please indicate the top o f the figure and the name o f the author o f the paper on the back. High quality h a lf-
Footnotes. A ll footnotes must be placed on the same page as the text and numbered continuously.
References. Full citation o f literature referred to should be given in an alphabetically arranged list at the end
the text. In the references use full names of journals if they are not to be found on the last page of AOH. In the
text they should be cited by the names o f the authors and the year o f publication, e.g. Doerfer and Tezcan
(1980), or if pages are referred to: Doerfer and Tezcan (1980, pp. 21-22). In case o f publications of the same
year use letters: D oerfer and Tezcan (1980a).
Reference to books:
Doerfer, G.-Tezcan, S. (1980), W örterbuch des Chaladsch. (Dialekt von Charrab). Budapest (Bibliotheca
Orientalis Hungarica X X V I).
Gervers, M . (1997), Nomads as gatherers and creators o f world culture. The artistic sources of the great fe lt
hanging from Pazyryk, Kurgan 5. Sum m aries o f congress papers. 7th International Congress o f M ongolists
1 2 -16 August 1997, Ulaanbaatar. p. 3. Ulaanbaatar.
Berta, Ä. (1994), Zum Wandel des auslautenden -G im Kiptschakischen. Journal o fT u rko lo g y 2:2, pp. 163-195.
Transcription. Th e authors should use the transcription internationally accepted in their respective field o f
studies. Please try to avoid the usage o f special character that is not to be found in W inW ord 95 or 98 Sym bol
True Type font. T ry to avoid Chinese, Arabic, etc. non Latin or Cyrillic characters. Use the Pinyin transcrip-
special symbols, please enclose a floppy containing the fonts (in case o f Chinese preferably Big5. G B or
Chinese Star).
Proofs. The author w ill receive the first proof. Please read it carefully and return it promptly, if necessary by
Reviews. General rules of scientific publication (see above) should be followed.
Addresses of the authors. We publish the mailing addresses o f the authors as they are given in the publish-
HU ISSN 00016446‫־‬

ActaOrientalia 52

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ActaOrientalia 52

Uploaded by

Copyright:

Available Formats

ACTA

VOL. 52 (1999) NO. 1

AKADÉMIAI KIADÖ, BUDAPEST

ADVISORY BOARD: Gy. HAZAI, Z s . K a k u k , Gy. KARA, M . MAROTH, F. TÖKEI

TECHNICAL EDITORS: Éva CsÂKi and Imre HAMAR

ACTA ORIENTALIA is a refereed journal. It is abstracted in Sociological Abstracts.

© Akadémiai Kiadô, Budapest 1999

VOL. 52 (1999) NO. 1

AKADÉMIAI KIADÖ, BUDAPEST

A ndräs RönA-Ta s , Chuvash and historical morphology .................................................................. 1

B éla Bartdk’s Anatolian folk music collection (J. S ipos) .................................................................... 97

Books and journals received ..................................................................................................................... 123

Addresses of the au thors ............................................................................................................................ 125

CHUVASH AND HISTORICAL MORPHOLOGY1

Key words: Historical morphology, ordinal suffix, Chuvash, lambdacism

0 0 0 1 1 9 9 9 © 5.00 $ /99/ 6446‫ ־‬Akadémiai Kiadô, Budapest

7. sic, sicë siccë

The origin of these two series is a debated question. According to Clauson

Acta Orient. Hung. 52, 1999

seven siësëmër siësër siëcësë

There is a discussion on the nature of these numerals. Ashmarin (1976, p. 44)

Acta Orient. Hung. 52, 1999

In Tuva we have -(X)k(X) (dörtkii ‘fourth’), and in Tofalar -(X)sk(X) (dörtüskü

Cluster Yakut/Dolgan Tofalar/Tuva Chuvash

16 On +ki in the ordinal numerals see Doerfer (1988, p. 115).

with the exception of Chuvash where it remained m. Levitskaja, Hovdhaugen and

Acta Orient. Hung. 52, 1999

Acta Orient. Hung. 52, 1999

Old South Siberian ns East Old Turkic nc

Yakut/Dolgan Tuva/Tofalar Chuvash All other Turkic languages

Acta Orient. Hung. 52, 1999

Would be this hypothesis chronologically acceptable? The exact time of the

1. +(X)m —>Chuv +(X)m ?

C+(X)m ( > Chuvash +m) V+s

The historical processes of the morphological feature can be formulated as

Acta Orient. Hung. 52, 1999

Adamovic, M. (1996), Ordinalia im Bolgarischen. Symbolae Turcologicae, Studies in honour of

Doerfer, G. (1988), Grammatik des Chaladsch. Wiesbaden.

Acta Orient. Hung. 52, 1999

Acta Orient. Hung. 52, 1999

ON COGNATESHIP OF OLD JAPANESE i/si

YOSH IZO ITABASHI

Key words: Japanese, Korean affixes, indefinite, definite, generic, specific

[1] Functions of personal pronouns

(1) 2nd person pronoun

1] ya ga katji bitji [Okinawa dialect: Uchima 1984, p. 443]

[2] Old Japanese and Ryukyu morphemes *i and *si

[A] All the functions of the morpheme i

® emphatic nominative case suffix

Acta Orient. Hung. 52, 1999

naramaro komaro-ra i sakasimanaru t02 m02 gara

2] aritigata arinagusame2te2 ikamed02 m 02 iFenaru imo i obosimisemu

® instrumental case suffix

Acta Orient. Hung. 52, 1999

® vocative case suffix

® locative case suffix

Acta Orient. Hung. 52, 1999

© unbound flipping suffix (N —» N)

Acta Orient. Hung. 52, 1999

As in the examples immediately above, the function of the suffix i is deic-

Acta Orient. Hung. 52, 1999

2) conditional(izenkei) forming suffix

Acta Orient. Hung. 52, 1999

[2] Old Japanese and Ryukyu morphemes i and si

ire---- < i [3rd person sing.] + *re [pre-Ryukyu]