Actaantiqua 28 PDF

ACTA ANTIQUA
ACADEMIAE SCIENTIARUM
HUNGARICAE
ADIUVANTIBUS
I. B O R Z S K , I. H A H N ,
Z S . R I T O K , . S Z A B , S. S Z D E C Z K Y - K A R D O S S
REDIGIT
J. HARMATTA
TOMUS XXVIII FASCICULI 1-4
AKADMIAI KIAD, B U D A P E S T
1980
ACTA ANT. HUNG.

ACTA A N T I Q U A
A J O U R N A L OF T H E H U N G A R I A N ACADEMY
OF SCIENCES
Acta Antiqua publish papers on classical philology in English, German, Fronch, Russian
and Latin.
Acta Antiqua arc published in yearly v o l u m e s of four issues by
AKADMIAI KIAD
Publishing House of t h e H u n g a r i a n Academy of Sciences

H-1064 B u d a p e s t , Alkotmny u. 21.
Manuscripts and editorial correspondence should bo addressed to
Acta, Antiqua
H-1303 B u d a p e s t P.O.Box 24
Subscription information
Orders should be addressed to
K U L T U R A Foreign Trading Company

H-1389 B u d a p e s t P.O.Box 14(1
or t o its representatives abroad
Acta Antiqua is indexed in Current Contents

ACTA A N T I Q U A
ACADEMIAE S C I E N T I A R U M
H U N G ARI CAE
ADIUVANTIBUS
I. B O R Z S K , I. H A H N ,
ZS R I T O K , . S Z A B , S. S Z D E C Z K Y - K A R D O S S
REDIGIT
J. HARMATTA
TOMUS XXVIII

A K A D M I A I KIAD, B U D A P E S T
1980
ACTA ANT. H U N G .
INDEX
P. Daffin : On Kaspapyros and the So-Called 'Shore of t h e Seythians' 1

H. W. Haussig : Die ltesten Nachrichten der griechischen u n d lateinischen Quellen
ber die R o u t e n der Seidenstrae nach Zentral- u n d Ostasien 9
B. A. LitvinskiyI. Ii. Pichikyan: Monuments of A r t f r o m t h e Sanctuary of Oxus
(Northern Bactria) 25
I. Borzsk: Zentralasiatischc Elemente in dem Alexanderroman 86
. Staviskiy : K a r a Tepe in Old Termez. A Buddhist Religious Centre of t h e K u s h a n
Period on t h e Bank of the Oxus 96
J. Wolski: Les sources de l'poque hellnistique et p a r t h e de l'histoire d ' I r a n . Diffi-
cults de leur interprtation et problmes de leur valuation 137
F. A. Davidovich : The First Hoard of Tetradrachmas of t h e K u s n a 'Heraios' 147
I. Gershevitch: The Colophon of the N O K O N Z O K I n s c r i p t i o n . . ' . . ' 179
J. Harmatta : K i d a r a and the Kidarite H u n s in Kamr 186
Ph. Gignoux: Titres et fonctions religieuses sasanides d'aprs les sources syriaques
hagiographiques 191
A. D.H. Bivar : Questions of Interpretation in the Inscriptions of the Sasanian Seals 205
K. Czegldy : Zur Geschichte der Hephthaliten 213
M. Grignaschi: L a chute de l'Empire Hephthalite d a n s les sources byzantines et
perses et le problme des Avar 219
I. Ecsedy: Western Turks in Northern China in the Middle of t h e 7th Century 249
Bo Utas: The Pahlavi Treatise Avdeh sahikh Sakistn or "Wonders and Mag-
nificence of Sistan" 269
M. Matth: Die X a l a g in den arabischen Quellen 269
I. Gershevitch: The Bactrian F r a g m e n t in Manichean Script 273
I. Gershevitch: B e a u t y as the Living Soul in Iranian Manischeism 281
W. Sundermann: Probleme der Interpretation manichisch-soghdischer Briefe. . . . 289
M. Maroth: Ptolemaic Elements and Geographical A c t u a l i t y in al-uwrizm's
Description of Central Asia 217
I. Frhlich : Historiographie et aggada dans le Liber A n t i q u i t a t u m Biblicarum du
Pseudo-Philon 353
T. Nagy: Die zwei Neratii Prisci und Neratius Priscus der J u r i s t 411
Cs. Tttssy : The Variants of t h e F r a m e Story of the S u k a s a p t a t i 437
t J . Horvth (7. oct. 19113. fv. 1977). (J. Harmatta) 466
D. D. Moukanos: 'O TQCITIO TOV elvai rwv fiaihj/iarixwv vrixei/ivv xax rv
xal rv 'Agiaxoxhrj (Ree. . Szab) 469
R . Liver: Die Nachwirkung der antiken Sakralsprache im christlichen Gebet des
lateinischen und italienischen Mittelalters. (Ree. I. Boronkai) 460
1 P. DAFFIN
ON KASPAPYROS AND THE SO-CALLED

'SHORE OF THE SCYTHIANS'
This paper has no claim for saying anything substantially new. I t s aim
is just to recall a few facts t h a t strangely enough have been completely over-
looked by recent writers on the subject.
I t is well-known t h a t K a s p a p y r o s is mentioned in ancient literature only
three times : once by Hecataeus of Miletus and twice b y Herodotus. L a t e r on,
towards the end of the first century A. D., the Periplus of the Erythrean Sea
mentions the vno r\ Karrvovoivrj.1 As KarrvovQlvrj does not make a n y sense,
Carl Mller proposed to correct it into *KaananvQt]vrj, while Herzfeld suggested
t h e reading *Kan(n)aovQlvrj? Both t e r m s were understood as adjectival forms
f r o m * Kaspapura or *Kappavur(a), t h e supposed I n d i a n original of Greek
Kaspapyros. This would then be a f o u r t h mention of, or rather allusion t o t h a t
city, but it only rests on conjectural emendations.
For their information about K a s p a p y r o s both Hecataeus and H e r o d o t u s
depended on the now lost Periplus of Skylax of K a r y a n d a . According t o Hero-
dotus, Skylax and others were e n t r u s t e d by King Darius with the t a s k of ex-
ploring the Indus down to its m o u t h . They did it and voyaging over t h e sea
westwards, they came in the thirtieth m o n t h to that place whence the E g y p t i a n
king (i.e. P h a r a o h Nekho, 610 595 B. C.) sent the Phoenicians afore-men-
tioned to sail round Libya. After t h i s circumnavigation Darius subdued the
Indians and made use of this Sea. 3
Taking literally the words of Herodotus, scholars have generally separated
t h e exploration from the conquest. 4 B u t H . Schiwek has conclusively demon-
1
Periplus, 48 (edited by H . FRISK : Le Priple de la Mer Erythre. Gteborg
1927. 16). For the date of the Periplus see A. DIHLE : Umstrittene Daten. Kln
Opladen 1965. 9 35; I d e m : Die entdeckungsgeschichtlichen Voraussetzungendes Indien-
handels der rmischen Kaiserzeit, in H . TEMPORINI and W. HAASE : Aufstieg und Nieder-
gang der rmischen Welt. I I . 9,2. Berlin New York 1978. 570 n . 74.
2
C. MLLER: Geographi Graeci Minores. I . Paris 1865. 183; E . HERZFELD:
Early Historical Contacts between the Old-Iranian Empire arid India. Fest. J. Ph.
Vogel. Leyden 1947. 183; I d e m : The Persian Empire. Wiesbaden 1968. 340. See also
H . G. RAWLINSON : Intercourse between India and the Western World. Cambridge 1922.
1617 n. ; J . I. MILLER: The Spice Trade of the Roman Empire. Oxford 1969. 89.
3
Herodotus, IV. 44 (edited and translated by A. D. GODLEY, vol. I I . London
New York 1921).
4
See, for instance, V. MARTIN : L a politique des Achmnides. Museum Helveti-
cum 23 (1965) 43, Ici encore l'exploration prcde et prpare la conqute. Also J . I .
MILLER : The Spice Trade of the Roman Empire. 248 252.
1 Acta Antiqua Academiae ScierUiarum Hungurirae 28, 1H80

2 P. D A F F I N
s t r a t e d t h a t it could n o t be so and t h a t in this case exploration and conquest

were only one thing. 5
HinduS, i.e. t h e basin of the Indus, first appears among t h e provinces of
D a r i u s ' empire in t h e hieroglyphic portion of one of the inscriptions of Tell el-
M a s k h t a and, at t h e s a m e time, in the hieroglyphic portion of t h e inscriptions
e n g r a v e d on the s t a t u e of Darius discovered a t Susa in 1972." According t o
H i n z all this epigraphic material can be d a t e d t o 496 B. C., namely to a couple
of y e a r s after the completion of the canal between Nile and R e d Sea whose
works were started a b o u t 510 B. C. and finished in 498 B. C.7 As it has always
a p p e a r e d reasonable t o connect the expedition of Skylax with the building of
t h e canal, t h a t expedition, and therefore t h e conquest of P a n j b andSindh, can
safely be dated around 510 B. C., or between 510 and 498 B. C.
According to H e r o d o t u s the starting-point of the expedition was the city
of K a s p a p y r o s in t h e ) yrj, or land of t h e IJxrvec,.8 I n all the MSS. t h e
a c t u a l reading is KaanrvQo, to be precise Kaanarvpov in t h e genitive case,
with t h e variants KOLTIVQOV and KaaraTivoov B u t confusion between pi (xaana-
TIVQOV) and tau (xacsna TVQOV) is quite normal in the transmission of Greek
t e x t s a n d as -UVQO/-TIVQOV is almost certainly t h e Indian word pur A, 'city', so
t h e r i g h t spelling of t h e name must probably have been , which is
t h e f o r m it has in H e c a t a e u s as we shall presently see, and t h e one I will use
here throughout. Paktues and Paktuik have been the subject of many specu-
lations into whose details it is not material t o go here. 1 0 Probably, they are ety-
mologically connected with Vedic Paktah.11 This, however, does not help too
5
H . SCHIWEK : D e r Persische Golf als Schiffahrts- und Seehandelsroute, Bonner
Jahrbcher. 162 (1962) 819.
C f . W. GOLNISCHEFF : Stle de Darius a u x environs de Tell el-Maskhoutah.
Recueil de travaux relatifs la philologie et archologie gyptiennes et assyriennes. X I I I .
1890. 99109; G. POSENER: La premire domination perse en Egypte. Le Caire 1936.
184186 ; A. T. OLMSTEAD : History of the Persian Empire. Chicago 1948. 145 n. 42 ;
E . H E R Z F E L D : The Persian Empire. 2 9 3 2 9 4 ; M . K E R V R A N , D . STRONACH, F . VALLAT,
J . YOYOTTE : Une s t a t u e do Darius dcouverte Suse. Journal Asiatique. 260 (1972)
253 266 ; J . YOYOTTE : L e s inscriptions hiroglyphiques de la s t a t u e de Darius Suse.
Cahiers de laD.A.F.I. 4 (1974) 1 8 1 - 1 8 3 .
7
W . HINZ : D a r i u s u n d der Suezkanal. Archologische Mitteilungen aus Iran.
N . S . 8 (1975) 1 1 5 - 1 2 1 .
8
Herodotus, I V . 44,2. See also I I I . 102,1.
9
See critical a p p a r a t u s in . HUDE'S edition, Oxford 19273, vol. I, ad locum; a n d
i n F . JACOBY : Die Fragmente der griechischen Historiker (henceforth FGrHist). I I I . 587.
10
A few bibliographical references will s u f f i c e : J . MARKWART : Untersuchungen
zur Geschichte von Ern. I I . Leipzig 1905. 119 ; H . G. RAWLINSON : Intercourse between
India and the Western World. 16 n. 2 ; G. MORGENSTIERNE : Pushto, Pathan and t h e
T r e a t m e n t of r + S i b i l a n t in Pashto. Acta Orientalin. 18 (1940) 141 - 142 ; H. W. BAILEY :
K u s a n i c a . Bulletin of the School of Oriental and African Studies 14 (1952) 430 431 ;
R . SIIAEER : Ethnography of Ancient India. Wiesbaden 1954. 19 n. 3 ; O. CAROE : The
Pathans. London 1964 2 . 33 37 ; A. D. H . BIVAR : Indo-Bactrian Problems. Numismatic
Chronicle. Seventh series 5 (1965) 95 n. 2 ; H . TREIDLER: Paktuik, in A. PAULY
G . WISSOWA W. KROLL K . WITTE : Realencyclopdie der Classischen Altertumswissen-
schaft (hereafter RE), Supplementband X . S t u t t g a r t 1965. Cols. 475 476.
11
Seo M. MAYRHOFER : Etymologisches Wrterbuch des Altindischen. I I . Heidel-
berg 1963. 183.

ON KASPAPYROS AND T H E SO-CALLED ' S H O R E OF T H E SCYTHIANS' 3
much in locating them on the map. All we can say is t h a t as Hecataeus p u t s

Kaspapyros in Gandhra, the Paktuik must have been included in t h e Gan-
dhric satrapy, whatever m a y have been t h e extent of t h a t satrapy in Darius'
time. 1 2
The position of Kaspapyros along the bank of a watercourse is beyond any
doubt. Herodotus says t h a t Skylax and his companions set out from t h e city
and sailed down the river towards the east and the sunrise till they came t o
the sea. 13 I t is also certain t h a t t h a t river cannot be the Indus, but only one of
its right-hand tributaries, most probably t h e Kbul-rd. This led Markwart t o
identify Kaspapyros with Puskalvat, the western capital of Gandhra, to t h e
north of the confluent of the K a b u l and Svt rivers. He also explained the Greek
name of the city with the assumption t h a t Puskalvat was then called *Kusu-
mapura, 'city of flowers', which through the Middle-Indic forms *Kus(u)va-
pura > *Kuspapura would have originated the Greek Kaspapyros.14 Other
scholars preferred, however, other identifications of the place and other expla-
nations of the name. Sir Alexander Cunningham traced Kaspapyros back t o
Sanskrit Ksyapapura through a P r a k r i t *Kassapapura; and as yap apura
is an ancient name of Multn in U t p a l a ' s commentary on Varhamihira's
Brhatsamhit, he accordingly identified the city with Multn. 1 5 His opinion was
shared b y Poucher and, at first, also by Herzfeld. 1 6 However, Professor P e t e c h
pointed out t h a t the identification of Kaspapyros with Multn does not tally
to the description given by Herodotus, besides the fact t h a t the mention of
Kyapapura as another n a m e of Multn is not to be found in the e x t a n t
text of Utpala, b u t only in a passage of it as quoted by al-Brn in the X l t h
century A. D. 17 Petech himself was of the opinion t h a t Kaspapyros was t o
be looked for in the neighbourhood of A t a k , or perhaps at Pesvar itself. 18
The identification of Kaspapyros with A t a k was proposed by Herzfeld a f t e r
he had given up the previously accepted identification with Multn. 1 9 The equa-
tion of Kaspapyros with P u r u s a p u r a / P e s v a r was due to Kramers and Olaf
12
See FGrHist, l a . 365,2426.
13
I V . 44,2 (GODLEY'S translation, see note 3 above).
14
See J . MARK WART : Untersuchungen zur Geschichte von Ern. I I . 242, 246 n. 3.
15
See A. CUNNINGHAM : Ancient Geography of India. Oxford 1871. 234 ; new edi-
tion, Calcutta 1924. 267. Cf. also . CH. LAW: Historical Geography of Ancient India.
Paris, n. d. 112.
16
See E . HERZFELD : Zarat hustra. Archologische Mitteilungen aus Iran 1 (1929
30) 93 9 4 ; I d e m : Die Magna Charta von Susa. ibid. 3 ( 1 9 3 0 - 3 1 ) 101 n. 1 ; I d e m :
Sakastn. ibid. 4 (1931 32) 10 n. 3 ; A. FOUCHER : Les satrapies orientales de l'empire
achmnide. Comptes rendus de l'Acadmie des Inscriptions 1938.347348; I d e m : La
vieille route de l'Inde. . Paris 1947. 194, 198, 215 n . 6, 252.
17
See L. PETECH: Northern India according to the Shui-ching-chu. R o m a 1950.
67 and footnote 1.
18
Ibidem.
19
E . HERZFELD : E a r l y Historical Contacts between the Old-Iranian E m p i r e
and I n d i a . 182183 ; I d e m : The Persian Empire. 340.
1* Acta Antiqua Acaderr.iac Scientiarvm Hungaricae 28, 1980

4 P. DAFFIN
Caroe. 2 0 B u t neither A t a k nor Pesvar suits Herodotus' words any better t h a n

Multn did, since Herodotus says t h a t in t h e first part of their voyage Skylax
a n d his companions sailed down the river towards the east and the sunrise,
a n d this can only refer t o a right-hand t r i b u t a r y of the Indus, most probably
t h e Kbul-rd. 2 1
Writing a few decades earlier t h a n Herodotus, but drawing from the same
source, Hecataeus p u t Kaspapyros in G a n d h r a ( Kaonnvgo, nofo )
a n d glossed the s t a t e m e n t with the words Zfxv&cbv ctxrrj. I purposely say gloss-
ed because quite certainly the words Zxv&v xxij do not go back to Skylax,
b u t are Hecataeus' own words explaining t h e location Kaspapyros had on his
m a p . For Hecataeus' Periegesis was nothing else b u t a description and ex-
planation of a m a p and cannot be understood unless one constantly bears
this fact in mind. 2 2
Hecataeus' Kaspapyros fragment has been handed down to us by Stepha-
nus Byzantinus whose MSS. do not supply, in this case, any variant reading. 2 3
The t e r m akt is not met with elsewhere in t h e extant fragments of the Peri-
egesis, b u t a f e w e x a m p l e s of its use can be collected from Herodotus. I n a num-
ber of cases (IV, 38 ; 39 ; 177) akt simply indicates a tongue of land projecting
into the sea and in IV, 41 it is applied to L i b y a : f\ xTt) fjrtg Atvrj xexkrjxai.
I n other passages akt is, more in particular, a headland, a promontory (even
a rocky one), a cape (VII, 33 ; 3 4 ; 183,3; 188,1; 191,2; I X , 120,4). Its second
meaning is t h a t of 'coast', 'shore' (VII, 45 ; V I I I , 95 ; compare Latin acta) and
in this sense akt is also found in proper names of seaside places, such as Katy
xrrj in Sicily (VI, 22,2 ; 23,2) and Aevxrj xrrj in Thrace (VII, 25,2). Addition-
al meanings, semasiologically related to those attested b y Herodotus, as for
instance 'river-bank' and 'edge', can be easily detected in passages of other
authors and any good Greek dictionary will show them.
Now w h a t is puzzling in the K a s p a p y r o s fragment is t h e fact t h a t there
akt is seemingly referred to not as a t r a c t of land running out into the sea nor
as a river-bank, b u t as something related t o a people : the Scythians. Of course,
the oddity of this usage may be only a p p a r e n t and due to our ignorance of the
whole context to which this particular f r a g m e n t belonged. At any rate t h a t
20
J . H . KRAMERS : Peshawar. Annual Bibliography of Indian Archaeology. X V .
1950. X V I I I , X X I ; O. CAROE: The Pathans. 3 0 - 3 3 a n d 441 n . 15.
21
H. G. RAWLINSON : Intercourse between India and the Western World. 16 17 n . ;
A. HERRMANN: K a s p a p y r o s . RE, X , 2. S t u t t g a r t 1919. Col. 2270, 5 2 - 5 6 ; FGrHist,
l a , 365,21 2 3 ; H . SCHIWEK : Der Persische Golf als Schiffahrts- u n d Seehandelsroute.
12 13.
22
See F. JACOBY: Hekataios. RE, VII. Stuttgart 1912. cols. 2667 2 7 5 0 ;
L . PEARSON: Early Ionian Historians. O x f o r d 1939. 2 8 ; FGrHist, l a . 328,47; H .
BENGTSON a n d V. MILOJI : in Grosser Historischer Weltatlas. I. Erluterungen. Mnchen
19583. Cols. 58-59.
23
See Steph. Byz., Ethnicorum quae supersunt, ed. A. MEINEKE. I. Berlin 1849.
364,8-9.

ON KASPAPYROS A N D T H E SO-CALLED 'SHORE OF T H E SCYTHIANS' 5
should have been a warning against any hasty solution of the problem. B u t it
was not so.
Unembarrassed by t h e aforesaid difficulties, Markwart offered a quite
plain and easy interpretation of Hecataeus' words. To him it was a m a t t e r of
course t h a t Kaspapyros being the starting-point of Skylax' voyage, Zxvdotv
xr) could only mean 'coast of the Scythians' or r a t h e r 'Sakas'. For 'Sakas'
Markwart meant the Sak H a u m a v a r g whom he located between G a n d h r a
and Bactria. The conclusion which followed was obvious Der Ausdruck
xrrj . . . hngt wohl d a m i t zusammen, da hier die Schiffahrt auf dem I n d u s
begann und jene S t a d t ein wichtiger Stapelplatz f r den Handel zwischen
Indien und den Lndern im Norden und Osten des Hindukus und P a m i r ,
den die Saken vermittelten, war. 24 In this way K a s p a p y r o s was promoted
to the rank of an i m p o r t a n t riverine port, without t a k i n g into account t h a t
since then it was never heard of again and completely vanished from t h e
scene of history, which is indeed an unusual lot for a wichtiger Stapelplatz. Less
t h a n two centuries later, Alexander did not meet with Sakas or with traces of
their past presence on his way from Bactria to India, nor he found in G a n d h r a
any such riverine trading-station as the one conceived of by Markwart. None-
theless, the legend spread on Markwart's authority t h a t in the V l t h century
B . C . Saka tribes were settled in North-West India and were facing K a s p a p y r o s
from the other side of the river. So Kaspapyros could be called by Rawlinson
a frontier city of Gandhra, on the Scythian borderland. 25
Widely accepted as it was, this view, however, was not shared by all
scholars. Those better acquainted with historical geography, as for instance
Albert Herrmann, openly rejected it. According t o H e r r m a n n , the expression
Hxvwv xrj could not be understood in the context of t h e fragment and there-
fore, he regarded the correction of xrij into am rj proposed by Wilhelm Sieglin,
as the only possible solution. 20 Now the adjective vrlog ( -rj, -ov), with its adver-
bial form vrlov, has a well definite meaning in Greek geographical terminology.
To use Wikn's words, Durch vrlov und dergleichen wird in der geographi-
schen Terminologie immer der Wohnort jenseits eines scheidenden Wassers
bezeichnet. 27 This s t a t e m e n t can be aptly illustrated b y three examples also
t a k e n from Herodotus. In I, 202 the people of the Massagetae, to the north of
the Araxes, is said to be vrlov, t h a t is to say opposite t o t h e Issedones. In I I ,
34,1 Egypt is said to be approximately vrh] to mountainous Cilicia, while in
24
J MARKWART : Untersuchungen zur Geschichte von Krn. I I . 140, 242.
25
H . G. RAWLINSON : Intercourse between India and the Western World. 17.
24
A. HERRMANN : Kaspapyros Unverstndlich ist hier der Ausdruck: Exvdwv
xx) . . . Sinngem wird der Satz erst, wenn wir mit Sieglin das Wort xxr'j in vxlrj
verbessern.
27
E . WIKN : Die Ansichten der Hollenen ber den Nordrand der Oikuinene vor
Pytheas. Fest. M. F. Nsson. L u n d 1939. 549.
Acta Antiqua Acadcmiae ScierUiarum Hungaroae 28 , 1980

6 P. D A F F I N
I I , 34,2 t h e city of Sinope is said to lie vriov to the mouth of the D a n u b e which
according to Herodotus was to be found on t h e northern coast of t h e Black Sea.
I t is clear t h a t in each of these three cases vrio or vriov serves to locate two
geographical points along a north to south vertical or, as we would say, on the
same meridian, without implying b y t h a t a n y close proximity between them.
T h e y are in fact far a w a y from one another and separated moreover by a water
which in t h e first example is the Caspian or Hyrcanian Gulf conceived as elon-
gated t o w a r d s the east, in the second one it is the Mediterranean, in the third one
the B l a c k Sea28 In accepting Sieglin's correction Herrmann was well aware of
this particular value of vrio and accordingly he believed t h a t t h e Scythians
and K a s p a p y r o s were facing one another in a north to south direction. Not even
in t h i s case, to be sure, was any close proximity implied, since for H e r r m a n n the
S c y t h i a n s of the f r a g m e n t were not t h e Sakas of Central Asia, b u t the regular
S c y t h i a n s of South Russia. According t o him on Hecataeus' m a p Scythia
s t r e t c h e d east till the point the Caspian Gulf joined the great outer sea, so t h a t
G a n d h r a was actually to be found right to the south of Scythia. B u t let us
q u o t e H e r r m a n n ' s own words nach der Hekataioskarte u n t e r den gegenber-
w o h n e n d e n Skythen nicht die Saken . . . , sondern tatschlich die Skythen zu
v e r s t e h e n sind, da zwischen G a n d h r a u n d den Skythenlande der Kaspische
Meerbusen zu denken ist. 29 And again Wenn wir uns . . . vergegenwrtigen,
d a H e k a t a i o s das Skythenland bis zu der Verbindung des Kaspischen Meeres
mit d e m Okeanos eingezeichnet hat,sdlich von demselben Meer aber Gandhra
u n d Indien, so liegt f r ihn Gandhra den Skythen gegenber. Mit den Sakai
h a t also das Zitat des Stephanos nicht d a s Geringste zu tun. 30
Strengthened b y H e r r m a n n ' s arguments, the correction proposed by
Sieglin was accepted b y J a c o b y in his edition of the Hecataeus' fragments.
I n t h e critical a p p a r a t u s he only suggested t h a t the original reading might have
been vrirj xerai, 'lies opposite'. 3 1 In fact, Sieglin's reading is h a r d l y satisfactory
unless a verb is understood after it.
Surely, Sieglin a n d H e r r m a n n h a d come near to, b u t not arrived at the
p r o p e r solution of t h e problem. To f i n d it out, it was indeed t h e merit of E .
H e r z f e l d . Starting f r o m t h e masterly reconstruction t h a t J a c o b y himself had
given of Hecataeus' m a p , Herzfeld did not think it necessary t o emend the t e x t
of t h e fragment a n d read it as it is. As Markwart h a d done before him,
H e r z f e l d took akt in its usual sense of 'coast' or 'coast line', b u t a t the same time
p o i n t e d out that on Hecataeus' m a p this coast line served as a sort of parallel ;
t h e r e f o r e the expression ZJxvwv / came to mean t h a t Kaspapyros had,
so t o say, the same degree of latitude as t h e Scythians of South Russia. Das
28
On the H y r c a n i a n Gulf separating Massagetae and Issedones see E . WIKN :
loc. cit.
29
A . HERRMANN : K a s p a p y r o s . C o l . 2 2 7 0 , 1 9 27.
30
A. HERRMANN : Sakai. RE. zweite Reihe I I . 1920. Col. 1772,20 42.
31
FGrHist, IA, 3 8 , 1 4 - 1 5 ; l a , 3 6 5 , 3 9 - 4 1 .

ON KASPAPYROS AND T H E SO-CALLED 'SHORE OF T H E SCYTHIANS' 7
Bruchstck Herzfeld concluded besagt nicht etwa, da die Skythen an

einem Meer, noch da Kaspapyros nahe dem Sitz der Skythen lag. 32
Herzfeld's argument should have settled the question once for all, ruling
out for ever any a t t e m p t to use t h e Hecataeus' fragment as a proof of the exis-
tence of Saka tribes in North-West India in the V l t h century B. C. B u t long-
cherished ideas are hard to die, more so when they suit so well one's precon-
ceived theories about the ancient ethnography of this or t h a t part of the world.
So Julius J u n g e , though writing some years later t h a n H e r r m a n n and Herzfeld,
practically reverted to the position of Markwart, only replacing the Sak H a u -
mavarg by the Sak Tigrakhaud. He did not take into account H e r r m a n n ' s
argument and only cursorily referred to t h a t of Herzfeld. Significantly enough,
he felt unable to decide between xr/ and vrir], as if a reading actually handed
down could be p u t on equal level with a conjectural one. 3 3
I n his posthumous work on the Persian empire, however, Herzfeld rightly
confirmed his previous statement mxrij must not be 'corrected' . . . in vrir],
for it is in Hecataeus' idiom a kind of parallel running along a coast line. Zxv&wv
shows t h a t Hecataeus' map p u t Scythia and P a k t y i k under the same latitu-
de. 34 B u t once again Herzfeld's voice was t h a t of one crying in the wilderness.
J u s t one year after t h e publication of Herzfeld's posthumous book, a
s t u d y on the Sak H a u m a v a r g by Professor Litvinskij appeared in F r a n z
Altheim's Festschrift. There can be read about Kaspapyros more or less t h e
same things already said by Markwart and Junge. Exv&wv s xrr] is rendered as
'harbour (gavan') of the Scythians', while the emended reading Exv&wv vrir]
is indifferently interpreted either as 'opposite side (storona protivopoloznaja) of
the Scythians', or else as vorota v stranu slcifov, 'gateway t o t h e country of t h e
Scythians'. Hence the unavoidable conclusion t h a t since t h e time of the first
Achaemenids there were Saka tribes bordering on India. 3 5
More recently the Hecataeus' f r a g m e n t has been anew examined b y
Professor Tucci who also shares Markwart's and Junge's opinions, being ap-
parently unaware of those of H e r r m a n n and Herzfeld. Professor Tucci seems t o
be also unacquainted with the correction proposed by Sieglin and with the dis-
cussion it raised. According to him akt is not a port, it is a shore with
easy access, a bay ; it indicates a bend of the Kbul river with ample a n d
shallow waters. Exv&wv xrrj means, therefore, shore of t h e Scythians ; to
which fatally follows the usual conclusion t h a t , since those times, Scythian
tribes were pressing on Gandhra. 3 6
32
E . HERZFELD : S a k a s t n . 10. Cf. F . JACOBY: H e k a t a i o s . Cola. 2 7 2 5 2 7 2 6 .
33
J . JUNGE : Saka-Studien. Leipzig 1939. 32 and n . 2, 50 a n d n. 5, 83 a n d n . 3.
34
E . HERZFELD : The Persian Empire. 338 n. 5.
35
B. A. LITVINSKIJ : Sak H a u m a v a r g . Fest. F. Altheim. I. Berlin 1969. 118 119-
36
G. T u c c i : On Swt. The D a r d s a n d Connected Problems. East and West. 27
(1977) 16-17.
1*ActaAntiqua Acaderr.iac Scientiarvm Hungaricae 28, 1980

8 P . DAFFIN: ON K A S P A P Y R O S AND T H E SO-CALLED ' S H O R E O F T H E SCYTHIANS'
Leaving it to professional Greek scholars to decide whether akt m a y be

so forced as to have t h e meaning of ' b a y ' o r o f ' b e n d of river with ample a n d
shallow waters', I would only remember t h a t more t h a n f o r t y years ago T a r n
h a d already ironically remarked t h a t Asia was getting very full of Sakas. 37
The Kaspapyros case is an instructive one. I t shows in a palpable way
t h a t m a n y times problems subsist only because their solution has fallen into
oblivion. I t also shows t h a t to reconstruct the story of a problem is, more of-
t e n t h a n not, to solve t h a t problem.
Rome.
Postscript. When this paper was read on the occasion of the Third
Conference on the Historical Sources for Pre-Islamic Central Asia, Professor
H e l m u t H u m b a c h kindly pointed out t h a t one more mention of Kaspapyros
is provided by the Tabula Peutingeriana where between Alexandria Bucefalos
a n d Palibotra a place is to be found by the name of Spatura (Segmentum
X I I , 3; cf. . Miller, Itineraria Romana, S t u t t g a r t , 1916, col. 801). Accord-
ing t o Professor H u m b a c h (if I rightly understood his words) Spatura should
be emented into caSpaiira-Kaspatyros. This is in fact an old hypothesis a n d
n o t t h e only one devised to explain t h e place-name Spatura. Wilhelm To-
maschek (Die Strassenzge der Tabula Peutingeriana", Sitzungsberichte der
Wiener Akademie der Wissenschaften, 102, 1883, p. 260) was of the opinion
t h a t Spatura corresponded to Simtura in the Cosmography of the R a v e n n a s
A n o n y m u s (II, 3, 2; cf. J . Schnetz, Itineraria Romana, I I , Lipsiae, 1940, p.
17) and, therefore, should be identified with Saltura. I n any case Professor
H u m b a c h ' s *Caspatura, like Mller's *\ and Herzfeld's *( ja-
ovglvr], is not an actually handed down reading, b u t only a conjecturally
reconstructed one.
On the same occasion Dr. A. D. H . Bivar suggested t h a t in the Heca-
t a e u s ' fragment t h e words Rxv&wv xry might be a gloss added by a later
copyist who knew t h a t N o r t h - W e s t India h a d actually been, for some time,
u n d e r the sway of Saka tribes (compare t h e 'Skythia' of Periplus 38 39,
a n d t h e 'Indoskythia' of Ptolemy, V I I , 1, 55). Should Dr. Bivar's hypothesis
prove right, it would strike a deadly blow to the belief in the existence of
S a k a tribes in N o r t h - W e s t India at the time of Darius t h e Great. For being
Zxv&wv xrrj a later gloss, Hecataeus would be left with the only words
KaannvQo, nli Favanixrj and there would be no question at all of a 'shore
of t h e Scythians' in the V l t h century . C. Quod erat demonstrandum.
3
' W . W. TARN: The Greeks in Bactria and India. Cambridge 1938 (19512). 287.
Acta Antiqua Academiae Scientiarum Hungaricae 28, 1980

H. W . H A U S S I G
DIE LTESTEN NACHRICHTEN DER GRIECHISCHEN

UND LATEINISCHEN QUELLEN BER DIE ROUTEN
DER SEIDENSTRASSE NACH ZENTRAL-
UND OSTASIEN
Die in diesem R a h m e n angesprochenen Berichte aus der rmisch-grieehi-

schen Antike sind der Forschung seit langem bekannt. Trotzdem hat ihre bishe-
rige Interpretation nicht die Beziehung ihrer Angaben auf die Routen der Sei-
denstrae nach Zentral- und Ostasien deutlich machen knnen. Der Grund
liegt darin, da hierfr die Erkenntnisse der Frhgeschichte nicht gengend u n d
die Beziehung ihrer Nachrichten auf die drei verschiedenen R o u t e n der Seiden-
strae berhaupt nicht beachtet wurden.
Hinzu kommt, d a man bisher nur eine einzige R o u t e der Seidenstrae
k a n n t e , deren E n t s t e h u n g man in die Mitte des ersten J a h r h u n d e r t s v. Chr.
setzte. Hierbei ging m a n von der Vorstellung aus, da erst von jenem Zeitpunkt
an, zu dem die chinesische E x p a n s i o n im Westen mit dem Pamir- und Alai-
Gcbirge die Grenze zur griechisch-iranischen Welt erreicht h a t t e , eine Verbin-
dung zwischen China u n d dem Westen mglich gewesen wre. E s zeigt sich aber
durch eine Neu-Interpretation des Quellenmaterials, da eine Verbindung zwi-
schen China und der westlichen W e l t von dem Bestehen einer gemeinsamen
vom Alai- und Pamir-Gebirge gebildeten Grenze unabhngig war.
Sie bediente sich in den J a h r h u n d e r t e n , die vorangingen, jener Vlker als
Vermittler, die je nach der Route der Seidenstrae entweder als Issedonen oder
Serer bezeichnet wurden. Die Chinesen, denen sie ebenso wie den Griechen u n d
Iraniern als Vermittler dienten, n a n n t e n sie Ye-chi oder Wu-'sun.
Die Nordroute und ihre Vorgeschichte
Sowohl die Bodenfunde wie die Nachrichten der schriftlichen Quellen

machen deutlich, da die lteste Verbindung zwischen diesen beiden Teilen der
Welt, die die Bezeichnung Seidenstrae verdient, sich nicht der Wege ber
die Psse des Alai- u n d Pamir-Gebirges bediente, sondern f r ihren Weg nach
China eine Route im Norden b e n u t z t h a t . Diese Route, die von Herodot u m
430 v. Chr. beschrieben wurde, l t sich auf Grund seiner Beschreibung bis in
den Westen der spteren chinesischen Provinz Kan-su verfolgen. Nach ihr be-
gann diese Strae an der Mndung des Don. Sie wandte sich zunchst nach
Norden, ehe sie d a n n nach Osten a b b o g und das damals von den P a r t h e r n be-
Acta Antiqua Academiae Scientiarum Hungaricae 28,1980

10 H. W. HAUSSIG
wohnte Gebiet erreichte, v o n dort benutzte sie einen im Norden des Tien-chan
entlang fhrenden Weg, der entweder ber die Turfanoase oder ber Hami in
den Westen der spteren chinesischen Provinz Kan-su f h r t e .
D a s Auffllige an der Beschreibung dieser Strae schien bisher ihr Abbie-
gen nach Norden zu sein, in ein Gebiet, das Herodot als Waldzone bezeichnet. 1
Dieses Rtsel lsen die Bodenfunde. Seit der Aufdeckung der Begrbnispltze
von Seima und Turbino im Westen des mittleren Ural in der Nhe jenes Weges,
der, der K a m a folgend, ber den Pass des Ural nach Sibirien f h r t , wei man,
da es zwischen dem U r a l u n d China in vorgeschichtlicher Zeit eine Handels-
verbindung gegeben h a t . E s sind Ringe aus J a d e , die dort gefunden wurden,
die in F o r m und Material mit jenen identisch sind, die aus China bekannt sind
u n d hier auch bei der Verehrung des Himmelsgottes durch den Knig eine Rolle
spielten. 2 D a J a d e im U r a l bisher nicht gefunden wurde, knnen die Ringe auch
vom Material her gesehen, n u r aus China stammen. H i n z u k o m m t , da die Gr-
ber nicht nur diese Jaderinge, sondern auch Bronzexte u n d Messer des gleichen
T y p s enthielten, der im China der spten Shang-Zeit zu finden ist. 3 Demnach
scheint schon am E n d e des zweiten J a h r t a u s e n d s v. Chr. ein Handelsverkehr
mit China bestanden zu haben.
J e d e r Handelsverkehr setzt eine Gegenseitigkeit voraus. Man wird also
fragen, was China als Gegenleistung f r seine Lieferungen an Jaderingen und
Werkzeugen aus Bronze aus dem Westen erhielt. F r den Ural kme hier das
bei der Bronzeherstellung wichtige K u p f e r in Frage. E s wurde damals im Ural
a b g e b a u t . Aber auch kostbare Steine knnen von dort ausgefhrt worden sein.
D a s war aber, wie vor allem die Rezeption aus dem Westen in China beweist,
nicht allesi Es seien hier n u r zwei Dinge genannt : die E r f i n d u n g des Wagens
u n d der Kopf des Lwen.
Die Rezeption des griechischen Streitwagens durch China
Der Zeitpunkt f r die Rezeption des Wagens in China lt sich aus den
bilderschriftlichen Zeichen der Shang-Zeit nachweisen. 4 D a das Zeichen zu
Beginn der Shang-Zeit mit dem A u f k o m m e n dieser Hieroglyphen vorkommt
u n d in der lteren Chou-Zeit weiter ausgebildet wird, mssen die Anfnge der
1
Herodot, Historien IV, 21 27. Den ersten Teil dieser Wegbesehreibung benutzt
Herodot dazu, u m die Ostgrenze des von skythischen S t m m e n in Sdruland bewohn-
t e n Gebietes abzugrenzen.
2
Vgl. hierzu W. WATSON : Cultural Frontiers in Ancient Asia. Edinburgh 1966.
S. 6960.
3
E s waren Bronzexte m i t Tllen aber vor allem Messer mit Tierdarstellungen,
die bis in die Zeit um 1500 v. Chr., die beginnende Shang-Zeit, hinabreichen. Bronzexte
des gleichen Typs fanden sich n a c h Watson auch im Tal des oberen Jenissei.
4
Das Zeichen fr Wagen kommt schon u n t e r den ltesten Zeichen der Hiero-
glyphen auf den Orakelknochen vor und besitzt dort die gleiche Gestalt wie auf den
ltesten Felszeichnungen.

e
a
I3
S'
Die R o u t e n der Seidenetrae

12 H. W. HAUSSIG
R e z e p t i o n in die zweite H l f t e des zweiten J a h r t a u s e n d s gesetzt werden, also

in die gleiche Zeit, in der die Jaderinge, die Bronzexte u n d die Messer nach
dem U r a l gekommen sind.
D e n Weg, den der Wagen nach China benutzt h a t , bezeichnen vor allem
die Felszeichnungen. 5 H i e r n a c h mu die Vermittlung des Wagens ebenso wie
der E x p o r t der Jaderinge ber die nrdliche R o u t e der Seidenstrae erfolgt
sein. Felszeichnungen begleiten vor allem das letzte Stck seines Weges nach
China, jene Strecke d u r c h das Gebirge, wo steile oder berhngende Steinwnde
Felszeichnungen mglich machten. Sie finden sich daher in T a n n u Tuwa, an den
Gebirgshngen des Altai u n d vor allem an den Felswnden im Changai-
Gebirge und weiter im Sden, bis dort, wo die Grenzen des alten China began-
nen. Dargestellt wurden Wagen mit zwei R d e r n , teils als Quadriga von vier
P f e r d e n , teils von zwei P f e r d e n gezogen. Von diesen beiden T y p e n scheint man
in China den mit zwei P f e r d e n bespannten Wagen bevorzugt zu haben. Ihm be-
gegnet m a n in chinesischen Grbern wie z. B. in dem von Shang-ts'un-ling aus
der P r o v i n z Honan aus d e m 8. bis 7. J a h r h u n d e r t v. Chr. 6 Die H e r k u n f t der
Quadriga, die in Ostasien vor allem durch die Felszeichnungen in der Mongolei
nachweisbar ist, 7 verleugnet nicht ihre H e r k u n f t aus der griechisch-mykeni-
schen Welt. Selbst in Einzelheiten, wie den acht Speichen der R d e r , gleichen
sie ihren europischen Vorbildern, die als Darstellungen auf Bronzeblechen in
Olympia und aus Schilderungen durch die Rias bekannt sind. Die bereinstim-
m u n g mit den griechischen Streitwagen gilt auch f r die F o r m der Anschir-
r u n g der Pferde und f r die Sttze des Lenkers auf dem Wagengestell, die ihm
das Festhalten whrend des Fahrens ermglichte. 8
Die Rezeption des griechischen Lwenkopfes als T'ao-t'ieh durch China
Auf dem gleichen Wege wie der Streitwagen wurde auch der Lwenkopf
nach China vermittelt. Die bermittlung erfolgte hier wahrscheinlich durch die
E p i s e m a t a , die meist aus Bronze gearbeiteten Schildzeichen. I n China, wo es
keine Lwen gibt, wurde er offenbar zur gleichen Zeit wie der Wagen rezipiert.
E s f i n d e t sich dort das Bild des von oben gesehenen Lwenkopfes, wie ihn z. B.
5
Vgl. Hierzu E . NOWQOROBOWA : Alte K u n s t der Mongolei (deutsche ber-
setzung 1980. Seite 79 99). F u n d e mit Wagendarstellungen vom K a u k a s u s , dem Pamir,
dem Altai und dem Changai-Gebirge lassen den Weg erkennen, den der Wagen von
E u r o p a nach China genommen h a t .
"Vgl. das Foto von ausgegrabenen Wagengrbern des 8. u n d 7. J a h r h u n d e r t s
v. Chr. aus der Zeit dor Chou-Dynastie bei WATSON : . . . Tafel 40. Die hier aufge-
n o m m e n e n Wagen wurden bei Shang-ts'un-ling in der Provinz H o n a n gefunden.
7
Vgl. NOWGORODOWA : Tafel 59. Die Zeichnung zeigt den gleichen halbrunden
Wagenkasten, wie er aus d e n griechischen Darstellungen b e k a n n t ist.
8
Die Darstellung zeigt einen Krieger, der seinen Streitwagen besteigt. Es
h a n d e l t sich um ein Bronzoblech aus der Zeit um 590 v. Chr. Vgl. Olympia von
M. ANDRONIKOS : Athen 1975. Tafel 35.

D I E ROUTEN DER SEIDENSTRASSE NACH ZENTRAL- U N D OSTASIEN 13
eine Lwenmaske auf einer T e t r a d r a c h m e von Rhegion aus der Zeit u m 430 v.
Chr. zeigt. 9 Die Mnze a h m t offensichtlich ltere Wiedergaben, wie sie auf
Schilden blich waren, nach. I n China wurde das griechische Vorbild des von
oben gesehenen Lwenkopfes zur T'ao-t'ieh Maske umstilisiert, 1 0 sie t a u c h t
zuerst auf den Bronzen der Shang-Zeit auf. Auch in diesem Fall lt sich der
Weg nach Ostasien auf der Nordroute der Seidenstrae verfolgen. So begegnet
m a n der Lwenmaske, wenn auch in einem Beispiel aus spterer Zeit, u n t e r den
F u n d e n aus sibirischen Kurganen als holzgeschnitztes Medaillon in der gleichen
Sicht von oben wie bei der Mnze aus Rhegion und dem chinesischen T'ao-
t'ieh. 1 1
Der Ural als Teilhaber am griechischen Chinahandel
Wagen und Lwenkopf kamen aus der mykenischen K u l t u r . D a m i t wird

die Frage gestellt, von wo aus diese Vermittlung nach China erfolgte. N a c h
Herodot begann die Nordroute der Seidenstrae an der Mndung des Don. 1 2
Dort befand sich nach dem lteren Plinius in frher Zeit eine Niederlassung der
Karer, 1 3 jenes Volkes, das in der spteren griechischen Zeit z.B. von T h u k y d i d e s
mit den Trgern der mykenischen und minoischen K u l t u r gleichgesetzt wurde. 1 4
Demnach h t t e es schon in der mykenischen Zeit eine Verbindung mit China
gegeben, die ihren Ausgangspunkt an der Mndung des Don besa u n d ihre
nchste Station an einem Handelsplatz, der auch von Hndlern aus dem U r a l ,
wo die chinesische J a d e und die Bronzewerkzeuge aus China gefunden wurden,
erreicht werden konnte. (Die Mglichkeit, da die griechischen K a u f l e u t e selbst
nach dem Ural gekommen sind, wird f r die mykenische Zeit schon d a d u r c h
ausgeschlossen, da weder der Wagen noch der Lwenkopf, wie das Fehlen von
entsprechenden Zeugnissen beweisen, auf diesem Weg China erreicht haben.)
Eine Verbindung der griechischen K a u f l e u t e mit dem Ural bestand auch in
my kenischer Zeit nur durch den Zwischenhandel. F r die Zeit Herodots wird
das indirekt durch seine Beschreibung der Nordroute der Seidenstrae b e s t t i g t .
Sie begann bei ihm an der Mndung des Don und f h r t e erst nach N o r d e n
durch das Gebiet der Sauromaten und d a n n in die Waldzone zur Stadt der Budi-
9
Vgl. L. U. M. LANCKORONSKI : Mythen und Mnzen. Mnchen 1968. T a f e l 29.
10
Vgl. K . SCHEFOLD: Der skythische Tiorstil in Sdruland. ESA 12 (1938)
Seite 59. SCH. vorwoist hier auch auf die Forschungen von Boroffka, der die W a n d e r u n g
des Motivs anhand der sibirischen F u n d e n verfolgt h a t .
" V g l . S. 1. KUDENKO : Frozon Tombs of Siberia, t h e P a z y r y k Burial of Iron-
Age Horsemen. bersetzt von M. W. THOMPSON. London 1970. Tafel 138. E s h a n d e l t
sich u m12 Holzschnitzerei von einem Sattolzoug aus der Sammlung Frolov.
Herodot IV, 21. E r spricht hier von der Spitze des Maietissees, mit d e m d a s
Asowscho Meer gemeint ist.
13
Plinius N a t . hist. IV. cap. 5 (7), 20 : Oppidum in Tanais quoque ostio. Tenuere
/initima primum Cares.
14
Thukydides I, 3.
2* Acta Antiqua Academiae ScierUiarum Hungaricae 28, 1980

14 H . W. HAUSSIG
ner, 1 5 von wo sie n a c h Osten abbog. 1 6 D e r Handelsverkehr n a c h China k n n t e

sich auf diesem Wege in mykenischer Zeit so abgespielt haben, da die griechi-
schen K a u f l e u t e v o n der Mndung des D o n ihre Erzeugnisse, wie z. B. Gold-
schmiedearbeiten, n a c h der S t a d t der B u d i n e r oder einem anderen damals d o r t
b e s t e h e n d e n H a n d e l s p l a t z brachten u n d h i e r f r dort von d e n Hndlern aus
d e m U r a l K u p f e r u n d kostbare Steine erhielten, mit denen sie dann von d e n
Issedonen Bronzewerkzeuge u n d Jaderinge a u s China k a u f e n konnten, die sie
f r d e n weiteren W a r e n a u s t a u s c h mit den K a u f l e u t e n aus d e m U r a l bentigten.
D e m n a c h wrden also die mykenischen H n d l e r , ber die neben anderem d e r
W a g e n u n d der L w e n k o p f nach China g e k o m m e n waren, auch den Transithan-
del f r Waren von u n d nach dem U r a l b e r n o m m e n haben.
E r s t mit d e m 6. J a h r h u n d e r t scheint auf diesem Wege auch der E x p o r t
v o n Seide aus China begonnen zu haben, m i t d e m dann auch die Ausfuhr von
A r b e i t e n griechischer Gold- u n d Silberschmiede und skythischer Bronzegie-
ereien einsetzte. A n diesem Handel war a u c h der Ural beteiligt.
Bei der Seide w a r e n es nicht n u r einfache Seidenstoffe wie die F r a g m e n t e
a u s d e m F r s t e n g r a b bei der Heuneburg 1 7 u n d die Seidenreste vom Kerameikos
in Athen, 1 7 a sondern auch kostbare Gewebe wie die als Goldenes Vlies be-
k a n n t e Seide, Ziel des v o n J a s o n gefhrten Zuges der Argonauten nach Kolchis.
Sie besa offenbar wie die chinesischen Seiden aus Kertsch u n d Lou-lan 1 8 eine
goldgelbe F o n d f a r b e . D a s gab zu ihrer Bezeichnung Goldenes Vlies Anla.
(Kolchis war d a m a l s E n d p u n k t einer aus d e m Tal des Araxes von der K s t e des
K a s p i s c h e n Meeres k o m m e n d e n Strae. 1 8 3 V o n dort gab es offenbar schon d a -
m a l s eine Verbindung ber dieses Meer, so d a auf diesem W e g eine der Statio-
n e n der Seidenstrae in Mittelasien erreicht werden konnte.) 1 8 b
F r diese H y p o t h e s e der E i n s c h a l t u n g des Ural in den Handel mit C h i n a
s p r i c h t , da aus einer erheblich spteren Zeit z. B. stlich des Ural am Tobol
eine groe Zahl griechischer Goldschmiedearbeiten gefunden wurde. 1 9 E s h a n -
delt sich um Psalien u n d Phalerae f r das Geschirr der P f e r d e u n d Grtelschnal-
len u n d Applikationen, die auf der K l e i d u n g getragen wurden. Sie alle sind von
griechischen Goldschmieden in einem Stil gestaltet, der sich ikonographisch eng
a n die berlieferung der Nomadenvlker anlehnte. Das, was erhalten ist,
15
Herodot IV, 21.
16
Herodot IV, 22. Nach Herodot bog die Strae, nachdem sie sieben Tagereisen
w s t e s Land passiert h a t t e , dann nach Osten ab.
17
J . J . HUNDT : Vorgeschichtl. Seidenfunde, J a h r b . des Rom.-Germanischen
Zentralmuseums, Mainz, J a h r g . 16, 1971, S. 59 71.
17a
Vergl. A n m . 17.
18
N. TOLL' : Z a m e t k i kitajskom selke n a juge Rossii. Seminarium Kondakovia-
n u m 1 (1927) p. 85, 92.
LEA Herodot I , cap. 201. Vgl. Hierzu H e r o d o t , Historien, ed. H. W. HAUSSIG,
d e u t s c h e bersetzung von A. HOENEIFEB, S t u t t g a r t , Kroner-Verlag, 4. Aufl. 1971,
A n m . 160 des K o m m e n t a r s zu Buch I.
18b
19
Vgl. Anm. 18a.
K. JETTMAR : Die frhen Steppenvlker. Baden-Baden 1964. S. 58.

D I E R O U T E N D E R S E I D E N S T R A S S E NACH Z E N T R A L - U N D OSTASIEN 15
s t a m m t zum grten Teil erst aus dem 5 . - 4 . J a h r h u n d e r t v. Chr. 20 Die E r h a l -

t u n g der heute vorhandenen Stcke ist nur einem Zufall, dem Interesse des
Zaren Peter des Groen zu verdanken. Durch ihn k a m e n sie in die S c h a t z k a m -
mer des Monarchen und spter in die Ermitage, whrend der grere Teil, d a s
was vorher gefunden worden war, eingeschmolzen wurde. Man darf v e r m u t e n ,
da darunter auch Funde aus lterer Zeit waren. Von dem Handel in mykeni-
scher Zeit unterschied sich jener, den Herodot beschreibt, nur in den W a r e n ,
nicht aber in dem von ihm benutzten Weg.
Herodots Bericht ber die Stationen der Nordroute der Seidenstrae
Da es auch in spterer Zeit hnliche Handelsbeziehungen wie in mykeni-

scher Zeit mit dem Ural gegeben h a t , besttigt indirekt Herodot, wenn er v o n
der H a u p t s t a d t der Budiner, die griechische K a u f l e u t e nach einer Reise von
15 Tagen erreichten, berichtet. 2 1 Sie lag hiernach schon in der Waldzone u n d
war daher ganz aus Holz gebaut. D o r t gab es eine eigene Niederlassung der
griechischen K a u f l e u t e mit Tempeln ihrer Gtter. 2 2 Offensichtlich waren b e r
diese H a u p t s t a d t der Budiner die griechischen Goldschmiedearbeiten in d a s
Gebiet von Tobolsk gebracht worden, in dem sie im 17. J a h r h u n d e r t g e f u n d e n
wurden. Herodot beschreibt den weiteren Verlauf dieses Handelsweges, der
durch das Gebiet der Thyssageten fhrend den W a r erreichte, der mit der Wolga
zu identifizieren ist. Von dort f h r t e er dann nach Mittelasien.
Herodot h a t die einzelnen Stationen dieser R o u t e mit den N a m e n jener
Vlker bezeichnet, in deren Gebiet sie lagen. Diese Stationen entsprachen R a s t -
pltzen, wo m a n nicht nur die Dolmetscher, sondern bei sechs von ihnen offen-
20
Z u r D a t i e r u n g K . JETTMAR : . . O. S. 198.
21
Herodot IV, 22.
12
Herodot IV, 108. Nach H e r o d o t bestand die 22 Stadien lange Mauer, die die
S t a d t der Budiner umgab ebenso a u s Holz wie alle Gebude der Stadt, darunter a u c h
die griechischen Tempel. Sie war von Dareios bei seinem Feldzug gegen die Skythen im
J a h r e 510 v. Chr. zerstrt u n d niedergebrannt worden (Herodot IV, 123). Dareios zog
d a n n nach Nordosten zu den Thyssageten bis zum W a r (Oaros)-Flu, wo er Befestigun-
gen errichtete. Die Bezeichnung Hwar ( = Wasser) ist als N a m e des Dnjepr u n d
des A m u d a r y a b e k a n n t . Sie lt sich ebensowenig identifizieren, wie die drei a n d e r e n
Flsse. Nach der E r w h n u n g der Thyssageten zu urteilen, k n n t e der Uralflu gemeint
sein. Dareios ist denmach bei der Verfolgung der Skythen einen Teil der Nordroute der
Seidonstrae entlang gezogen. Man gewinnt aus den Angaben Herodots den E i n d r u c k ,
d a der Knig nicht ber den Kaukasus, sondern ber die Sogdiana nach Persien zurck-
kehren wollte. D a sein Vater Hystaspes ber Ostiran herrschte, waren ihm wahrschein-
lich die sich dort a m Syr darya vereinigenden Nord- und Sdrouten dor Seidenstrae
bekamit u n d er h a t t e sie in seinen strategischen Plan einbezogen. Die acht von dem K n i g
errichteten Befestigungen sollten offenbar oine durch unwegsames Gebiet f h r e n d e
Strae blockieren, u m den verfolgten Skythen die R c k k e h r urunglich zu machen. Bei
diesem Feldzug k o n n t e sich der Knig bis zur D o n m n d u n g auf die Niederlassungen der
ihm unterworfenen griechischen S t d t e sttzen. Sie k a n n t e n auch den Wog nach Mittel-
asien, dem Gebiet der P a r t h e r . Der Grund fr den Abbruch des Feldzuges war, d a
offenbar einige dieser Stdte schon damals vom Knig abfallen wollten.

16 H. W. H A U S S I G
b a r a u c h die K a r a w a n e n f h r e r wechseln mute. 2 3 Das ergab sich aus den groen

E n t f e r n u n g e n . Den gemieteten K a r a w a n e n f h r e r n war der Weg hier nur zum
Teil v e r t r a u t . Die Dolmetscher m u t e n sogar siebenmal gewechselt werden.
B r a u c h t e m a n doch f r die Issedonen, denen m a n auf einem Handelsplatz
bei d e n Argyppaioi begegnete, ebenfalls einen Dolmetscher. 2 4 Die Verwendung
der Dolmetscher beweist, d a das Griechisch der K a u f l e u t e bis hin zu den
Issedonen wenigstens v o n einer kleinen G r u p p e Sprachkundiger verstanden
w u r d e . M a n k a n n daher d a v o n ausgehen, d a d a s Griechische f r den von Hero-
d o t beschriebenen Teil d e r Strae allgemein die Verkehrssprache war. D a n e b e n
scheint die griechische S c h r i f t in einzelnen F l l e n auch zur Schreibung der Lan-
dessprachen verwendet worden zu sein. So sind in E u r o p a kurze T e x t e in t h r a -
kischer Sprache e r h a l t e n , die mit griechischen Buchstaben geschrieben wur-
den. 24 " I n Mittelasien b e n u t z t e das s o g e n a n n t e Baktrische, die Sprache der
K u s c h a n a , ebenfalls die griechische S c h r i f t . 24b Daneben scheint m a n sich im
H a n d e l s v e r k e h r a u e r d e m noch einer bisher nicht gelesenen Schrift bedient zu
h a b e n , die mit den t r k i s c h e n R u n e n v o m Orchon hnlichkeit aufweist. Anga-
ben in dieser Schrift f i n d e n sich auf einem groen Silberbarren aus Ai K h a -
num. 24C F a s t alle bisher b e k a n n t e n Beispiele f r diese Schrift s t a m m e n aus Mit-
telasien.
I n Mittelasien erreichte die von H e r o d o t beschriebene Strae das L a n d
der P a r t h e r . 2 5 Zwar werden sie von i h m nicht namentlich g e n a n n t , aber die
A n g a b e n ber ihre F l u c h t von den kniglichen Skythen nach Mittelasien, die
f r die P a r t her aus P o m p e i u s Trogus b e k a n n t ist, lt an der I d e n t i t t des hier
e r w h n t e n Volkes m i t ihnen keinen Zweifel. Von den P a r t h e r n erreichte die
S t r a e die Argyppaioi. Sie bestritten n a c h H e r o d o t ihren L e b e n s u n t e r h a l t aus
den E r t r g e n der J a g d u n d dem S a m m e l n von Frchten. 2 6 I m R a h m e n einer
Schilderung ihrer Lebensweise beschreibt er das bei ihnen gebte Auspressen
23
Herodot IV, 24.
21
Herodot IV, 23.
24a
Goldschtze dor Thraker, Thrakische K u l t u r und K u n s t auf bulgarischem
Boden, K a t a l o g d. Ausstellung im Osterreichen Museum fi - angewandte K u n s t 4.3.
31. 5. 1975, Abb. 167. R i n g m i t Inschrift. H e u t e im Archologischen Museum Sofia.
246
Vgl. Foto bei J . AUBOYER : L ' A f g h a n i s t a n et son Art, 1968, Abb. 40.
240
Vgl- hierzu A. A . MOTAMEDI : Discovery of an Inscription in an Unknown
L a n g u a g e a t Ai-Khanum. Afghanistan. Vol. 33, Nr. 1, S. 45 48. Aufzeichnungen in
der in Ai K h a n um gebrauchten Runenschrift finden sich in verwandten Formen in den
beiden ersten J a h r h u n d e r t e n v. Chr. berall in Mittelasien. Die F u n d o r t e decken sich
m i t Stationen, die den mutmalichen Verlauf der Seidenstrae in diesem Gebiet bezeich-
nen. N a c h den Gegenstnden, auf denen sich die Aufzeichnungen in dieser Schrift befin-
den, gehen sie auf Hndler zurck. Neben dem Silberbarren von Ai K h a n u m ist hier die
Silberschale aus dem K u r g a n von Issyk in der Kazakischen SSR zu nennen. ber-
wiegend finden sich die Aufzeichnungen in diesen R u n e n auf Ostraka u n d auf Keramik,
die z u m Verkauf bestimmt war.
25
lustin. Trog. 41, 1,1. Parthi . . . Scytharum exules fuere. Schol. Berolinense in
Lucani Bell. Civ. I, 553 : Scythico sermone Parthi exules dicuntur. Weitere Erwhnungen
bei J o r d a n e s Get. u. Isidor Origin. I X , 3.
26
Herodot IV, 23.

der Frchte eines Pontikon genannten Baumes zu einem Brei, den sie Aschy
nannten. (Aschy h a t in den Turksprachen die B e d e u t u n g bitter. Man k n n t e
daher an eine Beziehung der Argyppaioi zu den spteren Turkvlkern denken,
die die Beschreibung ihres ueren und ihre Lebensweise in J u r t e n , die mit Filz
bekleidet waren, nahe legt.) 27
Bis zu den Argyppaioi, deren Wohnsitze m a n sich wahrscheinlich im Tal
des Iii vorzustellen h a t , kamen nach Herodot griechische Kaufleute von der
Mndung des D n j e p r u n d aus Stdten an der K s t e des Schwarzen Meeres. 28
D a s wird durch F u n d e von Mnzen des 4. J a h r h u n d e r t s v. Chr. aus Olbia u n d
Pantikapaion in der Dzungarei besttigt. 2 9
Der Schatten Chinas ber dem Ostteil der Nordroute der Seidenstrae
Die Strae f h r t e dann von den Argyppaioi weiter nach Osten. 3 0 I m

Gegensatz zu den Argyppaioi, die auch von griechischen Kaufleuten besucht
wurden, haben die Issedonen diese offenbar selbst bei den Argyppaioi besucht. 3 1
Man kann daher annehmen, da griechische K a u f l e u t e weder in das zuweilen
als Land der Issedonen bezeichnete eigentliche China noch in seine mit diesem
N a m e n ebenfalls bezeichneten Randgebiete gekommen sind.
Es gab also an diesem stlichen Teil der Seidenstrae schon einen Handels-
platz, den die Issedonen besuchten. Die Angaben Herodots ber die Issedonen
stammen, wie er bemerkt, nur aus skythischer Quelle. Die Skythen h a t t e n also
damals die Dolmetscher f r die Issedonen gestellt. Die Angaben, die sie ber-
mitteln, gehen nach zwei Richtungen. Einmal berichten sie ber die angeblich
im Norden der Issedonen wohnenden Dmonen u n d bernatrlichen Wesen. 3 2
Diese Angaben geben eigene Mythen der Issedonen wieder. Die zweite G r u p p e
27
Herodot bezeichnet die aus Zweigen zusammengesetzten J u r t e n als B u m e , die
m a n im Winter m i t Filz umhllte.
28
Herodot IV, 24.
29
Eine Abbildung der Mnzen findet sich bei E . H . MINNS : Scythians a n d Greeks.
Cambridge 1913.
30
Herodot IV, 24.
31
Herodot IV, 25 u. 20. Herodot betont zwar, da a u c h das stlich von den Argyp-
paioi liegende L a n d der Issedonen bekannt sei, weist d a n n aber in K a p . 26 darauf hin,
d a die Nachrichten ber die Issedonen aus skythischer Quelle stammen.
32
Herodot IV, 27 : Die hier zitierte Mythe der Issedonen, die von Einugigen
u n d den das Gold htenden Greifen spricht, zeigt hier ihro skythische H e r k u n f t d u r c h
das von Herodot zitierte skythische Wort f r die Einugigen: Arimaspen. A u c h die
altchinesische Mythologie kennt die Einugigen. Sie e r w h n t sie in Verbindung mit
der borgabe des Bogens und der Pfeile durch den Himmelsgott Tsn an den J a g d g o t t
Yin (W. MNKE : Die klassische chinesische Mythologio. S t u t t g a r t 1976. S. 279 Shao
Hao). Diese Mythe erinnert mit der E r w h n u n g des J a g d g o t t e s Yi (MNKE . . O. S. 344)
an die Legende vom kniglichen Schtzen Skythes, der den Bogen des Horakles u n d
d a m i t die Herrschaft ber das Reich dor Skythen erhlt (Herodot IV, 10). Wie L u k i a n
(1 10) zeigt, gaben die Griechen die Bedeutung von Skythes mit Toxaris Bogen-
schtze wieder. Skythes entspricht germanisch skutja = Schtze. Sie ist durch die
Bezeichnung ZXTTU bei Priskos, E x c . p. 125, 22 fr einen der engsten Beiater A t t i l a s
belogt.

18 H. W. HAUSSIG
der d u r c h Herodot b e k a n n t gewordenen Nachrichten ber die Issedonen bezieht

sich a u f ihre Bestattungssitten. 3 3 Hiernach gab es bei ihnen keine B e s t a t t u n g
der T o t e n im blichen Sinne. Das wrde erklren, warum es im Westen von
K a n - s u u n d im Tarimbecken seit dem Neolithikum bis zur chinesischen Erobe-
rung in d e m J a h r h u n d e r t vor der Zeitwende keine Grabfunde gegeben hat.
E i n e klare Vorstellung von den Wohnsitzen der Issedonen k a n n man aber
nicht a u s der von Herodot gegebenen Beschreibung dieser nrdlichen R o u t e der
Seidenstrae gewinnen. H i e r f r m u m a n die Erwhnung dieses Volkes durch
P t o l e m a i o s heranziehen. Seine Beschreibung desTarimbeckens, d a s er als Serike
bezeichnet, kennt im Nordwesten die Wohnsitze der Issedonen. 34 Die Issedonen
mssen also zu seiner Zeit dort beheimatet gewesen sein, wo die chinesischen
Quellen die Wu-sun u n d auch die Ye-chi kennen. 3 5 D a Ptolemaios f r seine
D a r s t e l l u n g wie die chinesischen Berichte Quellen aus der ersten H l f t e des
zweiten u n d aus dem ersten J a h r h u n d e r t v . Chr. benutzt, 3 6 mssen zu dieser
Zeit die Issedonen in der Hauptsache im N o r d e n desTarim u n d im Siebenstrom-
land g e w o h n t haben. D a s s t i m m t aber nicht f r die Zeit, von der Herodot be-
richtet. D a m a l s wohnten sowohl die W u - s u n wie auch die Ye-chi der chinesi-
schen Quellen, die nach den geographischen Angaben hei Ptolemaios mit den
Issedonen gleichzusetzen sind, noch im Westen der chinesischen Provinz Kan-su
u n d zu einem Teil auch in der Turfan-Oase. Von dort sind sie erst 176 v. Chr.
nach W e s t e n ausgewandert. 3 7 Demnach reicht die von H e r o d o t gegebene
Beschreibung der N o r d r o u t e der Seidenstrae von der Mndung des Don bis in
den Westteil der Provinz Kan-su.
Der Versuch einer Rekonstruktion der Sdroute der Seidenstrae
E i n e mit Herodots Bericht ber die Nordroute der Seidenstrae vergleich-

bare zusammenhngende Beschreibung gibt es f r die Sdroute nicht. Sie h a t
es, wie einige Fragmente beweisen, sicher einmal gegeben. Diese Teilstcke einer
Beschreibung, die verschiedenen Zeiten angehren, reichen aber nicht aus, um
ihren Verlauf im einzelnen rekonstruieren zu knnen. Zu den ltesten Fragmen-
t e n g e h r t eine nur z u m Teil erhaltene akkadische Keilschrifttafel aus dem 7.
33
Herodot IV, 26.
34
Bei Ptolemaios 431, ed. WILBERG heit e s : vno zavza /ik/oi zcv Kaatcav
'Ioorjove pya &vo mi vazoXixzsQoi avzv Ogavoi. Hiernach mssen die Issedonen
u n m i t t e l b a r westlich des hier mit der sogdischen Bezeichnung e r w h n t e n Tunhuang
g e w o h n t haben. Wo das im einzelnen im Tarimbecken gewesen ist, geht aus der Be-
m e r k u n g des Ptolemaios nicht hervor. Sicher ist, da die Wohnsitze der Issedonen
zu dieser Zeit im Westen von Kan-su gelegen haben mssen. Die Provinz Kansu, wo
die ehemaligen Wohnsitze der Ye-chi lagen, scheidet also fr diese Zeit aus.
36
E . ZRCHER: T h e Yeh-shih a n d K a n i s k a in Chinese Sources. S. 348 353 in
P a p e r s o n the Date of Kaniska, ed. A. L. BASHAM, 1968. Hiernach d r f t e n die chinesi-
schen Quellen etwa aus der gleichen Zeit s t a m m e n wie die des Ptolemaios.
37
E . ZRCHER : . . O., S. 366 bersetzt hier die sich auf die ehemaligen Wohn-
sitze dieser Vlkor beziehenden Angaben der chinesischen Quelle.

J a h r h u n d e r t v. Chr. 38 Was von ihr vorhanden ist, beschreibt einen Abschnitt,

der von einer S t a d t in Mesopotamien bis etwa nach E k b a t a n a reicht. Weitere
Teile, die m a n als ihre bis Mittelasien reichende Fortsetzung ansehen kann,
sind in den Stathmoi des Isidor v o n C h a r a x erhalten, ein Werk, das der parthi-
schen Zeit angehrt. 3 9 Hier sind wie bei Herodot die E n t f e r n u n g e n zwischen
den einzelnen Stationen angegeben. Nach seiner Beschreibung f h r t e die Strae
aus Mesopotamien ber E k b a t a n a bis Kyreschata ( = Kuru-kath = Stadt des
Kyros), 40 eine Festung, die sich schon in der Sogdiana befand. I n achmeni-
discher Zeit war sie der nrdlichste persische S t t z p u n k t im Osten Irans.
I m Norden der Sogdiana gab es damals die sogenannten spitzmtzigen
Saken 4 1 und im Westen von ihnen zwischen Aralsee u n d Kaspischem Meer
wohnten die Massageten, 42 gegen die Kyros der Groe in der Ebene der Daher
im J a h r e 529 v. Chr. gefallen war. 4 3 Von der Sogdiana aus erreichte die R o u t e
nach Plinius den Silis genannten Flu, 4 4 der von Strabo als Iaxartes ( = Perlen-
f l u ) bezeichnet wird. 45 Der N a m e Silis gibt die parthische Aussprache des
Flunamens, 4 6 der in seiner persischen Form Sir auf die Serer zu weisen scheint,
wieder. Die Bedeutung von Iaxartes = Perlenflu war auch den Chinesen
b e k a n n t , die sie f r ihre Bezeichnung des Flusses bernahmen. 4 7 J e d e r der
N a m e n dieses Flusses gehrt zu der Beschreibung einer anderen R o u t e der
Seidenstrae. Als Iaxartes wird er bei Strabo bezeichnet, der Teile der Sdroute
38
Bei E . HERZFELD : Iran in t h e Ancient East. 1941.
39
Die Stathmoi Parthikoi des Isidor von Charax folgen der alten Karawanen-
strae, die vor allem das alte parthische Zentrum in Nisa u n d Hekatompylos berck-
sichtigte.
40
Die Stadt, deren N a m e n BENVENISTE identifiziert h a t , scheint mit dem im
pTudd al 'lam (Transi, and explained by V. MINORSKY, 2 Aufl. ed. C. E . BOSWORTH,
1970, S. 115 u 354) erwhnten K u r k a t h an der Strae, die an der durch Ferghana in das
Tarimbecken fhrte, identisch zu sein. E s gehrte spter zum F r s t e n t u m Usrushana
u n d lag westlich des J a x a r t e s u n d a m Nordhang des Buttamangebirges. Hier befan-
den sich Gold- u n d Silberbergwerke. E . BENVENISTE, La ville Cyreschata. J o u r n a l
Asiatique (1943/1945), p. 163.
41
E s handelt sich bei ihnen u m die von den Persern saka tigraxauda spitzmtzige
Saken genannten Stmme, die sdlich u n d nrdlich des Syr d a r y a wohnten. Herodot
V I I , 64 beschreibt ihre die F o r m einer Tiara besitzende Kopfbedeckung. Die als Hunnen
bezeichneten Vlker in Mittelasien, die sich der gleichen K o p f b e d e c k u n g bedienten,
wurden nach dem Wort f r Tiara, xlagt, als kidaritische Hunnen bezeichnet. Man
b e r t r u g damit auf sie die von den Knigen der Kusehana als Herrscher ber diese Saken
g e f h r t e Bezeichnimg Kidara KuSana Sahi. Vgl. auch Yasht, I X , 30, wo Vischtaspa u m
den Sieg ber die Feinde mit dem spitzen Helm betet.
42
Herodot I, 204. Nach H e r o d o t wohnten die Massageten in jener Ebene, die sich
bis z u m Ostufer des Kaspischen Meeres erstreckte.
43
Die Notiz bei Berossos spricht von Daern.
44
Plinius N a t . Hist. VI, 16 s a g t : . . . includente jlumine laxarte, quod Scythae
Silirn vocant.
45
Strabo Geogr. X I , 8,2.
46
Mani bezeichnet das Gebiet der Serer nicht als CLIEOJC sondern mit der F o r m
CIAEOIC (Kephalaia. E d . B H L I G . S. 1 8 8 , 3 2 ) , der die bei Plinius berlieferten Namons-
form f r den Flu Silis entspricht.
47
Die Chinesen n a n n t e n ihn Tschen-tschu = wahre Perle (Chavannes, Documents
sur les Tou-kioue Occidentaux, P a r i s 2. Aufl. 1900, S. 9). Sie bersetzten d a m i t diese
Bezeichnung in ihre Sprache.

20 H. W. 1IAUSSIG
k e n n t . Silis heit er bei dem lteren Plinius, der ein Stck der Land- und See-
route nach einer indischen Quelle beschreibt. 48 Beide Quellen unterscheiden sich
d u r c h die Darstellung der Serer. Strabo, der um die Zeitwende schrieb, ver-
s t a n d u n t e r den Serern, wie seine Schilderung der Herstellung des nicht mit der
Seide identischen Stoffes, der aus dem Bast der Ramiepflanzen gewonnen wurde,
zeigt, 49 die Chinesen ; der gleiche Autor identifiziert an einer anderen Stelle mit
ihnen auch die Ye-chi in Kan-su. Plinius der ltere dagegen, der etwa ein
halbes J a h r h u n d e r t n a c h Strabo schrieb, sah in ihnen ein Volk mit den gleichen
krperlichen Merkmalen, wie sie nach den chinesischen Quellen die Wu-sun
besaen. 5 0
D e r Unterschied e r k l r t sich aus der Benutzung verschiedener Quellen.
S t r a b o geht ber Apollodoros von A r t a m i t a wahrscheinlich auf die Ephcmeri-
den jenes Eeldzugs z u r c k , den Seleukos I I . im J a h r e 232 v. Chr. unternommen
h a t t e , 5 1 um die mit der E r h e b u n g der P a r t h e r zusammenhngenden partikula-
ristischen Bestrebungen im Osten des Irans zu unterdrcken. Strabo war durch
diese Quelle in der Lage, sehr ausfhrlich ber die damals in Mittelasien woh-
n e n d e n Vlker zu berichten. Seine Informationen bezogen sich auf die zweite
H l f t e des 4. J a h r h u n d e r t s . Plinius der ltere dagegen s t t z t e sich auf eine
wesentlich jngere indische Quelle, die Beschreibung eines Teilabschnittes der
von d e r indischen K s t e kommenden R o u t e der Seidenstrae. Sie s t a m m t von
dem indischen K a u f m a n n Rachias. 5 2 Dementsprechend k a n n t e Strabo die Serer,
soweit er sie nicht b e r h a u p t mit den Chinesen gleichsetzte, noch in ihren Wohn-
sitzen, die sie bis 176 v. Chr. in der spteren chinesischen Provinz Kan-su be-
h a u p t e t e n . Plinius aber w u t e nur von ihren Wohnsitzen in ihrer neuen Heimat
im Nordosten des Tarimbeckens und im Siebenstromland.
Das Chinabild in den Nachrichten von der Sdroute der Seidenstrae
I m Gegensatz zu Herodot erwhnt Strabo 5 3 bei der Beschreibung der

N o r d r o u t e auch die Chinesen, wenn er von dem Verfahren der Serer zur Her-
48
Plinius, N a t . H i s t . , V i , 54 u. V, 88. Rachias war als Gesandter des Knigs von
T a p r o b a n e (Ceylon) u n t e r der Regierung des Kaisers Claudius n a c h R o m gekommen.
Sein Boricht sttzte sich auf die Erzhlung seines Vaters, der diese Strae bereits hatte.
49
Strabo XV, 693 rotavia teal r Erjnix x zivcov (p/.oioiv aivo/ivrj vaaov. E r be-
schreibt liier nicht die H e r s t e l l u n g der Seide aus den Kokons des in Maulbeerbaumplanta-
gen gezchteten Seidenspinners, sondern die eines Gewebes, das aus den Fasern der Ramie-
pflanzo gewebt wurde. Die Ramiepflanze wurde in China sdlich des Hoang-ho angebaut.
50
Plinius, N a t . Hist., V I , 88. Plinius erwhnt ihre rote H a a r f a r b e (rutilis comis)
u n d ihre blauen Augen (caeruleis oculis). Die W u - s u n besaen rote Brte und azur-
b l a u e (ts'ing) Augen (CHARPENTIER: Die ethnographische Stellung der Tocharei.
ZDMG 71 (1917) S. 374).
51
Vgl. HATTSSIG : Byzantinische Quellen bor Mittelasion in ihrer historischen Aus-
sage. S. 41 in Prolegomena to the Sourcos on t h e History of Pro-Islamic Central Asia,
e d . J . HARMATTA, B u d a p e s t 1979.
52
Vgl. Anm. 48.
53
Vgl. Anm. 49.

Stellung der von ihm mit der Seide verwechselten Stoffe aus dem Bast der R a m i e -
Pflanze spricht, denn weder von den Serern in Kan-su vor dem J a h r e 176 v.
Chr. noch von den damals nach dem Nordwesten des Tarimbeckens ausgewan-
derten Teilen dieses Volkes ist b e k a n n t , da sie sich mit der Herstellung dieser
A r t Stoff b e f a t hatten. 5 4 Die chinesischen Quellen, die sich in einer eigenen
Monographie mit den Wu-sun beschftigen, berichten d a v o n nichts. 55 D a die
Serer bei Plinius dem lteren nach ihrer Beschreibung mit den Wu-sun identisch
sein mssen, 56 gilt das auch f r sie. W e n n Strabo die Chinesen als Serer bezeich-
net, 5 7 wiederholt sich hier ein Vorgang, der auch bei der E n t s t e h u n g des
N a m e n s anderer Vlker zu beobachten ist, bei den I t a l e r n in Sditalien, nach
denen die Bewohner der ganzen Halbinsel genannt wurden, oder den Germanen
im Sdwesten Mitteleuropas, deren N a m e als Bezeichnung der brigen Stm-
me gebraucht wurde. Die Ursache f r die Verwendung dieser N a m e n ist ber-
all die gleiche. E s waren jene Vlker, denen die Griechen in Italien, Deutschland
oder China zuerst begegneten.
Die Mittlerrolle der Serer im Handel mit China

Whrend Herodot einen Handelsplatz bei den Argyppaioi als S t t t e des
Warenaustausches zwischen Griechen und Issedonen n u r voraussetzt, wird
Plinius der ltere auch hier deutlicher. Nach ihm erfolgte der Warenaustausch
mit den Serern am Ufer eines Flusses u n d dort in der F o r m eines stummen
Tauschhandels. 58 Dem scheint auf den ersten Blick Herodot zu widersprechen,
der voraussetzt, da die griechischen K a u f l e u t e mit den Issedonen durch sky-
thische Dolmetscher sprechen konnten. 5 9 Der Widerspruch klrt sich dadurch
54
E s gibt eine alte Erzhlung in Chotan, nach der eine Prinzessin aus K u t s c h a
das bostehende Verbot der Ausfuhr von Eiern des Seidenspinners dadurch umgangen
habe, da sie die Eier unter ihrer Kopfbedeckung ber die Grenze nach Chotan
brachte, wo m a n bis zu diesem Zeitpunkt weder Maulboerkulturen noch die durch sie
ernhrten Seidenraupen kannte. (Vgl- CHAVANNES, Doe., S. 126126). Die Erzhlung
keweist da es im Tarimbecken erst relativ s p t zur E i n f h r u n g der Seidenraupenzucht
gekommen ist.
55
Vgl- J - J . M. DE GROOT : Chinesische Urkunden zur Geschichte Asiens. Teil 2.
BerlinLeipzig 1926. S. 125 16. Aus den Berichten geht hervor, da dio Wu-sun
Nomaden waren u n d in J u r t e n wohnten. Man pflegte ihnen als Geschenke chinesische
Seiden zu bersenden.
56
Vgl. Anm. 50.
" V g l . Anm. 49.
58
Plinius N a t . Hist. VI, 88 u. Mela I I I , 59.
59
Herodot IV, 24. u. 27. Herodot erwhnt die Notwendigkeit von sieben Dol-
metschern fr die Reise von der D o n m n d u n g zu den Argyppaiorn. Da man auf dem Weg
zu ihnen n u r sechs Vlker berhrte, war offensichtlich der siebente f r das Gespch mit
den Issedonen bestimmt. Diose Dolmetscherttigkeit bezog sich auch auf die schriftli-
chen chinesischen Angabon, die sich auf den zum Kauf angebotenen Seidenrollen befan-
den. I n seiner Scholie zu Dionysios Perigetes heit es : Ol Erjeg ovxoi orjQtxv mo-
ovvxe (moxQovovxai xifiryia imygayavreg xo aaxxoig, xai /mono i m.yga<pij noiexai
x noxQoei. Die Angaben der Scholie werden durch den B e f u n d an den in Lo-lan ge-
fundenen chinesischen Seidenrollen besttigt. Sie enthielten nicht nur Angabon ber
den Preis, sondern auch ber Lnge, Breite und Gewicht. A. STEIN, Central Asian relies
of Chinas ancient silk trade in T'oung P a o 20 (1921) 131 ff.
Acta Antiqua Academiae Srientiarum Hungaricae 28, 1980

22 H . W. 1IAUSSIG
auf, d a Plinius hier, wie auch an anderer Stelle, eine indische Quelle benutzt. 6 0
Die I n d e r verfgten o f f e n b a r zu dieser Zeit ber keine Dolmetscher, durch die
sie sich mit den Serern verstndigen konnten. D a man den Syr darya als Silis
( = Sercrflu) bezeichnete 6 1 der N a m e des Flusses wie der N a m e Serer wird
bekanntlich in parthischer Aussprache s t a t t des r mit einem l gesprochen
k a n n m a n vermuten, d a der Warenaustausch mit den Serern im Ferghana-Tal
am oberen Syr d a r y a erfolgte.
Der kombinierte See- und Landweg der Seidenstrae
Dieser dritte W e g der Seidenstrae setzt sich, was den Seeweg b e t r i f f t ,

aus einer gyptischen u n d mesopotamisohen R o u t e zusammen, die beide nach
B a r y g a z a , einem H a f e n an der Mndung des N a m a d a i n d e n Indischen Ozean,
f h r t e n . Ihre Beschreibung ist in dem sogenannten Periplus des Roten Meeres
e n t h a l t e n , mit dem der Indische Ozean gemeint ist. Es handelt sich um eine aus
der Zeit um Christi G e b u r t stammende Darstellung, die sich auf den Schiffsver-
k e h r im Indischen Ozean bezieht. Sie e r w h n t den von B a r y g a z a ber Baktrien
nach Thinai fhrenden Handelsweg. Thinai62 bezeichnet hier die chinesische
H a u p t s t a d t , deren N a m e von der Bezeichnung Ts'in des westlichsten chinesi-
schen Staates gebildet ist.
Zusammenhngende Beschreibungen der drei Landwege dieser R o u t e der
Seidenstrae gibt es nicht. F r einen von ihnen, jenen durch Mittelasien, bleiben
als einzige Spuren die indischen geographischen Bezeichnungen von drei Gebir-
gen, die der Weg, der an der Mndung des Kabulflusses in den Indus seinen
Ausgang nahm u n d bis in das Ferghana-Tal fast bis zur Quelle des Syr d a r y a
f h r t e , berhrte. E s sind die Sanskritnamen f r den H i n d u k u s c h : Ottokorra,
f r den Pamir : Imaos ( = hima) und f r das Alai-Gebirge : Kaschia nach einem
in Sanskritquellen vorkommenden N a m e n f r ein Gebirgsvolk, das vielleicht
mit den von Herodot e r w h n t e n Argyppaioi identisch ist. 63 Auf diesem Weg ist
vielleicht die von Nearchos als Dermata Serika ( = Serische Hute) bezeichnete
Seide nach Indien gekommen. 6 4 (Nearchos war der Admiral Alexanders des
Groen.)
Von der zweiten R o u t e , die ber den W a c h d i r - P a in J a r k n d das Tarim-
becken erreichte, ist n u r ein kleiner Abschnitt mit dem Wach-Flu, vom Vikto-
80
Vgl. Anm. 48.
61
Vgl. Anm. 46.
62
Periplus Maris E r y t h r a e i (ed. Mller, I, 303, 23) : nagaxeirai v avri) nXi yea-
yeio /leyiorr) Xeyo/ivrj Qlvai, <p' r EQIOV xai vfjfia xai o&viov HqQixov EI
Bagvyaa i BxTQWv nefj tpQETai.
83
Die auch im Sanskrit belegte Bezeichnung Khasa fl - Kaschgar (Vgl. LEVI :
La cat alogue gographique des Yaksa dans la Mahmyr. J o u r n a l Asiatique 15 (1903)
S. 52 u. 102) scheint auf die gleich lautende Bezeichnung der Bergstmme durch die
I n d e r zurckzugehen (LEVI : Notes chinoises sur l'Inde. B R F E O [1905], S. 2 6 0 - 2 6 1 ) .
84
Strabo XV, 693 arjQix Q/.

ria-See bis Balch, durch eine Beschreibung bei Plinius bekannt ; sie geht auf
die Wegbeschreibung durch den indischen K a u f m a n n Rachias zurck. 8 5
Auf die Benutzung der dritten Route, die vom oberen Indus ber Gilgit
u n d den K a r a k o r u m p a nach Chotan in das Tarimbeckon fhrte, weisen nur die
vom ersten J a h r h u n d e r t n. Chr. an lngs der Strae erhaltenen Graffiti. 8 0 Aller-
dings wurde diese R o u t e wahrscheinlich schon 176 v. Chr. von den Sai-wang, den
Saka der Indischen Quellen, nachdem sie von den Ye-chi aus dem Siebenstrom-
land vertrieben worden waren, f r ihre Auswanderung nach Indien benutzt. 8 7
Man k o m m t also zu dem Ergebnis, da die E n t s t e h u n g der drei R o u t e n ,
der Nordroute, der Sdroute u n d des kombinierten L a n d - und Seewegs der Sei-
denstrae das Ergebnis einer sich ber mehrere J a h r h u n d e r t e erstreckenden
Entwicklung war.
Die Bedeutung der Zusammenfassung lterer Wegstrecken fr die Entstehung der

Seidenstrae
Whrend die Nordroute offensichtlich schon in vorgeschichtlicher Zeit

f r den Warenaustausch zwischen dem Westen u n d China benutzt wurde1,
scheint f r die Sdroute als Ganzes eine Verbindung erst durch die skythische
Westwanderung im 8. u n d 7. J a h r h u n d o r t v. Chr. entstanden zu sein. Bei der
dritten R o u t e der Seidenstrae, dem kombinierten See- und Landweg, h a t die
a m Ende des 6. J a h r h u n d e r t s v. Chr. erfolgende Anglioderung des Pandschab
an das achmenidische Persien durch Dareios den Groen die verschiedenen
See- und Landwege zu einer durchgehenden Handelsstrae zusammengefat.
U n t e r Dareios wurde durch den Griechen Skylax von K a r y a n d a zunchst der
Weg von Kaspapyros in Paktyia, 8 8 einer Stadt, die im Tal des Kabulflusses lag,
ber den K y b e r p a bis zum Indus erkundet.
Hierauf folgte d a n n die Portsetzung der Reise zu Schiff bis zur Mndung
des Indus und d a n n von dort ber den Indischen Ozean zu den H f e n an der
sdarabischen K s t e , von wo m a n gypten erreichte. 69 Die andere Abzweigung,
65
Plinius, N a t . H i s t . V i , 16. Oxus amnis ortus in lacu Oaxo. Plinius gibt d a n n kurz
verschiedene Stationen der Strae nach Balch (Bactrum) wieder.
66
Diese folgt im wesentlichen der heutigen Karakorumhochstrae. Die Graffiti
entlang der Strae wurden von JETTMAR fotografiert u n d von HUMBACH bearbeitet, der
seine vorlufigen Ergebnisse jetzt vorgelegt hat.
e
' Vgl. hierzu die bersetzung dos chinesischen Berichtes, der auf Chang-k'ien
zurckgefhrt wird, in der Gestalt des Ch'ien H a n Shu durch ENOKI (bei A. K . NARAIN :
T h e Indo-Greeks. Oxford 1957. S. 130). Hierzu m u m a n noch die bersetzung, die
G. HALOUN (bei J . JUNGE: Saka-Studion, Klio Beiheft 41. S. 96 nach dem Han-schu
K a p . 96) gegeben h a t . Hieraus ergibt sich, da die von d e n Ta Ye-ehi vertriebenen
Sai-wang ber die Hngebrckenpsse, also die Strae von K h o t a n ber den K a r a k o r u m -
P a und Gilgit, nach Indien gezogen sein mssen.
08
Herodot I I I , 102. Die hier erwhnten Sage von den goldgrabenden Ameisen
keimt auch die tibetische Uberlieferung.
89
Herodot IV, 44.
Acta Antiqua Academiae Scientiurum Hunqarirae 28, 1980

24 H . W . HAUSSIG: D I E R O U T E N D E R S E I D E N S T R A S S E NACH ZENTRAL- U N D OSTASIEN
die v o n der Mndung des N a m ad a aus die mesopotamischen H f e n erreichte,

h a t t e zwei J a h r h u n d e r t e s p t e r Nearchos an der Spitze der makedonischen
F l o t t e erkundet. 7 0 Beide Seeverbindungen w u r d e n in Teilen sicher schon frher
b e n u t z t , aber erst d u r c h Dareios bzw. Alexander zu einer durchgehenden Ver-
kehrs- u n d Handelsverbindung ausgebaut.
A u c h bei der S d s t r a e sind einzelne A b s c h n i t t e schon lange vorher f r
d e n Handelsverkehr b e n u t z t worden. D a s gilt f r die von H e r o d o t erwhnte
Strecke v o n Syrien nach Mesopotamien, auf der z.B. der W e i n e x p o r t aus Syrien
n a c h Babylonien d u r c h g e f h r t wurde, u n d die F o r t s e t z u n g dieser Strae von
B a b y l o n nach E k b a t a n a . 7 1
A u f diesem Wege k a m vielleicht die bei H e r o d o t als medische Kleider
bezeichnete Wildseide, die bei d e n Persern sehr beliebt war, 72 n a c h Mesopota-
m i e n . Dagegen wurden die F d e n , aus denen die Gewebe der Wildseide herge-
stellt w u r d e n , wahrscheinlich ber die b e r h m t e persische K n i g s s t r a e durch
A n a t o l i e n nach Sardes u n d E p h e s u s e x p o r t i e r t . Hier an der W e s t k s t e Klein-
asiens lagen die griechischen M a n u f a k t u r e n , die aus ihnen die b e r h m t e n
koischen Stoffe herstellten. 7 3
Bochum.
70
Vgl. Strabo XV, 693.
71
Horodot I, 194.
72
Herodot I, 135. Hier wird von dem I m p o r t modischer Kleidung durch die Perser
b e r i c h t e t . Dort scheint zuerst die Wildseido hergestellt worden zu sein, die offenbar zu
den sogenannten medischon Kleidern verarbeitet wurde.
73
Die Berichte ber die Herstellung der Seide in Kos gehen bis in die zweite Hlfte
dos d r i t t o n Jahrhunderts zurck. Damals k a n n t e m a n nur dio Wildseide, bei der dio
K o k o n s in den Wldern gosammelt wurdo. Die Angabe bei Plinius (Nat. H i s t . 76) kaim
sich d a h e r nicht auf Kos bezogen haben, sondern wahrscheinlich auf L y d i e n .

31 . A. L I T V I N S K I Y I . . P I C H I K Y A N
MONUMENTS OF ART FROM THE SANCTUARY

OF OXUS
(NORTH BACTRIA)*
T h e l a s t d e c a d e s h a v e b e e n s u c c e s s f u l in t h e f i e l d of s t u d y of t h e c u l t u r e
of C e n t r a l A s i a , e s p e c i a l l y of t h e c u l t u r e a n d a r t of B a c t r i a . T h e a r c h e o l o g i c a l
e x c a v a t i o n s r e m o v e d t h e c u r t a i n over this region. The f u n d a m e n t a l results h a v e
b e e n a c h i e v e d t h r o u g h t h e u n e a r t h i n g a n d s y s t e m a t i c e x c a v a t i o n s of t h e G r a e c o -
B a c t r i a n c i t y A i - K h a n u m in N o r t h A f g h a n i s t a n ( S o u t h B a c t r i a ) . T h e y r e v e a l e d
h e r e p r i n c i p l e s of c i t y p l a n n i n g , s o l u t i o n s of p l a n n e d c o m p o s i t i o n , t e c h n i q u e of
c o n s t r u c t i o n , m o n u m e n t s of w r i t t e n c u l t u r e , t h e G r e e k w a y of life of a n a n t i q u e
c i t y in A n c i e n t B a c t r i a , f a r a w a y f r o m t h e c o a s t of t h e M e d i t e r r a n e a n . 1 C e r t a i n
p a r a l l e l s w e r e f u r n i s h e d b y t h e e x c a v a t i o n s in S a k s a n o k h u r ( N o r t h B a c t r i a ) ,
w h e r e t h e c o m p l e x of a p a l a c e a n d a t e m p l e h a s b e e n u n e a r t h e d . 2 H o w e v e r ,
ABBREVIATIONS
AO . .
.

MAP .

-
.
CA .
. .
. .
. .
Archaeologischer Anzeiger.
AM Mitteilungen des Deutschen Archologischen Instituts.
Athenische Abteilung.
AIBL Acadmie des Inscriptions et Belles-Lettres.
BCH Bulletin do correspondance hellnique.
BSFN Bulletin de la Socit Franaise de Numismatique.
IsMEO Istituto Italiano per il Medio ed Estremo Oriente.
JASB Journal of the Asiatic Society of Bengal.
JDAI Jahrbuch des Deutschen Archologischen Instituts.
JHS Journal of Hellenistic Studies.
MDAFA Mmoires do la Dlgation Archologique Franaise en
Afghanistan.
ML W. H . Roscher: Ausfhrliches Lexicon der griechischen u n d
rmischen Mythologie.
PPS Proceedings of the Prehistoric Society.
RA Revue archologique.
4
P . BERNARD: Fouilles d'Ai K h a n o u m (Campagnes 1966, 1966, 1967, 1968).
MDAFA, X X I . Paris 1973.
2
B . A. LITVINBKIYH. MUHITDINOV : (
) CA 1969. No. 2, 166 foil.
Acta Antiqua Academiae Scientiarum Hun garicae 28, 1980

26 . A. LITVINSKIY I . . P I C H I K Y A N
as a m a t t e r of f a c t , j u s t a few m o n u m e n t s of Graeco-Bactrian represent

t a t i v e a r t (statues, reliefs, etc.) h a v e been discovered. F r o m the earlier,
A c h a e m e n i d , period in t h e Bactrian t e r r i t o r y we have interesting d a t a on cities,
s t r u c t u r e s of architecture, complexes of m a t e r i a l culture, b u t we have hardly
a n y f i n d s of m o n u m e n t s of art. The researchers, therefore, h a v e referred for
a h u n d r e d years t o t h e Graeco-Bactrian coins, collected in Afghanistan, in
N o r t h e r n India a n d in t h e southern p a r t of Central Asia, a n d t o the treasure of
t h e O x u s (in R u s s i a n professional l i t e r a t u r e : Amudar'inskiy klad), contain-
ing a larger q u a n t i t y of objects made of gold f r o m the Achaemenid and Seleucid
periods. The striking artistic quality, t h e expressive and fine portraiture a n d
characteristic r e p r e s e n t a t i o n of the Graeco-Bactrian kings on t h e coinages con-
vincingly showed t h e existence of a high level of m o n u m e n t a l a r t , especially of
s c u l p t u r e , i.e. of t h e original specimens copied b y the minters. This is also shown
b y t h e highly artistic m o n u m e n t s of fine a r t s contained in t h e stock of t h e
t r e a s u r e of the Oxus, f o u n d by local i n h a b i t a n t s in 1877. 3 T h e history of t h e
accidental discovery of t h e treasure of t h e Oxus and the i n t e r p r e t a t i o n of t h e
f i n d s contained in it h a v e been heatedly discussed u p t o t h e present time. All
researchers agree t h a t t h e treasure was discovered within t h e boundaries of
B a c t r i a , a t the Oxus (Amu-Darya). I n f a c t , it has definitively entered history
as t h e treasure of t h e Oxus (or in R u s s i a n as Amudar'inskiy klad).
The history of t h e archaeological s t u d y of the upper course of the Amu-
D a r v a on the right b a n k , t h a t is of t h e n a r r o w coastal strip, situated between
t h e A m u - D a r y a a n d t h e rocky ridge of t h e Teshik-Tash, e x t e n d i n g near to t h e
river, is therefore of interest, because it is almost indisputable t h a t the Amu-
D a r v a treasure was f o u n d in one of t h e t w o fortresses on t h e right bank, pro-
t e c t i n g t h e ancient crossing place, viz. t h e southern T a k h t - i K u b a d , opposite
t o t h e m o u t h of t h e Q u n d u z - D a r y a a n d t h e n o r t h e r n fortress Takht-i Sangin,
t h e Stone Platform (Kamennoye gorodishche according t o M. M. Diakonoff),
a t t h e confluence of t h e Vakhsh and t h e P a n j .4
N . Mayev a n d N . Pokotilo, who s t a y e d in t h a t region at t h e Takht-i
K u b a d for some years a f t e r the discovery of t h e treasure of t h e Oxus, and con-
siderably later also D . N . Logofet, published reports in f a v o u r of the discovery
of t h e treasure a t t h i s place (N. Mayev, N . Pokotilo), 5 and speak of the richness
3
. M. DALTON : T h e Treasure of t h e Oxus. London I I I . ed. 1964.
4
. M. DALTON : op. cit. ; E . V. ZEYMAL : -
. . 1 (28) (1962) 30. See E . V. ZEYMAL: -
. Leningrad 1979 ; M. M. DIAKONOFF : .
15 (1950) 183; idem:
() 19501951. 37 (1953) 254 ; idem: . Stalin-
u b a d 1956. 69.
5
F . ZHUKOV : -.
18. I I I 12 (1880) (N. A . MAYEV under t h e pseudonym F. ZHUKOV published some of
his scientific and belletristic works) ; N. POKOTILO : -
. . 25 . (1889) 480.

MONUMENTS OF ART FROM T H E S ANCTUARY OF OXUS 27
in gold of t h e b a n k of t h e river ( D . N . L o g o f e t ) . 6 The l a t t e r writes quite obscurely

a b o u t a nail-formed (cuneiform ?) inscription on the rocky wall, b u t he publishes
no photograph. Consequently, the a u t h e n t i c i t y of this report, of course, raises
serious doubts. E x c e p t t h e gold diggers, a b o u t whom we k n o w f r o m the r e p o r t s
of N. Mayev and N . Pokotilo, nobody h a d m a d e excavations a t Takht-i K u b a d .
At T a k h t - i Sangin t h e first exploratory trenches were m a d e in 1928 by t h e
members of the expedition of the N a t i o n a l Museum of t h e A r t of the Peoples
of the Orient B. P . Denike. M. M. Diakonoff found on a column the inscription :
To send t o the Moscow museum . . . , written with copying pencil, and he
reported a b o u t this. 7 T h e results of t h e works of this y e a r have not been p u b -
lished. 8 M. M. Diakonoff has left impressive lines on his s t a y at K a m e n n o y e
gorodishche, however, he did not regard t h e treasure of t h e Oxus as a uniform
complex, contesting t h e opinion of O. D a l t o n to be sure without a r g u m e n t a -
tion on t h e I r a n i a n origin of the objects contained in it. 9 Almost t h i r t y years
a f t e r the works of B. P . Denike, in 1956 A. M. Mandelshtam m a d e an explora-
t o r y excavation 11 m . b y 7 m., in a d e p t h u p t o 2 2,5 m . in K a m e n n o y e goro-
dishche, which did n o t furnish substantial results. 1 0 I n a series of his articles,
on the basis of the p o t t e r y discovered, he d a t e d the early layers of K a m e n n o y e
gorodishche t o Graeco-Bactrian times. 1 1 A. M. M a n d e l s h t a m did not connect
t h e explored place with t h e treasure of the Oxus.
Beginning with 1976 t h r o u g h 1982 the South T a j i k i s t a n Archeological
Expedition, led by B . A. Litvinskiy, Corresponding Member of the A c a d e m y of
Sciences of the T a j i k S S R , has been carrying on excavations in K a m e n n o y e
gorodishche. The T a k h t - i K u b a d t e a m of the Archeological Expedition of S o u t h
Tajikistan (headed b y I. R . Pichikyan) did not set itself t h e t a r g e t to i d e n t i f y
t h e site of t h e treasure of t h e Oxus. The selection of t h e place of the works a t
Takht-i Sangin ( K a m e n n o y e gorodishche) was connected with a systematic ex-
ploration of the cultural relations of t h e ancient peoples of Central Asia w i t h
t h e E a r E a s t , esp. with China, India, I r a n , Afghanistan a n d t h e Mediterranean
area. The antique p h a s e of these relations has been studied very little so f a r .
T h e location of the gorodishche on the ancient caravan r o u t e , t h e excellent pre-
8
D. N . LOGOFET : . 3- . .
I I I , . 1909.5769; M . . DIAXONOFF : . 255.
7
M. . DIAKONOFF : . . . 254 foil. I n 1977 we found in
t h e same place a big piece of a column, although without a n inscription.
8
B. P . DENIKE : . -
. I . Moscow 1927. 18; idem:
1927 . I . Moscow 1928. 4.
9
M. M. DIAKONOFF : . . . 255.
10
See . A. LITVINSKIYI. P . PICHIKYAN : - .
1976 1978. . Moscow 1980. 195.
11
A. M. MANDELSHTAM: . 136 (1966) 146. Idem :
( ).
98 (1964) 2 3 . I d e m : .
Leningrad 1975. 1 3 5 - 1 3 6 .
4* Acta Antiqua Academiae Scientiarum Hungaricae 28, 1980

28 . A. L I T V I N S K I Y - I . R. PICHIKYAN
s e r v a t i o n of the strong stone protective walls (not characteristic a t all of t h e u n -

b u r n t brick and mud architecture of t h e region), the p a r t s of antique columns
s c a t t e r e d on the surface of the e a r t h h a v e aroused interest f r o m immemorial
t i m e . However, the d i f f i c u l t approach t o t h e place, the distance f r o m the i n h a b -
i t e d places, the lack of drinking-water, rendered the excavations very complic-
a t e d , even if at the s a m e time the exploration of this gorodishohe was of great in-
t e r e s t b o t h for the historian and for t h e archaeologist, b u t it d e m a n d e d great ma-
t e r i a l a n d h u m a n e x p e n d i t u r e s a t t h e p e r m a n e n t works. 1 2 As f r o m 1976 f r o m
y e a r t o year, the progress of the works h a s been showing more a n d more conv-
incingly the perspectivity of research w o r k a t Takht-i Sangin. The archaeological
a n d stylistic parallels of certain f i n d s of t h e treasure of t h e Oxus show t h e
possibility of a more scientific u n d e r s t a n d i n g of the highly artistic objects con-
t a i n e d in it t h a t , t o g e t h e r with the new discoveries of the works of a r t , is impor-
t a n t f o r the history of culture of C e n t r a l Asia, Afghanistan a n d Iran. 1 3
T a k h t - i Sangin consists of u r b a n s t r u c t u r e s and m a n o r s surrounded from
t h e n o r t h and the s o u t h with protective walls in a distance of 1 kilometre. I n the
c e n t r e of the city t h e r e is t h e citadel (165 m . by 237 m.), s u r r o u n d e d with a m o a t
u p t o 3 metres deep, a n d with stone walls rising u p to a height of 6 metres even
t o d a y , with a thickness of 2 metres, a n d a t the corners f l a n k e d by bastions.
T o t h e west the gorodishche stands close t o the ridge of t h e Teshik-Tash, where
t h e necropolis is s i t u a t e d . Thus f r o m t h e west the f o r t r e s s is protected by a
m o u n t a i n o u s ridge, f r o m t h e east b y t h e A m u - D a r y a , and f r o m the north and
t h e s o u t h by a double line of protective walls. Thus, as a result of the strategi-
cally considered selection of the place, it was inaccessible f r o m the d r y land.
T h e n a t u r a l factors s t r e n g t h e n e d the efficiency of the structures of fortification,
s t r o n g even without t h e m . The p l a n n i n g of the citadel, in its eastern half, is
characterized by the existence of a big s q u a r e paved with big bricks (50 cm. b y
50 cm.), and in its western half there are structures of a palace a n d a s a n c t u a r y
rising 3 metres above t h e court. T h e ruins of these s t r u c t u r e s form three
m o u n d s : the central, t h e northern, t h e north-western a n d southern.
I t was clear f r o m t h e very b e g i n n i n g of the research work t h a t in the
c e n t r a l m o u n d (digging of 25 metres b y 30 metres) we are u n e a r t h i n g a sanctu-
a r y structure. 1 4 Now, on t h e basis of an inscription consecrated to t h e god
12
M. M. DIAKONOFF : . . . 265.
13
. M . DALTON : op. cit.; E . V . ZYMAL : ; T . I . ZEYMAL
E . V. ZEYMAL: . . . 40 46; . G. GAFUROV : . Moscow 1972. 78;
E . E . KUZMINA: . Moscow 1977. 47, 124; eadem : -
-
. . Moscow 1977. 16; . YA. STAVISKIY :
. . Moscow 1977. 74, 93 ; idem: -
. . Moscow 1977. 43.
14
I . R . I'icHiKYAN : (,
).
( ). Tashkent 1979. 8890.
Acta Antiqua Academiae ScierUiarum Hungarica e 28, 1980

Oxus this a t t r i b u t i o n has been fully confirmed. The nucleus of t h e palace-

s a n c t u a r y complex consists of t h e central q u a d r a t i c White Hall (144 square
metres) a n d t h e two lines of c i r c u m a m b u l a t o r y galleries, with a b r e a d t h of
a b o u t 3 metres, a n d the length of the southern section being 12 metres, t h a t of
t h e western section 21 (26) metres, and t h a t of t h e northern section, being t h e
reflection of t h e southern section and explored so far only on the surface, 12
metres. The central four-columned White Hall a n d t h e corridors are f o r m e d by
mudbrick walls on the natural subsoil, with a preserved height of u p t o 5.5
metres and a thickness of 3 metres, rising f r o m t h e subsoil u p t o t h e pre-
sent-day surface. The size of t h e bricks ( 5 0 x 5 0 x 1 5 cm) is unusual for t h e
s t r u c t u r e s of the K u s n a m o n u m e n t s of N o r t h Bactria. 1 5 E a c h wall was plaster-
ed three times and whitewashed several times.
The described planning of t h e temple of T a k h t - i Sangin (Fig. 1) can be
t r a c e d b a c k t o t h e I r a n i a n fire temples. A difference can be observed in t h e ar-
r a n g e m e n t a n d n u m b e r of the doorways, in t h e connection of t h e t h r e e inner
a m b u l a t o r y corridors among themselves, f r a m i n g t h e cell from the f r o n t a n d t h e
sides, and in other not important details of the u n i f o r m construction of t h e east-
ern fire temple. W e see an entirely identical planning in Persepolis, in t h e little
known fire temple, discovered b y E . Herzfeld. 1 6 Obviously the architects plan-
ning the s a n c t u a r y of Takht-i Sangin relied u p o n this, or another similar model.
Other architectural and planning solutions of t e m p l e s with four-columned cells
a n d three a m b u l a t o r y corridors near to those of Persepolis and T a k h t - i Sangin
are published b y K . S c h i p p m a n n in his book dedicated to the d e v e l o p m e n t of
t h e form of the I r a n i a n fire temples. 1 7 I t has t o be a d d e d t h a t the s a n c t u a r y in
t h e north-eastern p a r t of Dilberjinalso belongs t o this archaeological s t a n d a r d . 1 8
The roofing of the central White Hall ( n a m e d by us so because of t h e
colour of the white plaster and the white alabaster floor) rested on four columns.
Of the columns t h e torus-shaped bases of the e a s t e r n order have been preserved,
with removable tori on the two-step pedestal. The dimensions of t h e bases (1
m e t r e on t h e side of the pedestals, and 0,65 metres at the lower p a r t of t h e
columns) allow t o reconstruct t h e volumetrical solution of the central W h i t e
15
V. A.NILSEN : VVIII . Tashkent 1966. 204208 ;
G. A. PUGACHENKOVA : -
( ) CA 4 (1963) 74; eadem: . Tashkent 1966.
126127; eadem: . . T a s h k e n t
1978. 232; eadem:
. . I. I . Moscow 1976. 126; eadem: -. ( 1974 .).
, . 2. Moscow 1979. 7072.
10
A. GODARD : L'Art do I r a n . Paris 1962. 161, Fig. 166.
" K . SCHIPPMANN: Die iranischen Feuerheiligtmor. Boriin -New York 1971.
496, Fig. 83, 1 7 8 - 1 8 0 , Fig. 24.
18
1. T. KRUGLIKOVA : . 19701972 . P a r t I. Moscow. 16.
Unfortunately I . T. KRUGLIKOVA does n o t mention a n y t h i n g about the formation of t h e
hall With inner column (room 1), presuming t h a t t h e columns must stand outside a n d
not within t h e cell.

30 . A. L I T V I N S K I YI . . PICHIKYAN
H a l l q u i t e f i r m l y w i t h t h e original height of t h e c o l u m n s of n e a r l y 7 m e t r e s .
T h e r e m o v a b i l i t y of t h e t o r i f i n i s h e d on a r o t a t i n g millwork, s e p a r a t e l y f r o m
t h e p e d e s t a l s , t h e c h a r a c t e r of t h e w o r k i n g of t h e stone, h a v i n g analogies in
t h e A c h a e m e n i d a r c h i t e c t u r e 1 9 a n d in t h e G r a e c o - B a c t r i a n order of A i - K h a -
n u m , 2 0 s p e a k for t h e Seleucid or early G r a e c o - B a c t r i a n times as t h e d a t e of t h e
original c o n s t r u c t i o n , t o b e d a t e d on t h e b a s i s of t h e archaeological c h a r a c t e r -
istics n o t l a t e r t h a n t o t h e 3 r d c e n t u r y B . C .
T h e analysis of t h e f a s h i o n i n g of t h e m o n u m e n t a l altar, discovered in t h e
n o r t h - w e s t e r n corner of t h e W h i t e H a l l , c o n f i r m s t h i s early d a t e of t h e e s t a b -
l i s h m e n t of t h e s a n c t u a r y . T h e s t y l o b a t e of t h e a l t a r area, s t a n d i n g o n t h e sub-
soil, w i t h t w o bases for circular a l t a r s on t h e u p p e r p l a i n surface w i t h a d i a m e t e r
of 1 m e t r e , is a t t a c h e d d i r e c t t o t h e w e s t e r n w a l l of t h e hall a n d s t a n d i n g a t a
d i s t a n c e of 0.6 metres f r o m t h e n o r t h e r n wall in its full l e n g t h . T h e b r e a d t h a n d
l e n g t h of t h e s t y l o b a t e is close t o t h e s q u a r e , viz. 2.65 metres b y 2.5 m e t r e s b y
0.3 m e t r e s . T h e r e c t a n g u l a r s q u a r e s r u n n i n g a l o n g t h e p e r i m e t e r of t h e s t y l o b a t e
h a v e a n e d g e w i t h o u t o b l i q u e grooves t h a t p o i n t s t o a n early Hellenistic d a t e
of t h e e s t a b l i s h m e n t of t h e a l t a r and t o t h e Greek t r a d i t i o n of s t o n e - c u t t i n g
c r a f t . A s r e g a r d s its axes, t h i s s t r u c t u r e is p e r f e c t l y parallel w i t h t h e sides of t h e
s q u a r e f o r m e d b y t h e f o u r columns, a n d a b s o l u t e l y identical w i t h it r e g a r d i n g
t h e c h a r a c t e r of t h e f a s h i o n i n g of t h e s t o n e s . T h e u n i t y of t h e c o m p o s i t i o n a l
s o l u t i o n of t h e interior of t h e W h i t e H a l l a n d t h e q u a l i t y of e x e c u t i o n of t h e
m o n u m e n t a l a l t a r a n d of t h e columns p o i n t t o t h e e s t a b l i s h m e n t of t h e s a n c t u -
a r y of O x u s as a whole in G r a e c o - B a c t r i a n t i m e s . A close a n a l o g y t o t h e form
of t h e b a s e of t h e stone a l t a r is f u r n i s h e d b y t h e a l t a r of Dilberjin, w i t h dimen-
sions 2.47 m e t r e s b y 2.48 m e t r e s , a n d h e i g h t 0.6 m e t r e s . I t is t r u e t h a t t h r e e
r o w s of b l o c k s were p r e s e r v e d t h e r e (the u p p e r one is definitely t h e l a t e r t h e
A u t h o r s ) . 2 1 A n analogous a l t a r is k n o w n in S u r k h K o t a l in t h e i n n e r cell of t h e
t e m p l e of K a n i s k a . 2 2
T h e m o n u m e n t a l l y f o r m e d d o o r w a y s , t h e j a m b s of which are reinforced
b y t r i p l e pylons, set o n r e c t a n g u l a r s t o n e b a s e s give an e f f e c t of s t a t e l i n e s s
19
The bases with removable tori in the White Hall cannot be dated later than
t h e Hellenistic period. In the profile the tori discussed here are entirely analogous to the
three-stepped torus-like bases of the outer portico of the propylaeum of Ai-Khanum.
As a consequence of this the original construction of the central nucleus of the cultic
complex discovered by us is very likely contemporaneous with the propylaeum of Ai-
K h a n u m . E . F . SCHMIDT : Persepolis I. Figs 7579, Pis. 119, 122 ; A. B. TLLIA : Studies
and Restorations at Persepolis and Other Sites of F r s . I I . Rome 1978. Figs 4 5, 1215,
1 8 - 2 5 , 38, 39.
20
P . BERNARD : Chapitaux corinthiens hellnistiques de l'Asie Centrale dcou-
verts Ai Khanoum. Syria 45 (1968) 128, 129, Figs 8a, 10 ; P . BERNARD : Campagne de
fouilles 1969. AIBL, 1970, avriljuil. Paris 1970. 337, Eig. 29; ibidem 333, Fig. 28.
21
1 . T. KRUGLIKOVA : . 1970 1972 . 29, Figs 2029.
22
D. SCHLUMBERGER : Der hellenisierte Orient. Die griechische imd nachgriechi-
sche K u n s t auerhalb des Mittelmeerraumes. K u n s t der Welt. Baden-Baden 1969. 62,
Fig. 27.

MONUMENTS OF ART FROM T H E S ANCTUARY O F OXUS 31
similar to those of A i - K h a n u m a n d earlier m o n u m e n t s of eastern archi-

tecture. 2 3
The m a i n votive offerings, concentrated t o a d e p t h of 3.5 t o 4 m e t r e s
u n d e r the modern surface, were lying on the floors n o t f a r from the subsoil,
a t a distance of 0.05 t o 1 metre t h a t has t o range t h e m w i t h the first period of
construction. However, u p to t h e present time no p u r e Graeco-Bactrian layer
could be separated, since in t h e floors, pits and cultic burials the latest, u n -
d o u b t e d l y K u s n a material seems t o be d e t e r m i n a n t as t o t h e dating. This c a n
be explained b y t h e literal period of t h e life of the s a n c t u a r y , and also b y t h e
preservation of its functional meaning also after the c o n q u e s t of Graeco-Bactria
b y t h e K u s n a s . The early d a t e is confirmed only by t h e basic mass of the f i n d s ,
viz. b y highly artistic, and as a result of this fairly reliably datable, works of
fine arts of the Seleucid and Graeco-Bactrian periods.
Certain f i n d s of t h e Achaemenid period, earlier t h a n t h e time of construc-
t i o n of the s a n c t u a r y , purely Greek items of the classical t i m e , are well explained
b y t h e s t a t e m e n t s of the written sources on ancient weapons and votive m o n u -
m e n t s of different times t o be f o u n d in t h e temples. This picture is given b y t h e
comparisons with t h e excavations of other sanctuaries, 2 4 where the chronological
dispersion of t h e votive stock is determined by the careful relation of the sacri-
ficers t o the sacrificial offering t o t h e deity, characteristic of the religions of all
t i m e s and peoples. The robbing of t h e sanctuaries h a v i n g very precious treas-
ures, still in ancient times, most f r e q u e n t l y by people belonging t o foreign
tribes, or in very exceptional cases b y compatriots, is shown even more con-
vincingly by t h e results of the archaeological excavations, t h a n by the a n t i q u e
authors. I t has t o be s t a t e d t h a t t h e richness of the m a t e r i a l in Takht-i Sangin
is n o t characteristic of t h e u r b a n sanctuaries.
A normal w a y of preservation of offerings out of use both in Greece a n d
in t h e E a s t was their hiding in specially digged pits (bothroi) or in old worn-out
reservoirs, receptacles, cisterns, w a t e r t a n k s . R e g a r d i n g t h e character of t h e
storages of the finds, in Takht-i Sangin several types of cultic interring have been
23
R . NAUMANN : Architektur Kleinasions von ihren A n f n g e n bis zum E n d o dor
hethitischen Zeit, Tbingen 1971. 160171, Figs 194 217, 406 ; P . BERNARD : Fouilles
d ' A i K h a n o u m . Pis. 46, 66, 61 6 4 ; B . A . L I T V I N S K I Y X . M U H I T D I N O V : op. cit. 164,
Figs 3, 5.
24
. CARARANOS : Dodone et ses ruines. Paris 1878. Pis. X I I I X L I V , X L I X L ;
R . M. DAWKINS : The Sanctuary of Artemis Orthia at S p a r t a . London 1929. 239, 244,
Pis. CLXVII, 2, C L X X V , 14 ; 2 3 6 - 2 3 7 , PI. C L X I I ; G. DAUX : Chronique dos fouilles
et dcouvertes archologiques en Grce en 1963. BCH 90 (1966), 2, P . 847, Fig. 6 (a, b) ;
F . SCHMIDT : Persopolis, I I . Chicago 1957. 63, Fig. 8 (PT, 5, N 409 ; P T 4, N 409, N 449),
p . 99, F i g . 20 ( P T 6, N 578 c, P T U , 1114 a ) . O. M. DALTON : op. cit. I n t h e N o r t h e r n
Coast of the Black Soa : M. M. HUDYAK: . Leningrad 1962. 22, 26,
47, 5 4 ; E . I. LEVI: . 50 (1956) 41 ; E . I . LEVI: -
. I n : . Mos-
cow 1977. 96; E . I. LEVI: . . .
MoscowLeningrad 1964. 132, 133.

32 . A. L I T V I N S K I YI . . PICHIKYAN
discovered. These are d u m p s in d i s t a n t corners of corridors (Nos 1 4), castings

i n t o specially c o n s t r u c t e d pits (bothroi) (Nos. 1 5) and d u m p s behind a parti-
t i o n (favissa) (No. 1). T h e r e a f t e r t h e corridors were walled u p blocking the
d o o r w a y s with other bricks smaller in size, t h a n those of which t h e walls were
b u i l t (doorways Nos. 1, 2). I n the second period of construction t h e doorways
w e r e renovated, t h e corridors were s t r e w n t o 2 metres, leveled w i t h a s t a m p e r
a n d p a v e d , the walls were strengthened w i t h arch-butresses. T h e functioning
of t h e sanctuary c o n t i n u e d t h a t is t e s t i f i e d b y t h e bringing of horns of sacrificial
a n i m a l s into the corridors, viz. of stags, bulls, goats, rams, f o u n d on the pave-
m e n t without a c c o m p a n y i n g material. Soon afterwards t h e whole territory of
t h e s a n c t u a r y , t h e hall a n d the three a m b u l a t o r y corridors were filled u p with
s t i f f clay. The cultic votive deposits, a r r a n g e d in the area of t h e sacred sites be-
side, a n d in the s a n c t u a r y itself in specially digged pits (bothroi), 2 5 in favissas, 28
t h e uniqueness of t h e votive objects a n d t h e difference in t i m e of the cast off
m a t e r i a l , are widely k n o w n f r o m t h e excavations of the a n t i q u e monuments.
T h e offering of t h e h o r n s of the sacrificed animals has analogies in the most
a n c i e n t eastern sacred m o n u m e n t s , c o n t e m p o r a r y with t h e sanctuary of
Oxus.27
T h e description of t h e structure a n d dimensions of t h e b o t h r o i and cultic
i n t e r r i n g s is given in t h e most detailed f o r m in special archaeological reports,
p a r t l y already published. 2 8
I n the present work we c a n n o t publish all m o n u m e n t s of a r t discovered
b y us. Therefore, we discuss in detail certain finds of a m b u l a t o r y corridor No. 2,
a m i we only mention t h e other, b r i g h t e s t ones.
All the proposed m o n u m e n t s h a v e been discovered in t h e same d e p t h ,
in t h e last lower m e t r e of a m b u l a t o r y corridor No. 2 of t h e e x c a v a t e d temple
25
As was noted b y E . I. LEVI : The p i t s with cultic materials were concentrated
e i t h e r immediately in t h e neighbourhood of t h e temples, or on the borders of the temenos.
26
The structure of t h e favissa with p i t closed with a Doric capital see : H. A.
THOMPSON: Activities in t h e Athenian Agora. Hesperia 27, 2 (1968) 145; D. BURR:
A Geometric House a n d a Proto-Attic Votive Deposit. Hesperia 2 N o . 4 (1933) 542 ;
K . LEHMAN: S a m o t h r a k a . Hesperia 19 (1950) 12, Pl. 9 ; . WEICKERT : Typen der
archaischen Architektur in Griechenland u n d Kleinasien. Augsburg 1929. 78.
27
To the bothroi, pits for the t h r o w i n g of cultic deposits, there are so broad anal-
ogies in the antique world t h a t it is no use t o mention them. An analogous repository
of t h e horns of sacrificial a n i m a l s has been discovered recently in t h e southeastern corri-
d o r in Dedoplis Mindori (Georgia). The t h r o w i n g down is dated to t h e 1st century . C.
According to the a u t h e n t i c assertion of t h e researcher, this temple s t a n d s side by side
w i t h a multitude of t e m p l e a n d palace buildings of ancient and Hellenistic East. Y u . M.
GAGOSHIDZE : I . . . ( ).
161 (1977) 107.
28
. A. LITVINSKIYI. R . PICHIKYAN : -
( 1976- 1978 .) 195218; . A. LITVINSKIY I. R . PICHIKYAN: -
. 1980. N 7. 124
133 ; . A. LITVINSKIY - I. R . PICHIKYAN : . 1981. N
3. 87110. see also t h e articles of I. R . PICHIKYAN in the volumes entitled -
for t h e years 19761980.

MONUMENTS OF ART FROM T H E SANCTUARY OF OXUS 33
on T a k h t - i Sangin. 29 This, f i r s t of all, relates to cultic deposit No. 4, consisting

of different cultic a n d dedicated material cast on t h e floor in a depth of 0.1
m e t r e above the subsoil into the most d i s t a n t corner of t h e 21 metres long a m -
b u l a t o r y corridor No. 2. 3 0 The casting of the cultic objects took place in t h e 1st
c e n t u r y A. D., t h e y showed a synchronous character, b u t t h e main offering
gifts can be d a t e d to t h e Hellenistic period, and certain specimens to t h e 5 t h
c e n t u r y B. C.
L e t us t u r n t o t h e m o n u m e n t s of polar dating i.e. t o those of the earliest
a n d t h e latest dating, and also to those of fairly strict dating.
The sheath of t h e Achaemenid akinakes is m a d e of solid piece of ivory
(Fig. 2). The upper end with a side p a r t consisting of two semiovals has a circular
hole for the a t t a c h m e n t t o t h e belt. 3 1
On t h e f r o n t surface two scenes are engraved. The main, upper scene is
very expressive, viz. it is filled b y the large figure of a lion standing on its
h i n d paws, holding in its forelegs a small stag. The body of t h e king of the ani-
mals is represented in profile, and its h e a d in front-view with a little turning t o
t h e stag. The head of t h e lion is modelled clearly, expressively. The bulges of t h e
cheek and the forehead are projecting relief-wise, t h e chin is well cut, the eyes
are carefully fashioned, t h e nose is modelled with wrinkles on t h e upper surface
a n d with nostrils, and t h e whiskers are widely projecting on t h e sides. F r o m t h e
closed, snarling m o u t h big slantwise canines are protruding. The lion holds t h e
stag in its paws. The lion is standing on its fully straightened out hind p a w s ,
t h e right one being ahead (on a higher level), and the left one behind (on a lower
level). The legs of t h e animal are accurately represented, on t h e right leg even
the contracted pillows of t h e claws are shown. The tail is lowered between t h e
legs and twisted around t h e right leg. I t ends in a broadening, a tassel. T h e
powerful body of t h e beast of prey is strongly stressed with its paw stretched
ahead, with which it holds t h e stag. The hair on t h e back a n d the mane are
indicated with parallel slanting bulging stripes with outer curly ends. The g a p
between t h e stripes is filled u p with circular bulging hatches running at angles
to t h e basic ones. All this, although in a conditional and generalizing way, gives
the impression of a thick f u r .
29
The upper two construction periods, t h e second and the t h i r d (6 to 2 metres above
the subsoil) furnished altogether about 20 fragments of K u s n a p o t t e r y . I n corridor
No. 2, the second construction period, on a level of 2 metres above tho datum-point of
levelling, on 1 metre above t h e real subsoil, exactly corresponded to t h e material on t h e
synchronous level in corridor No. 1. Here there were found about ten whole horns of t h e
Oasella subgutturosa, r a m s a n d goats. On t h e floow, together with t h e horns, no o t h e r
finds were discovered.
30
Cultic deposit No. 4 in the northern end of the western circular gallery of t h e
temple, of corridor No. 2, consisted mainly of alabaster sculptures of different forms.
31
Tho whole sheath of akinakes made of ivory was cleaned a n d fixed by I . R .
PICHIKYAN and the restaurator N. TURLIGIN, A. Donish Institute of History of the Acad-
emy of Sciences of t h e Tajik SSR, inventory No. 1000/1091. On t h e outer surface there
are minor defects and crackles.
3 Acta Antiqua Academiae Scientiarum Hunqaricae 28, 1980

34 . A. L I T V I N S K I YI . . P I C H I K Y A N
I n the femural p a r t the legs, two ovals with a notch, conditionally stress-
ing the muscles, at t h e same time serve as a filling ornament, characteristic of
archaic art. In the f r o n t p a r t of the thigh are two twisted converging stripes,
w i t h circular thin lines between t h e m , similar to the horns of a goat. On t h e
f r o n t shoulder-blade there are twirls, b u t without notching, whose outlines,
besides these, are represented with a small bulging. The joints of the paw and
t h e bulging claws are fashioned in a detailed form.
A vertical stripe runs along t h e breast and the abdomen (it incurves be-
low) with a horisontal notch. B u t the h e a d of the lion is fashioned most clearly
a n d in a most detailed form. I t is represented in front-view with a slight turning
t o t h e right, to the stag. The figure of t h e stag is represented vertically in profile,
it is most expressive and realistic. The muzzle of the stag is turned towards t h e
lion, its feet are twisted and let down in a pose of obedience.
The lower broadening of the sheath (bouterolle), is filled up with the figu-
rine of a doubling u p beast of prey and with the head of a goat (Fig. 3). The
t r u n k of the attacking animal belonging to the felidae is highly stylized. The
carver paid the greatest attention t o t h e head of the goat, which is not only
dominating in its size, b u t it is represented also with a careful fashioning of
m a n y details. The side surfaces and t h e upper edge are ornamented in two
stripes with the Greek ornament of ovoli, viz. with an outer narrow one with
close-set perpendicular notches, and with an inner broad one, consisting of
Ionic ovoli, t u r n e d inside (bigger t h a n on the lower, narrow one).
The composition of both scenes is excellent. The figures of the animals of
p r e y and their victims are excellently drawn into the picture-plane, utilizing
each millimetre of the space. The sheaths, besides being highly artistic works of
a r t , at the same time reflect the synthesis of the components of Achaemenid
culture merging with the achievements of antique Asia Minor (Ionic ornament
f r a m i n g the representations), of ancient eastern symbology (in spite of the
zoomorphic physiognomy of the lion, it is unusually anthropomorphic, un-
doubtedly this is a mythological allegory, viz. the t r i u m p h a n t winner the
king or god and the obedient subject), and the animal style, to which the
lacerating scene on the bouterolle belongs, interpreted in its own way and com-
pleted with ornamental motifs.
The sheaths of Takht-i Sangin are undoubtedly sheaths of dagger-aki-
nakes. On the reliefs of Persepolis the form of the akinakes and its sheath borne
b y t h e Persians sharply differ from those of the Modes. 32 The akinakes of the
32
E . HERZFELD : Iran in the Ancient E a s t . London New York 1941. 266 ; E . F .
SCHMIDT : Persepolis, I. Structures. Reliefs. Inscriptions. Chicago, I I I 1953. (The Uni-
versity of Chicago Oriental Publications. Vol. X L V I I I ) 165166; E . F . SCHMIDT:
Persepolis, I I . Contents of t h e Treasury a n d Other Discoveries. Chicago. I. 11 1957. (The
University of Chicago Oriental I n s t i t u t e Publications. Vol. L X I X ) 99 ; B. GOLDMAN :
Aohaemenian Chapes. Ars Orientalis I I . Michigan 1957. 43 45.

MONUMENTS OF A R T FROM T H E SANCTUARY OF OXUS 35
Persians had a hilt broadening towards the pommel and to the guard. Its lateral
p a r t s were arch-formed and t h e pommel had the form of egg-shaped bulging.
The dagger is in a sheath, with a triangular pointed shape, and its upper end
on one side a long lateral projection, the upper and lateral line of which are
slightly bulging, and the lower one is festoon-like (two closed arches carved in
the plain surface of the projection). On the side opposite to the projection there
is a narrow projection, separated below from the main part of the sheath with
a thin triangular cutting. Such akinakes was carried on the middle of t h e ab-
domen, p u t under the belt. 3 3
The Median akinakes differs sharply from the more simple Persian one
(at a n y rate as concerns its sheath). The ornate Median akinakes had a s t r a i g h t
hilt, sometimes decorated with two closed longitudinal ribs on the centre, f r o m
which on the sides there were rows of bulging lozanges and indented triangles.
Its pommel is oval whetstone-shaped with an inner oval hollow. The sheath
was plain or richly ornamented. On t h e upper part of the sheath there is a bulg-
ing buttsrfly-shaped projection, a festoon-like cartouche, the upper surface of
which consists of two closed arches with a point in the centre, and the lower sur-
face also consists of two arches, b u t here a hollow corresponds to the point.
This cartouche is somewhat broader t h a n the adjoining part of the sheath and
its base projects below the sheath. From the cartouche a lateral projection
s t a r t s out also of figurai form, and its upper side terminates in a circle with a
hole. The base of the sheath, t h e bouterolle, is of the form of a trifoliate pal-
mette, or a triangle with rounded angles. The bouterolle is also bulging as com-
pared with the middle part of the sheath. 3 4 The sheath is plain, in which case it
has a longitudinal central rib (but even then the bouterolle is frequently decora-
tive), or decorated.
The most ornate sheaths were covered with whole rows of representa-
35
tions. The study of the reliefs shows t h a t not all sheaths were absolutely identi-
cal in form. Thus, the Mede who was the usher, represented on the relief of t h e
procession on the eastern stair of the apadna,36 bore a sheath the upper edge of
which had a nearly straight form with somewhat raised edges.
33
E . F. SCHMIDT : Persepolis. I . PI. 28, 6 1 - 5 2 , p p . 164165 ; G. WALSER : Die
Vlkerschaften auf don Reliefs von Persepolis. Berlin 1966. (Teheraner Forschungen,
B d . 2) Taf. 8, 11, 21, 36 etc.
34
E . F . SCHMIDT: Persepolis. I. Pl. 27, 57, 65, 120, 165, 166, etc. (in one case,
perhaps, frequently also the cross-bar, Pl. 65) ; G. WALSER : Die Vlkerschaften . . .
Taf. 8, 28, 31, 34, 35 ; A. B. TLLIA : Studies and Restoration a t Persepolis a n d O t h e r
Sites of Ears. R o m e 1972. (IsMEO R e p o r t s and Memoirs. Vol. XVI). Pis. C L X X / 1 1 0 ,
CLXXVII/111. See also the representation of the dagger of t h e Median type on tho silver
r h y t o n from Erebuni. B. N. ARAKELYAN : . CA 1
(1971) 146147, Figs 2, 3 ; V. G. LUKONIN : . Moscow 1977.
76.
35
. F . SCHMIDT : P e r s e p o l i s . I . P l . 120.
33
E . F . SCHMIDT : P e r s e p o l i s . I . P l . 37.
3* Acta Antiqua Academiae cientiarum Hunqaricae 28, 1980

36 . A. LITVINSKIY I . . P I C H I K Y A N
T h e sheaths of t h e Median akinakes, besides all similarities with those of

T a k h t - i Sangin, also h a v e considerable differences, viz. more lengthened p r o - ,
p o r t i o n s on account of t h e greater length of t h e middle part, t h e bulging charac-
t e r of t h e upper c a r t o u c h e , which forms a n independent p a r t of t h e sheath,
in t h e m a j o r i t y of t h e cases different outlines of t h e upper border a n d the boute-
rolle.
T h e gold sheath of akinakes with scenes of royal hunting on lions belongs
t o t h e stock of the O x u s treasure. 0 . M. D a l t o n dates it to t h e 6th century . C.
(even earlier datings are proposed). 3 7 I t slightly differs in form f r o m t h e akinakai
of t h e Persepolis t r e a s u r e . T h e upper p a r t of t h e guard has projections over
t h e line of the s h e a t h in b o t h sides. This g u a r d f i t s evenly (without a bulging
elevation) to the middle p a r t of the sheath. The latter approaches this sheath
t o t h o s e of Takht-i Sangin.
T h e reliefs render possible to fully imagine t h a t the Median akinakai were
carried (on the right side). 3 8 The akinakai are h a n g e d and fixed this way on the
representations of t h e peoples of Central Asia, with the Saka-Tigraxauda, 3 9 and
w i t h t h e Sogdians. 4 0 As f a r as it can be j u d g e d on the basis of t h e reliefs, t h e
d a g g e r s of the S a k a - T i g r a x a u d a had a small bulging point on t h e upper edge.
W i t h t h e Sogdians t h e u p p e r edge was s t r a i g h t with projecting borders, and
t h e h a n d l e of the dagger was antenna-like, in t h e form of a ring with sharpened,
c o n t r a c t e d ends. The S c y t h i a n sheaths are k n o w n in several specimens, kept in
t h e m u s e u m s of the U . S . S . R . A. I. Melyukova divides the sheaths of t h e Scyth-
ian daggers into t w o g r o u p s . She includes in t h e first group t h e sheaths with
a b r o a d e n i n g bouterolle in the lower end, a n d in the second g r o u p the
s h e a t h s without such broadening. 4 1 The bouterolle can be rhomboidal (Melgu-
n o v treasure), 4 2 f l a t t e n e d - o v a l (Kelermes), 4 3 circular (Solokha). 44 The sheaths
f r o m t h e Chertomiik t u m u l u s 4 5 and from T o l s t a y a mogila 46 are g r a d u a l ly broad-
e n i n g in their lower p a r t , t h e base is convex. These sheaths h a d gold covering,
u s u a l l y richly o r n a m e n t e d . I n one case (at t h e village D a r y e v k a , district of
3
' . M. DALTON : T h e Treasure of the Oxus with Other Examples of Early Metal
W o r k . I I I ed. London 1964. 911, Pl. I X ; E V. ZEYMAL : . 43 44.
33
E . F . SCHMIDT : P e r s e p o l i s . I . P I . 65 ; G . WALSER : D i e V l k e r s c h a f t e n . . .
Taf. 35.
39
E . F. SCHMIDT: Persepolis. I. PI. 3 7 ; G. WALSER: Die V l k e r s c h a f t e n . . .
Taf. 56, 57, 83.
40
E. F. SCHMIDT : Persepolis. I. PI. 43, below (holds in t h e hands) ; G. WALSER :
Die Vlkerschaften. Pis. 24, 69.
41
A. I. MELYUKOVA : . Moscow 1964. 61 62.
42
M. I. ARTAMONOV : . Leningrad 1966. Pl. I.
13
Ibidem, Pis. 6, 7.
44
Ibidem, Pl. 145 ; . I. MELYUKOVA : . 61, Pl. 17/3.
45
A . I . MELYUKOVA: . . ., 61, P l . 1 7 / 4 ; M. I . ARTAMONOV: -
. . ., Pl. 183.
46
. . Mos-
cow 1975; . . MOZOLEVSKY : . Kiev 1979. Figs 52 5 5 ; E . V. G'HER-
NENKO : . . Kiev 1975. 157, Fig. 5.

MONUMENTS OF A R T FROM T H E S ANCTUARY OF OXUS 37
Kiev) a bone bouterolle was found with engraved representation of the a n i m a l

style. 47 On the sheath there was a figurai projection above, on the lateral side.
The butterfly-shaped upper p a r t of the sheaths sometimes is present (Melgunov
treasure), while on other sheaths it appears rudimentarily (Kul-Oba), or it is
missing (Solokha). The Scythian sheaths, besides their typological nearness t o
the Median sheaths on the reliefs of Persepolis, differ from t h e m by the absence
of standardization and in a series of details.
R e t u r n i n g to the Achaemenid akinakai, it should be noted t h a t their
sheaths could be prepared of metal (gold, bronze, etc.) ; with metal sheathing
(gold, bronze, etc.) covering on wooden base ; of leather, of wood, or of bone,
of ivory. I n Persepolis a f r a g m e n t of the pointed end of the sheath of an aki-
nakes made of bone or ivory, has been found. 48 A whole series of bouterolles of
the Achaemenid circle made of bronze and ivory is known ,49 b u t whole sheaths
m a d e of ivory have not been found so far, 5 0 although the existence thereof is
shown by a Greek inscription (IG, II 2 , 1425, I, 75 78), which mentions in t h e
treasure-house of the Parthenon an iron akinakes with gold handle and sheath
made of ivory, with gilded (or gold) covering. 51 Supposedly it originates f r o m
Persian trophies.
According to a construction inscription of Darius in Susa (DSf), ivory
(Old Persian piru) was furnished t o Achaemenid Iran from Ethiopia, Arachosia
and I n d i a (DSf, 3.40 45).52 Carving in bone in Achaemenid I r a n continues
the Assyrian tradition, especially clearly expressed in the products from Nim-
rud, on the one hand, it does n o t continue, however, the l a t t e r directly b u t
through a series of intermediate branches (Ziwiye, etc.), 53 and on the other
hand, it continues Syro-Phoenician, E g y p t i a n and Greek traditions. 5 4 The large
collection of such objects found in Susa gives an idea of the Achaemenid articles
made of ivory, b u t there is no sheath of akinakes even among them. 5 5
47
A . I . MELYUKOVA : . . . , 61, P l . 19/2.
48
E . F . SCHMIDT : Persepolis, I I . 99, PI. 76/10.
49
B. GOLDMAN : Achaemenian Chapes. Ars Orientalis Vol. I I . Michigan 1957 ;
1967 ; P . BERNARD : A propos des bouterolles de fourreaux Achmnides. R e v u e archo-
logique 2 (1976) ; R . A. STTJCKY : Achmenidische Ortbnder. Archologischer Anzeiger
1 (1976); S. S. SOROKIN: . , X X X I V . Leningrad
1972. 76, 77.
50
D. . THOMPSON : The Persian Spoils in Athens. The Aegean a n d Near E a s t
Studies Presented to H . Goldman. New York 1966. 284.
51
P . BERNARD : A p r o p o s . . . 227, n . 1 ; R . A . STUCKY : A c h m e n i d i s c h e O r t -
bnder. 22, . 62.
52
R . . KENT: Old Persian. G r a m m a r . Texts. Lexicon. 2nd ed. New H a v e n ,
Connecticut 1953. (American Oriental series. Vol. 33) 143144. See also E . HERZFELD :
Die Magna Charta von Susa. AMI (1931) 67 68 ; E . HERZFELD : T h e Persian E m p i r e .
Wiesbaden 1968. 335.
63
B. GOLDMAN : Achaemenian Chapes. 5051.
34
A review see R . D. BARNETT : A catalogue of the N i m r u d Ivories with Other
E x a m p l e s of Near Eastern Ivories. London 1957. 123131; R . D. BARNETT: E a r l y
Greek a n d Oriental Ivories. Journal of Hellenistic Studies 68 (1948) 3 5 (with literature
on t h e question).
55
P . AMIET : Les ivoires Achmnides de Suse. Syria 49 (1972) 157191,319 337.

38 . A. L I T V I N S K I Y - I . R. P I C H I K Y A N
The scene on t h e surface of the akinakes of Takht-i Sangin shows two

figures : t h a t of a lion a n d t h a t of a stag (fallow deer). According to the antique
sources, lions occurred in Asia Minor and in the Near E a s t , notably in Mesopo-
t a m i a (especially in t h e area of the Euphrates). They were known in Armenia,
in P a r t h i a and in Arabia. The antique authors also knew about the spreading
of lions in India. 56 There is a report connected also with Central Asia, where ac-
cording to the narration of Quintus Curtius (VIII, I, 1417), Alexander the
G r e a t , turning down t h e help of his companion Lysimachus, personally killed
a lion of rare size with a single stroke. 57 There is mentioning of lions also in Hyr-
cania. 5 8 I n contemporary literature, however, the opinion has been spread t h a t
in reality in antiquity as well as in the Middle Ages no lions were living either
in B a c t r i a or in H y r c a n i a , tigers were believed to be lions. 59
I n the ancient a r t of Mesopotamia two species of lions were represented,
viz.: Felis leo Persicus, Persian lion, with a thick mane covering the neck and
t h e shoulders, and with thick hairs on the abdomen, and Felis leo Goojratensis,
I n d i a n lion, usually without mane, just like a cat. 6 0 The lion on the sheath of
T a k h t - i Sangin, as well as the majority of lions on the monuments of Achaeme-
id a r t in Iran, u n d o u b t e d l y belong to t h e species Felis leo Persicus.
The characteristic feature of the figure of the stag (fallow deer) on the
s h e a t h of Takht-i Sangin the small circles represent spottedness. In Meso-
p o t a m i a n art the spottedness is characteristic of one of the three breeds of
stags represented in it, viz. Cervus (Dama) Mesopotamiens Brooke.61 I t belongs
t o t h e genus Cervidae, subspecies fallow deers, and it is spread in Iran, Meso-
p o t a m i a , Asia Minor a n d Palestine. There is a dark stripe along the spine of the
animal, on the sides of which there are rows of longish, whitish spots. Another
characteristic of this subspecies is the shovel-like broadening of the upper
branches of the horns. 6 2
The spottedness can be observed also on representatives of the subspecies
real stags {Cervus Linneus). The Cervus elaphus Bactrianus Lydeker Bukha-
r a n stag (also called khangul, govas) also belongs to the subspecies of
these stags. I t is spread in the USSR along the A m u - D a r y a , and in earlier times
also along the lower course of the Syr-Darya, and also in Northern Afghanistan.
O n t h e young animals, and sometimes also on the grown up ones, along the
dark-brown stripe r u n n i n g on the spine, on the sides, f r o m the withers to the
tail, there is a row of grayish, circular spots, and there are similar spots on the
58
List of t h e sources : STEIER : Lwe. R E J X X V H b b d . S t u t t g a i t 1926 ; W. L .
BROWN : The E t r u s c a n Lion. Oxford 1960. 167.
57
lust., XV, 3, 7 9 ; Plin. Sec. N H V I I I , 64 ; Plut., Dem., 27.
58
STEIER: Lwe. R E X X V . Stuttgart 1926. 970.
59
M. E . MASSON : . . .
Vol. 200. Tashkent 1963. 1 5 4 - 1 5 6 .
80
A. SALONEN : J a g d und Jagdtiere im alten Mesopotamien. Helsinki 1976. 120.
81
A . S A L O N E N : op. cit. 128129.
82
. . FLEROV : . Moscow Leningrad 1952. 134.
Acta Antiqua Academiae Scientiarum Hunqaricae 28, 1980

r u m p and on the feet, especially on the haunches.The horns have usually five
branches each. 63
Thus, even if the artist represented a Mesopotamian stag, still he endeav-
oured to render the shovel-like form of the upper branch, w h a t does not exist
here. However, when in art the Cervus (Dama) Mesopotamia's Brooke was re-
presented, usually they endeavoured to render the shovel-like form of t h e
branch. By way of example we can mention the engraved representations of
such stags on the silver rhyton from the Karagodeuashkh tumulus, 6 4 the silver
s t a t u e t t e (handle ?) from Bukhtarma, 6 5 etc.
At the same time, of course, on this basis it is impossible to assert definite-
ly t h a t this is the representation of the Bukharan stag.
V. Malmberg has noted t h a t on the Attic vases only the ordinary stag can
be found, while on the vases of Rhodes exclusively the spotted stag (he calls it
mottled) occurs. Therefore, he presumes t h a t originally t h e mottled stag was
known only by the Greeks of Asia Minor (and those of the islands). 66 Still earlier
it was represented by the art of Mesopotamia. 67
The representation of the lion appeared in the art of E g y p t , of the eastern
Mediterranean, of Asia Minor, 68 and Mesopotamia early a n d very frequently,
viz. in scenes of royal hunting, laceration of herbivorous animal by a lion, etc.
Usually, in this scene the lion, living or killed, is in horizontal (or slanting) posi-
tion. 69 We, of course, do not propose to examine all subjects connected with t h e
lion, but only those, which iconographically (or semantically) have some simi-
larity with the group lion stag on the sheath of Takht-i Sangin. In the given
case the Mesopotamian material has a special meaning. Therefore we discuss it
in greater detail.
83
K . K . FLEROV: . . ., 1 6 6 1 7 2 ; N . A . BOBRINSKIY. .
FLEROV : Cervus. . . -
I . Moscow 1934. 21 23 ; V. G. GEPTNERV. I. TSALKIN :
. (Systematica a n d Zoography). Moscow 1947. 46 49.
M
V . MALMBERG: - , -
. MAP, No. 13, ., 1894. 141, Figs. 16, 17.
85
Ibidem, 166167, Fig. 26. This s t a t u e t t e obviously h a d Central Asian or N e a r
E a s t e r n origin. S. I . and N . M. RUDENKO : . Moscow 1949. 21.
88
V. MALMBERG : . . . 166 - 168.
87
A. S ACONEN : J a g d u n d Jagdtiere. 128129 (list of representations).
88
From Asia Minor a n d t h e Eastern Mediterranean t h e representations of lions
spread towards the West. As from the 7th century . C. t h e p r o d u c t s made of ivory
appeared in eastern a i t , so t h a t they apparently came about u n d e r t h e influence of t h e
art of Palestine a n d Phoenicia (W. L. BROWN : The Etruscan . . . 2). The representations
of lions made of ivory in Greek art fall also in this time, it is possible t h a t they go back
to L a t e Hittite art. (J. DRING : F r h e Lwen. Mitteilungen dos Deutschen Archologi-
schen Instituts, Athenische Abteilung, Vol. 76. Berlin 1961.). On t h e lion in Greek a r t
see G. M. A. RICHTER : Animals in Greek Sculpture. London 1930. 39, Pis. I V I I I .
89
On the possible utilization of tamed lions as hunting animals, on their utiliza-
tion in fight and for guarding see B. BRENTJES : Mensch und K a t z e im Alten Orient.
Wissenschaftliehe Zeitschrift der Martin-Luther-Universitt HalleWittenberg. Ge-
schichte-Sprachwissenschaft, Bd. XI/16. Halle 1962. (with a d e q u a t e interpretation of
t h e monuments of art).

On the t e r r a c o t t a mould from Mari there is a very early (in the opinion
of some researchers one of the earliest) representation of t h e lion with t h e
b o d y in profile and w i t h t h e head in front-view. Here t h e lion lacerates a
stag. 7 0 I n Alaca-Hyiik in t h e h u n t i n g scene the representation of the lion is
t h e same. 7 1
On the Mesopotamian seals of t h e t h i r d early-dynastic period (about
2500 2316) we know a l r e a d y representations of the lion in profile, standing on
its h i n d legs, with its h e a d t u r n e d t o w a r d s t h e spectator. 7 2 L a t e r , in the Sargonid
period (last third of t h e 3rd millennium . C.) such lions were represented in
profile. 7 3 We know lions standing on t h e i r hind legs on t h e neo-Assyrian a n d
Neo-Babylonian seals, represented in profile. 7 4
On the Neo-Babylonian seals a winged hero is holding in his h a n d s ,
s t r e c h e d on the sides a n d half bent, two lions by their tails, hanging with their
h e a d s downwards. 7 5
On t h e Syrian seals f r o m R a s - S h a m r a , 15th14th centuries, the motif of
a pulled down fallow deer a n d the lion (a pair of lions) t r a m p l i n g it, is usual. 7 6
On the ivory f r o m N i m r u d there are several representations, where a lion
lacerating a bull is represented in profile, with its h e a d in front-view t h a t is
t u r n e d towards the spectator. 7 7 This is, however, as a rule, a horizontal figure,
biting t h e bull with a t u r n . A n exception is an ivory t u b e , on which lengthwise
t h e r e are two relief figures of lions in a very complicated perspective, between
t h e m t h e r e is the vertical figure of a bull with its b o d y in profile. E a c h lion
presses its foreleg t o t h e b o d y of the bull (similarly t o t h a t of Takht-i Sangin).
T h e muzzles of t h e beasts of prey represented in front-view, are t u r n e d a t
r i g h t angles t o their bodies (and to t h e b o d y of the bull), t h e y bite t h e bull
opposite t o each other. 7 8 This scene of laceration is d a t e d t o the 8th cen-
t u r y B . C. 79
I n Assyrian a r t we know also other representations of lions standing on
t h e i r h i n d legs, holding some vessel. 80 A m o n g the reliefs f r o m K h o r s a b a d t h e r e
is a colossal figure of a hero (Gilgamesh ?), with his left h a n d pressing to himself
70
Syria 18 (1937) Pl. X I I .
71
H . BOSSERT : Altanatolien. Berlin 1942. Fig. 520.
72
H . FRANKFORT : Cylinder Seals. A Documentary Essay on t h e Art and Religion
of t h e Ancient Near E a s t . London 1939. 82, Pis. X I I , X V l / a .
73
H . F R A N K F O R T : op. cit. 8 2 8 3 , PL. X V I / d , f.
74
H . FRANKFORT : op. cit. P l . X X X V l / a , g.
75
H . F R A N K F O R T : op. cit. PL. X X X V I / 1 .
76
D u MESNIL DU BUISSON : Le m y t h e oriental de deux g a n t s du jour et de la
n u i t . Archaeologia I r a n i e a . Miscellanea in honorem R . Ghirshman. Leiden 1970. Pl. I ,
Fig. 6.
77
R . D. BARNETT : A Catalogue of t h e Nimrud Ivories with Other Examples of
N e a r E a s t e r n Ivories. London 1957. 73, Pl. X X X V I (S - 62), Pis. X L I I - X L I I I (S 72),
L X V H (S - 158 c, d).
73
R . D . BARNETT : op. cit. 73, 196, P i s . X L I I - X L I I I (S 72).
79
R . D . B A R N E T T : op. cit. 134.
80
R . D. BARNETT : A Catalogue of the Nimrud Ivories. 95, Fig. 32.

MONUMENTS OF A R T FROM T H E SANCTUARY OF OXUS 41
a small figure of a lion. 81 The head of t h e lion is t u r n e d towards the s p e c t a t o r ,

its b o d y is in profile. Unlike t h e lion on t h e sheath, t h e body is not vertical,
b u t inclined, while t h e hind legs are represented as if in motion, t h e loft leg is
higher and t h e right one is stretched down.
I n E g y p t i a n art the thick hair of t h e lion is represented quite realistical-
ly. 82 I n principle we see the same also on p a r t of the Assyrian representations,
although there also a tip appears in t h e back p a r t of this cover of hair. 8 3
On other m o n u m e n t s of Assyrian a r t , especially on t h e later ones, 7 t h 6 t h
centuries B. C., however, t h e conception becomes allegorical, in the f o r m of a
prolongation going out from t h e m a n e a n d curlingly t a p e r i n g off t o w a r d s t h e
end, reminding of a narrow wing or feather, 8 4 perhaps u n d e r the influence of
t h e iconography of t h e griffon.
On t h e reliefs in the throne-hall of the north-western palace in N i m r u d
t h e r e is a representation of a winged person, who in one of his hands holds t h e
small figurine of a stag (in a n o t h e r case, of a goat), a n d in his other h a n d ,
stretched ahead and upward, a twig (in a n o t h e r case a corn-ear). I n the opinion
of R . D . B a r n e t t , these figures h a d a magic destination. H e believes it possible
t h a t this is perhaps the demigod N I N . A M A S . K U . G A , who sacrificed a goat
(stag) t o gain health. The skeleton of a gazelle, u n d o u b t e d l y sacrificed a t such
a ceremony, was found in N i m r u d d u r i n g t h e excavations of 1950.85 On t h e b o d y
of t h e stag there were oval circles. 86 This stag, just like on the sheath of T a k h t - i
81
J . B. PRITCHARD : The Ancient N e a r E a s t in Pictures Relating to the Old Testa-
m e n t . Princeton 1954. Fig. 615; A. PARROT: Assur. Die mesopotamische K u n s t vom
X I I I vorchristlichen J a h r h u n d e r t bis zum Tode Alexanders des Grossen. Mnchen 1972.
Figs 36, 38.
82
H . J . KANTOR : The Shoulder Ornament of Near E a s t e r n Lions. J N E S 6 (1947)
Fig. 5.
83
A. PARROT: Assur. Fig. 38 (Khorsabad 8th century B. C.) ; Fig. 65; R . D.
BARNETT: Assyrian Sculpture in t h e British Museum. Toronto 1975. Figs 103104,
108, 127, 130; A. H . LAYARD : Discoverios in the Ruins of Nineveh and B a b y l o n .
London 1863. 278 (Arban) ; G. PERROTCH. CHIPIEZ: Histoire de l'art d a n s l'anti-
quit. T. I I . Paris 1882. Pl. between pp. 540 a n d 541. A similar interpretation of the sculp-
tures of lions of the 8th century . C. in L a t e H i t t i t e art, the influence of Assyrian tradi-
tion on whose iconography is doubtless, is given b y : E . AKURGAL: Die K u n s t der
H e t t h i t e r . Mnchen 1976. Pis. 135136.
84
A. PARROT : Assur. Fig. 64 (Nimrud, 9t,h century B. C.) ; Fig. 221 (A) (Babylon,
7th 6th centuries B. C.) ; R . D. BARNETT : Assyrische Palastreliefs. Pl. V I (Babylon,
7th 6th centuries . .), Figs 70, 74, 94, e t c . ; R . D. BARNETT: Assyrian Sculpture.
Figs 32, 110, 118, 121, 178.
85
R . D. BARNETT: Assyrische Palastreliefs. Prague 1962. 1213, 28 (note 25)
Pis. 1 2. See also J . B. PRITCHARD : T h e Ancient Near East in Pieturos Relating t o t h e
Old T e s t a m e n t . Princeton 1954. Fig. 614; Cf. R . D. BARNETTM. FOLKNER: T h e
Sculptures of ASSur-Nasir-Apli I I ( 8 8 3 - 8 5 9 . C.), Tiglat-Pilesar I I I ( 7 4 5 - 7 2 7 . C.),
E s a r h a d o n (681 669 . C.) from the Central a n d South-West palaces at N i m r u d . Lon-
don 1962. Pis. L X X V I C X X V I . A list of similar representations see : A. SALONEN :
J a g d u n d Jagdtiere im alten Mesopotamien. Helsinki 1976. 128129.
88
L i Mesopotamia there were three kinds of stags. One of t h e m is the Cervus dama,
also called russety stag, which differs, besides the colour of t h e fur and the character
of t h e horns, also by the presence of spottedness. (A. SALONEN : J a g d und Jagdtiere . . .
128-129).
Acta AiUiqua Academiae Scientiarum Hungaricae 28, 1980

42 . A. L I T V I N S K I Y - I . R. PICH IKY AN
S a n g i n , is represented in a free loose pose. T h e person holds t h e stag, putting his

h a n d between its forelegs, possibly, pressing it t o himself hanging. There is here
a p a r t i a l iconographie a n d semantic parallel (just partial and n o t full) with t h e
scene on the sheath.
I n K h o r s a b a d a relief was discovered, where a king sacrificing a goat is
represented (8th c e n t u r y . C.). The king is holding the small animal in his h a n d
on t h e level of his b r e a s t putting his h a n d between the forelegs of the animal. 8 7
F o r the a p p e a r a n c e of the subject on t h e sheath of T a k h t - i Sangin a n d
for its original semantics, the representation on a Neo-Babylonian seal is also
interesting, which b y E . P o r a d a is included in t h e group of seals of the 9 t h 7 t h
centuries . 88 On t h e m the fight between a hero and a lion for a bull is re-
presented. The hero seized the bull by t h e h i n d leg and lifted it with his straight
s t r e t c h e d left h a n d . W i t h his left foot he t r a m p l e d on the neck and the head of
t h e animal, pressing t h e m to the e a r t h . T h e forelegs of the bull are bent. To t h e
l e f t f r o m this group, t h e enraged lion is s t a n d i n g on its hind legs and moving
t o w a r d s the hero. T h e head and the lower (hinder) p a r t of t h e b o d y of the lion
are represented in profile, and the u p p e r (fore) p a r t of the b o d y in three q u a r -
ters. I t stretched one p a w ahead and touches with it the hind leg (being above)
of t h e bull, quasi pressing it, while it lifted t h e other paw u p . The tail is t u r n e d
u p w a r d . Its head a n d t h e lower (hinder) p a r t of its body are represented realis-
tically, while the u p p e r (fore) p a r t of its b o d y a n d the forelegs are represented
human-like, b u t t h i s impression does n o t change the toes with claws. I n the re-
p r e s e n t a t i o n of t h e lion the a n t h r o p o m o r p h i z a t i o n is clear. This is not simply
a lion, b u t a man-lion.
Among t h e L u r i s t a n bronzes, middle of t h e 8th century middle of t h e
7 t h century . C., t h e r e are representations of lions standing on their hind legs
w i t h their bodies in profile and their heads in three quarters. The hind (lower)
legs are parted, t h e foreleg (the second one is not seen) is lifted up. Along t h e
b a c k of the animal t h e r e is a stripe with transversally grooved hatching. T w o
s u c h lions are sideward from the h u m a n person striking a blow with a dagger.
T h e lions put t h e i r forelegs on t h e shoulders of the man. I n general, the f l a n k -
ing lions for t h e L u r i s t a n bronze are no exceptional subject, t h e characteristic
is in t h e position of t h e head. 8 9 The s u b j e c t of t h e Neo-Babylonian seal described
above, found a reflection on the L u r i s t a n bronze, but the place of the hero hold-
ing a n d t r a m p l i n g o n a bull is occupied here b y the lion. 90
8
' H. FRANKFORT : The Art and Architecture of the Ancient E a s t . London 1970.
174, P I . 199 ; A . PARROT : A s s u r . F i g . 4 3 .
88
E . PORADA : Mesopotamian Art in Cylinder Seal of the Pierpont Morgan Library.
N e w York 1947. 66, Fig. 89.
88
E . PORADA: Alt-Iran. Baden-Baden 1962. 65; P . R . S. MOOREY: Some
E l a b o r a t e l y Decorated Bronze Quiver P l a q u e s Made in Luristan. c. 750650 . C. I r a n
Vol. X I I I . London 1975. 21 22, Fig. 1.
90
P . R . S . MOOREY : op. cit. 2 6 2 7 , F i g . 7.
Acta Antiqua Aeademiae Seientiarum Hunqaricae 28, 1980

Switching over t o Achaemenid a r t , we remark first of all t h a t the connec-

tion of t h e iconography of the lion in this a r t with the Assyrian iconographie
t r a d i t i o n has been established already long ago. 81 Lions were represented on
stone reliefs, on glazed tile panels, on stone and bronze sculptures, on works of
toreutics and lithoglyptics, on textile products, etc. These are single figures of
lions, processions of lions, scenes of f i g h t of a lion with herbivora, victory of a
king over a lion, etc.
Lot us restrict ourselves to t h e subjects connected with the lion on seals.
This is t h e scene of h u n t i n g of t h e king (Darius), travelling on a war chariot on
h u n t i n g lions ; 9 2 the king is m o u n t e d on a stag, he is h u n t i n g on a lion, 93 kills it
with a dagger on foot in single fight. 9 4 I n t h e glyptics of t h e Achaemenid t i m e
we f i n d t h e motif of t h e king seizing t h e lion and pressing it to himself b y t h e
neck of the animal. 9 5 Sometimes t h e king holds in his h a n d s one lion e a c h .
He sized each of t h e m by the hind leg, a n d t h e y are hanging head downwards. 9 6
The l a t t e r motif is repeated on m a n y seals. 97 Sometimes t h e king holds t h e lions
hanging head downwards by the tails. 9 8 On another series of seals the king holds
in his horizontally stretched hands lions by their shoulders, 99 etc. We know t h e
representation on an Achaemenid cylindrical seal, on which a t u m b l e d s t a g is
a t t a c k e d by lions f r o m two sides. 1 0 0 On seal-stamps a goat is a t t a c k e d b y a
lion. 1 0 1 A comparison of the figure of t h e lion on the s h e a t h of Takht-i Sangin
with t h e lions in Assyrian and Achaemenid a r t shows its nearness especially t o
the Achaemenid tradition. Let us pause on this in greater detail.
The appearance of the lion a n d the lion-griffon, s t a n d i n g vertically on
the hind legs, is not u n f r e q u e n t in Achaemenid art, e.g. in Persepolis in t h e
scenes, where the hero is fighting with a lion (griffon). On this both forelegs are
shown. 1 0 2 Reliefs are also known, where t h e hero seizes and presses the s t a n d i n g
91
G. LAMPRE : La representation d u lion Suse. MDP T. V I I I . Paris 1905. 159
176 ; . M . DALTON : T h e t r e a s u r e . 8 9 .
98
H . FRANKFORT : Cylinder seals . . . PI. X X X V l I / d .
93
H . FRANKFORT : C y l i n d e r s e a l s . . . PI. X X X V I I / M .
94
J . BOARDMAN : Pyramidal S t a m p Seals in the Persian E m p i r e . Iran Vol. V I I I .
London 1970. Fig. 2/g ; 11/77, 83, 85, Pis. 4/82, 84, 86; 3 2 - 3 3 .
95
SPA VII. 1967, Pl. 123/L.
99
SPA V I I . 1967, Pl. 124/L.
97
F. SARRE: Die K u n s t des Alten Persien. Berlin 1923. Pl. 52 below; D. J .
WISEMAN : Cylinder Seals . . . PI. 105 ; E . F . SCHMIDT : Persepolis. I. PI. 3/3 ; 4/6 ;
J . BOARDMAN : Pyramidal Stamp Seals . . . PI. 5/108, 110.
93
J . J . BOARDMAN : Pyramidal S t a m p Seals . . . PI. 5/107.
99
G. A. EISEN: Ancient Oriental Cylinder and Other Seals with a Description
of the Collection of Mrs. W. H. Moore. Chicago, I I I 1940. (The University of Chicago
Oriental Institute Publications. Vol. X L V I I ) 54, Pl. XI/101.
100
. PORADA : The Collections of t h e Pierpont Morgan Library. Vol. I. Washing-
ton 1948. (Corpus of Ancient Noar Eastern Seals in North American collections. T h e
Bollingen series. Vol. X I V ) 105/N 835.
101
E . PORADA : T h e Collection . . . 106SN 843 ; J . BOARDMAN : Pyramidal S t a m p
Seals . . . PI. 3/55.
101
E . HERZFEI.D : Iran in the Ancient E a s t . London New York 1941. Pl. L X V ;
E. F. SCHMIDT: Persepolis. I. Pis. 114 117, 1 4 4 - 1 4 6 , 195.

44 . A. L I T V I N S K I Y - I . . P I C H I K Y A N
lion, turning its h e a d back. 1 0 3 The hind (lower) legs are also represented j u s t
like on the sheath, viz.: t h e left leg is d r a w n almost back. On t h e relief from t h e
necropolis of X a n t h u s t h e head of the lion standing on the hind legs in profile is
t u r n e d towards t h e spectator. 1 0 4
A winged lion standing on its hind legs is represented on a bulging pla-
q u e t t e made of ivory (Susa). 105 On the gold cup from the treasure of the Oxus 1 0 6
t h e r e are six pairs of lions standing on their hind legs. The lions are represented
in profile, with their heads in front-view. T h e hind legs are somewhat p a r t e d ,
t h e b o d y is not entirely vertical, one foreleg is drawn down and the other u p ,
t h e tail is curved ringwise. E . Herzfeld presumes t h a t this cup was prepared in
U r a r t u . . M. D a l t o n , polemizing with him, points out t h a t t h e Assyrian t y p e
of lions continued in E a r l y Achaemenid art, eventually through the mediation
of t h e Medes. He d a t e s this cup to the 6th century B. C. 107 As has been shown,
on t h e Achaemenid seals the king or hero frequently fights with one or t w o
lions, or lion-griffons. The figures of the animals are represented usually stand-
ing on their hind legs, while the body and t h e head is in profile. 108
I n the scenes of laceration on the reliefs of Persepolis profile horizontal
figures of lions are embossed, with their heads turned in front-view. 1 0 9 On t h e
lions of these scenes there are common f e a t u r e s with the lion of the sheath of
T a k h t - i Sangin, which with its right foreleg quasi presses t h e stag in the plane.
T h e action of the foreleg is represented absolutely identically also with the lions
lacerating a bull on t h e eastern staircase of t h e apadna of Persepolis, 110 on t h e
m a i n staircase of t h e Council-hall, 111 in t h e palace of Darius, 112 and in t h e
palace of Xerxes. 1 1 3 T h e difference is t h a t on t h e reliefs the leg is not stretched,
b u t half bent and as a whole more realistic. At the same time, not only the pose
coincides, but also t h e minor iconographie details. I n Persepolis and on t h e
s h e a t h of Takht-i Sangin on the legs there are four fan-like pared toes, their
length is unproportionately big, the bulges of the joints are represented identi-
cally, the right projection-toe is conceived in t h e same way in the form of a pin
on a disk. The representation of the back p a r t of the elbow joint (of arched in-
flexions) also coincides. The lions from t h e Council-hall, very likely, are nearest
103
E . F . SCHMIDT : I . P I . 147 ; A . PARROT : A s s u r . F i g . 2 4 4 .
104
M. DIEULAFOY: L ' a r t a n t i q u e de la P e r s e . 3. P a r i s 1885. P l . X V I .
105
P . AMIET : L e s ivoires . . . Syria (1974) 188189, Fig. 21.
' O . M . D A L T O N : op. cit. P l . V I I I / 4 7 .
107
O . M . D A L T O N : op. cit. 8 9 .
108
SPA V I I . 1967. P l . 123/A, C, E , G, L , M ; D . J . WISEMAN : Cylinder Seals of
W e s t e r n Asia. L o n d o n . P i s . 100, 101, 108, 1 1 5 ; E . F . SCHMIDT: Persepolis. I I . P i s .
3/1 2 ; 4/5: 5/10,11; 11/37; 13/60; E. R. GOODENOUGH : J e w i s h symbols in the
G r a e e o - R o m a n p e r i o d . Vol. V I Toronto 1958. (Bollingen Series, X X X V I I ) 41, Fig. 36.
,09
M . DIEULAFOY: L ' a r t antique de la P e r s e . 3. Pis. 19, 20, 62, 66, 122/13, 153,
161, 166, 189.
110
E . F . SCHMIDT : P e r s e p o l i s . I . P i s . 19 2 0 .
111
Ibidem. Pis. 62, 66, 69.
112
Ibidem. P i s . 132/B, 152, 153.
113
Ibidem. Pis. 161, 166, 169.
Acta A ntiguu Acaemiae Scientiarum Hungaricue 28, 1980

MONUMENTS OF ART FROM T H E S ANCTUARY O F OXUS 45
to t h a t of Takht-i Sangin, they are brought near also b y t h e identity of t h e oval

on the f r o n t shoulder-blade, and also t h a t the tail is d r a w n between t h e legs.
The position of the foreleg of the lion is also represented in the same way, as it
cuts into the lying gazelle, on the stone plaquette in t h e collection of the Orien-
tal Institute, Chicago. 114 I t could be presumed t h a t t h e carver preparing t h e
sheath copied the f r o n t part of the body of the lion f r o m the reliefs, b u t t h e
other iconographie details differ considerably.
The two variants of the representation of the hair under the a b d o m e n ,
developed already in Assyrian times, continues to be varied in the Achaemenid
period. The variant in the form of almost straight or slightly bent feather-like
wings occurs very frequently. I n Persepolis, on the reliefs, this wing b e n t
upwards, inside has such intermediate part, scaled filling up, as the mane. 1 1 5
The wing is sometimes drawn with a groove, a lineal contour. 1 1 6 The concep-
tion of the hair under t h e abdomen on the tile panel in Susa, 117 and in the apa-
dana of Artaxerxes Mnemon (404459 B. C.), 118 is close to the Takht-i Sangin
conception. However, all these representations of the hair under the a b d o m e n
in the form of a wing differ from the representation on the sheath in the f o r m
of an oval, especially in the absence of the sharply t u r n e d up end, and in t h e
inner filling in. There are, however, some monuments w i t h a closer conception.
Thus, the sewn-on gold label from the excavations in Persepolis represent-
ing a lion going to t h e right, 1 1 9 is interesting because t h e hair under the a b d o m e n
is represented by a narrow oval, on the inner field of which, with distances f r o m
the margins, there is a steep-transversal hatching. On another analogical gold
ornament (from the collection of the Oriental Institute of the Chicago Univer-
sity), 120 the engraver did not understand this detail a t all and transformed it
into a pointed p a l m e t t e with central stem and hatches perpendicular to t h e m
(from both sides).
On the woolen textile-breast-plate from the 5th t u m u l u s of Pazyryk, on a
light brown base, there are 15 proudly striding lions. T h e hair under the ab-
domen is in the form of an almost straight or slightly bent wing-shaped f e a t h e r
going backward with ends turning upward. I t joins t h e m a n e and has the s a m e
114
H . FRANKFORT : A Persian Goldsmith's Trial Piece. J N E S 9 N2 (1950) 111 112
Pl. I I I .
115
E . HERZFELD: I r a n . . . Pl. L X V , Fig. 362; E . F . SCHMIDT: Persepolis. I
Pis. 20, 28, 68, 69, 146 ; A. B. TILIA : Studies . . . 1972. Pis. G ' V I I I - C I X .
A . GODARD : D i e K u n s t . . . P l . 67.
117
R . D. BARNETT: Assyrische Palastreliefs. Pl. X X ; R . GHIRSHMAN : P e r s e .
Fig. 193. E x a m p l e for a more realistic interpretation of the sculptures of Susa./SPA V I I .
1967. P I . 115 E ; G . LAMPRE : L a r e p r s e n t a t i o n . . . P l . I X .
118
Ibidem.
119
E . F . SCHMIDT : P e r s e p o l i s , I I . p . 77, F i g . 1 4 / B .
120
H . J . KANTOR: Achaemenid Jewelry in the Oriental I n s t i t u t e . J N E S 16 N 1
(1957) 7 8, Pl. V/A. Cf. the interpretation of this detail on t h e r h y t o n from H a m a d a n :
E . PORADA : Alt-Iran. Die K u n s t in orientalischer Zeit. B a d e n - B a d e n 1962. 163 (Tafel) ;
V . G . LUKONIN : . . ., 53.

46 . A. LITVINSKIYI . . PICHIKYAN
colour. S. I. R u d e n k o , n o t without a n y reason, compares t h e m with the Achae-

m e n i d representations, especially with t h o s e f r o m Susa, and d a t e s t h e m t o t h e
5 t h c e n t u r y B. C . m T h e textile lions f r o m P a z y r y k , on t h e basis of the form of
t h e wing-like feather, are very close t o t h a t of Takht-i Sangin.
On the open-work disk from the t r e a s u r e of the Oxus w i t h t h e representa-
t i o n of a sphinx t h e h a i r under the a b d o m e n , in the form of a narrow smooth
wing-like feather, is s h a r p l y bent a t t h e end. 1 2 2 On another disk with the repre-
s e n t a t i o n of a lion-griffon t h e hair u n d e r t h e abdomen is n o t represented, b u t
o n t h e wing there is a stripe with transversally grooved surface. 1 2 3 According t o
. M. Dalton, its d a t e is t h e 5th c e n t u r y B . C. E . V. Z e y m a l writes t h a t it is,
possibly, somewhat l a t e r . The transversal grooving in t h e f o r m of notching on
t h e n a r r o w feathers, wings, etc., is k n o w n f r o m the materials of t h e collection
of o b j e c t s made of ivory f r o m Susa (Achaemenid period). 124 On t h e other hand,
t h e r e t h e hair on t h e b a c k is represented w i t h such notching 1 2 5 (cf. t h e position
of t h e mane along t h e spine on t h e lion of the sheath of T a k h t - i Sangin).
I t s h o u l d be kept in m i n d t h a t the n o t c h e d stripe along t h e spine of the animal,
as h a s been said above, is already on t h e representations of t h e L u r i s t a n bronzes
f r o m t h e middle of t h e 8th century middle of the 7th c e n t u r y . C., and t h e
r e p r e s e n t a t i o n of t h e h a i r under the a b d o m e n is also similar on t h e m o n u m e n t s
of t h e Middle E l a m i t e period f r o m Susa (about t h e 1 2 t h 1 1 t h centuries
B.C.).126
There are m a n y analogies in the representation of t h e m u s c u l a r system of
t h e hind leg in t h e f o r m of longitudinally b e n t ovals. On t h e woolen textile
f r o m t h e 5th t u m u l u s of Pazyryk 1 2 7 on t h e hind legs of t h e lions t h e r e are pecul-
i a r small half horseshoe-like p a t t e r n s , sometimes converging i n t o small circles.
I n certain lions on t h e reliefs a t Persepolis, 1 2 8 on t h e hind leg t h e r e are two
d e e p e n e d grooves : one of these is parallel t o t h e rear edge of t h e upper p a r t of
t h e leg and the inner one is in the beginning parallel t o t h e f i r s t one, t h e n it
t u r n s inside, becoming horizontal. On t h e lions on the gold c u p of t h e treasure
of t h e Oxus, on t h e h i n d leg there are oval figures. 129
121
S. I. RUDENKO : -
. Moscow 1968. 76, 78, Fig. 71. See also S. I . RUDENKO:
(middle of t h e 1st millennium . .). Moscow 1961.
19 2 0 , Fig. 12.
122
. M . D A L T O N : op. cit. P I . X I I , 2 6 , 2 8 .
123
Ibidem. 14.
124
P. AMIET : Les ivoires . . . Syria 49 (1972) 188, Fig. 50. Also on some seals:
J . BOARDMAN : P y r a m i d a l S t a m p Seals . . . PI. 1/8.
125
P. AMIET : Les ivoires . . . Syria 49 (1972) 184, PL. V/2.
126
E. PORADA: A l t - I r a n . . . 62 (Tafel).
427
S. I. RUDENKO : . . . Fig. 71.
128
129
. M. DALTON : T h e T r e a s u r e . . . P l . V I I I / 4 7 .

I n this respect t h e representations of animals on t h e tile panels in Susa are

most interesting, viz. t h e procession of lions, 130 the winged bulls, 1 3 1 t h e grif-
fons, 1 3 2 t h e sphinxes, 1 3 3 and other m o n u m e n t s from Susa. 134 I n t h e lions on t h e
b a c k shoulder-blade there are two ovals, viz. a small one inside and a bigger
one a t t h e edge of the hinder p a r t of t h e body. The same applies t o t h e lion-
griffons. Unlike those of Takht-i Sangin, these ovals are parallel. Obviously,
t h e changed poses of t h e body (instead of the moving animal, t h e one s t a n d i n g
on t h e hind legs), induced the artist t o change the m u t u a l position of t h e ovals.
The enigmatic horned design on t h e hind leg of t h e lion on t h e s h e a t h is
also elucidated b y the representation f r o m Susa, viz. it originates f r o m , or
r a t h e r is, a highly stylized representation of an ovally-pointed cartouche of t h e
lion on t h e Susa representations. The representation of t h e muscular system on
t h e shoulder-blades with brackets a n d with pointed ovals was applied in t h e
6 t h 3 r d centuries also on the figures of other animals, viz. of stags, 1 3 5 bulls, 1 3 6
goats, 1 3 7 etc.
E . P o r a d a , examining the details of a bronze weight in t h e form of a lion
f r o m Susa (presumably f r o m the 5th c e n t u r y B. C.), kept in the, L o u vre, notes
the representation of t h e muscles on t h e fore shoulder in t h e f o r m of a p o i n t e d
oval (with bulging inside), which f o r m s lower a noose. H e r e m a r k s t h a t this is
a specific feature of t h e Achaemenid style. 1 3 8 Actually, t h e schematic represen-
t a t i o n of t h e muscular system of t h e fore shoulder-blade of t h e lion in t h e f o r m
of an open oval with a noose is very widely spread on the m o n u m e n t s of Achae-
menid a r t . Let us mention a few e x a m p l e s . This is the practice on t h e lions of
the stone p l a q u e t t e f r o m t h e collection of t h e Chicago Oriental I n s t i t u t e ; 139
on some lions of t h e Persepolis reliefs, 1 4 0 etc.
On t h e Achaemenid gold o r n a m e n t f r o m theTeheran Museum of Archaeol-
ogy, 1 4 1 there are two lions represented in profile with t h e i r b a c k s t u r n e d t o w a r d s
each other, with t h e head of one t u r n e d towards the s p e c t a t o r . The muscles
of t h e shoulder-blades of one of t h e m are represented in t h e f o r m of t h e figure
130
R . D. BARNETT : Assyrische Palastrelief. Pl. X X ; R . GHIRSHMAN : Perse . . .
Fig. 193.
131
R . D. BARNETT : Ibidem. Pis. X I X I I ; R . GHIRSHMAN : Perse . . . Fig. 192.
132
R . D. BARNETT : Assyrische Palastreliefs. Pl. X I V ; R . GHIRSHMAN : Perse . . .
Fig. 191 ; SPA V I I . Pl. 77/a.
133
R . D. BARNETT : Assyrische Palastreliefs. Pl. X X I V .
134
F . SARRE : Die K u n s t des alten Persien. Berlin 1923. Pl. 37.
135
O . M . DALTON : T h e T r e a s u r e . . . PL. X X / 3 6 .
136
H . FRANKFORT : A Persian Goldsmith's . . . Pl. I I I ; R . D. BARNETT : Assyri-
sche Palastreliefs . . . Pis. X I X I I .
137
M. P . GRYAZNOV: . Leningrad 1968. Pl. 41 ; .
M . DALTON : T h e T r e a s u r e . . . Pl. X/24.
133
E . PORADA : Alt-Iran . . . 161 below.
139
H . FRANKFORT : A Persian Goldsmith's . . . Pl. I I I .
140
141
7000 Years of Iranian Art Circulated by the Smithsonian Institution. 1964
1966. 89, Fig. 456.

48 . A. LITVINSKIYI . . P I C H I K Y A N
e i g h t , and of t h e o t h e r in t h e form of a p o i n t e d oval with a noose below. On t h e

shoulder-blades of t h e figures of lions in t h e Susa procession of lions,142 t h e r e
is a twirl in t h e f o r m of a drop-like oval, one side of which continues in the form
of a hook, which in its contours is absolutely analogous with the figure of t h e
t w i r l on the lion of t h e T a k h t - i Sangin s h e a t h .
On the g a r m e n t of t h e king on t h e relief in t h e m a i n hall of the palace of
D a r i u s in Persepolis, t h e r e were a series of carelessly engraved lions and other
representations (in t h e opinion of E . Schmidt, this is a p r e p a r a t o r y drawing for
t h e colouring). 143 T h e stretched oval with noose a n d circle on the fore shoulder-
b l a d e are absolutely analogous with those of T a k h t - i Sangin.
The half b r a c k e t on the foreleg of t h e lion of T a k h t - i Sangin has an ab-
solutely accurate analogy in such details of t h e griffons of t h e bracelets of t h e
t r e a s u r e of the Oxus, which are d a t e d b y . M. D a l t o n t o t h e 5th century t h e
f i r s t half of t h e 4 t h c e n t u r y B. C. 144 There is such a half bracket also on t h e
foreleg of the griffon on t h e silver r h y t o n f r o m Armenia, which is also d a t e d t o
t h e 5th century B . C. 145 On the forelegs of t h e figures of sphinxes on t h e tile
reliefs in Susa t h e r e are also pointed half brackets. 1 4 6 Finally, the framing of t h e
edge of the side of t h e sheath consisting of Ionic ovoli has an analogy in t h e
c a r v e d bone of t h e Achaemenid period f r o m Susa. 147
To achieve m a x i m a l expressive power, the artist, a u t h o r of the sheath of
T a k h t - i Sangin, u n i t e d t h e plastic volume of t h e figures with a multitude of
e n g r a v e d details, placed on t h e same. This m e t h o d is characteristic also of t h e
specimens m a d e of ivory from Susa. 148
Thus, there were vertical profile representations of lions with heads in
front-view already in Assyrian a r t . T h e representation of t h e abdominal hair
in t h e form of a n a r r o w , bent wing (feather) going b a c k w a r d f r o m the mane was
wide-spread. Together with the representation of t h e lion lacerating a herbivore
animal, there a p p e a r s t h e representation of t h e king, of a hero, or a demigod,
holding a goat or a stag, and this is connected with ritual scenes. The small
figure of the stag r e m i n d s of the representation on the sheath. This scene, where
t h e powerful (and mighty) king, hero, or demigod is holding an obediently
motionless t i n y s t a g or goat, is the s e m a n t i c archetype of t h e Takht-i Sangin
142
R . D. BARNETT : Assyrische Palastreliefs. Pl. X X ; R . GHIRSHMAN : Perse . . .
Fig. 193.
143
E . F . SCHMIDT : Persepolis. I. Pl. 142 ; A. B. TILIA : Studies and Restorations
a t Persepolis and O t h e r Sites of Ers. I I . (IsMEO R e p o r t s a n d Memoirs. X V I I I ) . R o m e
1974. Pl. X X X I I I , F i g . 3 9 - 4 0 .
144
. M . DALTON : T h e T r e a s u r e . . . 32 33, PL. 1 / 1 1 6 . E. V. ZEYMAL : -
. . . 65 also agrees with this dating.
145
. M . DALTON : op. c. 4 2 4 3 , P l . X X I I .
146
R. D. BARNETT: Assyrische Palastreliefs. Pl. X X I V . This detail, very likely,
a p p e a r s already in t h e Assyrian style : A. H . LAYARD : Discoveries in the Ruins of Nine-
veh a n d Babylon, London 1873. 278.
14
' P . AMIET : Les ivoires . . . Fig. 66.
148
Ibidem.

MONUMENTS OF ART F R O M SANCTUARY OF OXUS 49
scene, while t h e d e v e l o p m e n t led t o t h e s u b s t i t u t i o n of t h e h u m a n f i g u r e f o r t h e

r e p r e s e n t a t i o n of a lion, b u t b y t h i s t h e lion shows s h a d e s of a n t h r o p o m o r p h i c
f e a t u r e s . This b e c a m e possible as a r e s u l t of t h e i d e n t i f i c a t i o n of t h e k i n g w i t h
t h e lion, or b y t h e a p p e a r a n c e of t h e k i n g in t h e f i g u r e of a lion.
T h e comparisons m a d e a b o v e p e r m i t t o d e f i n i t e l y include t h e scene o n
t h e s h e a t h of T a k h t - i Sangin into t h e circle of t h e w o r k s of A c h a e m e n i d ( b u t ,
m a y be, n o t imperial, b u t provincial) a r t of t h e 5 t h 4 t h centuries . C., w i t h
t h e p r e d o m i n a n c e of t h e parallels p o i n t i n g t o t h e 5th c e n t u r y , a n d some f e a t u r e s
c o n n e c t t h i s scene w i t h t h e A s s y r i a n t r a d i t i o n .
T h e analysis of t h e bouterolles is facilitated considerably b y t h e d e t a i l e d
special works of t h r e e researchers, viz. B . Goldman, 1 4 9 R . S t u c k y , 1 5 0 P . B e r -
n a r d , 1 5 1 b y t h e p u b l i c a t i o n of t h e Persepolis reliefs in t h e m o n u m e n t a l w o r k of
E . Schmidt, 1 5 2 a n d b y t h e well i l l u s t r a t e d edition of G. Walser. 1 5 3 T h e c o m p a r i -
son of t h e typological t a b l e , necessary a t t h e careful e x a m i n a t i o n of t h e s t y l i s t i c
e v o l u t i o n of t h e r e p r e s e n t a t i o n s o n t h e lower p a r t of t h e s h e a t h s w a s f a c i l i t a t e d
b y t h e collection of t h e illustrative m a t e r i a l b y P . B e r n a r d , k i n d l y f u r n i s h e d t o
him b y B . G o l d m a n a n d R . S t u c k y . T h e r e f o r e , t h e n u m e r a t i o n of t h e b o u t e r o l l e s
will be given according t o t h e new e d i t i o n thereof m a d e b y P . B e r n a r d .
I n t h e articles of Soviet scholars t h e previous period is well r e p r e s e n t e d ,
a n d also t h e S c y t h i a n style of t h e w o o d e d steppe, a n d t h e Scytho-Siberian s t y l e ,
w h i c h is i n t e r e s t i n g f o r t h e e l u c i d a t i o n of t h e genesis of t h i s f o r m a n d t h e r e p r e -
s e n t a t i o n s embellishing it. 154 I n t h e p r e s e n t case we are r e s t r i c t e d t o t h e a l r e a d y
P. Bernard 1979 R. A. Stucky 1976 B. Goldman 1957
No. 1, fig. 4 No. 2, fig. 4 fig- 1, 1

No. 2, fig. 5 No. 3, fig. 6 fig- 1, 4
No. 3, fig. 6 No. 4, fig. 6 fig- 1, 2
No. 4, fig. 7 No. 10, fig. 12
No. 5, fig. 8 No. 7, fig. 9
No. 6, fig. 9
No. 7, fig. 10 No. 6, fig. 8
No. 9, fig. 11 No. 6, fig. 7 fig- 1, 7

No. 10, fig. 12 No. 9, fig. 11 fig- 1, 6
No. 11, - No. 1, fig- 3
No. 12, fig. 13 No. 8, fig. 10
No. 13, fig. 14 and 1 fig. 2 fig- 1, 3

No. 14, fig. 16 fig. 1 fig- 1, 6
No. 16, fig. 16 fig- 1, 9
149
B. GOLDMAN : Achaemonian Chapes. 43 54.
150
R . A . STUCKY: A c h m e n i d i s c h e O r t b n d e r . A A 1976. 1 3 2 3 .
151
P . BERNARD : A propos des bouterolles . . . 227 246.
168
E . F . SCHMIDT : P e r s e p o l i s . I . P i s . 27, 29, 30, 32, 34, 38, 40, 4 2 , 4 4 , 46, 4 8 , 6 1 ,
52, 57, 67, 70, 110, 111, 153, 155.
143
G . W A L S E R : op. cit. F i g . 8 , 1 8 , 3 1 , 3 4 , 3 5 , 5 6 , 5 7 , PL. I .
154
A . I . SHKURKO : . . . 31 3 3 ; N . L . CHLENOVA : -
. Moscow 1967. 118 foil., 127 ; M. I .
A R T A M O N O V : ... 12; S . S . S O R O K I N : ibidem. 7677.
4 Acta Antiqua Aeademiae Scientiarum Hungarieae 28, 1980

developed pure A c h a e m e n i d style of representation reaching, in f a c t , the mul-

tiple development on 16 bouterolles.
Agreeing with t h e opinion of P . B e r n a r d on t h e Achaemenid time of all
bouterolles published b y him, let us follow their evolution in order to separate,
in t h e framework of t h e d a t e s proposed b y him, a t least stylistically, the earliest
ones f r o m the latest ones. This requires a comparison of t h e known Achaemenid
bouterolles with t h e s h e a t h of Takht-i Sangin for the more accurate definition
of t h e dating of t h e l a t t e r .
The first two bouterolles (Nos. 1 a n d 2), 155 f r o m t h e stylistic point of view,
are t h e earliest. According t o P . Bernard t h e y belong t o the Achaemenid period
a n d were prepared b y Achaemenid masters, or b y E g y p t i a n ones in the period
of t h e first Achaemenid conquest of E g y p t . The characteristics of the represen-
t a t i o n of the ungulate on t h e m are as follows : t h e head of t h e goat is engraved
in profile to the left, t h e rounded off horns are clearly seen, t h e ears are repre-
s e n t e d realistically, u n d e r a line there is t h e p e c t i n a t e beard, t h e neck is stylized
i n t o a pectinate, fir-like ornament. As for t h e representation of the predatory
a n i m a l of the species of felidae, on bouterolle No. 1 a p r e d a t o r y animal below is
a t t a c k i n g , which j u d g e d b y the encircling m a n e , is a curling u p lion. The whole
figure of the p r e d a t o r y animal is represented very realistically, t h e hind leg and
t h e foreleg are bent, t h r e e ribs are shown. The long tail is behind the hind leg.
I n t h e opened m o u t h t w o canines are clearly seen, an upper a n d a lower one.
O n bouterolle No. 2, of t h e figure of t h e p r e d a t o r y animal only the long, cat-
like b e n t tfeil has r e m a i n e d . The whole b o d y divided into two segments, is
represented by concentric grooved arches. Beginning with bouterolle No. 2,
n o t only the stylization of the body of t h e p r e d a t o r y animal can be observed,
b u t also the division of t h e bouterolle into three segments facing each other,
filled u p with o r n a m e n t a l motifs.
Bouterolle N o . 3 is very characteristic. 1 5 6 Here t h e ear of the goat, going
i n t o t h e circle f o r m e d b y the horns, is stylized into a rudimentary p a l m e t t e .
O n t h e following bouterolles, according t o t h e stylistic typology of P . Bernard,
i n s t e a d of the horns a n d the ear only a p a l m e t t e remained. However, on
bouterolle No. 3 t h e b o d y of the p r e d a t o r y animal shows an intermediate form
of stylization nearer t o t h e realistic r e p r e s e n t a t i o n of N o . 1. N o t only the cur-
ling feline tail is clearly visible, b u t also t h e realistic relief of t h e hinder p a r t of
t h e predatory animal, including the half of t h e abdomen, while the f r o n t p a r t
155
The n u m b e r s of t h e collections a n d t h e Figs, are given according to the article
of P . BERNARD : A propos . . . 227. Cf. also Ancient Art in the Norbert Schimmel Col-
lection, E d . by O. WH. MUSCARELLA. Mainz 1974. N 166 bis.
Bouterolle No. 1 : Louvre MNo. 1376, Collection Clot-Bey, Fig. 4.
Bouterolle No. 2 : Louvre MNo. 1377, Collection Clot-Bey, Fig. 5.
166
Bouterolle N o . 3 according to P . Bernard's typology corresponds to No. 2
according to B. G o l d m a n . Bouterolle No. 3 : Louvre MNo. 1379, Collection Clot-Boy,
Fig. 6.

MONUMENTS O F A R T FROM T H E S ANCTUARY O F O X U S 51
was transformed in t h e lower segment with a concentrically grooved surface, and

with a hole in the middle for t h e s t r a p or ribbon t o f i x t h e sheath to t h e waist.
This gave to B. Goldman an i n d u c e m e n t t o grant a d v a n t a g e t o bouterollo No. 3
in t h e evolution series and t o place it before bouterolle N o . 2 and even before
bouterolle No. 4, w h a t is absolutely correct, if we base t h e typology on t h e
degree of stylization of the figure of the predatory animal. Thus, if t h e stylistic
priority of bouterolle No. 1 is doubtless, t h e n bouterolles Nos. 2, 3 and 4 are
v e r y close. On bouterolle No. 2 t h e head of the goat is very archaic, onbouterolle
N o . 3 the body of t h e lion preserved the original iconographie basis in a great
degree. Here t h e idea of the a r t i s t obviously appears endeavouring t o represent
t h e two figures more convincingly, separated by lower o r n a m e n t a l motifs, b u t
n o t losing for him t h e original semantics. A t the same t i m e , the transition t o
ornamentalization is obvious, viz. bouterolles Nos. 4 a n d 5 are in style near
t o bouterolle No. 3.
On bouterolles Nos. 4 and 5 1 5 ' the heads of the goats are represented with
horns stylized in a p a l m e t t e , a n d t h e hinder p a r t of t h e p r e d a t o r y animal is
covered by concentric grooved notches.
Bouterolles Nos. 6 and 7 158 differ f r o m bouterolles N o s . 3 t o 5 in t h e chang-
ing of the horns into a developed p a l m e t t e , so unique t h a t it is difficult to f i n d
o u t its genesis. The ears and t h e twirl of the horns are those of a r a m . However,
t h e heads of t h e goats with beard represented with notches are very realistic.
On bouterolles Nos. 9, 10 and Nos. 12 t o 14 the p a l m e t t e of the developed
t y p e , decorating the head of t h e goat in different stylistic executions, o r n a m e n -
tally is opposed t o the already independent grooved circles forming a triangle.
I t is t r u e t h a t in the knowledge of t h e previous stylistic representations t h e tail
and hinder p a r t of t h e p r e d a t o r y animal can be guessed. Besides, on all t h e
e n u m e r a t e d specimens, the tail b e t r a y s t h e feline origin of t h e predatory animal,
it is so t o say its mark. 1 5 9
And finally, bouterolle No. 15, the representation of t h e sheath on t h e
relief of the R o y a l gate (Audience-hall) of Persepolis is t h e logical realization of
t h e whole typological series, where, according to the opinion of P . B e r n a r d ,
t o t h e p a l m e t t e representing t h e horns of t h e goat, resembling to the bull, an
accurately similar p a l m e t t e is opposed, instead of the hinder p a r t of t h e p r e d -
157
Bouterolle No. 4 : Louvre MNo. 1378, Collection Clot-Bey, Fig. 7. Bouterolle
No. 5 : Louvre, AO-24673, Fig. 8.
158
Bouterolle No. 6 : Louvre MNo. 1380, Collection Clot-Bey, Fig. 9. Bouterolle
No. 7 : British Museum No. 132925, Fig. 10.
159
Bouterolle No. 8 : British Museum No. 5428, given w i t h o u t representation.
Bouterollo No. 9 : Ashmolean Museum without No., Fig. 11.
Bouterolle No. 10 : Ashmolean Museum, Fig. 12, bronze specimen, all t h e pre-
vious Nos., 1 9 were m a d e of bono.
Bouterolle No. 11 : originates from Syria.
Bouterolle No. 12 : New York, Collection H . Schimmel, F i g . 13.
Bouterollo No. 13 : Persepolis, representation in the A p a d n a , Fig. 14.
Bouterolle No. 14 : Persepolis, Audience Hall, Fig. 16.

52 . A. L I T V I N S K I Y - I . R . P I C H I K Y A N
a t o r y animal. 1 6 0 On t h i s specimen nothing has remained f r o m t h e representa-

t i o n of t h e predatory animal. Of the scene of laceration t h e h e a d of the goat has
r e m a i n e d , losing its specific accuracy. I t is t r u e t h a t t h e o r n a m e n t a l l y repre-
s e n t e d beard has been preserved also here, this is the only link t h a t connects
t h i s representation with t h e previous representations of t h e goat. However, t h e
neighbouring representation No. 14 in the same hall has preserved tho tail and
t h e m a s k of the p r e d a t o r y animal, and the outstretched f o r m of t h e head of t h e
g o a t . I n a s m u c h as bouterolles Nos. 14 a n d 15 are u n d o u b t e d l y contemporary,
it h a s t o be presumed t h a t t h e master preparing No. 15, realizing this innova-
t i o n , did not recognize t h e necessity to follow t h e artistic rule so scrupulously,
or d i d not u n d e r s t a n d it.
Summing u p t h e analysis of the representation on t h e bouterolles, let us
t u r n t o the conceptions represented b y specific research in this question.
B . G o l d m a n and R . S t u c k y examined t h e representations on the bouterolles
as scenes of laceration. 1 6 1 P . B e r n a r d , pointing out the erroneous character of
t h e opinions of his predecessors, explains their error w i t h t h e misunder-
s t a n d i n g the territorial a n d stylistic origin of bouterolle No. 1, found in E g y p t .
Both scientists believe P . B e r n a r d writes t h a t it was b r o u g h t there from
t h e north-eastern regions of I r a n , or f r o m t h e northern p a r t of Mesopotamia in
t h e pre-Achaemenid period. Thereafter, as t h e y presume, t h e head of the lion
disappeared, its b o d y a n d t h e head of its victim formed a whole, which was
t r a n s f o r m e d into a t u r n i n g animal. I n the course of time, t h e h e a d of the goat,
as is presumed, was t r a n s f o r m e d into t h e head of a y o u n g bull, but without
horns. 162
The opinion of P . B e r n a r d , differing f r o m the conceptions of the preceding
researchers, was f o r m u l a t e d in the following way : We, on our part, believe
t h a t t h e origin of t h e d r a w i n g a n d its development are d i f f e r e n t . A more careful
s t u d y of some iconographie details and places, where some bouterolles were
f o u n d , especially t h a t (No. 1 the author), which B. G o l d m a n and R . Stucky
believe t o be the oldest, m a k e us t h i n k t h a t t h e basis of t h e d r a w i n g was already
t h e t u r n i n g animal, t h e lion, neither the goat, nor both t o g e t h e r , b u t the y o u n g
bull, a n d the representation of the lion a t t a c k i n g the goat, on bouterolle No. 1
is a local variant, characteristic of the E g y p t i a n type, created on the basis of
t h e Achaemenid prototype. 1 6 3
The evolutionary scheme set u p b y P . Bernard, in his opinion, bringing
a b o u t t h e series of bouterolles Nos. 1 a n d 2 with t h e lion a t t a c k i n g the goat,
h a s t o be examined in two aspects. His first contradiction can be summed u p
in t h e disputation of t h e species of the lacerated animal. T h e second one can be
180
Bouterolle No. 15 : Persepolis, Audience Hall, Fig. 16.
161
B . G O L D M A N : op. cit. 4 3 ; R . S T U C K Y : op. cit. 13.
162
P . B E R N A R D : op. cit. 230.
163
Ibidem.
Acta Antiqua Academiae ScierUiarum Hungarica e 28, 1980

summed u p in the inclusion of bouterolles Nos. 1 and 2 into t h e Achaemenid

period. His second, u n d o u b t e d l y correct, contradiction p e r m i t s us to discuss
t h e whole evolutionary series in t h e framework of t h e Achaemenid period.
W e are interested in t h e first two bouterolles, in which all researchers, including
also P . B e r n a r d , see the goat lacerated b y the lion. The s h e a t h of Takht-i Sangin
connects t h e early, complete representation of bouterolles Nos. 1 and 2 w i t h
the later, more stylized ones. The o r n a m e n t in the form of a h e a r t , which is on
t h e head a n d within which a flower (Nos. 3, 12), or a p a l m e t t e (Nos. 7,915)
is placed, are the stylized representations of t h e ear, and n o t of t h e horns, as is
s t a t e d b y R . Stucky, 1 6 4 P . B e r n a r d writes this stylization can be f o u n d
f r e q u e n t l y in the representational a r t of both the steppe peoples and in t h e
representations of the articles belonging t o t h e Ziwiye treasure. A n d if we ex-
clude Nos. 1 and 2, t h e n nothing remains for t h e confirmation t h a t this is t h e
head of some kind of capripeds. 165 The same opinion is m a i n t a i n e d by S. S.
Sorokin a n d V. G. L u k o n i n . l e e T h u s , t h e bouterolle of T a k h t - i Sangin, according
t o this characteristic, s t a n d s between bouterolles Nos. 1 a n d 2 according t o t h e
n u m b e r i n g of P . B e r n a r d .
The t w o paired lines of y a r n s with twisted ends, which cover the neck
a n d the b a c k of the animals, indicate in t h e opinion of P . B e r n a r d as fol-
lows : This double representation of yarns, arranged this w a y on the back of
t h e animals in such fixed form, is characteristic exclusively of t h e representa-
tions of t h e e m p t y - h o r n e d animals of t h e Neo-Assyrian and Achaemenid pe-
riods. 167 I t is possible t h a t these twirls (from two to five) bordering the lower
sphere of t h e bouterolle f r o m t w o sides, the first sphere, on all bouterolles,
covered w i t h groovings, are also dividing elements between t h e spheres t o give
t h e m a t r i p a r t i t e division. Being a r u d i m e n t of twirls represented on the boute-
rolle of Ziwiye, 168 t h e y belong already t o t h e o r n a m e n t a l motifs. B y the w a y ,
these twirls are not always present, on t h e m a j o r i t y of bouterolles t h e y do n o t
exist, viz. on bouterolles Nos. 1 5, 7, 10, 12. On the bouterolle of T a k h t - i
Sangin t h e y are definitely o r n a m e n t a l .
E x a m i n i n g the bouterolle of T a k h t - i Sangin in this connection and in t h i s
evolutionary series, it has t o be noted t h a t a f t e r the most masterly E g y p t i a n
bouterolles Nos. 1 2, it occupies u n d o u b t e d l y the third place, and if we rely
u p o n t h e representation of t h e p r e d a t o r y animal, it occupies t h e second place.
The head of t h e goat is represented with the ear, a point directed upwards.
164
R . A . S T U C K Y : op. cit. 1 8 .
185
P . B E R N A R D : op. cit. 2 3 6 2 3 7 .
166
S. S. SOROKIN: op. cit. 7677; V. LUKONIN :
- . . .
X L 1 I . 50.
4
" P . B E R N A R D : op. cit. 236.
188
R . GHIRSHMAN: Tombe princire do Ziwiy ot le d b u t de l'art animalier
scythe. Paris 1979. Pig. 1.

54 . A. L I T Y I N S K I Y I. R. PICHIKYAN
T h e goat with beard a n d o u t s t r e t c h e d narrow muzzle, is j u s t as u n d o u b t e d ,

as on bouterolles Nos. 1 a n d 2. On the b o d y of the p r e d a t o r y animal the charac-
t e r of stylization can be seen, although u n d o u b t e d l y in a smaller degree t h a n on
bouterolles Nos. 2 and 3. T h e p r e d a t o r y animal has lost only the head, t h e legs
are also represented in a stylized form, b u t t h e y have n o t y e t been t r a n s f o r m e d
i n t o concentric grooves. According t o P . B e r n a r d we h a v e t o do with one of the
earliest specimens of t h e stylistic evolution, if not with t h e second one, t h e n
with t h e third one, d e p e n d i n g on, which of t h e animals is accepted b y us gener-
alizing in this scheme.
T h e evolution proposed by P . B e r n a r d appears t o be not only doubtless,
b u t also more characteristic and advantageous, t h a n t h e table of development
b y B . Goldman, which is smaller in its composition. P . B e r n a r d gives preference
t o t h e h e a d of the v i c t i m over the stylization of t h e b o d y of the p r e d a t o r y
animal, t h e degree of which is less i m p o r t a n t in the comparisons.
T h e parallelism of t h e decrease of t h e specific characteristics on t h e goat
a n d t h e predatory a n i m a l shows t h a t in t h e early Achaemenid period t w o ani-
m a l s are represented. T h e decrease of t h e specific characteristics in the course
of stylization takes place so gradually t h a t it connects alone all bouterolles
e n u m e r a t e d by P . B e r n a r d into a unifie united evolutionary chain, where the
differences of the bouterolles are less conspicuous t h a n the similarity among
them.
T h e representations on the R o y a l gates in Persepolis can serve as an
e x a m p l e of the u n i t y of t h e style, in different degrees of stylization, where two
serially different representations, No. 14 and No. 15, most differing in t h e evo-
l u t i o n a r y series, are accurately contemporaneous. This speaks in favour of the
d a t i n g of all bouterolles t o the Achemenid period as proposed by P . B e r n a r d .
A n o t h e r example is t h e bouterolles of t h e collection of Clot-Bey originating
f r o m E g y p t . They n o t only belong actually to the Achaemenid style, b u t , in all
probability, belonged t o the Achaemenid warriors occupying E g y p t . This is
c o n f i r m e d by the investigation of the weapons of t h e a r m y of Cambyses found
in E g y p t .
W h a t was the i d e a of the representation of the s h e a t h of Takht-i Sangin?
L e t us outline very selectively the circle of conceptions a n d beliefs connected
with t h e animals a c t i n g in this scene. T h e hunting on lions with two-wheeled
chariots was pursued in E g y p t , in Asia Minor, in Syria, in Assyria and in Achae-
menid Iran. The Assyrian kings c o n s t a n t l y h u n t e d on t h e m in the jungles of
Mesopotamia, where lions were very numerous. An Assyrian king proclaimed
t h a t he, on foot and on a two-wheeled chariot, killed 920 lions, another king
modestly declared t h a t he killed 450 lions. 169
1,9
R . D. BARNETT : A Catalogue of t h e Nimrud Ivorios.
Acta Antiqua Aeademiae SeierUiarum Hungaricae 28, 1980

MONUMENTS OF ART FROM T H E SANCTUARY OF OXUS 55
The royal h u n t i n g on lions was in Assyria a p a r t i c u l a r form of royal

sport. On the other h a n d , it was also ritual. Sources f r o m t h e year 570 ( ?) r e p o r t
t h a t a t the time of his yearly festival t h e god N a b himself set out for the h u n t -
ing-ground t o h u n t on lions. The king by this d e m o n s t r a t e d (with a definite
chronological interval) his divine hypostasis and a t the s a m e time his e a r t h l y
power and a p t i t u d e t o perform his e a r t h l y duties.
Another explanation is also possible. Aur, Nergal, N i n u r t a and I s t a r
were deities, who appreciated t h e h u n t i n g on lions and helped the king d u r i n g
its time, t h u s the lion itself was t h e m o s t sacred of t h e animals. 1 ' 0
I n Assyrian times, judged f r o m t h e seals and t h e m o n u m e n t s of a r t , t h e
lion was the animal of t h e king. 171 I n t h e chariot of I s t a r seven lions were h a r -
nessed. 172 According t o a t e x t , the m o n s t e r L a m a s t u had t h e face of an e n r a g e d
lion. 173
On the colossal sculpture f r o m t h e temple in K a l a h t h e r e is a long votive
inscription of ASsur-nasir-apla t o t h e goddess .
Because still in the Sumerian period the s t a t u e of t h e lion bore the n a m e
of t h e king, it can be t h o u g h t t h a t t h e temple sculpture a t t h e same time per-
sonified the king and t h e goddess I s t a r , the patroness of t h e lions. 174
I n L a t e - H i t t i t e a r t t h e masculine deity and t h e king sometimes were
represented sitting or standing on a pedestal, flanked b y a pair of lions, 175 or
right on a lion. 176 I n a H i t t i t e t e x t t h e representation of t h e deity is described
as being m a d e of silver and travelling m o u n t e d on a wooden lion. 177 On a H i t t i t e
rock relief t h e representation of t h e god is shown with lion's fore p a r t s s u b s t i t u t i n g
his shoulders and his arms, and with t w o figurines of lions h a n g i n g head first along
t h e b o d y of t h e god. The lion's c h a r a c t e r of the god is r e n d e r e d this way. 178
T h e iconographie materials show t h a t in the a n c i e n t world, including
Mesopotamia, the lion received a solar symbol. 1 7 9 The lion was the a t t r i b u t e also
of t h e Syrian solar deities. I t was t h e animal of A s t a r t e , K e d e s and A t a r g a t i s .
I n Hieropolis and E m e s a t h e solar d e i t y Gennaios (GNN') was worshipped b y
t h e Syrians in the form of a lion. 180 Therefore, in the A n c i e n t E a s t the killing
1,0
R . D. BARNETT : Ibidem. 70 71 ; R . D. BARNETT : Assyrische Palastreliefs. 24.
171
A . SALONEN : J a g d . . . 122.
172
Ibidem. 209.
1,2
Ibidem. 224.
174
E . R . GOODENOUGH : Jewish Symbols in the Graeco-Roman period. Vol. V I .
Toronto 1968. (Bollingen Series. X X X V I I I ) 39.
175
E . AKURGAL: Die K u n s t der H e t t h i t e r . Mnchen 1976. Pis. 109, 126 127.
174
E . R . G O O D E N O U G H : op. cit. V o l . V I I , 4 3 , F i g . 4 1 .
177
H . GOLDMAN : A Crystal S t a t u e t t e from Tarsus. A O E H . New York 1952. 131,
with reference to a work not accessible to us.
178
E . R . G O O D E N O U G H : op. cit. P l . V I I , p . 30 ; 43, F i g . 38.
179
A. VOLLORAFF-ROES : The Lion with Body Marking in Oriental Art. J N E S 12
(1963) 4 4 - 4 6 .
180
R . D. BARNETT : A Catalogue of the Nimrud Ivories . . . 71 ; E . R . GOODE-
NOUGH: op. cit. 44 45. On the lions in Syrian art and in religions, especially s e e :
SEYRIG : La triade hliopolitane et los temples de Baalbek. Syria 10 (1929). (The lion
was the incarnation of all members of t h e divine trinity.)
Acta Antiqua Academiae Scientiarum Hungarinic 28, 1980

56 . A. L I T V I N S K I Y - I . R . PICHIKYAN
of t h e lion, in the ritual sense, must have been a sacral act of the connection of
t h e king with this deity. The lions from t h e tomb of Antiochus I from Comma-
gene are the symbol of the king, of his divine nature and, especially, of his solar
hypostasis. 1 8 1
The sacral, in particular the solar, symbology, connected with the lion,
is known in the religious conception of m a n y peoples of antiquity. Thus, the lion
of Nemea, with which Heracles fights, according to the tradition, is believed
t o have appeared from the sky or from the moon. The lion is the sacred animal
of Apollo of Patara. In the lions the sun shines Seneca said (when
t h e sun at the verge of the summer arrives at the sign of the Lion, this is
t h e hottest time of the year). With t h e Berbers and the Egyptians the lion
is compared with the sun. At the same time it is the symbol of the star Venus,
t h e attribute of Eros, Bacchus, Cybele, Astarte, the Syro-Phoenician Atargatis,
t h e Punic Baal and Tanit.
I n Egypt, Syria, and in the whole Graeco-Roman world the head of the
lion was connected with the sources of water. In Athens, Ephesus, Olympia, and
in m a n y other places the lion's head with opened mouth was placed into the
springs and the water flew through it, 182 and in Assyria the sculptured heads
of lions served as the guards of the springs. 183 In ancient Greece and in Rome
t h e lion is the symbol of power, particularly t h a t of the emperor. Still another
hypostasis of the b o n i s chthonic,it is especially clearly shown in the conception
on the lion as the guard of the graves (spread in the Graeco-Roman world and
also beyond its boundaries). 184
I n the ancient Indian epic the lion appears as the vehicle of physical
s t r e n g t h , power, greatness, etc. 185 The comparison of the ruler with the lion,
t h e meaning of the word lion, as hero, intrepid, knight, was wide-spread
in medieval Tajik and Persian literature. 186
I n Mithraism the lion was identified with the nature of power and energy
of bodily and earthly existence. The lion is the fullest personification of the soul
in a bodily cover. At the same time it is the natural principle of fire and its
reason, the source of internal heat, giving motion and energy to others. This
181
K . HUMANNO. PUCHSTEIN: Reisen in Kleinasien m i d Nordsyrien. Berlin
1890. Pis. X I X X L .
182
E . R . G O O D E N O U G H : op. cit. 5860.
183
G. RERROTCH. CHIPIEZ: Histoire de l'art dans l'antiquit. I I . Paris 1884.
F i g . 311.
184
Detailed references a n d sources s e e : E . KAGAROV : , -
. . 1913. 216 217 ; . KELLER : Die Antike Tier-
welt. I . Leipzig 1909. 4 5 6 1 ; STEIER : Lwe R E , X X V . S t u t t g a r t 1926. 983 984;
E . R . G O O D E N O U G H : op. cit. 6062.
183
S. L. NEVELEVA : . -
. Moscow 1979. 103.
186
M.-N. . OSMANOV : - IXX . Moscow.
1974. 95.

animal is compared with J u p i t e r annihilating the e a r t h l y giants, t h e lion is t h e

p l a n e t a r y symbol of Jupiter. 1 8 7
The lion was t h e travelling companion of the f o u r - a r m e d goddess, whose
n a m e is n o t y e t known t o us, b u t who was worshipped in N o t h e r n Asia. H e r
representations are everywhere f o u n d on seals, on jewellery and on works of
toreuti'cs.
I n t h e F a r E a s t and in Central Asia the image of t h e lion comes f r o m
B u d d h i s t iconography, t h e lions are t h e usual guards or t h e carriers of t h e
B u d d h a , t h e lion is t h e p e r m a n e n t companion of t h e d e i t y of highest wisdom
Manjusri. 1 8 8
The reports, which we have regarding the beliefs connected with t h e s t a g ,
with t h e peoples of Asia Minor a n d t h e Mediterranean, are considerably smaller
in volume. I n Mesopotamia t h e s t a g was not only a wild animal, b u t it was also
k e p t as a domestic animal in t h e temples. The milk of t h e s t a g was e.g. a sacri-
ficial food of t h e goddess Ningirsu. Besides this, t h e horns of t h e stag a n d its
other p a r t s were used as medicines. 189 According to t h e iconographie historians,
the stag is t h e sacred animal of t h e H i t t i t e masculine d e i t y . The symbol of t h e
t h i r d m e m b e r of t h e holy t r i n i t y , K a r h u h a s a , was t h e h o r n or pair of horns of
t h e stag. 1 9 0 According t o a H i t t i t e t e x t f r o m t h e end of t h e 8th century B. C.,
t h e god, p a t r o n of t h e fields, stood on a stag. 1 9 1
The stag, especially t h e s p o t t e d stag, in the period of classic Greece, was
held mainly an Asiatic animal. I t s s p o t t e d hide was t h o u g h t t o be t h e reflection
of the star-lit sky a t night. Therefore, t h e stag was t h e a t t r i b u t e of the moon
goddess, who was t h o u g h t t o be A r t e m i s (and also A t h e n e ) by the Greeks.
The stag occupied a significant place in t h e cult of Artemis, in her iconography,
including coin iconography, as well as in her poetic descriptions. The stag was
also the symbol of the night and darkness, of the evil principle, which is de-
feated by t h e solar lion, t h e representative of light a n d giver of happiness. 1 9 2
I n ancient I n d i a t h e stag, t h e fallow deer, was held the s y m b o l of helplessness. 193
I n popular imagination t h e Ossetic god, Afsati was an old m a n with a h e a d
decorated with stag-horns. 1 9 4
187
L. A. CAMPBELL: Mithraic Iconography and Ideology. Leiden 1968. ( t u d e s
prliminaires aux religions orientales d a n s l'Empire Romain. T. 11) 68, 287, 309.
LES N . Y. DIAKONOVA :
. (1961) 266267; eadem : .
10 (1969) 88.
189
A . S A L O N E N : op. cit. 161.
190
R . D . BARNETT : A C a t a l o g u e . . . 73.
191
L. ROST : Zu den hethitischen Bildbeschreibungen. Mitteilungen des I n s t i t u t e s
f r Orientforschung der Deutschen Akademie der Wissenschafton zu Berlin 8 (1961)
179180 ; E . PORADA : Of Deer, Bolls a n d Pomegranates. Archaeologia Iranica. Miscel-
lanea in honorom R . Ghirshman. Leiden 1970. 116.
192
E . KAGAROV : . . . 267 ; O . K E L L E R : op. cit. 277.
193
S. L . NEVELEVA : . . . 106.
194
V. I. ABAEV : - . . . . , I I I . 14.
g* Acta Antiqua Academiae Scientiarum Hungaricae 28, 1980

58 . A. L I T V I N S K I Y I . . PICHIKYAN
On the sheath of T a k h t - i Sangin t h e lion and the stag do not appear

isolated, b u t in a scene, where the stag is seized by the lion, in which the incom-
mensurability of their size is stressed. This is the scene of laceration and at the
same time, the scene of t h e t r i u m p h of the power of the lion and t h a t of the
defencelessness of the s t a g . I n the a r t of the Ancient E a s t there is a frequent
variation of the scene, where the lion lacerates another herbivorous animal,
a bull. The fight between t h e lion and t h e bull, the scene of laceration of the
bull b y the lion, is regarded by the research as having a mythological or ritual
meaning. However, the concrete interpretations proposed b y the different re-
searchers differ considerably from one another. Thus, in the light of the Canaan-
ite m y t h s , they presume t h a t the lion and the bull, each one of t h e m , personifies
a certain deity fighting with another, a t t h e same time the scene personifies the
alternation of the seasons. 195
Publishing the Assyrian bronze disk with the scene of the laceration of
t h e bull by the lion (the lion is on the back of the bull), the u m b o of the shield
of Sargon I I (721 705 . .), E . P o r a d a presumes t h a t the Assyrians originally
regarded the lion as an enemy. Thereafter the representation of the king of the
animals was transformed into the representation of the conqueror and finally,
into the king himself. T h e t r i u m p h of t h e lion over the bull had a symbolic
character. Therefore t h e lion on the disk is shown considerably bigger t h a n t h e
bull. This corresponded t o t h e ideology of both the customer and the a r t i s t .
This meaning of the motif explains, why these representations are found on t h e
embroidered tunic of Assurnasirpal (9th century . C.). 196
W. H a r t n e r and R . Ettinghausen suppose t h a t this motif originally had
an astrological meaning a n d this interpretation was preserved still in the Achae-
menid period, for example a t the time of t h e ceremonies of Nouruz in Persepolis.
R . Ettinghausen, at t h e same time, points out t h a t on the reliefs of the stair-
cases of the apadna of Persepolis there are no royal representations, and the
scene of the fight of t h e lion with the bull is placed side by side with the inscrip-
tion, where Xerxes praises him realm. This scene is found n o t less t h a n sixteen
times. R . Ettinghausen, following E. Herzfeld, arrives at t h e conclusion t h a t in
Persepolis, together with the original astral meaning, this scene represents a
heraldic emblem.
R . Ettinghausen also underlined t h e evolution of this scene, viz. on t h e
Assyrian disk the lion floats over t h e bull, while on the Achaemenid reliefs it
really lacerates the herbivorous animal. 197 I n the opinion of W. Hartner t h e
195
R . D . H A R N E T T : op. cit. 72.
196
E . PORADA : An Assyrian Bronze Disc. Bulletin of the Museum of Fine A r t .
Boston 48 (1950). Cited a f t e r : W. HARTNERR. ETTINGHAUSEN : Oriens 17 (1964) 168.
197
E . HERZFELD : I r a n in the Ancient E a s t . London New York 1941. 251 252 ;
W . HARTNERR. ETTINGHAUSEN : The Conquering Lion. The Life Cycle of a Symbol.
Oriens 17 (1964) 169. I n Achaemenian a r t t h e scene of laceration of a bull by a lion was
t h e emblem of some cities, Biblos, Kition a n d Tarsus. (R. D. BARNETT : A Catalo-
gue . . . 73.)

MONUMENTS OF ART FROM T H E SANCTUARY OF O X U S 59
scene of the fight of the lion with t h e s t a g has to be regarded as an equivalent

of t h e scene of the f i g h t of t h e lion with t h e bull. 198
I t has t o be added t h a t the solar symbology of t h e s t a g was well known
in t h e I n d o - E u r o p e a n consciousness. I n Russian folklore t h e sun is a glowing
s t a g . I n t h e Tales of r e c u r r e n t summers t h e fawns fall f r o m t h e sky. I n t h e
Scandinavian beliefs t h e s t a g is t h e son of the Sun, 199 four s t a g s correspond t o
t h e four cardinal points. 2 0 0 Of course, t h e possibility of entirely different inter-
p r e t a t i o n s should not be excluded either. W e mention e.g. an Ossetic folk-tale
motif : t h e heroine appears before the hero in the form of a fallow-deer to t u r n
t h e r e a f t e r into a beauty. 2 0 1 T h e fallow deer is the image of feminine helplessness
also in the ancient I n d i a n epic, where t h e victory over t h e weak e n e m y resem-
bles t o t h e victory of t h e lion over a t i n y animal, and t h e victory over a m i g h t y
e n e m y resembles to t h e victory over a big animal (a bull). I t should also be k e p t
in mind t h a t the king in t h e ancient I n d i a n epic is compared with the lion. 202
W e do n o t know e x a c t l y w h a t conceptions are reflected b y t h e scene on
t h e sheath of Takht-i Sangin, b u t it is clear t h a t t h e y belong t o t h a t group of
conceptions, which was outlined above.
I n conclusion we note t h a t the lion on the sheath n o t only seizes and holds
t h e stag, b u t if we t a k e t h e s h e a t h as a whole, with t h e bouterolle, t h e n also
t o g e t h e r with it so t o say t r a m p l e s on t h e contracting goat, and although t h e
p r i m a r y imago on the bouterolle, as a scene of laceration, in t h e iconographie
relation was already almost lost, this representation very likely continued t o be
understood as the sign symbolizing such a scene.
V. V. I v a n o v and V. N. Toporov discovered the all I n d o - E u r o p e a n concep-
t i o n on t h e embedding of ritual symbols into each o t h e r , connected with the
world (or sacred) tree, b y which t h e world tree stands a t t h e beginning of t h e
chain and d e a t h is a t t h e end. 2 0 3 Conceptions of the same t y p e have been dis-
covered in ancient I n d i a a n d in Central Asia. 204 And on t h e s h e a t h of T a k h t - i
Sangin, a p p a r e n t l y , we h a v e t o do with a continuously n a r r a t i v e series of
sacrificial-ritual conceptions (acts), in which the scenes of t h e f i g h t and victory
of t h e lion over the stag (goat) s u b s t i t u t e t h e stages of t h e f i g h t of t h e e a r t h l y
ruler (or divine hero) with their rival (rivals), of the f i g h t , which ends with
t r i u m p h and apotheosis, when the rival, in the form of a s t a g , is defeated and
humiliated.
198
W. HARTNER: The Earliest History of the Constellations in t h e Near E a s t and t h e
Motif of t h e Lion-Bull Combat. J N E S 24 N 1 2 (1965) 16 ; PH. AOKERMAN developed a
similar conception, see : PH. ACKERMAN : Iconography. SPA I . 1967. 389 391.
199
V. I. ABAEV : - I I I . 14.
. IVANOVV. N. TOPOROV: -
. Moscow 1974. 28.
201
V . I . ABAEV : -. . . 14.
202
S . N - N E V E L E V A : op. c. 1 0 3 , 1 0 6 .
203
V . V . I V A N O V V . N . T O P O R O V : op. cit. 35.
204
B. A. LITVINSKIY. I. ZEYMAL :
s t u p a . . 1981.
Acta Antiqua Academiae Scientiarum Hungar icae 28, 1980

60 . A. L I T V I N 8 K I Y I. R . P I C H I K Y A N
If we accept t h e explanation of V. I. Abaev, according t o which the ethno-

n y m Saka originates f r o m the reconstructed Old Iranian *saka- stag, and t h e
opinion (of I. M. Artamonov, N. L . Chlenova and others), according to which
t h e stag was the most wide-spread t o t e m of the Sakas, 2 0 5 t h e n it would be
t e m p t i n g to concretize this scene, as t h e symbolically reflected victory of t h e
Achaemenid king, or of his (Bactrian ?) governor over the Central Asian Sakas.
However, the explanation sak stag cannot be regarded as t h e only possible20
a n d more or less firmly established explanation 2 0 7 t h a t renders t h e above men-
tioned interpretation problematic.
The sheath of T a k h t - i Sangin, with regard to its form, is close to the gold
coverings of the sheaths originating f r o m the treasure of t h e Oxus (No. 22).
T h e y coincide at an accuracy of millimetres both regarding their size (27.6 cm)
a n d t h e time of their manufacture, viz. 6th century t u r n of the 6th and 5th
centuries . C. I n the territory of Tajikistan, Uzbekistan, Afghanistan, P a k i s t a n
a n d northern India, similar articles, close with regard t o their time, have n o t
been discovered. This confirms the opinion on the provenance of the treasure of
t h e Oxus in the upper course, on t h e right bank of the A m u - D a r y a and it is
possible t h a t this speaks in favour of the origin of t h e treasure from temple
treasuries. And t h e material ivory , a whole series of specific details of
t h e form and conception of the representation, point t o t h e local, B a c t r i a n
m a n u f a c t u r e of these sheaths. So f a r sheaths of similar t y p e were known only
f r o m t h e representations on the reliefs of Persepolis, while now as a result of
t h e excavations the a r t of the world has been enriched with an excellently
preserved, highly artistic original specimen made of ivory from the 6th5th
centuries B.C.
The circumstance t h a t the E g y p t i a n bouterolles are very close to t h a t of
T a k h t - i Sangin, is n o t t h e only example for the width of t h e area of spreading of
t h e imperial style in t h e Achaemenid period. The same phenomenon is seen b y
us a t the study of t h e architectural monuments, especially in the spreading of
t h e torus-shaped Achaemenid bases from E g y p t and Palestine to Central
Asia. 2 0 8
All the 15 bouterolles published by P . Bernard, are d a t e d to the period
524404 . C., t h e epoch of the f i r s t hegemony of t h e Persians, t o the time of
t h e first conquest of E g y p t . B y this t h e dating in the b r o a d sense, to the 5th
c e n t u r y seems t o us t o be most acceptable. I n this case t h e f a c t is also very
i m p o r t a n t t h a t we f i n d the closest analogies of the form of sheaths on the ear-
205
V . I . ABAEV : op. cit. 1 2 1 6 .
206
. A. LITVINSKIY : . Moscow 1972. 151 162.
207
. SZEMERNYI : Four Old I r a n i a n Names : S c y t h i a n - S k u d r a - S o g d i a n - S a k a .
W i e n 1980. AW Phil.-hist. Kl. Sitzungberichte 371. B d . P . 41.
208
1. R . PICHIKYAN : (, ).
. Moscow 1979. 6761.

MONUMENTS OF ART F R O M T H E SANCTUARY O F OXUS 61
liest specimen, No. 22, of the treasure of t h e Oxus and on t h e early representa-
tions of Persepolis. 209
As far as the chronology is concerned, according t o t h e continuous evolu-
tionary scheme, proposed by P . B e r n a r d , it seems t o be very convincing t o
include in it also the sheath of Takht-i Sangin. The unbroken evolution, t h e
similarity of the representations and of t h e stylistic marks, first of all, show t h a t
the bouterolle of Takht-i Sangin and, consequently also t h e sheath, cannot be
d a t e d later t h a n t h e 5th century B. C. A n d if we take i n t o consideration, as it is
accepted in archeology and in history of a r t t h a t the stylistic marks most fre-
quently are also chronological characteristics, then it becomes an obvious f a c t
t h a t the sheath of Takht-i Sangin has t o be dated to t h e first half of the 5th
century B . C . , and b y no means later.
The ornament f r a m i n g the upper p a r t of the sheath also speaks in f a v o u r
of this or an even earlier dating. I t is d a t e d to the transitional period (6th cen-
t u r y B.C.) of the Minor Asian tonguelet ornament of Ionic ovoli with the initial
stage of the intermediate stems, 2 1 0 hardly observed only in the lower p a r t of
t h e stripe between the upper representation and the bouterolle. The braiding
above the dividing ornamental stripes above the bouterolle belongs to the Minor
Asian and Iranian early archaic ornamental motifs 2 1 1 borrowed from Asia
Minor, and used by no means later t h a n t h e first half of t h e 5th century B. C.,
similarly as the representations, ornamentation and f o r m of the sheath as a
whole has also to be d a t e d .
In cultic deposit No. 4, together with the sheath a votive stone altar was
found, with a bronze sculptured figure of Silenus Marsyas playing on an two-
branched flute (Fig. 4), placed on it.
The first Ancient Greek inscription on the territory of Asia Minor reads
as follows :
Ev%r)v
vfhjxev
'AxQoadtxr]
"Oon
Atrosokes dedicated his vow to Oxus (Fig. 5). I n t h e open altar the syn-
thesis of Graeco-Bactrian culture is realized with unusual sharp clearness. The
208
O . M . D A L T O N : op. c. N 2 2 ; E . F . SCHMIDT : P e r s e p o l i s . I . N 6 5 A . B , D.,
PI. 120, 121. Unlike the late sheath of Artaxerxes I I I , where t h e bouterollo is trifoliate,
s e e : E . F . SCHMIDT: Persepolis. I I I . PI. 75 (18).
2,0
C. WEICKERT : Das lesbische K y m a t i o n . Leipzig 1913; W . B. DINSMOOR :
T h e Architecture of Ancient Greece. London I960. 13, 25, 29. 44.
211
R . GHIRSHMAN : T o m b e . . . P l . I X , 6, 8, X I I , 7 ; CH. WILKINSON : V e s s e l s
from Ziwiye Terminating in Animal Heads. Proceedings of the l V t h International Congress
of Iranian Art and Archaeology. P a r t A, April 24 - M a y 3, I960. Teheran 1967. 2982,
Fig. 1046.
Acta Antiqua Academiae Scientiarum Hungar icae 28, 1980

local sacrificer of fire, Atrosokes (his n a m e f r o m Old I r a n i a n is deciphered a s

burning with the sacred fire, or useful t o t h e deity of fire or mostly having
t h e force of the deity of fire) dedicates a small altar of Greek f o r m with a figu-
r i n e of t h e Greek Silenus, a Bacchic deity, w i t h an Ancient Greek inscription in
f o r m a n d language t o t h e local deity of river Oxus (the Greek transcription of
t h e V a x s A m u - D a r y a ) . W e have t o do n o t with a superficial mixing, or t h e
a m a l g a m a t i o n of t h e f o r m s , b u t with a deep process leading t o t h e mutual pene-
t r a t i o n of the world view of the B a c t r i a n s a n d t h e Hellenes. T h e votive inscrip-
t i o n s are characteristic of t h e sacred groves, t h e Greek t e m e n e , t h e y are t o be
f o u n d also in the a r e a of temples, b u t t h e y are always addressed to the deity t o
w h o m t h e temple is consecrated. Consequently, also the s a n c t u a r y of T a k h t - i
S a n g i n , as it becomes clear f r o m t h e inscription, was consecrated t o Oxus, t h e
h i g h e s t deity of t h e r i v e r and of the p r i m e v a l element of w a t e r . Therefore, t h e
i n f o r m a t i o n included in t h e four words is v e r y significant. T h e ancient Greeks
r e n d e r e d the h y d r o n y m Vaxs with t h e word Oxus (by which the local in-
h a b i t a n t s a t t h a t t i m e did n o t call only t h e river Vaxs, b u t also the whole
A m u - D a r y a ) . The c o m m o n n a m e a p p e a r e d quite late. 212 On t h e K u s n a coins
t h i s d e i t y is represented with t h e inscription 01 (Vaxs). Thus, in
a n t i q u i t y not only one of t h e main rivers of Central Asia w a s denoted by t h i s
w o r d , b u t also the d e i t y connected with it. 2123 I n the sacred hierarchy it occu-
p i e d a high place, because in t h e inscription on a gem (kept in the C a l c u t t a
m u s e u m ) the phrase Vaxs, the sole god appears. The great Khorezmian
scholar Birni in t h e 11th century s t a t e d t h a t Vaxs is t h e n a m e of the angel
p l a c e d (to supervise) t h e waters, especially river Jaihn (Amu-Darya, t h e
a n c i e n t Vakhsh). 2 1 3 U p t o t h e twenties w i t h the Tajiks a n d t h e Uzbeks m a n y
beliefs connected w i t h this river were preserved.
The figure of SilenusMarsyas s t a n d i n g on the base of t h e altar is well
k n o w n in the Greek p a n t h e o n . SilenusMarsy as, playing a two-branched f l u t e ,
is s t a n d i n g in full-size on t h e small bronze base, cast together with the figurine.
T h e small base is m o u n t e d on the upper plain surface of t h e a l t a r with the help
of a hammered o u t slit, soldered with lead. Silenus holds t h e f l u t e in his b e n t
h a n d s . His left foot is slightly s t r e t c h e d a h e a d . The legs are represented f r o n -
t a l l y , somewhat geometrically and generalized. The big h e a d sharply separates
in profile together w i t h t h e p r o t r u d i n g abdomen. The shoulders are in propor-
t i o n very narrow, only slightly projecting f r o m the line of t h e head. F r o m t h e
c h i n t o the omphalos along t h e s t e r n u m t h e r e is a deep incision separating t h e
big, hanging breasts (Fig. 5).
212
J . MARKWART : Wehrot und Arang. Leiden 1938. 351.
212A
J . HARMATTA : Cusaniea. Acta Orient. H u n g . 11 (1960) 198 200; J.ROSEN-
FIELD: The Dynastic A r t s of the K u s h a n s . Berkelev and Los Angeles 1967. 92, PI.
VIII/155.
213
BIRUNI ABURAYKHAN : . Tashkent 1957. 258.

The big, bald skull is f r a m e d b y long locks. The coiffure directly goes o v e r
t o t h e short, broad beard, fashioned, like the locks, generalized, b y deep, s h o r t
incisions. On the forehead there are two wrinkles : t h e one is vertical, deep,
t h e o t h e r is short, horizontal. The small eyes are deep-set. The cheeks are p u f f e d -
u p f r o m tension. The huge, jug-eared head, with b r o a d nose, the stressed ugly
figure, are characteristic of the o u t w a r d of Silenus. The grasp of
t h e n a k e d masculine body, the representation of t h e realistic Hellenistic m y t h o -
logical figure in the traditional m a n n e r , the caricaturing of the face, b e t r a y t h e
h a n d of a Greek master. The face of Silenus belongs t o t h e caricatured p o r t r a i t
representation of Socrates, f a v o u r e d in Hellenism.
The altar combines Hellenistic and B a c t r i a n traditions. Language, for-
m u l a a n d script of the inscription are Greek, the contents B a c t r i a n . On t h e
altar, dedicated t o the B a c t r i a n d e i t y Oxus Vax, is set the Hellenic sculp-
t u r a l figure of t h e Greek deity Marsyas whose f u n c t i o n was a m o n g others t h e
protection of the watercourses. T h u s , one and t h e same semantic idea got
twofold expressions : an epigraphic (written) expression (Oxus VaxS)
and a sculptural one (Marsyas). This reduplication increased n o t only t h e
sacred force of the altar b u t it also bore witness t o t h e cultural synthesis a n d
drew attention to the culturally a n d perhaps ethnically mingled Graeco-Bactrian
environment.
The epigraphic analysis of t h e inscription, 2 1 4 t h e comparison with t h e
m o n u m e n t a l inscriptions of the 3rd century . C. a n d financial dipinti of t h e
2nd c e n t u r y . C. discovered in A i - K h a n u m , as well as with t h e later inscrip-
tions of Surkh K o t a l , permit t o d a t e the dedication on t h e altar t o t h e middle
of t h e 2nd century . C. t h a t is t o t h e last decades before t h e fall of Graeco-
Bactria. 8 1 5
I n cultic deposit No. 4 t h e basic mass of the f i n d s consisted of polychrome
clay sculpture. The large q u a n t i t y of finds made of alabaster and u n b u r n t clay,
u n d o u b t e d l y has t o be included in local, Bactrian a r t . Their materials are so
breakable t h a t their local m a n u f a c t u r e does not raise a n y doubts. Besides t h e
forms and difference of the m a t e r i a l , t h e heads, torsos and smaller details of t h e
figures can be divided according t o t h e style of m a n u f a c t u r e and the chronology
into three entirely independent schools.
214
V. A. LIVSHITS agrees with t h e dating proposed by us. Adhering to our opinion,
we q u o t e with gratitude the views of t h e national epigraphists, specializing on the inscrip-
tions of Olbia, Chersoncsus and the Bosporus. E . I. LEVI, A. I. BOLTUNOVA, N. S. BELOVA
and V. V. RUBAN are inclined to an earlier date, viz.: 3rd century turn of t h e 3rd
century-beginning of t h e 2nd century B. C. J u . G. VINOGRADOV believes t h a t , accoiding
to tho epigraphic d a t a , the offering of Atrosokes should be d a t e d to t h e middle of t h e 2nd
century . C.
215
L. ROBERT : Les inscriptions. P . BERNARD : Fouilles d'Ai K h a n o u m . MDAFA
X X X . Paris. 617 520, PI. 11 ; P . BERNARD: Pratiques financires d a n s la Bactriane
hellnise. B S F N 34 Pl. I I ; H . HUMBACH : Baktrische Sprachdenkmler. I I . Wiesbaden
1967. Pis. 1721, 22 above.

T h e sculpture of t h e Seleucid times, produced according t o Greek tradi-

tions of Asia Minor belongs to the f i r s t school. The second, Graeco-Bactrian
school is characterized b y specific local t r a d i t i o n s . The t h i r d school, t h e most
n u m e r o u s one, includes sculptures p r e p a r e d in the P a r t h i a n a n d Indo-Gan-
d h a r i a n style. At t h e s a m e time it is clear t h a t the sculptures were previously
crushed, and t h e r e a f t e r b r o u g h t and t h r o w n into the d i s t a n t n o r t h e r n dead-end
of corridor No. 2, t o g e t h e r with the inscription dedicated t o Oxus, t h e sheath of
akinakes, the silver h a n d bigger t h a n t h e n a t u r a l size, the Scytho-Siberian
n e p h r i t e representation of t h e wolf-dragon, and the 15 stone disks, because it
was seen already before t h e restoration t h a t all the sculptures were f r a g m e n t a r y
a n d only one whole a l a b a s t e r statue could be p u t together t h a t of an eastern
s a t r a p , a n d the rest were only separate p a r t s .
T h e head of a beardless man with r o y a l hair-band or diadem, representing
t h e f i r s t school (Fig. 6), is t h e best with r e g a r d t o style and t h e workmanship of
p r e p a r a t i o n . The clay is local, yellowish b r o w n in colour, with inclusions of sand
a n d limestone. The h e a d was formed s e p a r a t e l y from the body. The blue-black
locks, held together w i t h a pink fillet, d r o p on t h e forehead and t h e temples with
b e n t , elegant plaits t h i n n i n g towards t h e ends, divided with a modelling tool
into several parts. T h e pupils of t h e deep-set eyes are not shown plastically,
t h e y were painted. T h e curls of the locks stress t h e b e a u t y of t h e s t r a i g h t fore-
h e a d . I n spite of the s m a l l dimensions (its scale is half of t h e n a t u r a l size) the
head looks m o n u m e n t a l . T h e head is modelled with hand. The m a s t e r was un-
d o u b t e d l y a Greek, or a strongly Hellenized Bactrian, a c q u a i n t e d with the best
t r a d i t i o n s of the Greek sculptors Scopas a n d Lysippus, widely known f r o m orig-
inal works and copies t o t h e local m a s t e r s creating the Hellenistic sculptural
koine. The sculptor modelling the h e a d of a large s t a t u e was familiar with the
Seleucid specimens of Asia Minor, b o t h in m o n u m e n t a l and minor sculpture,
including the coin representations of t h e early Hellenistic rulers. The affinity of
t h e general appearance, as well as of t h e conception of the small details, viz.
t h e hair-do, the forehead, t h e eyes, t h e oval face, b e t r a y the nearness of t h e
schools, of t h a t to which t h e master of n o r t h e r n Bactria, more accurately t h a t
of t h e Takht-i Sangin, belonged, with t h e best workmanships of P e r g a m u m ,
Priene, Magnesia on t h e Maeander, working in t h e 3rd c e n t u r y . C. The head
is a p o r t r a i t . On the basis of the h a i r - b a n d or diadem worn b y t h e diadochoi
a n d epigons, it represents t h e governing d y n a s t .
T h e work of t h e second local school deserves special interest. I t is repre-
sented b y the sculpture of t h e s a t r a p robed in kyrbasia (Fig. 7). The masculine
h e a d is represented in front-view, its scale is one third of the n a t u r a l size. The
face is slightly b r o a d e n i n g below. The nose is thin, high, with a slight sinking
on t h e ridge and with a moderate conversity towards t h e tip. The nostrils are
fashioned with arches. T h e deep-set eyes are outlined with linear black ovals,
p o i n t e d a t the outer edge a n d rounded off a t t h e nose. I n the middle of the upper

MONUMENTS OF A R T F R O M T H E S ANCTUARY O F OXUS 65
line of the oval there is a circular spot, t h e pupil. The arches of the eyebrows are
indicated by black lines reaching over t h e lateral edges of t h e nose. The m o u t h
is n o t big. P a r t l y , especially t h e u p p e r lip, is covered b y moustache and b e a r d .
T h e lowerlipis straight, the small,semioval upper lip f i t s t i g h t t o it. The lips are
p a i n t e d with red colour. The whole lower p a r t of the face is f r a m e d by thick, black
hair, into which the moustache a n d t h e beard melt. The broad moustache is
p a r t e d in the middle, it t u r n s into t h e beard, which begins below t h e cheeks. The
beard in the beginning hangs along t h e contours of the face broadening down-
wards, and below it is pulled t o g e t h e r wedgewise. The hairs in t h e moustache
are represented with several oblique notches. The finishing of the beard is m u c h
more detailed. The naturalness of t h e representation is enhanced b y t h e relief-
like shaping of the beard as compared w i t h the plane of t h e face. Above tho low
forehead a bulging and almost s m o o t h stripe of locks is hanging, clipped. Higher
u p t h e hair-dress, the kyrbasia begins. I t consists of a surrounding, smoothly
projecting cylinder or band, above which the h a m m e r e d semispheric or conic
p a r t projects. Around the b a n d t h e s m o o t h plates-face-guards hang down, cov-
ering the lateral parts of the h e a d a n d reaching to the level of the chin. Below
t h e y have round outlines. The modelling is very accurate and convincing.
The sculpture is painted : the b e a r d , t h e moustache, the eyebrows, the oval of
t h e face, the eyes and tho pupils are black, the face is brown, the lips are ver-
milion, the head-dress is white. This sculpture is masterly, although it is a
laconically scanty portrait representation of a man, imperatively and a t t e n t i v e -
ly looking before himself. The h e a d is m a d e of white alabaster, it is hollow in-
side. A s t u d y of the reversed side creates t h e impression t h a t a t the preparation
of t h e relief the mass was so t o say c r a m m e d into the form. Meanwhile, into t h e
necessary places f r o m the reversed side additional lumps of alabaster were p u t
inside. I t is hoped t h a t the s u b s e q u e n t restoration will p e r m i t the publishing of
t h e whole figure. The dimensions of t h e f r a g m e n t are as follows : height 12 cm.,
b r e a d t h 7.4 cm.
K h a l c h a y a n furnished a whole series of masculine heads, including also
heads with beard and moustache a n d e v e n with similar head-dress. However,
w i t h regard t o style t h e y differ f r o m t h a t of Takht-i Sangin. I t is especially
i m p o r t a n t t h a t the common oval of t h e face in K h a l c h a y a n becomes narrower
downwards, while in Takht-i Sangin, on t h e contrary, it is broadening below.
I n this respect the nearness t o t h e sculpture of H a t r a a n d t o t h e sculptures f r o m
t w o P a r t h i a n centres can be followed. The question of t h e P a r t h o - B a c t r i a n
m u t u a l relations in the field of a r t a n d architecture has been elaborated repeat-
edly and in a detailed form. The sculpture of Takht-i Sangin showing traces of
P a r t h o - B a c t r i a n mutual relations in t h e field of art, s u p p o r t s this thesis. A t t h e
same time it expands our conceptions a b o u t the appearance of B a c t r i a n sculp-
t u r e as a whole. As for the d a t i n g , t h e broad dating t o t h e 3rd 2nd centuries
B. C. seems t o be most acceptable.

66 . A . L I T V I N S K I Y I . . P I C H I K Y A N
One of the n u m e r o u s sculptures, t h e alabaster sculpture representing a

B a c t r i a n , of the t h i r d local school (Pig. 8), is very interesting. The head, t h e
a r m s a n d legs are p a r t l y broken off. T h e m a n was represented in motion. T h e
right leg is ahead, t h e l e f t one is k e p t behind. The clothing is local, Indo-
B a c t r i a n , in the form of a long, pleated chiton girded with a sash, fastened with
a k n o t w i t h two long e n d s hanging down. Of t h e iconography, t h e stretched pro-
portions with narrow w a i s t , with r a t h e r t h i n thighs and n a r r o w shoulders, t h e
a c c u r a t e , almost graphic conception of t h e dimensions a n d t h e sure, somewhat
generalized representation of the form, are characteristic.
T h e comparison w i t h t h e K u s n a iconography shows some points of con-
t a c t w i t h the image of K a n i s k a eternized on coins. On t h e coins of K a n i s k a ,
t h e s t a n d i n g king is in girded k a f t a n , sharply, almost triangularly broadening
in t h e lower half. The line of the lower edge of t h e g a r m e n t is straight or wavy,
festoon-like tripartite, r o u n d e d off in t h e centre. On t h e lower half sometimes
folds are shown : t w o rows of horizontal arches closed on t h e axle line. 216 The
similarity of the s t a t u e t t e discussed, with this image is insignificant.
T h e same can be said also a b o u t t h e sculpture f r o m t h e K u s n a t i m e s
belonging to the M a t h u r a school. The belt on t h e s t a t u e of K a n i s k a from M a t h u -
ra 217 is decorated with plaques. I t surrounds t h e u n d e r - g a r m e n t , on which t h e
long, open k a f t a n is s p r e a d . Under t h e belt arch-shaped folds are indicated with
engraving, turned d o w n w a r d s . The lower edge is s t r a i g h t . T h e lower edge is
similar also on the s t a t u e of the K u s n a prince f r o m Mathura, 2 1 8 where a tassel
is h a n g i n g down f r o m t h e broad belt on t h e axle line. The s t r a p fastened in t h e
centre w i t h two tassels hanging down is rare in the a r t of M a t h u r a . Such fasten-
ing of t h e belt is found on earlier m o n u m e n t s of I n d i a n a r t , belonging to the 2nd
c e n t u r y . C. t h e beginning of our era, in B h a r h u t , B h a j a , Pitalhor, Sanchi,
etc. 2 1 9
Certain parallels are observed also with the P a r t h i a n c u l t u r e . T h e b e l t w i t h
t w o hanging tassels is u s u a l in P a r t h i a n a r t . B y way of e x a m p l e we can mention
t h e s t o n e reliefs f r o m B i s u t u n with t h e representation of a P a r t h i a n prince
(1st 2nd centuries A . D.), 2 2 0 where t h e line of the lower edge consists of t w o
arches closed in the c e n t r e , so t h a t between t h e m a t i p t u r n i n g downwards is
f o r m e d . Such character of t h e belt is established in P a r t h i a n a r t already in t h e
216
J . M . R O S E N F I E L D : The Dynastic A r t s of t h e Kushans. Los Angeles 1 9 6 7 .
PI. 11 (30, 38); P. G A R D N E R : The Coins of t h e Greek a n d Scythic Kings in Bactria
a n d I n d i a in the British Museum. Chicago 1959. (reprint) Pis. X X V I X X V I I .
2
I ' J . M . R O S E N F I E E D : op. cit. F i g . 2, 94.
218
Ibidem. Figs. 8, 13, 21, 22.
e
219
A . M. L O T H : L a vie publique et prive dans l'Inde ancienne II sicle a v a n t
e
J.-C. VII sicle environ. Fase. V I I . Planches. Paris 1979. (Publications du Muse
G u i m e t . Recherches et d o c u m e n t s d ' a r t et d'archologie. VI) Pis. I I I (1 4), IV (1, 2),
V I I (1).
220
R . G H I R S H M A N : I r a n . P a r t h i a n s a n d Sasanians. Paris 1 9 6 2 . Fig. 6 6 . Such
fixing of the belt is found in the women's costume of the P a r t h i a n s . Ibidem, Fig. 103.

1st c e n t u r y B. C. (reliefs from Nimrud-Dag). 2 2 1 This fashion was s p r e a d also

in P a l m y r a , a t a n y r a t e in the 2nd1st centuries B. C.222 I t s sources, a p p a r e n t -
ly, are still in Achemenid art, especially in its works, reflecting the direct in-
fluence of Greek art. 2 2 3
T h e s t a t u e t t e of Takht-i Sangin, on t h e basis of its proportions a n d , p a r t l y ,
its g a r m e n t , is p r o b a b l y nearest t o t h e statue of the s t a n d i n g aristocrat f r o m
Dalverzin-Tepe. 2 2 4 A certain exchange of experiences is discovered also f r o m
t h e sculpture of t h e sitting ruler f r o m Khalchayan. 2 2 5 However, also these a n a l -
ogies h a v e a very a p p r o x i m a t e a n d p a r t i a l character. On t h e basis of compari-
son w i t h them, it can be confirmed a t a certain degree of s u r e t y t h a t t h i s s t a t -
u e t t e belongs t o t h e circle of t h e works of B a c t r i a n a r t . I t s dating, h o w e v e r ,
contains significant complications. A t t h e present phase of research we d a t e it
within t h e broad limits, from t h e 3rd c e n t u r y B. C. t o t h e beginning of o u r e r a .
If one also t a k e s i n t o account t h e resemblance to the ancient sculpture f r o m
Asia Minor t h e n t h e d a t i n g to the f i r s t half of this period seems t o be preferable.
Thus, in cultic deposit No. 4, besides t h e great variety in the m a t e r i a l of
t h e works of art, and t h e richness of cultural representational sources, we get
t h e widest chronological range of t h e objects cast together, viz. from t h e 5th
c e n t u r y B. C. t h r o u g h the 1st c e n t u r y A. D. The m o n u m e n t s of a r t are in an
excellent condition of preservation (with t h e exception of t h e various a l a b a s t e r
sculpture t r a n s p o r t e d in a broken condition), they have been preserved really
in m u s e u m condition. The questions raised by this complex are so e x t e n s i v e
t h a t all the m o n u m e n t s require a special edition, each of which is u n i q u e in
this territory, and a whole series also in world culture.
The great significance of t h e treasure of Takht-i Sangin, p a r t i c u l a r l y , on
t h e basis of the results of cultic deposit No. 4, consists in t h e fact t h a t t h e y con-
t a i n t h e dynamics of development of t h e a r t of Bactria, u n k n o w n so f a r , of t h e
second half of t h e millennium B . C., f r o m the Achaemenid style t h r o u g h t h e
Graeco-Bactrian t r a d i t i o n to the development of the K u s n a schools of a r t .
The d a t e of the deposit, on the basis of t h e last find, is in t h e neighbourhood of
t h e beginning of our era.
One of the m o s t significant f i n d s in the temple was t h e discovery of t h e
s h e a t h of a miniature votive m a k h a i r a embellished with t h e relief r e p r e s e n t a -
t i o n of Alexander of Macedonia depicted as Heracles. The head of Alexander is
221
Ibidem, Figs 79 80. See also B. GOLDMAN : P a r t h i a n s a t Gandhra. E W 28
(1978) Fig. 6.
222
R . GHIRSHMAN op. cit F i g s . 84, 91.
223
A. U. POPE : A Survey of Persian A r t . VI. Teheran LondonNew York 1967.
( R e p r i n t ) . PI. 108 (A C). Especially: A. SHAPUR SHAHBAZI : T h e Irano-Lycian Monu-
m e n t s . Teheran 1976. (Institute of Achaemenid Research Publications. II) 138, PI.
LXXVIII.
224
G. A. PUGACHENKOVA: . Moscow 1979. Pis.
129130.
225
G. A. PUGACHENKOVA: . Moscow 1971. 6768, P l . 68.

68 . A. L I T V I N S K I Y - I . . P I C H I K Y A N
r e p r e s e n t e d in f r o n t - v i e w slightly t u r n e d t o t h e left and hardly perceptibly

l e a n e d t o the left s h o u l d e r . The look of t h e widely opened eyes without t h e
d r a w i n g of the pupils is directed above t h e onlooker. The r o u n d oval of the face
w i t h t h e chin drawn i n t o it is contracted w i t h a helmet representing t h e h e a d
of a s c a l p e d lion. The u p p e r j a w of the lion g a r l a n d s the forehead, while t h e t w o
h a l v e s of the lower j a w f r a m e the cheeks of t h e hero in t h e form of face-guards
of t h e helmet. The s t r e a m i n g , thinly modelled twirls-locks of the mane are
b e h i n d t h e head of A l e x a n d e r , and on his chest there are the legs of the lion,
f a s t e n e d in the knot of Heracles (one of t h e legs and a p a r t of t h e hide on t h e
l e f t shoulder have b e e n preserved, the right leg a n d the k n o t have been lost).
B e c a u s e of abrasion, t h e nose and the lips are somewhat blurred (Fig. ).228
T h e contour of t h e face of the Alexander of Takht-i Sangin is oval, unlike
t h e r o u n d types of h e a d s sculptured b y S c o p a s . T h e wide open eyes are set close
t o e a c h other and n o t v e r y deep, b u t the massive muscles of t h e eyebrows a t t h e
ridge of the nose are p r o j e c t i n g ahead, n o t hanging above t h e eyes t h a t still
gives t h e expression of p a t h o s to the face, characteristic of all portraits of
A l e x a n d e r , showing t h e i r a f f i n i t y and u n i t i n g t h e m on t h e basis of this one
m a r k . I t is also noted t h a t t h e furrows on t h e forehead and on the ridge of the
nose a n d the squinting of t h e eyes, characteristic of the late Hellenistic period,
a r e missing.
As was correctly r e m a r k e d by K . G e b a u e r , t h e question of the a t t r i b u t i o n
of e a c h portrait of A l e x a n d e r , has the only perspective of solution, if we examine
it f r o m t h e view-point of t h e whole existing material. 2 2 7 Alexander the G r e a t
w a s represented b y t h e b e s t known sculptors of Greece, viz. Lysippus, Leo-
c h a r e s , E u t h y crates, E u p h a n o r . At present t h e iconography of the portraits of
A l e x a n d e r is well elaborated, 2 2 8 and t h e l i t e r a t u r e on t h e a t t r i b u t i o n of his
226
The fragment w a s glued together by I . I t . PICHIKYAN of several vertically
s t r a t i f i e d scales of ivory, f o u n d in different p a r t s of t h e iron conglomerate. I n v . No.
1134/1091, Institute of H i s t o r y A. Donish of t h e A c a d e m y of Sciences of the Tajik SSR.
T h e conglomerate consisting of over 2000 arrow-heads a n d other iron objects of arma-
m e n t , was recorded by u s a s cultic deposit No. 2 in t h e most distant north-western corner
of t h e 20 metres long a m b u l a t o r y corridor (No. 2). Deposit No. 2 was discovered in a
d e p t h of 3.5 metres f r o m t h e present-day surface, 0.5 0.65 metres above the real sub-
soil. O n t h e glued f r a g m e n t of bouterolle the dimensions of the p o r t r a i t bust of Alexander
a r e a s follows : height : 33 m m , breadth : 28 m m . T h e representation of the head has
b e e n preserved completely, viz. height : 20 m m , eyes : 13 m m , nose : 3 mm. On t h e
o p p o s i t e side of the half of t h e bouterolle there a r e traces of a slit of a vertical incision
u s u a l for the functional a n d votive-decorative m o u t h of sheaths of daggers, found in
large quantities in t h e cultic deposits of the temple. See I. R . PICHIKYAN :
. CA 4 (1980) 202 212.
227
. GEBAUER: A l e x a n d e r b i l d n i s u n d A l e x a n d e r t y p u s . A M 6 3 / 6 4 (1939/9) 32.
228
The works of sculptors of the 4th c e n t u r y . C. are reviewed in detail in the
w o r k s of the eminent Soviet researcher O. F . Valdgauer. See O. F . VALDGAUER : -
IV . . . Idem :
. Leningrad 1938. 150 foil. . SCHREIBER: Stu-
dien b e r das Bildnis A l e x a n d e r s des Grossen. Leipzig 1903 ; W. HOFFMANN : Das lite-
r a r i s c h e P o r t r t Alexanders des Grossen im griechischen und rmischen Altertum. Leip-
zig 1907 ; F. WEBER : A l e x a n d e r der Grosse im Urteil der Griechen und Rmor bis an
Acta Antiqua Academiae Scientiurum Hungarirae 28, 1980

p o r t r a i t s is incredibly comprehensive, which facilitates our work considerably.

I n a s m u c h as t h e attribution of t h e p o r t r a i t a t T a k h t - i Sangin does n o t give
complications, therefore besides t h e analogies it is necessary t o establish its
place in iconographie typology.
The successful classification according t o the creative m a s t e r s has been
done b y O. F . Valdgauer. His work could be continued with a broadening classi-
f i c a t i o n according t o schools, because t h e authorship of t h e m a j o r i t y of t h e in-
definable portraits of Alexander is impossible t o establish. I n t h i s plan, in a
special monographic research, all k n o w n works should be grouped according t o
t h e m e t h o d of J . Beazley, applied b y him in t h e s t u d y of A t t i c vase p a i n t i n g ,
and n a m e d the masters a f t e r t h e n a m e s of t h e most characteristic portraits, 2 2 9
b u t t h i s is n o t a t all our t a s k .
T h e classification according t o schools of sculpture in t h e geographic r a n g e
has b e e n m a d e by M. Bieber and K . Gebauer, b u t their works do n o t excel w i t h
a comprehensive fullness. 2 3 0 The excellent work of K . G e b a u e r arranges t h e
coinages and sculptures with t h e representations of Heracles and Alexander
k n o w n t o t h e author, according t o regional schools. I n his opinion t h e s c u l p t u r a l
works of t h e Alexandrian school collected b y him, can be set against a l t o g e t h e r
only 12 marble works of the whole r e m a i n i n g Mediterranean. 2 3 1 However, it is
not so. I n the last years t h e archeological journals have published new a t t r i b u -
tions t h a t are completed on t h e basis of t h e s t u d y of museum and p r i v a t e col-
lections, 232 and also b y archaeological excavations, which can be testified b y
t h e excavations in Aigai, 233 and in T a k h t - i Sangin. The classification with t h e
selection of the R o m a n idealized p o r t r a i t a n d the so called group of R o m a n his-
torical p o r t r a i t s has earned already justification in the iconography of Alexan-
der t h e Great. 2 3 4 There are also typological groups, based on t h e m o s t diversi-
fied m a r k s , including t h e emotional characteristics. Thus, 0 . Y a . Neverov, clas-
sifying t h e deification of the creator of t h e v a s t Hellenistic m o n a r c h y according
t o t h e phases of the apotheosis of t h e ruler (what in itself contains a v e r y sensi-
die Konstantinische Zeit. Giessen 1909; . GEBAUER: op. cit.; V. v. GRAVE: Z u m

Herrscher Philippos I I . u n d Philippos I I I . von Makedonien. AA 3 (1973) 236 249, Fige
, 6 ; M. BIEBER : E i n idealisiertes P o r t r t Alexanders des Grossen. J D A I 40 (1925)
167; M. BIEBER: The Portraits of Alexander t h e Great. P P S 93 (1949) 349; M. BIE-
BER : T h e Sculpture of t h e Hellenistic Age. New York 1966 ; G. KLEINER : D a s Bild-
nis Alexanders des Grossen. J D A I . 66/66 (1960/1961) 206 ff.; G. BERTELLI : Alexand-
res I I I di Macedonia. Enciclopedia I t a l i a n a F o n d a t a di Giovanni Treeccani. I . R o m e
1957.
229
J . D. BEAZLEY : Attic Black-Figure Vase-Painters. Oxford 1963.
230
M . B I E B E R : op. cit. ; . G E B A U E R : op. cit.
231
. G E B A U E R : op. cit. 3 3 .
232
M. A. ZAGDAUN : Collection P a u l Canellopoulos. , I I I , 1979. 411, Figs.
29-32; H . G. NIMEYER : A l e x a n d e r k o p f i n S e v i l l a . A A 1978. I . 106 116.
233
M. ANDRONIKOS : .
, .
7 (1979) 3 foil. .
234
M . A . Z A G D A U N : op. cit.; H . G . N I M E Y E R : op. cit.

ble idea, b u t not a new one), 235 sees the evolution also in the style of the p o r t r a i t
of Alexander. This is w h a t leads 0 . Ya. N e v e r o v t o t h e typology based on emo-
t i o n a l principles. 236 There are classifications t h a t are a r b i t r a r y , difficult t o de-
f i n e , based on t y p o l o g y of principle u n k n o w n t o us. 237
The classification according to kinds a n d genres of a r t was realized only
b y numismatists, selecting f r o m the whole complex of representations, t h e re-
p r e s e n t a t i o n s in profile on the coins.
As a m a t t e r of f a c t , there is no generalizing iconographie typology of t h e
i m a g e of Alexander in representational a r t , since each researcher s t a r t e d o u t
f r o m portraits, in t h e course of time increasing considerably, which were a t
t h e i r disposal, or which were more or less close t o t h e m , or within their reach.
T h i s is also u n d e r s t a n d a b l e . The diversity of t h e material, the chronological
differences, the stylistic characteristics, t h e n o n - a d e q u a c y of the age and the
s u b j e c t , give complications for a single, methodologically reliable typology of
t h e representations of Alexander.
We propose t h e typological characteristic of the representations of Alex-
a n d e r the Great according t o s t a t u a r y subjects, in which all antique representa-
t i o n s would be included, regardless of t i m e , place of preparation, material, a n d
t h e a u t h o r creating t h e p o r t r a i t of the g r e a t s t r a t e g i s t . Following this m e t h o d ,
t h e same works of a n c i e n t sculptors-portraitists m a y be included in different
typological groups, viz.:
1. Alexander, t h e child.
2. Alexander, t h e palaestric, warrior.
3. Alexander, t h e philosopher, t h e s t a t e s m a n .
4. Alexander, t h e deified hero, or god.
5. Alexander, an idealized image.
I n each typological group it is proposed t o follow t h e chronological divi-

sion f r o m t h e p o r t r a i t s prepared during his lifetime t o t h e Hellenistic copies,
a n d f r o m the l a t t e r t o t h e R o m a n historical p o r t r a i t s . A f t e r t h e m o n u m e n t a l
s c u l p t u r e , i f i t exists, in each group t h e small sculpture, glyptics and coin re-
p r e s e n t a t i o n s follow, consequently, t h e m a t e r i a l of t h e sculptural works in
m a r b l e , limestone, a n d bone precedes t h e copies in t e r r a c o t t a and m e t a l s (be-
c a u s e original bronze masterpieces are u n k n o w n so far).
According t o t h e idea, t h e first group of t h e p o r t r a i t s f r o m the early y o u t h
s h o u l d be the earliest. As a m a t t e r of f a c t , a g r e a t p a r t of this group belonged
235
K . G E B A U E R : op. cit. 2.
238
. YA. NEVEROV : . VI. CA 2
(1971) 88 foil.
237
. S C H R E I B E R : op. cit. ; G . K L E I N E R : op. cit.

among the R o m a n copies t h a t is t h e y were executed in the R o m a n style of

portraiture. 2 3 8
The second group is Alexander, t h e palaestric, or t h e nude Alexander w i t h
spear (doryphores), going back to t h e praised statue by Lysippus known f r o m
w r i t t e n sources. 239 However, in t h e copies the figure of Alexander is altered b y
t h e too free conception (sometimes in reversed, reflected representation). Two
groups of similar statues are known, made of bronze and of marble. 2 4 0 Their char-
acteristics are : the polycletic presentation of the figure depending on the degree
of qualities of the copyist (chiasmus and motion in statics), heroic naked-
ness of t h e young athletic body, and a very generalized conception of the face
t h a t permits us to include them in t h e group of warriors. Alexander, the warrior
was represented in armour, in helmet, sometimes with shield (Alexander on
horseback in full military equipment). However, the main difference of all re-
presentations of Alexander the warrior is the emotional s a t u r a t i o n of the char-
acteristic of the image of the hero, or his idealized presentation. All copies are
very different in character, executed in different material, and as a whole t h e y
strongly fall behind the originals, prepared by sculptors from nature. W i t h t h e
exception of the mosaic of Pompeii, 2 4 1 their majority consist of considerably
reduced replicas of t h e kind of t h e s t a t u e t t e Alexander on horseback, 242 f r o m
Herculaneum not permitting a judgement of similarity with original as a result
of t h e generalization of the face. Very likely, the copy of t h e p o r t r a i t by Lysip-
pus, perhaps, a mounted statue, going back to the gold medallion of Abukir, 2 4 3
is t h e best specimen of this group.
The third group is Alexander, t h e philosopher. The degree of surety in t h e
a t t r i b u t i o n of these representations of Alexander is different in antique history
of a r t . There is no doubt t h a t t h e signed herma in the Louvre (collection of
Azar) represents Alexander the Great. However, in the beginning it was a t t r i -
b u t e d by L. Curtius to Leochares, 244 b y Th. Schreiber to Lysippus, 2 4 5 and t h e n ,
much more convincingly, it was re-attributed by 0 . F . Valdgauer to E u p h a -
nor. 246 I t has a close analogy in t h e British Museum, also erroneously included
among t h e copies from the sculpture of Lysippus. 247 The signed herma, accepted
a t present as the initial one, is valued very low, possibly undeservedly, on ac-
count of the bad preservation of t h e monument. M. Bieber writes straight t h a t
238
T H . S C H R E I B E R : op. cit. 4 1 , P l . 1 ( B ) ; . G E B A U E R : op. cit. P i s . 8, 12.
239
T H . S C H R E I B E R : op. cit. 2 1 7 , F i g . 2 7 .
240
. F . VALDGAUER : ., 165, 152, Figs 69, 62 ; . SCHREI-
BER : op. cit. Pis. VI, X I .
241
C . B E R T E L L I : op. cit. 240.
242
Ibidem. 357.
243
H . DELBRUECK: Zwoi P o r t r t s . J D A I . 40 (1925) 12, F i g . ; V. v. GRAVE:
op. cit. 2 5 4 , Fig. 5 .
244
H . G . N I M A Y E R : op. cit.
245
T H . S C H R E I B E R : op. cit. 2 8 4 0 , P l . 1 4 .
248
O. F . VALDGAUER : . 164165.
247
. G E B A U E R : op. c. 97, P l . 14 ( 3 ; 4).

72 . A. LITVINSKIY I. . PICHIKYAN
t h i s is a very b a d copy, 2 4 8 a n d 0 . F. Valdgauer points o u t t h e deformation of t h e

eyes, m a d e by t h e c o p y i s t . And still, this representation shows in a sufficient
degree n o t only t h e l i t e r a r y image of Alexander, t h e wise s t r a t e g i s t and states-
m a n , created by t h e a n t i q u e authors, b u t it also reflects t h e excellent charac-
t e r i z a t i o n b y 0 . F . Valdgauer : Alexander is t h e pupil of Aristotle, well versed
on t h e great Hellenic t e a c h e r s and poets. Alexander is a hero and great conquer-
or, realizer of t h e e x p e c t a t i o n s of Hellenism a b o u t t h e submission of the E a s t ,
t a l e n t e d with genial o r g a n i z a t o r y qualities. 249 As Th. Schreiber remarks, Alex-
a n d e r is represented o n t h e p o r t r a i t in m a t u r e age. W e should add t h a t in him
t h e wisdom of the s t a t e s m a n is stressed, Alexander, t h e philosopher should be
r a n g e d b y the p o r t r a i t s of t h e Louvre and t h e British Museum w i t h this t y p e of
iconography.
I n the e x a m i n a t i o n of the f o u r t h typological group t h e main a t t e n t i o n
will b e paid only to t h e appearance of Alexander in t h e image of Heracles, since
t h e r e s t of the h y p o s t a s e s , being not so wide-spread, a d d l i t t l e in comparison
w i t h t h e representation discussed here.
T h e f i f t h typological group, t h e largest one, consists mainly of copies
p r e s e n t i n g the idealized Alexander. 2 5 0 I n reality, Alexander was very handsome
(Arr., Anab. V I I , 26,1), b u t t h e sculptors of Hellenism, a f t e r his d e a t h endeav-
o u r e d t o invest him in addition also with f e a t u r e s of heavenly b e a u t y . J u d g e d
f r o m t h e number of t h e copies, this t r e n d of development stood nearest to t h e
t a s t e of t h e R o m a n age.
R e t u r n i n g t o t h e f o u r t h typological group, we r e m a r k t h a t t h e researchers
propose different n u m b e r s of specimens of t h e deification of Alexander. 0 . Y a .
N e v e r o v , touching t h i s question to some e x t e n t , mentions only t h e main icono-
g r a p h i e types of this g r o u p , viz. Alexander, in t h e image of Heracles, Dionysus,
A m m o n and Helios. 251 K . Gebauer gives a more general, b u t also incomplete
list in t h e form of a f o o t - n o t e without t h e illustration of t h e examples, viz.
Alexander, Helios, a Genius, a Dioscur, Ammon-Carneios, H e r m e s Diony-
sus, Heracles, Ares, Asclepius, River god, S a t y r , Attis. 2 5 2 Alexander-Zeus (Jupi-
ter) 2 5 3 is missing in b o t h cases, and Alexander in elephant h e l m e t , serving, ap-
p a r e n t l y , as a model of t h e representation of t h e Graeco-Bactrian coinage of
Demetrius, 2 5 4 is n o t m e n t i o n e d either.
248
M . BIEBER : E i n idealisiertes P o r t r t . . . 181.
249 V A L D G A U E R : . . . 1 6 5 .
250
. S C H R E I B E R : op. cit. 2 5 , P i s . 1 1 - D , I I I - F , I V , V , V I I ; T H . W I E G A N D :
E i n n e u e s A l e x a n d e r p o r t r t . J D A I 14 ( 1 8 9 9 ) 1 ; M . B I E B E R : E i n i d e a l i s i e r t e s P o r t r t . . .
F i g s . 1 6 , P i s . 6 8 ; V . v . G R A V E : op. cit. P l . 2 4 ( 9 1 2 ) ; . G . K L E I N E R : op. cit.
P i s . 1 4 , 5 , 7, 1 4 .
251
0 . J a . N E V E R O V : op. cit. 8 8 f o i l .
252
. G E B A U E R : op. cit. 1.
253
T H . S C H R E I B E R : op. cit. 1 1 3 , p l . V I I - O .
244
A. CUNNINGHAM: Relics f r o m A n c i e n t Persia in Gold, Silver a n d Copper.
J A S B 24 (18881) 172; B . STAWISKI : K u n s t d e r K u s c h a n . Leipzig 1979. Fig. 29.

L e t us s t o p a t t h e representation of Alexander in t h e image of H e r a c l e s .

T h e representations in profile of Heracles-Alexander and Alexander in lion
helmet d o m i n a t e over the s t a t u a r y r e p r e s e n t a t i o n s of p o r t r a i t s of other groups,
because t h e y are represented b y a v a s t m a t e r i a l of numismatics on n u m e r o u s
series of coins f r o m t h e time of Alexander, of t h e diadochoi, especially of Lysi-
machus, and t h e epigons. The comparison of t h e coin representations of H e r a c -
les, of Alexander in t h e image of Heracles, of Mithridates in t h e image of Alex-
ander-Heracles a n d Commodus in t h e image of Heracles, shows the evolution of
t h e represented image. Originally Heracles was represented on the coins w i t h
t h e apotrope, t h e lion's hide, reliable for a t t r i b u t i o n , i.e., t h e head of Heracles,
represented in profile, was garlanded with a lion's mask. Similar r e p r e s e n t a t i o n s
of Heracles in profile are widely known on t h e coins of t h e 5th4th centuries,
on t h e gold s t a t e r f r o m Tarentum, 2 5 5 on t h e t e t r a d r a c h m a s of Thasos, 256 on t h e
double s t a t e r s of Dicaios (Thrace). 257
Heracles on t h e t e t r a d r a c h m a s of Alexander personifies a manly w a r r i o r
skilled in fights (frequently with broken ridge of t h e nose, with h y p o t h e t i c a l l y
bulgings of t h e arches above t h e eyebrows and t h e cheeks, like a f t e r a p a n c r a -
t i u m j u s t finished). 2 5 8 I t is n o t for nothing t h a t all representations of b o x e r s
show a great similarity with his virile and s o m e w h a t tired and gloomy a p p e a r -
ance.
Alexander was t r a n s f o r m e d into Heracles still in his lifetime, n o t o n l y
on t h e coins t h r o u g h the mechanic changing of Heracles w i t h the p o r t r a i t of
Alexander in profile. The repeated and obviously not accidental mentioning of
Alexander's a f f i n i t y with Heracles by t h e authors, a b o u t his endeavour t o
ascend, and perform a heroic deed, w o r t h y of his legendary ancestor, or t o sur-
pass him, including not only t h e a t t a c k i n g of insurmountable obstacles a n d
t a k i n g u p f i g h t alone with numerous enemies, b u t also t h e direct single f i g h t
w i t h a lion, m e n t i o n e d above (Curt., V I I I , 11 14). The complicated process of
t r a n s f o r m a t i o n of Alexander on t h e coins is very accurately described b y K .
Gebauer : On t h e t e t r a d r a c h m a s of Alexander K . Gebauer writes t h e
observer, a t desire, can recognize Alexander in t h e image of Heracles, because
Alexander personally assimilated himself t o this god.259
However, t h e m a j o r i t y of t h e researchers, including also K . Gebauer, see
t h e decisive m o m e n t in t h e iconography of Alexander simultaneously on all
coins exactly in 327 B. C. This is shown b y t h e s t u d y of t h e coinage of A m p h i -
253
K . L A N G E : Gttor Griechenlands. Meisterwerke Antiken Mnzkunst. Berlin
(without year). Fig. 04.
238
Ibidem. Fig. 61.
^ Ibidem. Fig. 60.
238
Monnaies. Grecques-Romaines. Byzantines-Gauloises, Franaises-trangres.
Drouet rive Gauche-Salle, n. 6. Astole-France, 75007. Paris 2122 juin 1979. N. 31 39.
239
K . G E B A U E R : op. cit. 2 , 1 1 , 1 2 .

74 . A. L I T V T N S K I Y I . . P I C H I K Y A N
polis, Tarsus, Sidon, Pella, Tyre, Babylon, Alexandria, and other smaller series.
I n t h e opinion of K . Gebauer : Alexander t h e n arrived a t t h e recognition of t h e
m a n o f t h a t time as a s u p r e m e power, on which the fate of e v e r y t h i n g depends.
T h e legend of his fabulous results in I n d i a r e a c h e d the W e s t in this year ; a t t h e
s a m e t i m e r u m o u r s were spread t h a t he himself appeared sometimes in t h e form
of Heracles. Therefore, t h e engraver m a y h a v e ventured t o individualize t h e
h e a d of Heracles, as t h e p o r t r a i t of Alexander. Since then not a single represen-
t a t i o n of Heracles exists in a strict sense. I f t h e t y p e is somewhere i n t e r r u p t e d ,
t h e n t h e new face, a t least in some m i n u t e difference, always appears since
then. 2 6 0 E v e n if the concepts of Nietzsche m i x with naivety in t h e j u d g m e n t s of
K . G e b a u e r , t h e m a i n t r e n d of development in the iconography of Alexander
is p r e s e n t e d correctly. A t the m o m e n t we are interested in the d a t e of t h e
decisive moment of this evolutionary series. M. Bieber showed an interesting
scheme of t h e t r a n s f o r m a t i o n of Alexander i n t o Heracles on t h e coins. Cos, as
well as other city-states of t h e disintegrating great empire, between t h e years
300 a n d 200, minted t h e idealized head of Alexander in t h e image of Heracles,
i n s t e a d of t h e bearded Heracles. 2 6 1 I t can be presumed a t a high degree of pro-
b a b i l i t y t h a t in t h e f o r m a t i o n of the i c o n o g r a p h y of Alexander in t h e image of
Heracles, not t h e coin representations themselves played a significant role, b u t
t h e m o n u m e n t a l sculptures, one of whose authors was, m o s t probably, t h e
f a v o u r i t e courtsculptor, Lysippus. The s c u l p t u r e could come into existence soon
a f t e r t h e successful single fight with t h e lion, when Alexander was not f a r f r o m
M a r a c a n d a , pushing aside Lysimachus, s t a n d i n g behind t h e king, e n t e r e d into
single f i g h t and slained t h e lion of exceptional size, similarly t o Heracles, with
one stroke. 2 6 2 The s t a t u a r y representations of Alexander in t h e image of Heracles
are convincingly shown b y the originals of t h e period of Hellenism, 2 6 3 and b y his
r e p r e s e n t a t i o n in lion h e l m e t on the sarcophagus of Alexander. 2 6 4 B u t t h e diver-
sity itself of the coin representations shows t h a t various s t a t u a r y specimens were
utilized. 2 6 5 The mentioned spreading of t h e representations of Alexander on coins
of his numerous successors, diadochoi a n d epigons, was brought a b o u t by t h e f a c t
t h a t , according t o t h e very accurate r e m a r k of 0 . Y a . Neverov, Alexander be-
c a m e the necessary component of d y n a s t i c cult, which was formed in t h e em-
pires of t h e Ptolemies, t h e Seleucids, a n d t h e Attalids. The image of Alexander
t h e G r e a t surrounded with mystic aureola furnished so t o say t h e visibility of
legitimacy t o their power. 266
280
Ibidem. 11 12.
281
M . B I E B E R : E i n idealisiertes P o r t r t . . . 1 7 8 , Fig. 1 0 .
282
Q. C U R T I R U F I : H i s t . Alex., V I I I , I , 11 17. Cf. above p. 38, . 59.
283
. B U S C H O R : D a s hellenistische Bildnis. Mnchen 1 9 4 9 . Fig. 2 0 .
284
E . A K U R G A L : Ancient Civilizations a n d R u i n s of Turkey. Istanbul 1969. Pl. 23.
265 y . N E W E L L : Alexander Hoards. N u m i s m a t i c Notes a n d Monographs. N 19.
New York 1923. Pis. I - V I I I .
288
O . Y A . N E V E R O V : op. cit. 88 foil.
Acta Antiqua Aeademiae Scientiarum Hunqaricae 28, 1980

M O N U M E N T S O F A R T F R O M T H E SANCTUARY O F O X U S 75
Parallel with the general t r e n d of development of t h e r e p r e s e n t a t i o n s of

Alexander, an additional masked t r e n d of the images of Alexander-Heracles
and Alexander-Ammon appears a l r e a d y on the coins of t h e diadochoi. T h e Se-
leucids and t h e Ptolemies borrow f r o m t h e i r predecessor t h e image of A m m o n ,
and within two hundred years a f t e r Alexander, the image of Alexander-Heracles
is revived again by another great s t r a t e g i s t of a n t i q u i t y , Mithridates E u p a t o r
and a f t e r nearly five centuries a l r e a d y in the genre of t h e farce b y t h e
lenient R o m a n emperor Commodus, for t h e reflection of t h e share of the clan of
the Antonini in t h e patronage of Heracles. Mithridates E u p a t o r borrowed t w o
s t a t u a r y t y p e s of Alexander the G r e a t . His predecessor was Lysimachus. 2 6 7
The circumstance is interesting t h a t n o t only coin representations o f C o m m o d u s
have been preserved, b u t also his s c u l p t u r e d p o r t r a i t in lion's hide, with t h e club
of Heracles a n d the apples of the Hesperids in his hands. However, t h e image
of the emperor, created by the court sculptors entirely corresponds to t h e j u d g -
ment of L a m p i d i u s , viz. the expression of t h e face is dull, like t h a t of a d r u n k -
ard, and his speech is indolent. 268 T h e coins o f C o m m o d u s (182 192) icono-
graphically r e p e a t his monumental s c u l p t u r a l prototype, just like the coins of
Alexander. On t h e m Commodus is represented in left profile, the face with t h i n
aristocratic nose, small m o u t h and m o d e s t l y cast down eyes, reflects i n n e r
apathy 2 6 9 t h a t sharply distinguishes him f r o m Alexander.
I n addition to t h e iconographie analysis according t o t h e similarity of t h e
representations, for t h e identification of t h e portrait f r o m T a k h t - i Sangin, t h e
logical m e t h o d of exclusion can also be applied. Of the four representations in
lion helmet, considering the site of discovery, only Alexander remains as a real
competitor. This is shown by the canonical character of t h e idealization of t h e
appearance, b y t h e open juvenile face, n o t characteristic of Heracles. Mithri-
d a t e s and Commodus do not f i t in, because t h e time of m a n u f a c t u r e of t h e
m a k h a i r a is earlier t h a n their a p p e a r a n c e in t h e historical arena. 2 7 0
According t o t h e general similarity of t h e representations, of the idealizing
tendency of t h e 3rd century B. C., t h e p o r t r a i t from T a k h t - i Sangin of Alexan-
2.7
Ibidem.
2.8
Monnaies . . . N 6 4 ; N . N . B R I T O V A N . M. L O S E V A N . A. SIDOROVA :
. . Moscow 1975. 67, Fig. 112.
2.9
. L A N G E : op. cit. Fig. 141.
270
I n fact our attribution of t h e relief representation of Alexander the G r e a t in
the image of Heracles did not raise d o u b t s in a n y of the Soviet. A f t e r the lecture o n t h e
antique treasures of Takht-i-Sangin in 1980 in Moscow ( , -
, . . . ), in Leningrad (. ,
) and in Dushanbe ( . . . ) the speakers
confirmed t h e opinion on the doubtlossness of t h e attribution. We wish to e x t e n d o u r
sincere g r a t i t u d e to O. Y A . N E V E R O V , to Y T J . E . C H I S T Y A K O V and t o V. D. B L A V A T S K I Y
for t h e valuable advices. N . A. F R O L O V A a n d A. N I K I T E N extended help in the work
on the numismatic collection of the S t a t e Historical Museum (Gosudarstvenniy istori-
cheskiy Muzey GIM, Moscow), kindly furnishing photographs of the coins with t h e
representation of Alexander in lion helmet.
Acta Antiqua Academiae Scientiarum Hungar'ae 21, 1980

der is similar not only to the images of t h e strategist on t h e coins, but it is also
close to the sculptures of the school of Alexandria and Cos of the Hellenistic
period, 2 7 1 partly reaching to the copies of t h e R o m a n times. 272
T h e juvenile p o r t r a i t of Alexander f r o m the museum of the Acropolis of
A t h e n s can serve as a good analogy of Alexander from Takht-i Sangin. There is
a similarity in t h e s o f t appearance of t h e eyes, in the butterfly-like from of t h e
lips, 273 in the round oval of the face with t h e chin drawn into it. The Athenian
p o r t r a i t is counted t o t h e copies from t h e work of Leochares t h a t was conse-
c r a t e d after the b a t t l e of Chaironeia (338 . C.) in the Philippeion. V. G r a v e
writes t h a t it looks like a Greek original of t h e Late-Hellenistic period. 274
As it appears, t h e differentiation according to the material cannot be a t
all determinative, because the sculpture m a d e of ivory, on account of t h e f r a -
gility of the material, is preserved under v e r y favourable circumstances. T h e
known sculptures of t h e Hellenistic times 2 7 5 have recently been completed b y
t h e p o r t r a i t heads f r o m t h e presumed t o m b of Philip of Macedonia in Aigai,
one of which represents Alexander. 270 The a t t r i b u t i o n of t h e portrait head f r o m
Aigai is reliable, viz. t h e sculptor representing the young Alexander with t h e
head lifted up, presented t h e whole gallery of the similar iconography of t h e
Hellenistic period a n d t h e Roman imitations. The marble head from the Alex-
andrian museum No. 3405, dated to t h e 3rd century . C., possibly, really of
t h e circle of Lysippus, 2 7 7 in nearest according to the artistic characteristics a n d
conception to t h e representation of Alexander from Aigai.
The heads f r o m t h e so called sarcophagus of Alexander of Macedonia,
representing Alexander in the scene of f i g h t i n g and in t h e scene of hunting, are
very close to the p o r t r a i t from Takht-i Sangin. I n our opinion, these representa-
tions are unanimously a t t r i b u t e d by all researchers to Alexander with justifica-
tion. 278 The bending of t h e heads with t h e forehead ahead add dramatism a n d
a f i r m will to victory to these, images t h a t is seldom found in the iconography
of t h e strategist, whose head is usually represented slightly lifted up and slightly
t u r n e d to the left, as on the gigantic, f a m o u s replica f r o m Pergamum, 2 7 9 a n d
on other representations including also t h e p o r t r a i t from Takht-i Sangin. T h e
Heracles of P e r g a m u m is also very close t o this type, personified by G. K r a h -
mer as Mithridates. T h e identification of t h i s statue by G. Krahmer with t h e
2,1
K. G E B A U E R : op. cit. 35, 95, Pis. 6 (1,2), 9.
272
Ibidem. 48, 63, Pis. 14, 15.
273
B . A S H M O L E : Demeter of Cnidus. J H S . 7 1 ( 1 9 5 1 ) 1 3 - 2 8 .
274
V. v. G R A V E : op. cit. 246, 249, Figs. 5, 6.
275
M . B I E B E R : T h e Sculpture . . .
278
M . A N D R O N I K O S : op. cit. Photograph on t h e cover of the journal.
277
K . G E B A U E R : op. c. 3 5 , P I . 6 ( 1 , 2 ) .
273
M. BIEBER: E i n idealisiertes P o r t r t . . . Figs. 7, 8 ; G . M . A. RICHTER:
Sculptures and Sculptors of t h e Greeks. New H a v e n 1 9 3 0 . 7 6 , Fig. 1 7 6 ; E. AKURGAL :
op. cit. Pl. 2 3 .
273
M . B I E B E R : E i n idealisiertos P o r t r t . . . Fig. 11.

M O N U M E N T S O F A R T F R O M T H E SA N C T U A R Y O F O X U S 77
image of Mithridates E u p a t o r seems t o be absolutely inconsistent. 2 8 0 Heracles

is considerably more similar t o Alexander f r o m t h e sarcophagus of Alexander
(both are in lion helmet, t h e expression and the similarity of t h e faces are strik-
ing), and even more to the gigantic head of Alexander f r o m P e r g a m u m , as-
cribed t o Lysippus, 2 8 1 t h a n to t h e p o r t r a i t of Mithridates E u p a t o r , and even to
those of A t t a l u s I I or A t t a l u s I I I , as it was t h o u g h t b y t h e great specialists in
a n t i q u i t y , W i n t e r and L. Curtius. 2 8 2 W e have to do here r a t h e r directly w i t h t h e
influence of t h e portraits of Alexander on Hellenistic sculpture. I n a s m u c h as
t h e P e r g a m u m Heracles from t h e n o r t h e r n hall of t h e s a n c t u a r y of A t h e n a is
connected with t h e sculptural g r o u p of Prometheus, we cannot m a i n t a i n t h e
a t t r i b u t i o n of this sculpture t o Alexander, however, t h e unusual similarity
m u s t be s t a t e d . The gigantic head f r o m P e r g a m u m is connected with t h e P e r g a -
m u m Heracles not only by t h e common P e r g a m u m school, b u t also by t h e ab-
solute similarity of the general appearance, as well as of t h e certain f e a t u r e s of
t h e face, viz. t h e representation of t h e forehead, t h e nose, t h e lips, t h e d e p o r t -
m e n t of t h e head, and also t h e presentation of the juvenile appearance, a n d t h e
presence of t h e predatory, lion-like pathos, characteristic, as K . Gebauer re-
m a r k s , of a large number of original marble heads, which can be c o m p a r e d
r a t h e r on the basis of t h e presence of pathos, t h a n on t h e basis of o u t w a r d simi-
larity. 283
T h e p o r t r a i t f r o m T a k h t - i Sangin of Alexander t h e G r e a t unites a whole
series of works both sculptural a n d m i n t e d on coins, where Alexander is repre-
sented in the image of Heracles in lion helmet. Inasmuch as in t h e p o r t r a i t f r o m
T a k h t - i Sangin of Alexander obviously an unknown original is copied, which
became t h e p r o t o t y p e for numerous replicas and copies of t h e 3rd c e n t u r y B . C . ,
a n d in t h e p o r t r a i t discussed idealizing features are felt, characteristic of early
Hellenism, we d a t e the representation of Alexander t o t h e broad f r a m e s of t h e
t h i r d c e n t u r y . T h e proposed d a t e is supported by the comparison with identical
coin mintings f r o m 300 200 B. C. This fully corresponds to t h e d a t i n g charac-
teristic of t h e m i n i a t u r e Greek v o t i v e makhaira, which t h e relief embellishes.
T h e conceptions mentioned f i r m l y p o i n t to t h e d a t e of t h e 3rd c e n t u r y B . C.
I t is difficult to overestimate t h e significance f r o m t h e view-point of t h e
history of culture, because of t h e high importance of t h e purely Greek p o r t r a i t ,
lifting t h e curtain of t h e enigmatic Graeco-Bactrian a r t . This highly artistic
work is t h e first Graeco-Bactrian p o r t r a i t of Alexander, representing t h e image
of t h e king and hero Iskander, robed in t h e victorious insignia, as he was pre-
served in t h e Central Asian popular traditions so far.
2H0
G. I Y R A H M E R : Eine E h r u n g f r Mithridates VI. E u p a t o r in P e r g a m u m . J D A I
40 (1925) 182, 2 0 0 - 2 0 5 , Figs 2, 7, 11, 12.
21,1
M. B I E B E R : Ein idealisiertes P o r t r t . . . Figs 7, 11.
282
L. C U R T I U S : Bronzestatuette eines Heracles in Neapel. Walter Amolung zum
sechzigsten Geburtstag. Leipzig 1928. 65, Fig. 7.
283
K . G E B A U E R : op. cit. 22.

Cultic deposit N o . 3, t o be found in t h e centre of t h e southern third of

corridor No. 2, with a d i a m e t e r of 1.5 metres, on a level of 1 m e t r e (10 cm. above
t h e real subsoil), consists mainly of c u t t i n g a n d t h r u s t i n g weapons, whole iron
a k i n a k a i , spear-heads, coverings of ivory s h e a t h s of Greek swords : xiphe and
m a k h a i r a i , of a gilded bronze face-guard of a Greek helmet, and of two plates
w i t h scenes of h u n t i n g f r o m t h e 2nd1st centuries . C. A b o u t these we have
t o s p e a k in detail, because these plates d a t e cultic deposit No. 3, as the latest
ones.
This arsenal e n t e r s t h e history of military affairs of t h e period of Hellen-
ism, a l t h o u g h certain single m o n u m e n t s are d a t e d t o t h e 5th c e n t u r y . C., and
o t h e r s are definitely l a t e r , f r o m t h e 2nd c e n t u r y . C., and even from t h e 1st
c e n t u r y B . C . B u t f i r s t of all let us stop a t c e r t a i n representations, because t h e y
are r a r e s t even for t h e considerably e l a b o r a t e d Greek mythology.
One handle of a xiphos is o r n a m e n t e d on b o t h sides, t h e outer and t h e
i n n e r one, with e x a c t l y identical high relief representations of two s t a t u a r y
g r o u p s , viz. Heracles, throwing down Silenus (Fig. 10 a, b), Heracles is standing
in f u l l height, his right h a n d holding a k n o t t e d club, lifted u p to administer a
s t r o k e ; with his l e f t h a n d he seized Silenus b y t h e long hair, and pulling him
d o w n , pressed him t o his thigh. Heracles with his leg p r e v e n t i n g Silenus f r o m
r u n n i n g away, s t e p p e d on his thigh and pressed him t o t h e e a r t h . Silenus,
s t a n d i n g on one k n e e in canonical pose of r u n w i t h b e n t knee, is full of desire t o
f r e e himself from t h e g r a s p of t h e mighty enemy. Resisting, with his left hand
he s t r i v e s to t a k e off t h e h a n d of Heracles, holding him b y t h e hair and causing
p a i n t o him. On Heracles t h e r e is a cloak m a d e of lion's hide, t h e paws of t h e
lion a r e fastened a t his neck in t h e Heraclean knot. Silenus, as always, is nude.
H i s a t t r i b u t i n g m a r k , t h e goat's ear, is clearly seen, directed upwards. B o t h
f i g u r e s are athletic, w i t h anatomically correct representation of t h e trained,
well developed and relief-like projecting muscles. The d y n a m i s m of the f i g h t is
r e n d e r e d by t h e m o t i o n of Heracles, lifting t h e club to a stroke, b y the position
of t h e legs of Silenus striving to rise, and t h e i r strong bodies, pressed t o g e t h e r
in f i g h t , consisting of bundles of muscles. On t h e basis of t h e style of representa-
t i o n , t h e handle of xiphos should be d a t e d t o a period f r o m t h e beginning of t h e
4 t h c e n t u r y to early Hellenism.
T h e third relief on t h e ivory s h e a t h o r n a m e n t i n g t h e bouterolle of t h e
G r e e k sword, represents a f a n t a s t i c creature, a female figure with a serpent's, or
fish-tail, with horse's legs and with bird's wings (Fig. 11). I n her l e f t hand she
holds a paddle vertically, and in her right h a n d a round o b j e c t , a shell, an apple
or a stone, ready t o be t h r o w n . The loose long locks s t r e a m in thick curls.
T h e face with t h e long s t r a i g h t nose and w i t h t h e high forehead, with big eyes,
w i t h carefully elaborated upper eyelid, with r o u n d brows rendered with incision,
a n d pupils indicated w i t h dots, is wilful a n d a t t h e same t i m e beautiful. A smile
h o v e r s about her lips. T h e round relief here a n d there goes over t o a sculptural

M O N U M E N T S O F A R T F R O M T H E SA N C T U A R Y O F O X U S 79
solution, a n d here and there, intentionally in half relief on t h e base of t h e

smoothly polished bone. This is d e t e r m i n e d b y t h e given round form of t h e
bouterolle, a n d also b y t h e fragility of t h e material. The b e n t and curled t a i l of
t h e m o n s t e r ends in a noose and fins, t h e wings are represented in a swing, t h e
horse's legs in r u n : t h e right one before t h e l e f t one. However, t h e d y n a m i s m
still is missing in the whole figure. T h e given planeness, characteristic of t h e
representation as a whole, can be observed in all p a r t s of t h e body. T h e i m p r o b -
able j u n c t i o n s of t h e various elements of t h e creature are rendered fairly suc-
cessfully. T h e details are very carefully elaborated, viz. the scales of t h e tail,
t h e f e a t h e r s of t h e wings, t h e hoofs, o r n a m e n t e d with incisions, all these are
rendered with a naive carefulness, b u t spiritlessly. On t h e other hand, t h e fe-
male torso and t h e horse's legs are e x e c u t e d masterfully, t h e perspective short-
ening of t h e complicated joints n o t existing in n a t u r e is rendered artistically.
The work b e t r a y s the h a n d of an experienced m a s t e r , b u t t h a t of a c r a f t s m a n .
T h e a t t r i b u t i o n of this personality can be twofold, although it is monosyl-
labic. I t is clear t h a t we h a v e to do with t h e n y m p h of t h e water element, close
to t h e Mediterranean Scylla or female hippocampus. The Scylla was r e p r e s e n t e d
with fish-tail and with wings, with t h e a t t r i b u t e paddle and with shell. I t is con-
nected with t h e representation from T a k h t - i Sangin also by t h e transition f r o m
t h e female torso to t h e fish-tail f r a m e d by palmettes. The energetic swing of t h e
wings, t h e loose locks and t h e aggressive swing of the right hand are also anal-
ogous. However, in t h e case of t h e Scylla, especially in t h e representation on t h e
cover of t h e mirror from t h e t u m u l u s of A r t y u k h o v , two dogs appear f r o m u n d e r
t h e curl, personifying t h e monstrosity and t h e f u r y of t h e Scylla. On t h e repre-
sentation f r o m T a k h t - i Sangin t h e y are missing. W e feel t h a t on t h e A r t y u k h o v
Scylla t h e r e were also horse's legs, p a r t of a leg with hoof has been p r e s e r v e d
under t h e figure, l e f t to us fragmentarily. 2 8 4 T h e second a t t r i b u t i o n , close t o t h e
first one, is similarly probable : this is t h e hippocampus, t h e n y m p h of t h e
water element, or as it is called otherwise, centaurotritonis. 2 8 5 In b o t h cases,
t h e Greek sea deity could be t r a n s f o r m e d into a n y m p h , t h e companion of t h e
god of river Oxus. The d r y , and a t t h e s a m e t i m e careful style of t h e work does
n o t p e r m i t its d a t i n g earlier t h a n t h e 2nd century, more accurately it should be
d a t e d t o t h e f i r s t half of t h e 2nd c e n t u r y B . C.
Together with other, purely Greek finds, t h e Greek weapon embellished
with representations, found in t h e d e p t h of Central Asia, far from t h e Medi-
t e r r a n e a n coasts, shows t h e fact of t h e Hellenization of t h e culture of B a c t r i a
and t h e significance of t h e antique elements in t h e synthesis of Graeco-Bactrian
art.
281
M . I . M A K S I M O V A : . Leningrad 1 9 7 9 . Fige on p p . 3 1 , 89.
285
B. S A U E R : Hippokamp. ML I (2) 26732677 ; S . K E I N A C H : Repertoire de la
statuaire Grecque et Romaino. I . Paris 1906. 95 68, 113, 429, 430.

80 . A. L I T V I N S K I Y - I . R . P I C H I K Y A N
T h e scenes of h u n t i n g on the walls of t h e casket made of ivory (21.6 b y

6.2 b y 0.7 cm), h a v e furnished very valuable representations, viz. d o c u m e n t a r y
p o r t r a i t s , costumes, a r m a m e n t , horse-furniture, species of horses, sitting of t h e
horsemen, t h e irresistible blow of t h e Y u e h - c h i h or early K u s n a warriors a n d
conquerors of t h e Graeco-Bactrian K i n g d o m (Fig. 12).
Two horsemen in full gallop (one t u r n i n g back) are shooting f r o m bows
on wild animals. T h e field is full of r u n n i n g , wounded and already killed ani-
mals. A she-panther, already wounded b y t w o arrows in t h e breast, defending
h e r cubs, tries t o j u m p on t h e horseman, t w o small p a n t h e r ' s cubs are beside
her. T h e hunters h a v e killed several m o u n t a i n r a m s with spirally curving horns,
hares and foxes are r u n n i n g away. T h e e n g r a v e r carefully and artistically en-
g r a v e d t h e smallest details. From t h e compositional point of view, t h e scene is
solved with great skill. The horsemen gallopping in t h e opposite directions, are
a r r a n g e d a t t h e edges, b u t this does n o t e u t t h e picture into two p a r t s , because
b o t h horsemen are shooting in one direction, t h e y are united also by t h e animal,
filling two p a r t s of t h e field between t h e horsemen. A t the same time, we h a v e
t o n o t e t h e s h a r p c o n t r a s t between t h e direct a n d vivid, very accurate rendering
of t h e m o v e m e n t s of t h e beasts and animals, especially the wonderful rendering
of t h e gallopping horses, and the figures of t h e horsemen, who are represented
as if t h e y were n o t s i t t i n g on horses f l y i n g in gallop, b u t in court thrones. I n this
t h e following of t h e canon showed itself, w h a t a t f i r s t was discovered on this
m o n u m e n t , and l a t e r can be followed on wall paintings and other representa-
t i o n s of early medieval Central Asian a r t . As a whole, d y n a m i s m is t h e m a i n
a r t i s t i c d o m i n a n t f e a t u r e of this composition. The scene of hunting is one of t h e
m o s t favourite m o t i f s of t h e art of t h e E a s t . I t developed and was diversified in
t h e a r t of P a r t h i a n a n d Sassanian I r a n .
The panels f r o m Takht-i Sangin (together with the p o r t r a i t and ethnic
similarity found in t h e m with the sculptures a n d representations of K u s n a
princes f r o m K h a l c h a y a n and Dalverzin) 2 8 6 are m a d e excellent works of a r t b y
t h e d o c u m e n t a r y character of their p o r t r a i t characteristics.
Two circumstances are of interest. I n T a k h t - i Sangin, a fortress p r o t e c t i n g
t h e ford t h r o u g h t h e A m u - D a r y a , which was neither a capital, nor a big city,
m o r e sheaths of daggers f r o m the times of Alexander the G r e a t were found, t h a n
in t h e whole Greek world. 287 Secondly, t h e fortress and temple have not only
n o t been destroyed b y t h e passing of t h e K u s n a s , b u t on t h e contrary, t h e y
continued to flourish, judged from t h e rich K u s n a , Indian and P a r t h i a n sacri-
ficial gifts a n d coins.
286
G. A . PUGACHENKOVA : . 1971. Pis. 43, 5 5 , 6 5 , 68, 78
112.
287
The list of t h e material known to us see : I. R. PICHIKYAN : . . .
Acta Antiqua Aeademiae Scientiarum Hungarirae 28, 1980

MONUMENTS OF A R T F R O M T H E S ANCTUARY O F OXUS 81
T h e panels from Takht-i Sangin with the representation of Kusnas on

hunting are d a t e d to t h e time a f t e r t h e conquest of Graeco-Bactria b y t h e
Yueh-chih t h a t is to t h e end of t h e 2nd1st century B. C.
T h e discovery in Takht-i Sangin in t h e year 1979, of single m o n u m e n t s of
t h e Achaemenid period and of a m a s s of works of Graeco-Bactrian a r t , p e r m i t
a t t h e present time to draw some preliminary conclusions :
1. T h e treasure of t h e Oxus, u p t o t h e present t i m e t h e main m a t e r i a l on
t h e history of Bactrian a r t of t h e pre-Hellenistic and Graeco-Bactrian times
has so f a r special chronological a n d stylistic parallels only in T a k h t - i Sangin.
This s u p p o r t s the opinion of scholars, localizing the site of the treasure in t h e
u p p e r course of the A m u - D a r y a t o T a k h t - i K u b a d or T a k h t - i Sangin.
2. T h e Graeco-Bactrian culture, more accurately its establishers, in their
creation relied on the broad circle of i m p o r t e d or trophied works of Greek and
Achaemenid a r t which directly or indirectly influenced t h e synthesis of t h e
Graeco-Bactrian style.
3. T h e basic impulse to t h e Hellenization of a r t and culture of B a c t r i a
a f t e r 329 B . C. was given by the establishment of new Hellenic cities a n d for-
tresses, in t h e architecture of which t h e r e was a considerable local and e a s t e r n
influence.
4. I n Hellenistic times, in t h e period of the accession of B a c t r i a to t h e
s t a t e of t h e Seleucids, and a f t e r t h a t a t t h e time of t h e independent Graeco-
B a c t r i a n s t a t e , the contacts with t h e Greek cities of t h e Mediterranean were
regular. T h e subsequent pouring of Greeks to Central Asia a f t e r t h e Graeco-
Macedonian conquest is not excluded, w h a t is shown by the schools of a r t de-
veloping under the direct influence of t h e contemporary specimens of t h e a r t of
Seleucid Asia Minor.
5. T h e m u l t i t u d e of finds of a r m a m e n t Greek, Saka-Scythian and Sar-
m a t i a n - K u s n a permits to a p p r o a c h t h e m with a more scientific conception,
as a result of a considerable q u a n t i t a t i v e increase of material of a r m a m e n t of
t h e peoples of Central Asia from t h e wide chronological range of t h e 6th c e n t u r y
B. C. 4 t h century B. C.
6. T h e multitude of the most d i f f e r e n t a r m a m e n t , including o r n a t e a n d
highly artistic specimens, of Greeks, B a c t r i a n s and nomads, shows t h e strict
necessity of t h e times when the helmets, shields, swords, spears, javelins a n d
arrows served not only as indispensable instruments of fight, b u t also as valu-
able sacrificial gifts.
7. T h e numerous articles m a d e of ivory, the m u l t i t u d e of goods m a d e of
this r a r e a n d expensive material show n o t only the close political a n d economic
relations with India. The analysis a n d a t t r i b u t i o n of the representational works
m a d e of ivory points to t h e probability of t h e local Bactrian m a n u f a c t u r e of t h e
basic mass of t h e articles, which a t t h e same time f r e q u e n t l y correspond t o
Greek s t a n d a r d s , types and conceptions.

8. Certain f i n d s f r o m different times point to t h e constant connections

w i t h t h e Saka-Scythian world of the steppe.
9. The i n t e r m e d i a t e Graeco-Bactrian t r a d e with the Altai, India, Iran,
P a r t h i a and with t h e cities of the coastal region of Asia Minor, promoted t h e
flourishing of jewellery and bone carving c r a f t , w h a t is testified by the Takht-i
Sangin treasure. T h e t r a d e with works of representational a r t played a p a r t in
t h e cultural relations, in which the Hellenic goods had a general and wide
demand.
10. The formation of anything architectural and artistic Eastern-Hellen-
istic koine in the v a s t territory promoted t h e development of representational
a r t and t h e flourishing of local schools. I n this phase the ethnic, religious a n d
cultural components of Greeks and Bactrians amalgamated into a unique and
n o t reproducible Graeco-Bactrian phenomenon.
11. The functioning of the Graeco-Bactrian temple a t t h e Oxus in t h e
K u s n a period and t h e protecting relation to t h e existing religious tradition,
show n o t only t h e religious tolerance of t h e K u s n a s , b u t also the symbiosis of
Greek, eastern-Zoroastrian and K u s n a conceptions. 288
12. The components of the art of Central Asia of the beginning of the 1st
millennium, a f t e r t h e finds at Takht-i Sangin appear to be t h e symbiosis of
t r e n d s of different character in style, of t h e synthesis of primary eastern, intro-
duced Hellenic, G a n d h a r a amalgamated Graeco-Indian and Bactrian arts.
13. The achievements of Bactrian civilization became possible not as a
result of its seclusion, b u t on the contrary, as a result of t h e presence of broad
a n d diversified cultural and economic relations. T h a t mighty creative synthesis
developed exactly u n d e r these circumstances, which b y itself gave strong im-
pulses to the development of culture.
14. The unique treasures of Takht-i Sangin have enriched the ancient
culture of Tajikistan a n d Central Asia with a vast collection of significant works
of a r t . However, t h e significance of these finds is much broader, viz. we h a v e
t o do with clear and unambiguous evidences of t h e close m u t u a l relations be-
tween Central Asia, t h e countries of t h e F a r E a s t and the Mediterranean.
15. The discovery, a t Takht-i Sangin, of a whole series of excellent antique
m o n u m e n t s of world a r t visibly complete t h e reports of the antique authors on
t h e whole cultural a n d historical phase of t h e peoples of Central Asia a n d
Afghanistan.
16. The sculptures of Takht-i Sangin m a d e of the local clay and alabaster,
of a material so fragile and unsuitable for transportation, point to the local
m a n u f a c t u r e of t h e highly artistic m o n u m e n t s of art, which has its subsequent
development in t h e K u s n a and early medieval period.
288
B. A . L I T V I N S K Y I . R . P I C H I K Y A N : . (One of the aspects
of antique influence on Central Asian culture.) 1979/2 89109.

M O N U M E N T S O F A R T F R O M T H E S ANCTUARY O F O X U S 83
17. On t h e right bank of t h e Oxus Amu-Darya, as well as on t h e l e f t

b a n k of t h e river, Greek was not only w r i t t e n , b u t it is now clear t h a t it w a s also
spoken and Greek deities were worshipped. The power of t h e antique t r a d i t i o n
was so great in t h e K u s n a period t h a t t h e hypothesis according to which t h e
elements of this culture could exist also a f t e r the fall of t h e K u s n a s t a t e seemed
t o b e quite probable. The getting of f u r t h e r materials in this direction is very
i m p o r t a n t , because t h e d e m o n s t r a t i o n of t h e traditions of antique a r t a n d cul-
t u r e in medieval Central Asia would change the whole conception of t h e a p p e a r -
ance and development of science a n d culture in the Mussulman East. 2 8 9
Moscow.
289
G . V. S H A Y M U X H A M B E T O V A : -
. Moscow 1979. (With literature on the question.)

I. BORZSK
ZENTRALASIATISCHE ELEMENTE
IN DEM ALEXANDERROMAN
Die in einer kaiserzeitlichen Hippokrates-Biographie berlieferte Ge-

schichte von Hippokrates und Perdikkas, dem Sohne des Makedonenknigs
Alexanders I . bietet eine genaue Parallele zur zarten Sage von Antiochos u n d
Stratonike, in welcher die Rolle des klugen Arztes der berhmte Erasistratos,
anderswo der noch berhmtere Galenos, in einer orientalischen Fassung dersel-
ben Geschichte aber der arabische Galenos, d.h. Avicenna spielt. Ob es sich
dabei u m eine sptere bernahme der Hippokrates-, bzw. Erasistratos- oder
Galenos-Anekdote handelt, oder m a n die Wanderungen und Wandlungen der
Sage in einer umgekehrten Richtung sich zu denken h a t ? Unseren Gedanken-
gang* haben wir mit einem Hinweis auf eine andere anmutige Sage, d.h. auf
die rhrende Liebesgeschichte von Zariadres 1 und Odatis (Athen.XI1 c.35 =
J a c . F G r H i s t 125, frg.5) geschlossen, von welcher Chares von Mytilene ver-
sichert, sie sei v ra [oToglat, in den persischen Geschichtsbchern aufgeschrie-
ben, u n d welche im Orient weit u n d lange verbreitet sich wie in einem
m a t t e r e n Abbild in einer bei Aristoteles (frg.503, p.499 Rose) und I u s t i n u s
( X L I I I 3,8 ff.) zu lesenden phokisch-massaliotischen Sage wiederholt. 2
Eigentlich bewegen wir uns in demselben Gedankenkreis, wenn wir uns
nun der I n t e r p r e t a t i o n einer anderen Perdikkas-Geschichte zuwenden. E s han-
delt sich u m Perdikkas I., den Helden des herodoteischen Berichts ber den
Ursprung des makedonischen Herrscherhauses (VIII 137 ff.). Eine b e k a n n t e ,
o f t behandelte und grndlich miverstandene Erzhlung, ber deren Sinn m a n
sich viel den Kopf zerbrach. Nach Salamis, anllich des Botendienstes, den
Alexander I. der Philhellen, zu gleicher Zeit persischer Vasall auf Mardo-
nios' Gehei nach Athen t u t , erzhlt Herodotos, wie Perdikkas, der Vorfahr
siebenten Gliedes dieses Alexanders, den makedonischen Knigsthron erlangt
h a t t e . Drei Shne desTemenos entflohen aus Argos und nahmen in der makedo-
nischen S t a d t Lebaia Dienst beim Knig. Der lteste h t e t e die Pferde, der
* Der Hinweis gilt unserem Vortrag Avicennas Qnn im westliehen rztekanon,

gehalten anllich des Avicenna-Millenniums in Budapest am 1.10.1980 (bisher nicht er-
schienen).
1
D. h. Zairiwairi, der mit goldfarbenem Harnisch, vgl. K . Z I E O L E R W . H I N Z :
R E Zariadres 2325.
2
Vgl. E . R O H D E : Der griech. Roman und seine Vorlufer. 3 Leipzig 1914, 51.
Ada Antiqua Academiae Scientiarum Uungaricae 28, 1980

86 I. BORZSK.
zweite die Rinder, der jngste, Perdikkas, das Kleinvieh. Als sich ein Brotprodi-
gium mehrmals wiederholte, dachte der Knig, da es ein gttliches Zeichen
(xga) sei und auf e t w a s Groes hindeute (qsooi pey TI). E r befahl also den
Temeniden, sein L a n d zu verlassen, sie aber sagten, es sei billig, da sie ihren
L o h n als etwas Geschuldetes bekmen (xv piaftov . . . dnoXaovxeQ ovx l-
vai ). Als der Knig von dem Lohn hrte, sagte er in seiner Verblendung (eoXa-
rjg yev/tevo): Mioftv vjxlv yw v/xcov iov rove dnoico/ti, das ist der L o h n ,
eurer wrdig, den gebe ich euch, d a m i t zeigte er auf die Sonne am Boden, die
d u r c h das Rauchabzugsloch ins H a u s hineinschien. Die beiden lteren s t a n d e n
fassungslos d a (eoxadav xnenXrjy/tvi), der K n a b e aber s a g t e : Aexopeda, <
aoiXev, x iol (wie Alexander d.Gr. vor Issos : yopai xr\v xXrjva),3 u n d
m a c h t e mit dem Messer, das er bei sich h a t t e , einen Schnitt rings um die Sonne
auf dem Estrich des Hauses (negiygcpEi r f j /nayagr]. . . rv rjXiov), neQiyqxpa
xv xXnov XQI ovadyEvo xov fjXov dnaXXdaaexo . . . , d. h.: dann schpfte
er dreimal von d e m umrissenen Sonnenfleck in seinen Busen und ging davon
(mit seinen Brdern). E r s t zu spt stellt es sich heraus, da Perdikkas das ihm
Angebotene mit Sinn u n d Verstand angenommen habe (ovv vcp . . . Xdoi xd
i/tEva), vergeblich s a n d t e der Knig den Temeniden Reiter nach, um sie u m -
zubringen : der Grenzflu schwoll so gewaltig an, da die Verfolger nicht m e h r
hindurch konnten. Dieser K n a b e war Perdikkas, der so die Knigsherrschaft an
sich gebracht h a t t e .
S t a t t eine kritische historiographische bersicht zu geben, die die Er-
gebnisse des d a r a n verschwendeten Philologenwitzes zeigen wrde, nehmen
wir zuerst mit d e m Wenigen frlieb, was in den alten K o m m e n t a r e n als ge-
sichert gelten k a n n . I m skythischen Logos des Herodotos liest man innerhalb
der vier Traditionen von den Ursprngen der Skythen zwei Versionen, diejenige
der Skythen (IV 5 7) und der pontischen Hellenen (8 10), die sich mit der
eben besprochenen makedonischen Knigsgeschichte vergleichen lassen. Von
d e n drei Shnen des Targitaos konnte die vom Himmel herabgefallenen golde-
nen Gerte ( /jvaea noir) pax a ) nur der J n g s t e anfassen, so da die lteren Br-
der auf die H e r r s c h a f t verzichteten u n d sie dem J n g s t e n bergaben, und des-
gleichen war von den drei Shnen des Herakles, die ihm die skythische piôndq-
&Evo geboren h a t t e , nur der jngste S k y t h e s imstande, die herakleischen Auf-
gaben zu vollbringen : so blieb er im Lande, u n d die Knige der Skythen s t a m -
m e n von ihm ab.
Einen u n b e t r e t e n e n Weg schlug neuerdings H e r m a n n Kleinknecht in
seinem postum erschienenen Aufsatz Herodotos und die makedonische Urge-
schichte 4 ein. Seine D e u t u n g des Sonneschpfens aus der iranisch-persischen
wzranA-Vorstellung darf in der T a t als ein bedeutungsvoller interpretatori-
3
Luk., De lapsu in salut. 8 ; vgl. H e r o d . V I I I 115 oder I X 91, oder etwa Tae. A n n .
I I , 133 mit K o m m .
1
Hermes 94 (1966) 134 ff.

87I.BORZSK:Z E N T R A L A S I A T I S C H EE L E M E N T E I N DEM A L E X A N D E R R O M A N
scher Fund 5 betrachtet werden. Diese Trouvaille wollen wir hier nicht in allen
Einzelheiten rekapitulieren. Mit vollem Recht wies Kleinknecht die unverant-
wortlichen Erklrungen ab, die im herodoteischen Bericht eine geschichtlich
wertlose Grndungssage, eine verdunkelte und entstellte Mrchendichtung
oder gar ein Kuriosum erblickten, mit welchem man sich nicht weiter zu be-
schftigen brauche usw. Dieses Volksmrchen e n t p u p p t sich als eine ausge-
sprochene Knigsgeschichte, die Sonne aber, deren goldene Strahlen der zum
Knigtum Auserwhlte in seinen Busen schpft, m u der iranische Licht-
glanz der kniglichen Majestt, die Knigsglorie (H. H . Schaeder), also das
yvardnah sein, das als Versinnbildlichung und U n t e r p f a n d hchster, gottge-
wollter Herrschaft in besonderer Weise dem irdischen Knige eignet, der
dadurch als rechtmiger Herrscher legitimisiert, geschtzt, in seinem Verstand
zum Regieren erleuchtet und zum Sieg ber seine Feinde befhigt ist. Es ist eine
gttliche Gestaltwerdung des sonnenhaften, sieghaften Glcks, das ein Wesens-
merkmal echten Knigtums ist. 0
Kleinknecht kannte freilich auch Altheims Ausfhrungen, der aber die
F o r m - und Motivverwandtschaft mit der Perdikkas-Geschichte noch nicht ge-
sehen h a t . I m Sasaniden-Kapitel seines Buches Gesicht vom Abend und Mor-
gen7 erzhlt er in seinem verfhrerisch-geschmeidigen Stil den Mythos des
Reichsgrnders und der Reichsgrndung dem mittelpersischen Knigsbuch
nach, wie Ardaslr, Sasans Sohn, in Kenntnis der Weissagungen der Sternkundi-
gen seinem Herrn Ardewan mit dem Mdchen, das um das Geheimnis der Macht-
ergreifung wute, sowie mit einer K r o n e und vielen K o s t b a r k e i t e n entfloh. 8
Leute, die dem sie verfolgenden Knig begegneten, sagten, des Morgens seien
beide Reiter vorbergekommen, schnell wie der Wind, u n d ein dicker Widder sei
ihnen gefolgt. Als das sich wiederholt, f r a g t der Knig den Obersten der Magier,
was es bedeute, und er erhlt die A n t w o r t ; Der Widder ist der Glanz der k-
niglichen Herrschaft ; weh h a t er Ardasir nicht erreicht... Am nchsten Tage
berichten die Karawanleute schon davon, da mit einem von den beiden Reitern
zugleich ein groer Widder zu P f e r d e sa, d.h. da der Glcksglanz der k-
niglichen Majestt Ardaslr erreicht hat. U n d weil das %vardnah bei Ardair war,
gewann dieser den Sieg, t t e t e Ardewan und nahm dessen Tochter zur F r a u .
Das yvardnah als gttliche Wesenheit verleiht dem Helden Gelingen. 9
Die griechisch-baktrischen Knigsmnzen stellen dieses %varjnah in der Ge-
s t a l t der Tyche dar ; in mittelpersischer Schrift wird es durch ein aramisches
Ideogramm (GDH) wiedergegeben, das ebenso Glck heit, yyannah als Wid-
der hat auch in der griechischen Heldensage sein Gegenstck : 10 in Pelops'
5
Bemerkung des Herausgebers G. M L L E R : . O . 134.
KLEINKNECHT : 139.
' Fischer-Bcherei 79 : F r a n k f u r t a. M. Haniburg 1955.
8
Ausfhrlicher bei K L E I N K N E C H T : 1 4 3 .
Zum Folgenden vgl. A L T H E I M : 108.
10
Vgl. KLEINKNECHT : 139.

88 I. BORZSK.
Nachfolge wurde die H e r r s c h a f t ber Mykene durch ein von Hermes dem Atreus
geschenktes L a m m mit goldenem Vlie gesichert (vgl. Eur., El. 705 ff. ; Or. 998
ff.).* Gold und himmlischer Glanz ist auch dem i v a n n a h wesenhaft zu
eigen, . . . sein N a m e ist dem der Sonne (hvar-) verwandt. Feuriges Gold,
vom Himmel herabgekommen, spielte auch in der besprochenen skythischen
Sage die Rolle, die sonst dem ivannah z u k o m m t : dadurch, da Skythes das
glhende Metall anfassen kann, ergreift er sein Glck, erweist sich als gottge-
wollter Herrscher.
U n d jetzt wollen wir zeigen, da es nicht nur im mittelpersischen Knigs-
buch, d. h. in jener Zusammenstellung der heimischen berlieferung, die kurz
vor d e m Untergang des Sasanidenreiches niedergeschrieben wurde und dieses
als ein nationales Vermchtnis berdauert hat, 1 1 sondern hie und da auch im
berchtigten Alexanderroman berreste uralter mythischer Vorstellungen gibt.
D a b e i denken wir an eine Episode, die selbst die besten K e n n e r dazu bewog,
ber die r o m a n h a f t e Alexander-berlieferung im allgemeinen recht ungnstig
zu urteilen. Alexander sucht auf Ammons Gehei, der ihm im Traum in der
G e s t a l t des Hermes erschien, als sein eigener Bote verkleidet den Dareios auf,
u m ihm zu melden, d a Alexander d a ist, u n d zu fragen, wann er den Krieg be-
ginnen will (II 14). Der erzrnte Dareios f a t den Boten bei der H a n d und f h r t
ihn in seinen Palast, was Alexander als ein arj/uelov ya&ov b e t r a c h t e t und legt
sich im Palast als erster zur Tafel des Dareios nieder. Die Perser aber (15) sahen
Alexander voll Verwunderung ber seine kleine Gestalt an, wuten sie doch
nicht, ort v pLXQ) yyetcp xvyr]; ovgavov jv oa (da in einem kleinen Gef der
Glanz eines himmlischen Glckes verborgen war). 1 2 Als sie nun dem Wein str-
ker zusprachen, h a t t e Alexander folgenden Einfall : ooov axxpov; ekaev,
acoev xov xXnov eaAev. Als darauf Dareios ihn f r a g t e : W yevvae, TIQO xxavxa
yxonLei xaxaxei/tevo jii einvov, ersann Alexander eine Ausrede, die die Per-
ser e r s t a u n t anhrten. Bald aber sah sich Alexander e r k a n n t , so da er mit den
goldenen Bechern in seinem Busen (eycov v xol xknoi . . . xov yovaovg cfxv-
(pov: B ' L ; x ygvcfiov . . . Kroll) aufsprang, ging heimlich hinaus, dort
schwang er sich aufs P f e r d , um der drohenden Gefahr zu entgehen. Man verfolgt
ihn bis an einen F l u , der zugefroren war : Alexander r i t t bers Eis. Als das
P f e r d das gegenberliegende Ufer erreicht hatte, t a u t e der Flu auf ; das P f e r d
k o n n t e gerade noch seinen Reiter ans Ufer schleudern und wurde von der Str-
m u n g fortgerissen. Die Verfolger konnten den a u f g e t a u t e n Flu nicht mehr
berschreiten.
* Korr.-N: Ziegen als regni auctores in 1er makedonischen Knigsgeschichte bei

l u s t . V I I 1, 9.
11
ALTHEIM : 96.
12
Der Text nach B ' und L (ed. van Thiel) ; bei KROLL heit es : ort v . .
xyv/ia Tvyrj <ovgavov > tjxiv, vgl. den app. crit.

89I.BORZSK:Z E N T R A L A S I A T I S C H EE L E M E N T E I N D E M ALEXANDERROMAN
Man meint, dem Baron von Mnchhausen zuzuhren b e m e r k t R .

Merkelbach dazu, 1 3 nicht ohne allen Grund. J a , der Held des Romans be-
n i m m t sich manchmal wenig heldenhaft, 1 4 oder wie Ad. Ausfeld 1 5 seinerzeit ber
die alexandrinischen Zge dieses Alexanders schrieb : s t a t t echten Helden-
t u m s wird Alexander im R o m a n mit Eigenschaften ausgestattet, die den ale-
xandrinischen Handelsleuten vertrauter sind als Knigsstolz und Heldensinn.
E r ist klug u n d schlau wie sie, versteht zu beschwatzen und zu betrgen ; bricht
gelegentlich sein Wort, stiehlt auch wohl einmal... F a t a l e r Diebstahl, den of-
fensichtlich nicht nur die modernen Forscher, sondern bereits die K o m p i l a t o r e n
der hybriden Alexander-Geschichte miverstanden h a t t e n . U. E. h a n d e l t es
sich um ein berbleibsel einer uralten Ideologie der Knigsherrschaft, deren
uere Formen f r sptere B e t r a c h t e r ihren wahren Sinn lngst verloren haben.
Herodotos vermochte noch den mythischen Sinngehalt der Perdikkas-Geschich-
te als Knigsgeschichte16 aufzufassen u n d zu veranschaulichen, die Reprsen-
t a n t e n der spthellenistischen Geschichtsklitterung s a m t ihren Lesern nicht
mehr : sie werden in einer tiefsinnigen Szene der translatio imperii,17a wie der zu-
k n f t i g e makedonische Weltherrscher der goldenen Trinkschalen des persischen
Knigs der Knige gleichsam seines /vardnah h a b h a f t wird, nur etwas
Groteskes, Sensationelles empfunden haben.*
I n derselben Erzhlung f i n d e t man noch weitere Einzelheiten, die geeignet
sein d r f t e n , unsere oben skizzierte These zu erhrten. Dazu mssen wir ein
wenig ausholen.
I n einer Anmerkung von G. Widengrens lehrreichem Buche Iranisch-
semitische Kulturbegegnung in parthischer Zeit (Kln Opladen 1960, S.68,
238), dessen K e n n t n i s ich der freundlichen Zuvorkommenheit H. H u m b a c h s
verdanke, wird u. a. auf H . W. Bailey (Zoroastr. Problems in the 9 t h Cent.
Books, Oxford 1943, 32) hingewiesen, der aufgrund einer Textstelle aus dem
* Korr.-N.: Interessant ist die Erklrung der gelungenen Flucht im sog. Wiener
P r o s a t e x t (Paraphrase von H . J . G L E I X N E R : Das Alex.-bild der Byzantiner. Diss.
Mnchen 1961, 82): Die Torwchter berlistete Alexander, indem er die goldenen Becher
als Zeichen des Perserknigs ausgab.
13
Die Quellen des griech. Alexanderromans. Mnchen 1964. 67.
14
W . K R O L L : R E Kallisthenes 1711.
15
Der griech. Alexanderroman. Leipzig 1907. 236.
16
Und als berlieferung von einer Herrschaftsgrndung,deren weltgeschichtliche
Perspektiven zu Alexanders I . Zeiten noch niemand ahnen konnte! ber H e r o d o t s
Leistung vgl. K L E I N K N E C H T : 1 4 6 : ( Die Perdikkas-Erzhlung ist) durch u n d durch
herodoteisch : Weit davon entfernt, berlieferung unverstanden oder nicht voll ver-
standen weiterzugeben oder gar selbst zu entstellen, hat die eigene reife Erzhlergabo
u n d D e u t u n g s k r a f t die fremde Tradition in der griechischen F o r m erst das sein lassen,
was sie vom Hauso aus ist : ein Stck Urgeschichte. Als Parallele (wie die m y t h i -
schen Bezge einer S t a d t g r n d u n g bei Pindar, P y t h . I spter nicht mehr verstanden
wurden) m a g hingewiesen worden auf J . T R U M P F S Interpretation : S t a d t g r n d u n g und
Draehonkampf. Hermes 86 (1958) 129 ff., insbes. 140.
" yg). W. G O E Z : Translatio imperii. Tbingen 1 9 5 8 . passim, und unseren Versuch
(mit dem gleichen Titel) in der Festschrift f r K. Mart (Id s trtnelem, B u d a p e s t
1974), 33.

90 I. BORZSK.
D e n k a r t (ed. Sanjana, V I I I , 56 58) schildert, wie ein Licht von dem yvardnah
der M u t t e r Zarathustras auf den von ihr gewandelten Weg fllt. Ihr yvarmah
h a t sich beim E i n t r i t t in sie wie ein Feuer g e z e i g t . . .
N u n liest man im Alexanderroman (nach der Handschrift L, in der ber-
setzung H . van Thiels, I I 15,6) : Als Alexander . . . sich e r k a n n t sah, . . .
sprang auf mit den goldenen Bechern in seinem Gewand (v ro xfaioi) u n d
ging heimlich hinaus. D o r t schwang er sich auf sein Pferd . . . , und als er am
Tor einen Perser a u f W a c h e fand, ttete ihn u n d floh aus der Perserstadt. (7)Da-
reios . . . schickte b e w a f f n e t e Perser aus, u m ihn zu fangen. Alexander aber
s p o r n t e sein Pferd u n d erhellte ihm den Weg (iev&vvcov avxov xrjv v), denn
es war tiefe Nacht u n d der Himmel finster. (Seine Verfolger strzen in der Dun-
kelheit in die Abgrnde.) 'Ahavqo qv axrjo ef ovgavov qpatq
vtwv [tvo . . . (9 : N a c h der Rckkehr berichtet man dem Groknig vom
evxvyrj/ia Alexanders.)
Noch interessanter als die zitierte Stelle selbst ist der U riterschied zwischen
dem Wortlaut der H a n d s c h r i f t L und dem von W. Kroll nach der recensio ve-
tusta herausgegebenen T e x t , wo die offensichtlich nicht mehr verstandene, ge-
wissermaen noch m y t h i s c h e Erzhlung sinngem ergnzt, d. h. ihres ur-
sprnglichen Sinnes vllig beraubt wurde (p. 84,11 ) : xx xwv aatXelmv
lmv riva noarjv naatpvfaidoovxa a / yelna yovxa, xavxa aaxa
aqayiaae xov avna . . . (15) 'Ahavgo vu xov eov orj&ov/tEvoQ . . .
nooea/./.E rrjv nevxrjv xai tpwxaywyet rv nwXov rjv yn vv ahela . . . fiv
yq ycov nevxrjv csxi)Q nhaynev avxm . . . U. E. unterliegt es keinem
Zweifel, da die Version L (ohne die Ergnzungen van Thiels) ursprnglicher
ist, als die sog. recensio vetusta. Die Verballhornungen zeigen, da man mit der
l i c h t h a f t e n N a t u r des Besitzers des Glcksglanzes nichts anzufangen wute
u n d die unverstndlichen Hinweise erklren zu mssen glaubte.
Eine Bemerkung W . Krolls 1 ? b zeigt, d a die soeben behandelte Szene auch
diesem hervorragenden Kenner des Alexanderromans zu bedenken gab.
I n Wirklichkeit lt L e o ( I I 15 p. 91 Pfister) den kniglichen Gast genauso ver-
f a h r e n und dann fliehen : Alexander . . . vidit quendam Per sen tenendo in manu
facculam tollensque earn illi percussit eum, ascendit equum et [h]abiit. . . Alexan-
der portans in manu facculam, tenens iter rectum (erat enim obscura nox), inse-
quentes eum cadebant in foveas. Umso mehr gewinnt eine Stelle der darauffolgen-
den Kampfschilderung ( I I 16 p. 93 Pf.) besondere Bedeutung : . . . tunc ascen-
dit Alexander equum, qui dicitur Bucefalo, et occurrit et stetit in medio ante omnes.
Videntes eum (sc. Alexandrum) Perses dubitabant... ire super eum (sc. Alexan-
drum) eo, quod d i v i n i t a s c o o p e r i e b a t u r e u m . . . , desgleichen
in der Leo sehr naheliegenden syrischen Version (laut der deutschen berset-
17b
W. K R O L L : R E Kallisthenes 1711 : Fr die Moral des Autors ist es bezeich-
nend, da der Knig als Gast an Dareios' Tafel die kostbaren Becher einsteckt ; Leo
hat da.8 gendert.

91I.BORZSK:Z E N T R A L A S I A T I S C H EE L E M E N T E I N DEM ALEXANDERROMAN
zung von V. Ryssel) : Weil Gtterkraft auf ihm ruhte. Leider hilft dabei weder
die recensio vetusta (II 16,4 p. 86 Kroll nqoazyyiaai rovrw ovsl rjvvaro t
rrjv v avxw 'fatrrjra'f) ,vc noch die auch hier verballhornte Handschrift L
(II 16,6 ngoaeyyloai TOVTCO r 1 ovel vvaro) weiter, und man sieht
auch hier nur die Anstrengungen der spteren R e d a k t o r e n , die unverstnd-
lichen Elemente der berlieferten Erzhlung durch schulmeisterliche Einschieb-
sel, Glossen o. . zu erklren.
Nach diesem nachtrglichen E x k u r s d r f t e unsere Hypothese noch weni-
ger gewagt erscheinen, laut welcher es in der r o m a n h a f t aufgebauschten Alexan-
der-Geschichte gewisse rudimentre %vardnah-Vorstellungen gegeben haben
mu, deren mythischer Sinn im Wandel der J a h r h u n d e r t e vllig der Vergessen-
heit anheimfiel, deren Spuren aber hie und da selbst in den hoffnungslos t r b e n
Redaktionen des Alexanderromans aufgezeigt werden knnen.
U n d nun mit einem kleinen Sprung mchte ich die Geburtslegende
Dschingis Khans, wie sie in der Geheimen Geschichte der Mongolen 18 zu lesen
ist, besprechen. Da rechtfertigt Alanqo'a die bernatrliche Empfngnis ihrer
drei Kinder vor ihren zwei anderen, aus der E h e mit Dobun morgen (D. dem
Klugen) hervorgegangenen Shnen : Nacht fr N a c h t k a m ein goldglnzender
Mann durch die Dachluke der J u r t e herein. E r streichelte meinen Bauch, und
sein Strahl senkte sich in meinen Leib. Wenn er hinausging, kroch er hinaus auf
dem Sonnen- oder Mondstrahl wie ein gelber Hund. Alanqo'a nimmt dies zum
Beweis, da sie Shne des Himmels geboren hat : Wie knnt ihr . . . sie mit
schwarzkpfigen, gewhnlichen Menschen vergleichen ? Wenn sie einmal Kni-
ge ber alle sind, dann wird das gemeine Volk sie erkennen. 19 Wir haben nicht
vor, das riesige Material, das in Zusammenhang sei es m i t dem Danae-Mythos,
sei es mit Gilgames oder mit gewissen christlichen Legenden mehrfach vorgelegt
u n d interpretiert wurde, bei dieser Gelegenheit noch einmal zu registrieren.
N u r einige Bezge dieses Materials wollen wir hervorheben, um die Verbreitung
des Alexanderromans, bzw. den mythischen Gehalt gewisser Bestandteile dieses
Romans zu zeigen.
Den Zusammenhang des Danae-Mythos und der Geburtslegende Dschin-
gis K h a n s mit einer Einzelheit der Perdikkas-Geschichte hielt auch weiland
I. Trencsnyi-Waldapfel in Evidenz : das Gottesurteil in der Ursprungssage
der Kirgisen, 20 durch welches Altyn Bels Tochter die Auserwhltheit des Dschin-
17c
coni. KROLL (coll. Syr. Leo) ; equidem aXxrjxa malim, cf. LIDDELL-
SCOTT .18 v. iaX I I . 3 .
Hrsg. von E . H A E N I S C H : 'Leipzig 1948. 3 f. u n d 209; P . P E L L I O T : Histoire
sccr. des Mongols. Paris 1949. 121 f f . ; L . L I G E T I : A mongolok titkos trtnete. Buda-
pest 1962. 11.
" V g l . I . T R E N C S N Y I - W A L D A P F E L : Der Danae-Mythos im Osten u n d Westen.
Unters, zur Religionsgesch. A m s t e r d a m 1 9 6 6 . 1 9 2 ff. ; G. B I N D E R : Die Aussetzung dos
Knigskindes. Moisonheim a. Gl. 1964. 220 f. u n d Enz. des Mrchens. I I I . 263 ff. (Danae),
20
Vgl. VV. R A D L O F F : P r o b e n der Volkslit. der t r k . S t m m e Sd-Sibiriens. I I I .
Petersburg 1870. 82 ff.

92 I. BORZSK.
gis g e g e n b e r dessen drei H a l b b r d e r n d a r t u t , e r i n n e r t e ihn a n die genealo-

gische Sage des m a k e d o n i s c h e n H e r r s c h e r h a u s e s insofern als die S o n n e n s t r a h l e n
a u c h hier f r den a u s e r w h l t e n P e r d i k k a s eine h a n d g r e i f l i c h e R e a l i t t u n d zu-
gleich Vorzeichen seiner H e r r s c h a f t bedeuten. Als erster h a t t e J . G. F r a z e r 2 1
e i n e a n d e r e E i n z e l h e i t des griechischen M y t h o s m i t d e r i n n e r a s i a t i s c h e n S a g e
verglichen.
W a s die i n n e r a s i a t i s c h e V e r b r e i t u n g der Alexander-Sage b e t r i f f t , so will
m a n eine wichtige B e o b a c h t u n g v o n E . H e r z f e l d kurioserweise noch i m m e r
n i c h t z u r K e n n t n i s n e h m e n . I n P l u t a r c h s A l e x a n d e r - B i o g r a p h i e (c.2) liest m a n ,
d a O l y m p i a s als B r a u t P h i l i p p s einen T r a u m h a t t e : E s w a r ihr, als wenn bei
e i n e m G e w i t t e r der B l i t z ihr in d e n L e i b schlge u n d d u r c h diesen Schlag helles
F e u e r h e r v o r b r c h e usw. H e r z f e l d h a t n u n erwiesen, d a die z u r V e r h e r r l i c h u n g
D s c h i n g i s K h a n s k o n s t r u i e r t e mongolische L e g e n d e aus Griechenland (d. h.
a u s d e r griechischen A l e x a n d e r - b e r l i e f e r u n g ) s t a m m t . 2 2 D e r N a m e Alongoa
( A l a n - q o ' a , Alan-koa) i s t n i c h t s a n d e r e s als eine d u r c h falsche P u n k t s e t z u n g
e n t s t e l l t e F o r m f r griech. Olympia(s) ; z u r V e r d e r b nisp() ~ / q, die bereits
in d e r a m H o f e D s c h i n g i s - K h a n s gelesenen G e s t a l t des I s k a n d a r - n m a h vor-
lag, 2 3 vergleiche m a n die F o r m e n F i l i f u s Filiqus F e i l a q u s oder F a i l a q u s
( s t a t t Philippos) oder K a i d a f e ( s t a t t K a n d a k e ) usw. D e m e n t s p r e c h e n d ist die
g e w h n l i c h e b e r s e t z u n g Alan die Schne (Alanqava, Along-goa) eine sehr
n a h e l i e g e n d e mongolische E t y m o l o g i s i e r u n g , die nichts d a r a n n d e r t , d a die
m o n g o l i s c h e Dschingis-Legende u n d e n k b a r o h n e ihr griechisches Vorbild ist. 2 4
D a m i t ist aber n i c h t so einfach b e s t e l l t . A l a n q o ' a s N a m e k a n n getrost als
e n t s t e l l t e F o r m des griech. E i g e n n a m e n s O l y m p i a s e r k l r t w e r d e n , desgleichen
d r f t e n einige E i n z e l h e i t e n d e r Dschingis-Legende aus der Alexander-berlie-
f e r u n g hergeleitet w e r d e n , es gibt a b e r ganz v e r b l f f e n d e P a r a l l e l e n , deren ur-
s p r n g l i c h e F a s s u n g b z w . H e i m a t n i c h t so leicht erschlossen w e r d e n k a n n . D a z u
n u r ein einziges Beispiel, t r a v e r s les ges.
K t e s i a s , der als L e i b a r z t u n d neugieriger Grieche am persischen Knigs-
hof Gelegenheit genug h a t t e , u m o r i e n t a l i s c h e Sensationen a n z u h u f e n , e r z h l t
i n d e n S e m i r a m i s - K a p i t e l n des D i o d o r ( I I 16 f.), was die e h e m a l i g e K n i g i n f r
e i n e Kriegslist gegen die E l e f a n t e n d e r I n d e r ersann : aus d e r abgezogenen
H a u t v o n schwarzen Ochsen lie sie T r u g b i l d e r (eXa) v o n E l e f a n t e n verferti-
g e n , welche von K a m e l t r e i b e r n g e t r a g e n d u r c h ihre R i e s e n z a h l die I n d e r in die
F l u c h t h t t e n jagen sollen. E i n hnliches S t r a t e g e m k e h r t i m A l e x a n d e r r o m a n
( I I I 3) in einer g r o t e s k e n F a s s u n g wieder : d e r listenreiche A l e x a n d e r l t alle
21
The Golden Bough, London 1913. 74 f.
22
Alongoa. Der Islam 6 (1916) 320 ff.
23
H E R Z F E L D : 3 2 4 ; vgl. E D . M E Y E R : K L . Schriften I I . Halle 1 9 2 4 . 4 3 6 , 4 ; ebenso
J . E . B E R T E L S : R o m a n ob Aleksandre. M.-L. 1 9 4 8 . 2 6 , 2 . Vgl. noch N . P O P P E : Eine
mong. Fassung der Alex.-Sage. Zeitschr. f. D. Morg. Ges. 32 (1957) 105 ff.
21
HERZFELD: 327.

ZENTRALASIATISCHE E L E M E N T E IN DEM A L E X A N D E R R O M A N 93
ehernen Standbilder (vQivxa xakxov), die er bei sich h a t t e , und die erbeute-
ten Panzer (xaxacpgyjaaxa) der Soldaten stark erhitzen, so da das Metall gl-
hend wurde (in der syr. Variante : die S t a t u e n h a t t e n feurige Kohlen in sich),
und sie vor der Schlachtreihe aufstellen. Als die Tiere des Poros auf die S t a t u e n
lossprangen, verbrannten sich die Rssel, flohen b e s t r z t zurck und waren
nicht mehr zu gebrauchen.
Interessant ist zu sehen, wie die Kriegslist der Semiramis und des klugen
Alexander nach tausend J a h r e n bei den Mongolen wiederkehrt. Wie man in der
tatarischen Reisebestehreibung des Piano Carpini 25 liest, lie Dschingis K h a n
whrend seines indischen Feldzuges Puppen von Menschengre aus K u p f e r
anfertigen, setzte sie in Sttel auf Pferde, t a t Feuer hinein u n d setzte hinter jede
dieser kupfernen P u p p e n einen Mann mit Blasebalg aufs P f e r d . . . Die Mnner
warfen etwas Unbekanntes ins Feuer [Firdausi lt Alexander Erdl gegen
Poros verwenden] und fachten dann mit ihren Blaseblgen die Flamme mchtig
an . . . Fr. Risch (a. 0.) wies mit R e c h t d a r a u f h i n , da Piano Carpini das ur-
sprngliche innerasiatische Sagenmaterial mit antiken Requisiten d. h. mit
denjenigen der Alexander-berlieferung bereichert h a t t e . Die ursprngliche
mongolische Fassung f i n d e t man in Rasid-ed-Dins J a h r b c h e r n : 2 0 d o r t wird
der Feind durch Popanzen aus Filz, die hinter einem jeden Reiter sitzen, ge-
tuscht.
Und das ist keine Fiktion : die Verwendung solcher Popanzen, um die
Gegner durch ihre dahinsausende Menge zu erschttern, gehrte in der T a t zur
Taktik der Mongolen. D a f r haben wir den Bericht eines Augenzeugen aus der
Zeit der mongolischen Invasion im J . 1241. Rogerius, E r z d e c h a n t von Growar-
dein erzhlte in seinem Carmen miserabile (c. 27, Szentptery, S R H I I 568)
u. a., wie die Leute seines Bischofs per Tartaros ins Garn gelockt wurden : cum
(Tartari) haberent equos plurimos et ipsi pauci existerent, . . . fecerunt larvas et
monstra quam plurima, ea super equos vacuos, tanquam si essent milites, ordinan-
do, et equos illos . . . paucis cum servientibus dimiserunt mandantes eisdem, ut cum
Hungaris ingrederentur in pugnam, ipsi . . . procederent versus eos . . . Tartari
numero pauciores terga eis (sc. Hungaris) dare fingentes . . . retrocedere incepe-
runt. Et dum ipsi cum larvis exierunt. . . , Hungari opinantes eis (d. h. sibi) in-
sidias factas terga dederunt . . . , Tartari autem etc. (richteten ein frchterliches
Blutbad an).
Das war im 13. J a h r h u n d e r t . Ob ich mich zum Schlu an ein Leseerlebnis
aus meiner Kinderzeit wiedererinnern darf ? In einer alten illustrierten F a m i -
lienwochenschrift guckte ich staunend einen interessanten Stahlstich an, der
bewegliche Reiterattrappen darstellte, wie sie Anfang dieses J a h r h u n d e r t s im
25
Hrsg. von F R . B I S C H : Leipzig 1 9 3 0 . 1 2 3 .
28
Vgl. . I . S M I R N O W A - I L I . P A N T C R A T O W A. A. SEMENOW : R . Sbornik letopisei
1/2. M.-L. 1952. 221 ff.
Acta Antiqua Aeademiae Scientiarum liungaricae 28, 1980

94 I. BORZSK: Z E N T R A L A S I A T I S C H E E L E M E N T E I N DEM A L E X A N D E R R O M A N
russisch-japanischen Krieg verwendet wurden. Auch diese Einzelheit drfte uns

nachdenklich stimmen : innerhalb der romanhaften Alexander-berlieferung
gibt es auch wertvolle Tradition, deren Elemente aufzuhellen, wie sie gewisse
Wechselbeziehungen zwischen Osten und Westen spiegeln, eine lohnende Auf-
gabe nicht nur fr einen klassischen Philologen sein wrde.
Budapest.

101 . STAVISKIY
KARA TEPE IN OLD TERMEZ

A B U D D H I S T R E L I G I O U S C E N T R E OF T H E K U S H A N P E R I O D
ON T H E B A N K OF T H E OXUS
A v a s t site, occupying a n a r e a of over 500 h e c t a r e s , is s i t u a t e d in t h e

n e i g h b o u r h o o d of m o d e r n T e r m e z , a regional c e n t r e of S o u t h U z b e k i s t a n ,
o n t h e r i g h t b a n k of t h e a n c i e n t O x u s (now A m u - D a r y a ) . I t is t h e s i t e of Old
T e r m e z (Fig. 1), which a t t r a c t e d t h e a t t e n t i o n of travellers a l r e a d y in t h e 19th
a n d t h e beginning of t h e 20th c e n t u r i e s . 1 I t b e c a m e t h e principal o b j e c t of
research m a d e b y t h e f i r s t m a j o r archaeological e x p e d i t i o n of Soviet scholars t o
C e n t r a l Asia in 1926 28 t h e e x p e d i t i o n of t h e Moscow M u s e u m of E a s t e r n
C u l t u r e s (now t h e S t a t e Museum of O r i e n t a l Arts), h e a d e d b y prof. B . P . De-
nike, 2 which w a s followed in 1936 38 b y t h e large T e r m e z Archaeological Com-
p l e x E x p e d i t i o n (TACE) u n d e r p r o f . M. E . Masson. 3
E x c a v a t i o n s carried o u t b y t h e t w o expeditions r e v e a l e d t h a t Old T e r m e z
w a s t h e site of a t o w n d e s t r o y e d in 1220 b y t h e a r m i e s of Chenghis K h a n .
I t w a s also f o u n d o u t t h a t t h e r e w a s a s e t t l e m e n t t h e r e going b a c k as f a r as t h e
3rd 2nd centuries B.C., when B a c t r i a 4 w a s u n d e r Greek rulers. T h e t o w n reach-
ed i t s h e y d a y a t t h e t i m e of t h e K u s h a n E m p i r e , which, in t h e f i r s t c e n t u r i e s of
o u r e r a u n i t e d u n d e r its dominion t h e p r e s e n t territories of A f g h a n i s t a n , N o r t h -
ern P a k i s t a n , N o r t h e r n I n d i a , S o u t h e r n regions of t h e Soviet U z b e k a n d T a j i k
r e p u b l i c s a n d E a s t e r n T u r k i s t a n (Sinkiang).
I n 1928 A. S. Strelkov of t h e M u s e u m of E a s t e r n Cultures e x p e d i t i o n
f o u n d t h r e e filled-up caves on t h e e a s t e r n slope of a t h r e e - t o p hill rising in t h e
n o r t h - w e s t e r n corner of t h e site, close t o t h e b a n k of t h e A m u - D a r y a . T h e caves
1
M. E. : 1936 (
, I . 2); , 1941, 5122
2
, (. I), ., (1927), . II, . (1928).
3
See : . I 1936 ; . II - . , I, . II,
, 1945.
4
I share the view held by most of the Soviet scholars and m a n y Western research-
ers t h a t Bactria, as a historico-cultural entity, included not only the loft bank of Ainu-
D a r y a up to the Hindu Kush on the South (the present Northern Afghanistan), b u t also
the right bank of this great river, i.e., southern regions of the present Uzbek and Tajik
Soviet republics.

g
Sa
a
H
>
M
KJ
Fig. 1. Town-site of Old Termez. Schematic plan

KARA T E P E I N OLD TERMEZ 97
in his view, m i g h t h a v e been small B u d d h i s t temples of t h e pre-Moslem (prior

t o 7 t h 8 t h centuries A.D.) period. 5
T h e t h r e e - t o p hill, on w h i c h A. S. Strelkov discovered t h e caves, w a s
k n o w n locally as Black Hill ( K a r a t e p e in Uzbek), t h e e p i t h e t black referring
n o t t o i t s colour which w a s as of all t h e o t h e r hills here s a n d y light yellow, b u t t o
its s o m b r e p a s t : according t o legend a m u l t i t u d e of people h a d once perished
o n t h i s v e r y s p o t . T h e hill is k n o w n in scientific l i t e r a t u r e u n d e r t h a t n a m e .
I n 1934 36 t h e caves f o u n d b y A. S. Strelkov (Nos. 13) were investi-
g a t e d b y G. V. P a r f e n o v , d i r e c t o r of t h e Termez M u s e u m , w h o m a d e t h e de-
s c r i p t i o n of t h e caves and m e a s u r e d t h e m . Since 1936 e x c a v a t i o n work on K a r a
t e p e w a s carried on b y t h e T A C E : V. I . K e s a y e v a n d A. S. S t r e l k o v cleared
t w o m a n h o l e s in cave No. 2, which h a d f o r m e d a f t e r t h e ceiling c a v e d in ; B . B.
P y o t r o v s k j y s t u d i e d t h e r e m a i n s of t h e m u d brickwork on t h e hill's surface. 6
I n t h e 1937 season, a g r o u p of t h e T A C E archaeologists h e a d e d b y E . G.
P c h e l i n a d r o v e t h r e e e x p l o r a t o r y t r e n c h e s on t h e e a s t e r n slope of t h e hill n o r t h
of t h e caves 1, 2 a n d 3. I n one of t h e t r e n c h e s , remains of a small g r o u n d t e m p l e
were discovered, in a n o t h e r a p a s s a g e leading inside o n e of t h e caves a n d a n
a d j o i n i n g cave sell, a n d in t h e t h i r d a passage leading i n t o a n o t h e r cave.
F o u r K u s h a n copper coins (two of t h e m issues of K a n i s h k a ) a n d red-slip
p o t t e r y were f o u n d in t h e course of t h e excavations, w h i c h g a v e E . G. P c h e l i n a
s u f f i c i e n t g r o u n d t o d a t e her discoveries t o t h e K u s h a n period. D a t a pointing
t o t h e existence on K a r a t e p e of n u m e r o u s cave and g r o u n d s t r u c t u r e s also led
h e r t o t h e conclusion t h a t t h e hill w a s actually t h e site of a n a n c i e n t B u d d h i s t
m o n a s t e r y . 7 I n 1938, G. A. P u g a c h e n k o v a , t h e T A C E a r c h i t e c t , m e a s u r e d t h e
caves E . G. P c h e l i n a h a d been e x c a v a t i n g and drew t h e i r p l a n s a n d architec-
t u r a l cross sections. 8 A f t e r t h a t , e x c a v a t i o n work on K a r a t e p e was discontin-
ued.8
I t has been carried on since 1961 b y t h e joint archaeological e x p e d i t i o n
sponsored b y t h e i n s t i t u t i o n s u n d e r t h e U S S R Ministry of C u l t u r e . T h e expedi-
t i o n w a s originally f o r m e d b y t h e O r i e n t a l D e p a r t m e n t of t h e S t a t e H e r m i t a g e
6
E. . : -
-
(. I). - , ,
1964, 83.
Ibid., 8385 ; M. . : , 39.
7
. . , -. . -
, . 4, ., 1946, . 5256;
the same author, . . . , 85 99.
8
See H . F I E L D and E . P R O S T O V : Excavations in Uzbekistan. 1 9 3 7 1 9 3 9 . Ars
Islamica vol. I X , 1942 pt. 1 2, 144, Figs 6 8 ; . . , Jl. . :
. 1965. 5253, Fig. 20; . . :
. In : 1971
1972, , 1972, 6 9 - 7 0 .
" E . G . P C H E L I N A summed up the results of the 1 9 3 7 excavations in her Master of
Sciences thesis which she maintained in 1944 a t tho historical department of Moscow
State University. (See E. . : -. ).
7* Acta Antiqua Academiae Scientiarum Hungaricue 28, 1980

98 . STAVISKIY
M u s e u m . Since 1962, t h e State Museum of Oriental A r t s has also t a k e n p a r t in

its organization. I n 1968 t h e above-mentioned i n s t i t u t i o n s were joined b y t h e
U S S R Central R e s e a r c h L a b o r a t o r y on t h e Conservation a n d Restoration of
o b j e c t s of art ( W C N I L K R ) , and in 1977 by t h e S t a t e Museum of the History
of Religion and A t h e i s m (GMIR).
Besides archaeologists and restorers, m a n y other specialists (on oriental
languages, history of a r t , architecture, etc.) t a k e p a r t in t h e annual excavation
seasons of the joint expedition. I t s f i n d s are studied b y researchers from t h e
sponsoring institutions, and from t h e I n s t i t u t e of Oriental Studies of the U S S R
A c a d e m y of Sciences a n d its Leningrad branch, f r o m t h e Academies of Sciences
of t h e T a j i k and U z b e k republics, as well as by specialists f r o m Hungary, India,
t h e U S A and West G e r m a n y . The results of t h e research work done by this v a s t
collective have been published in over a hundred articles in Soviet and Western
publications. A special collection of articles is also p u t o u t devoted to the work
of t h e expedition (Materials of t h e J o i n t Expedition). T h e four issues already
published cover t h e period from 1961 to 1973, 10 and t h e f i f t h which is now
being prepared for t h e press will deal with the 1974 1977 seasons. 11
B U D D H I S T COMPLEXES OF KARA T E P E
I N T H E L I G H T OF T H E 1 9 6 1 - 7 7 EXCAVATIONS
The excavation work of our predecessors and e x c a v a t i o n s of 1961 77

covered only one of K a r a tepe's t h r e e tops t h e South (or, t o be more exact its
exposed slope) r e m i n d i n g of a horseshoe in plan (Fig. 2). 12
On the e a s t e r n p a r t of t h e slope t h r e e caves (Nos. 1, 2 and 3) were dis-
covered in 1928 a n d 193436. N o r t h of t h e m , two more caves were later found
b y E . G. Pchelina. E x c a v a t i o n of t h e l a t t e r caves was carried on by our expedi-
tion. These are caves P - I and P - I I ( P for t h e Russian peshchera cave, w i t h
t h e numbers d e n o t i n g t h e order in which t h e caves were discovered during o u r
1961 64 seasons).* Six other caves are being e x c a v a t e d on t h e northern slope :
caves 1, 2, 3 (or P - I I I ) and 4 of t h e complex and P - I V a n d P V caves. T h u s
t h e r e are eleven caves t h a t we know of on t h e southern t o p on less t h a n half of
its exposed slope. Consequently, we h a v e every ground t o believe t h a t t h e r e
m i g h t have been 25 30 caves all told on t h e t o p .
10
Materials of t h e joint expedition: I -
, 1964; II - -
, 1969; III - -
, ., 1972; IV - , 1975 (And
f u r t h e r -, I; -, II, -, III; a n d -, IV).
11
For brief d a t a on Kara-tepe excavations in 1974 1977 see
1974 , M., 1975, 512-513; 1975 , M., 1976,
535 536; 1976 , ., 1977, 543 - 5 4 4 ; -
1977 , ., 1978, 538 - 539.
12
In 1963 a cave was incidentally found on the northern t o p of the hill. However,
its p l a n and purpose is still to be discovered by special excavations.

KARA T E P E IN OLD TERMEZ 99
Fig. 2. Plan of the excavations of the southern t o p of K a r a tepe
As excavations and studies of t h e microrelief of t h e hill slopes have shown,

t h e K a r a tepe caves were n o t isolated b u t constituted p a r t of a complex which
also included a large square (or rectangular) courtyard a n d other ground struc-
tures. Such mixed ground and cave complexes might be s i t u a t e d next t o each
o t h e r or separated b y v a c a n t plots. T h e excavations of 1961 1977 brought o u t
more or less distinctly four complexes , , and D.*
A AND COMPLEXES
The A a n d complexes, isolated a t a later period, were originally incor-

p o r a t e d into a single group of s t r u c t u r e s on the southern t o p . This group in-
cluded three courtyards which were placed next to one another on a t e r r a c e
along t h e eastern slope from n o r t h t o south.
I n the middle of t h e n o r t h e r n c o u r t y a r d the remains of a s t u p a m a d e of
m u d brick and r a m m e d clay (pakhsa) a n d faced with tiles of white lime were
uncovered. S t u d y of t h e remains of t h e s t u p a and its facing, a f r a g m e n t of its
cupola, and of t h e stone umbrellas (chatra), as well as t w o identical representa-
tions of a s t u p a scratched on a wall in cave 2 (on t h e eastern slope) p r o m p t e d
* The subjoined charts give Russian letters to denote cave temples a n d complexes :
for P , and , , , for , , C, D respectively.

100 . S T A V I S K I Y
1 2 m
Fig. 3. A s t u p a a t Kara tepe. Reconstruction by I. L u n k o v a
our architect I. P . L u n k o v a to trace a graphic reconstruction of t h e s t u p a (Fig.

3). T h e s t u p a stood in an open c o u r t y a r d exposed to rain a n d snow. To p r o t e c t
i t f r o m being washed a w a y special water-consuming devices were made a t t h e
foot of t h e stupa a n d t h e water brought b y spring cloud-bursts was evidently
d r a i n e d from the y a r d t h r o u g h a ceramic t u b e (koubour) b u i l t into the base of
t h e e a s t e r n wall of t h e c o u r t y a r d .
Unlike the n o r t h e r n courtyard, t h e central and t h e southern ones were
bordered with porticos (aiwans). Their wooden columns placed on stone Attic-
t y p e bases supported f l a t wooden ceiling which protected t h e greater p a r t of
Fig. 4. The c o u r t y a r d of the complex. Reconstruction b y I . Macheret
Aria Antiqua Academiae Scientiarum Hungaricae 28, 1980

K A R A T E P E I N OLD T E R M E Z 101
t h e courtyard f r o m the sun, rain or snow. Only t h e central p a r t of the c o u r t y a r d

was open b u t its level was lower t h a n t h a t of the aiwans and t h e water casually
accumulated t h e r e did not flood t h e porticos (Fig. 4).
Fig. 5a. The staircase leading f r o m the southern courtyard up hill
The western aiwans of t h e central and southern c o u r t y a r d s adjoined t h e

vertically cut off slope of the hill, in which niches were hewn f r a m i n g e n t r a n c e s
t o P - I I and P - I caves. Each cave h a d t w o entrances s i t u a t e d close to t h e s o u t h -
western and north-western corners of t h e courtyards. I n t h e central c o u r t y a r d
t h e r e was one more niche in the middle of t h e western aiwan, in which p r o b a b l y
a big B u d d h i s t s t a t u e stood.
To the n o r t h of t h e northern entrances into t h e caves in b o t h c o u r t y a r d s
cave cells were discovered : one in t h e southern and t w o in t h e central c o u r t -
y a r d , one above t h e other. N a r r o w entrances with steps descending into t h e
cells led to t h e m f r o m t h e aiwans. T h e similar passage with steps connected t h e

102 . STAVISKIY
t w o cells in the central c o u r t y a r d . The u p p e r cell and t h e one in the southern

c o u r t y a r d had couches (sufas) along t h e walls c u t out, as t h e cells themselves,
in s a n d s t o n e (Fig. 22). T h e r e were flights of stairs above t h e cells t h a t led from
t h e n o r t h e r n aiwans of t h e central and s o u t h e r n courtyards u p t o the upper
p a r t of t h e hill side or u p t o t h e upper premises (Fig. 5).
Fig. 5b. The staircase leading f r o m t h e central courtyard up hill
All these stairs, as well as the few steps from t h e aiwans to t h e lower,
c e n t r a l , p a r t of t h e c o u r t y a r d s were paved with plates of b a k e d brick, evidently
t o p r o t e c t sandstone a n d mudbrick steps f r o m damage.
There was a c h a m b e r between t h e central and t h e southern courtyards.
Originally it gave access t o both yards, t h u s connecting t h e A and complexes.
L a t e r , however, it was isolated from t h e c o u r t y a r d s : f r o m t h e southern b y a
p l a s t e r e d mud-brick wall, and from t h e central b y a mud-brick partitioning.
I t w a s undoubtedly a place of worship. T h e remains of a square dais faced with

blocks of white lime were uncovered in t h e centre of t h e chamber. Floor space

around t h e dais was also paved with f l a t plates of white lime. Besides, f r a g m e n t s
of stone umbrellas were found here which confirmed t h e assumption t h a t a
votive s t u p a (probably a diminutive replica of the n o r t h e r n courtyard s t u p a )
stood in t h e centre of this relatively small chamber.
To t h e east of it (in t h e south-eastern corner of t h e central courtyard) was
a n o t h e r chamber, most probably a m o n k cell. True, t h e r e was no couch in it,
b u t its owner might well make himself comfortable on m a t s behind t h e clay
wall overlaid with plaster (alabaster) a n d painted, which was built near t h e
e n t r a n c e a n d partitioned off the n o r t h - e a s t e r n corner of t h e cell.
Another ground s t r u c t u r e was e x c a v a t e d to t h e s o u t h of the s o u t h e r n
c o u r t y a r d , which E . G. Pchelina, being t h e first to discover it, defined as a
B u d d h i s t temple. I t consisted of a cell a n d three corridors surrounding it (east-
ern, southern and western). The s o u t h e r n aiwan of t h e southern c o u r t y a r d
which adjoined t h e cell on t h e f o u r t h side and communicated with it and t h e
corridors closed t h e ring, so to say. T h e floor level of the cell was lower t h a n t h a t
of t h e aiwan and the e n t r a n c e in to t h e cell had a high threshold m a d e of blocks
of white lime. The blocks were originally adorned with a contour flower design
a n d later plastered over and painted red. T h e p a t h leading to t h e e n t r a n c e was
p a v e d with white lime plates (Fig. 6). T o t h e east of t h e e a s t e r n corridor t h e r e
was a small cell with a couch made of m u d bricks.
The P - I and P - I I cave3, which were evidently cave temples, in general
outline repeated t h e ground temple s t r u c t u r e . Each consisted of a cell and v a u l t -
ed corridors t h a t encircled it (Fig. 24). As mentioned above, each cave t e m p l e
h a d a cell n e x t to one of its entrances which housed a m o n k , a guard and a t t e n -
d a n t of t h e Buddhist s a n c t i t y . I n detail, however, the P - I a n d P - I I cave temples
differed b o t h from t h e ground temple a n d from one a n o t h e r .
Thus, as distinct f r o m the ground temple, the cave temples had four cor-
ridors each. The cell of t h e P - I temple h a d a floor space of over 28 sq.m., high
(not less t h a n 4 metres) vertical walls a n d , evidently, a wooden ceiling, while
t h e floor space of the central p a r t of t h e P - I I cave was u n d e r 12 sq.m. and it h a d
a v a u l t e d ceiling cut in sandstone. T h e e n t r a n c e into t h e P - I cell was in t h e
eastern wall and t h a t of t h e P - I I cell in t h e northern wall. I n t h e north-west-
ern corner of t h e P - I t e m p l e there was a small cave, m o s t probably a store-
room. I n t h e P - I I cell excavations revealed (in the southern wall, and southern
p a r t s of t h e adjoining walls) small sockets f r o m timbering of a wooden dais, t h e
like of which m a y be f o u n d in B u d d h i s t cave temples of E a s t e r n T u r k i s t a n
(Sinkiang) and on which Buddhist s t a t u e s and oil lamps evidently stood. I t s
f r o n t (northern) surface might have been adorned with reliefs. 1 3
13
See A. G R N W E D E L : Altbuddhistische Kultsttten in Chinesisch Turkistan.
Berlin 1912. 3 1 6 - 3 1 7 , Figs 636, 636a.

104 . S T A V I S K I Y

KARA T E P E I N OLD T E R M E Z 105
I t should also be n o t e d t h a t walls in almost all ground and cave premises

of t h e A and complexes were painted over the p l a s t e r (mostly with red p a i n t ) .
I n niches, cells and corridors of cave temples the red panels o f t e n hid ornamen-
t a l geometric or floral designs. I t is only a t the e n t r a n c e s in to the ground tem-
ples (the one with t h e small s t u p a in t h e southern a i w a n of t h e complex court-
y a r d and t h e temple in t h e southern aiwan of the A complex courtyard) t h a t
f r a g m e n t s of compositions depicting Buddhist personages and d o n a t o r s were
found.
The A and complexes were, evidently, connected in some way with t h e
ground structures t h a t were built above the P - I a n d P - I I temples. W e m a y
merely conjecture t h e existence of such a structure a b o v e the P - I temple for it
has not been b r o u g h t o u t y e t . However, in the course of t h e excavation of t h e
P - I cell m u d brickwork was uncovered over the t e m p l e ' s sandstone walls, ap-
p a r e n t l y t h e remains of t h e walls of t h e ground s t r u c t u r e . The s t r u c t u r e above
t h e P - I I ground t e m p l e has been fully excavated (Fig. 7). I t s lower p a r t was cut
in t h e hill's sandstone slope, with t h e southern and n o r t h e r n walls coinciding in
p l a n with t h e walls of t h e southern and northern corridors of the cave temple,
which m a y t e s t i f y t o their simultaneous construction.
The s t r u c t u r e over t h e P - I I temple was s i t u a t e d n o t far from t h e southern
t o p of t h e hill. I t was entered not f r o m below (from t h e complex) b u t f r o m
above, a short flight of stairs leading down the slope of t h e southern t o p into
t h e premises. On t h e right of t h e stairs, inside the a n t e c h a m b e r there was a big
jug (khum) dug into t h e ground, and on t h e left, also sunk in t h e ground a
m u d brick p l a t e with holes in it, a kind of wash-stand.
The jug was, a p p a r e n t l y , filled w i t h water for r i t u a l ablutions, which were
performed over t h e wash-stand. Though merely a symbolic operation, it
nevertheless called for certain measures to be t a k e n t o drain water o u t of t h e
building, and, w h a t was probably most i m p o r t a n t , f u r t h e r away from t h e tem-
ple below. A r e c t a n g u l a r reservoir was m a d e for t h e p u r p o s e under t h e wash-
stand plate, connected with t h e system of eight k o u h o u r s which were laid
towards the n o r t h e r n o u t e r wall of t h e building a n d d r a i n e d water beyond it
onto t h e v a c a n t lot on t h e northern slope of the s o u t h e r n hilltop. I t is peculiar
t o note t h a t t h e inclination of the drainage system w a s 3, i.e., fully conformed
t o t h e modern s t a n d a r d s .
The K a r a tepe a n t e c h a m b e r with the wash-stand brings to m i n d t h e
G a n d h a s a l a mentioned in B u d d h i s t t e x t s , though t h e analogy m a y be slightly
far-fetched. Besides t h e a n t e c h a m b e r , t h e r e were six m o r e rooms in t h e building,
one of t h e m an elongated chamber stretching f r o m s o u t h to north along t h e
western side of t h e building (Fig. 8). A t t h e n o r t h e r n wall of t h e chamber t h e r e
was a wide sufa plastered with clay a n d alabaster. T h e room was probably a
K a t h i k a s a l a where people gathered t o be t a u g h t a n d t o discuss themes con-
nected with B u d d h i s t t e x t s . If so, t h e monk-teacher (probably even B u d d h a s i r a

106
. STAVISKIY
Fig. 7a. The s t r u c t u r e above the P - I I cave-temple a) pia;
Fig. 7b. The s t r u c t u r e above t h e P - I I cave-temple b) cross-sections

Fig. 7. The structure above the P - I I cave-temple c) reconstruction by I. L u n k o v a
d h a r m a k a t h i k a himself, who was mentioned in a K a r a tepe inscription on

ceramics) 14 occupied t h e sufa, and in f r o n t of him, on m a t s , s a t his pupils or
monks.
Of special interest were e x c a v a t i o n s of a small room n e x t to t h e a n t e -
chamber, to t h e south of its eastern p a r t . Originally a communicating passage,
an extension of t h e antechamber, it l a t e r became a cell. Before it was t u r n e d
into a cell, it had three entrances : o n e from the a n t e c h a m b e r , another f r o m
t h e lecture-room (Kathikasala), a n d t h e third from t h e aiwan-type s t r u c t u r e
fringing t h e building on the east. T h e r e was an altar a t t h e western wall of t h e
chamber an a t t a c h e d niche with a shell-shaped back wall, in which a f l a m e ,
eternal or kindled f r o m time to time, b u r n e d . The back wall and t h e b o t t o m of
t h e niche were scorched and t e m p e r e d a n d the hollow for t h e f l a m e a t t h e bot-
14
See below.

Fig. S. The western chamber of the structure above P - I I
K A R A I N OLD T E R M E Z 109
torn filled to t h e brim with pure grey ash. The a l t a r was used during a long pe-
riod, for we have found traces of over 18 repairs a n d reconstructions (Fig. 9).
Subsequently t h e room with t h e a l t a r was reconstructed : t h e e n t r a n c e
from t h e lecture-room was walled, a clay screen was p u t before t h e e n t r a n c e
f r o m t h e antechamber, and, besides, against t h e western wall of t h e room,
a sufa was built, in which f r a g m e n t s of the altar niche were immured. The a l t a r -
room was t h u s t u r n e d into an abode, more c o m f o r t a b l e and rich t h a n t h e cells
next to the P - I and P - I I temples, t h e A complex g r o u n d temple and t o t h e
complex premises with t h e s t u p a . The s t a t u s of t h e owner of the former a l t a r -
room was a p p a r e n t l y much higher t h a n t h a t of t h e m o n k s inhabiting t h e above-
mentioned cells. H e might have been a t u t o r or B u d d h i s t preacher (probably
B u d d h a s i r a d h a r m a k a t h i k a we have already spoken about), who a c t u a l l y
had t h e whole of six or seven-room building a t his disposal.
N a t u r a l l y t h e explanation given here of t h e p u r p o s e of the premises de-
scribed,including the ground s t r u c t u r e near the hilltop, is no more t h a n a h y p o t h -
esis, b u t it is based on t h e latest d a t a , and, therefore, has the right t o life.
The existence and functioning for a long period of a fire altar in a B u d d h i s t
s t r u c t u r e is significant. The fire a l t a r with its a t t a c h e d niche m a y definitely
traced to t h e fire-places found in t h e Graeco-Bactrian t o w n of Ai-Khanoum. 1 5
However, it would be wrong to conclude t h a t it was a warming device. F i r s t ,
unlike the fire-places in A i - K h a n o u m which are r a t h e r common there a n d m a y
even be found in neighbouring premises, t h e K a r a t e p e a l t a r is t h e only one
discovered on t h e site so far. Second, traces of f r e q u e n t repairs t e s t i f y t o t h e
f a c t t h a t it was an object of special concern. And, t h i r d , w h a t gives s u f f i c i e n t
ground to speak of its ritual n a t u r e is its similarity to fire a l t a r s of t h e K u s h a n
period found in G y a u r - K a l a (Khorezm) 1 6 and in S a m a r k a n d (excavations u n d e r
A. I. Terenoshkin), 1 7 as well as to altars of a later period from T o p r a k K a l a
in Khorezm 1 8 (3rd4th centuries) and from P e n j i k e n t in Soghd 19 (6th 8th
centuries). I t is probable t h a t Bactria-Tokharistan of t h e K u s h a n period was
t h e first region in Central Asia t o use the Graeco-Bactrian-type fire-places 2 0 as
13
See Fouilles d' A K h a n o u m , I/MDAFA, t. X X I / P . 1973, 283, 284, CHAIBL
1969, P., 1970, 318; CHAIBL - 1970, P., 1970, 307, C H A I B L 1974, P., 1974, 284,
C R A I B L 1975, P., 1975, 172; . . : : -
. 1977. 153 a n d Fig. 21, , b. and Fig. 95,2 on p. 206.
10
. . . . : -.
- , . II, , 1958, 359 360.
17
See . . :
. . , IX, , 1953, 27 29.
18
. . : -
1949 1953 . - . II.,
19
See . : . -
. 124. .-. 1964, 67 72; . . : -
, 1973, 14.
20
I t should bo remoinbored, naturally, that domestic hearths in antiquity (in
t h e Hellenic world, as well as in the East) were also connected with worshipping, of t h e
fire as a home p a t r o n god, etc.

Fig. 9. The altar of the structure above P - I I a) a t the earliest stage
Fig. 9. The altar of the structure above P - I I b) a t the latest stage
112 . S T A V I S K I Y
Fig. 9. The altar of t h e s t r u c t u r e above P - I I c) a t the three principal stages.

Reconstruction b y I. Lunkova
a l t a r s . I n t h a t period, t h e borrowed Hellenic traditions were undergoing trans-

f o r m a t i o n under t h e i m p a c t of ethnic, historical and cultural factors. Altars of
t h i s k i n d were used b y f i r e worshippers and in temples of t h e local Mazdean
cults, a n d also in early B u d d h i s t s t r u c t u r e s .
According to t h e d a t a acquired in T o p r a k K a l a and P e n j i k e n t , the tradi-
t i o n of building such a l t a r s t o be used b y local cults survived u p to t h e period of
A r a b conquest. As t o t h e K a r a tepe fire altar, its f a t e m a y serve as evidence of
t h e change in a t t i t u d e of t h e B u d d h i s t c o m m u n i t y (or, a t least, t h e Buddhist
c o m m u n i t y of the K u s h a n Termez) t o f i r e worshipping as early as t h e 3rd 4th
centuries. At first t r e a t e d with respect (or tolerance) it was finally destroyed
a n d on its ruins a sufa was made, s o m e w h a t demonstratively, we m a y presume,
u p o n which sat or slept t h e high-placed o c c u p a n t of the room.
T h e fact t h a t t h e ground s t r u c t u r e was placed above t h e Buddhist cave
t e m p l e is peculiar a n d m a y be the object of special s t u d y .
AND D COMPLEXES
T h e excavation of t h e and D complexes on the n o r t h e r n slope of t h e

s o u t h e r n hilltop of K a r a t e p e is far f r o m completion. One of t h e caves of the
complex P - I I I (now cave 3) was discovered in the course of exploratory dig-

ging in 1962, a n d t h e D complex cave P - V in 1964. B u t systematic ex-

cavations of t h e complex have been going on since 1971, and of the D com-
plex since 1976.
T h e D complex is, evidently, similar in its composition and planning t o
t h e A and complexes. The central p a r t of its c o u r t y a r d was also s i t u a t e d on
a lower level and surrounded along its perimeter with aiwans. The wooden col-
u m n s of t h e aiwans were placed on stone Attic-type bases. T h e southern ai w a n ,
which has been completely excavated, adjoined, like t h e western aiwans of t h e
A and complexes, t h e vertically cut off slope of t h e s a n d s t o n e hill and was
similar to those aiwans : there were vaulted niches a t its b o t h ends w i t h en-
t r a n c e s to the P - V cave, and in t h e centre there was a big niche for a B u d d h i s t
s t a t u e . On t h e right of t h e western e n t r a n c e to the cave t h e r e was the e n t r a n c e
t o t h e cave cell of a monk a t t e n d a n t . The construction of t h e cell was not com-
pleted, however. H a v i n g hewn o u t t h e vaulted passage w i t h steps leading d o w n
f r o m t h e c o u r t y a r d into t h e cell and p a r t of t h e cell, t h e builders discontinued
t h e work, most probably, from fear of a possible caving-in, for they h a d over-
e s t i m a t e d t h e height of t h e hill a t this particular spot. T h e entrance was t h e n
walled with m u d bricks, plastered with clay and p a i n t e d w i t h red p a i n t . T h e
central niche (as distinct from t h a t of t h e complex) h a d a hiding-place a
small cave behind t h e thin sandstone back wall of t h e niche.
T h e P-V cave (most probably a cave temple), like t h e P - I and P - I I t e m -
ples, h a d two entrances, and four vaulted corridors. W h e t h e r t h e corridors sur-
rounded a sandstone-filled space or a cell as in the P - I a n d P - I I temples is y e t
t o be seen. I t will be m a d e clear by f u r t h e r excavations. T h e r e was a g r o u n d
s t r u c t u r e above t h e P-V cave, too. However, excavations h a v e revealed only
two rooms, t h e n a t u r e and purpose of which remains u n k n o w n .
T h e complex, s i t u a t e d between the and D complexes, differed sub-
s t a n t i a l l y from t h e l a t t e r two (and t o a lesser degree f r o m t h e A complex). O n l y
p a r t of t h e c o u r t y a r d has been u n d u g as yet, and it is, therefore, too early t o
d r a w conclusions. B u t as far as one could judge, there were no aiwans in i t , a t
least no t r a c e of columns has been found in t h e southern p a r t of the c o u r t y a r d .
A n d w h a t is more, instead of a single cave temple with a cell a n d four corridors,
t h e complex h a d four (or five) isolated caves. Three of t h e m ( 1 , 2 and 3) were
long and narrow chambers with vaulted ceilings. Caves 1 a n d 3 ended in a niche
(there was a small cell, evidently a hiding place, behind t h e niche in cave 1).
Cave 2, judging b y its s i t u a t i o n , is similar in character b u t its excavation h a s
n o t y e t been completed. The planning of cave 4 is n o t y e t clear so far. I t h a d
either an offshoot or (and this is more probable) it c o m m u n i c a t e d with t h e cave
corridor t h a t led t o t h e westernmost (the sixth) vaulted niche of the complex
c o u r t y a r d . I n t h e l a t t e r case cave 4 was h-shaped in p l a n (with a niche a t t h e
end of t h e eastern corridor, extending it to t h e south f r o m i t s juncture with t h e
southern corridor), its t h r e e sides (corridors) f r a m i n g t h e f i f t h vaulted niche of

114 . S T A V I S K I Y
t h e courtyard, or v a u l t e d aiwan (Fig. 25). Such aiwans were t y p i c a l not so m u c h

f o r Central Asia, where aiwans with columns were more c o m m o n , b u t for t h e
N e a r E a s t and I r a n , particularly in t h e P a r t h i a n and S a s a n i a n periods. There
w e r e six such niches in t h e complex c o u r t y a r d along its s o u t h e r n fringe.
The purpose of t h e complex caves is n o t y e t clear. H o w e v e r , their gener-
al s t r u c t u r e and t h e f a c t t h a t there were no sufas or a n y t r a c e s of hearths in
t h e m are all evidence t h a t t h e y were n o t dwelling premises. T h e y were h a r d l y
u s e d as storerooms or for a n y other household purposes. I t is m u c h more likely
t h a t t h e caves were places of worship a n d t h e niches in their walls held statues,
reliquaries or B u d d h i s t relics.
W h a t we m a y definitely assert, however, is t h a t t h e K a r a tepe caves and
complexes in their s t r u c t u r e did n o t follow a strict p a t t e r n . Though identical
i n m a n y respects, t h e y differed in details.
TRACES OF RECONSTRUCTIONS, A BURIAL GROUND AND

MORE RECENT L A Y E R S ON KARA T E P E
I n the course of excavations, E . G. Pchelina and, l a t e r , our expedition

c a m e across traces of numerous reconstructions in all t h e f o u r complexes de-
s c r i b e d above. N o t going into details, we m a y assert t h a t these reconstructions
(some of them m e n t i o n e d earlier) which m i g h t have been explained by t h e need
t o m a k e repairs or t o p r e v e n t caving-in, a n d sometimes b y reasons not clear t o
us, never brought a b o u t changes in t h e c h a r a c t e r and p u r p o s e of t h e K a r a t e p e
complexes as a B u d d h i s t religious centre. N o t e should be t a k e n , though, of one
i n s t a n c e of a reconstruction of a d i f f e r e n t order, t h a t of t h e c e n t r a l niche in t h e
w e s t e r n aiwan of t h e complex.
This grand niche between t h e t w o entrances to t h e P - I I cave temple, like
t h e similar one in t h e southern aiwan of t h e D complex, m o s t probably held
a b i g Buddhist s t a t u e (as we pointed o u t earlier). However excavation of t h e
n i c h e first revealed a mud-brick wall, b u i l t over a layer of debris, pieces of m u d
b r i c k and sand a n d screening i t f r o m t h e courtyard, and, t h e n , when t h e niche
itself was uncovered, a h e a r t h circular in plan. I t was b u i l t w i t h m u d bricks
o v e r clay and ganch plaster fallen f r o m t h e arched upper p a r t of t h e niche and
n o t i n the middle of i t b u t closer t o its l e f t - h a n d (southern) corner. This was,
e v i d e n t l y , a fire a l t a r : t h e h e a r t h was b u r n e d out inside a n d filled with fine
w h i t i s h ash, which contained no pieces of coal, bones, or f r a g m e n t s of p o t t e r y
a l w a y s found in domestic ovens. I t s position too shows t h a t it was used for
religious and not domestic purposes. W e m a y well believe t h a t a f t e r the com-
p l e x was destroyed or fallen to desolation a fire altar was b u i l t in t h e niche
w h i c h had previously held a B u d d h i s t s t a t u e . I t was m a d e h a s t i l y , out of m a t e -
r i a l s t h a t were a t h a n d and symbolized t h e victory of some o t h e r cult (most like-
ly Zoroastrian) over B u d d h i s m . W h o walled t h e niche w i t h t h e fire a l t a r and

why is not k n o w n . B u t whoever he might be he did not destroy this fragile

symbol of fire worship, b u t r a t h e r protected it f r o m destruction or desecration.
The period of desolation of the K a r a t e p e site is connected with t h e layer
discovered directly over t h e floors of all ground a n d cave structures. T h e layer
was n o t even. I n t h e caves i t was thinner t h a n in t h e ground premises, a n d
thicker a t t h e e n t r a n c e s into t h e caves (50 60 cm. above the floor level) t h a n
inside (510 cm). I t s m a x i m u m thickness exceeded 80 cm. The layer included
d u s t , blown-in s a n d and debris plaster fallen down from walls and ceilings
and sometimes f r a g m e n t s of t h e upper p a r t of s a n d s t o n e or mudbrick walls or
of t h e ceilings. R a i n s or r a i n y seasons l e f t m a r k s on t h e clay over t h e layer or
inside it.
I t is not unreasonable to suppose t h a t n u m e r o u s inscriptions and drawings
scratched on t h e plastered walls of the caves a n d their entrance niches refer t o
t h e period of desolation. Some of these graffiti, in f a c t , m a y be dated b y a m u c h
l a t e r period, t h e Moslem Middle Ages (most p r o b a b l y t h e 10th 12th centu-
ries). However, t h e n a t u r e and linguistic peculiarities of t h e inscriptions m a k e
i t possible to distinguish between those left b y t h e Moslems and those m a d e
earlier, in the period immediately following t h e d e s t r u c t i o n or decline of t h e site.
Above t h e period of desolation layer E . G. P c h e l i n a and our e x p e d i t i o n
discovered, more t h a n once, h u m a n bones. A n u m b e r of burials found g a v e u s
ground to believe t h a t finds of single bones were also remains of burial grounds,
r a t h e r t h a n a result of some casualty.
The burials in t h e desolated caves and ground premises of K a r a t e p e
sometimes contained ten or more corpses laid in rows a n d even several rows one
above the other. I t m a y be t h a t t h e upper rows in some cases presented l a t e r
burials, b u t it seems t h e greater p a r t of the burials were simultaneous affairs,
most probably connected (according to t h e special analysis of the bones m a d e
by paleopathologist Z. B. Altman) 2 1 with some epidemics, and not w i t h mass
slaughter, as t h e legend explaining the sinister n a m e of K a r a tepe has it.
I n t h e b e t t e r preserved burial we also f o u n d p o t t e r y , remains of clothes,
a d o r n m e n t s and other things (bracelets, rings, ear-rings, a gold p e n d a n t , a
brooch, an iron knife, mirrors, etc.). The objects t h a t went into the burials were
sometimes quite valuable. I t should be noted t h a t a coin or even two were o f t e n
p u t into t h e h a n d of t h e deceased. W h a t might be added is t h a t t h e e n t r a n c e s
i n t o a l l t h e premises, where burials or their r e m a i n s occurred, had been walled.
I n such cases we invariably discovered, above t h e period of desolation layer,
the remains of unplastered crudely built walls laid o u t of mud bricks or t h e i r
fragments. Sometimes stone blocks t h a t had faced t h e stupa, or p r o b a b l y t h e
terrace, were used for t h e purpose, and once even a capital of an o r n a m e n t a l
pilaster.
21
3. . :
- -. III, 1972, 104-110.

116 . S T A V I S K I Y
A p a r t from group burial grounds of t h e t i m e immediately following t h e

period of desolation (most probably t h e walls a n d ceiling were still i n t a c t then),
e x c a v a t i o n s of some cells and chambers b r o u g h t out, in the layers high above
t h e f l o o r level, single h u m a n skeletons or bones of the Moslem Middle Ages
period. T h e y are all in a poor s t a t e of preservation, therefore one could only
guess w h e t h e r these are burials or casualties.
I n cases when it is possible to t r a c e t h e stratification of t h e caves' filling,
it becomes apparent t h a t t h e layers with skeletons and bones of t h e 10th 12th
c e n t u r i e s contain also p o t t e r y of t h e same period and coincide in t i m e with t h e
d e s t r u c t i o n of the walls a n d ceilings of t h e caves. I t is through gaps f o r m e d a f t e r
t h a t t h e later visitors (up to A. S. Strelkov a n d G. . Parfenov, for t h a t m a t t e r )
e n t e r e d filled or half-filled caves.
MAJOR F I N D S
T h e K a r a tepe excavations b r o u g h t o u t wall paintings, stone a n d stucco
(ganch in local terminology) sculpture, t e r r a c o t t a s and pottery, m e t a l w a r e and
coins, inscriptions on p o t t e r y and g r a f f i t i on t h e wall of t h e caves a n d their
e n t r a n c e niches.
WALL PAINTING
I n almost all of t h e K a r a tepe cave or ground premises e x c a v a t i o n s un-

covered traces of red or black p a i n t on t h e walls, and, in a n u m b e r of cases,
t r a c e s of ornamental colour design : geometric or, rarely, flower p a t t e r n s or
panels (usually m a d e with red paint) on w h i t e ganch plastering^
A representation of a h u m a n figure was f i r s t found here in 1937 b y E . G.
P c h e l i n a who discovered (on t h e wall of t h e A complex ground temple, a t t h e
e n t r a n c e into the cell) a f r a g m e n t of wall p a i n t i n g showing two feet in a fro..tal
position w i t h toes t u r n e d outward and a border of a robe above t h e m coloured
r h y t h m i c a l l y with spots of white, light blue, blue, pink, red a n d brown. 22
T h e p a i n t i n g has n o t survived, b u t , besides t h e description, we h a v e a black-
a n d - w h i t e photograph of i t . J u d g i n g b y t h e footwear (soft l e a t h e r boots) and
t h e b o r d e r of a rich robe, we m a y presume t h a t w h a t E . G . P c h e l i n a f o u n d was
a f r a g m e n t of a painting depicting a K u s h a n nobleman as a B u d d h i s t d o n t o r .
T h i s view was b o r n e o u t in t h e course of our excavations in 1971 71.
F r a g m e n t s of multi-figure compositions were found in the s o u t h e r n aiwan of
t h e complex temple c o u r t y a r d . One of t h e m represented a group of donators,
t h e o t h e r the B u d d h a with monks. Like t h e wall painting discovered b y E . G.
P c h e l i n a , these compositions adorned t h e wall space next to t h e e n t r a n c e into
a p l a c e of worship, in this case t h e c h a m b e r w i t h t h e votive s t u p a .
T h e f r a g m e n t of t h e f i r s t composition f o u n d on the wall west of t h e en-
t r a n c e showed the feet of seven donators : f i v e men in soft l e a t h e r boots laced
22
E. . : . . ., -. I. M. 1964. 89.

w i t h s t r a p s and buckled a t t h e side, and two women in long robes from u n d e r

which only their toes showed (Fig. 10). The donators stood in a line, in some
cases only slightly covering one a n o t h e r . Their feet are placed in a f r o n t a l posi-
tion (as on the f r a g m e n t found b y E . G. Pchelina). I n t h e s a m e strictly f r o n t a l
position are represented the feet a n d t h e robes of t h e women. Drawing on t h e
analogies in representations of d o n a t o r s and other K u s h a n personages in sculp-
ture 2 3 and in imperial K u s h a n coinage (the figure of a K i n g before the altar) 2 4
we m a y assume t h a t a rigid strictly f r o n t a l posture was typical for d o n a t o r s on
t h e K a r a tepe wall paintings. I t should be noted t h a t t h e heads of t h e d o n a t o r s
(or a t least of some of them) are shown in profile (as those of Kings on t h e
K u s h a n coins). I t m a y be concluded f r o m a f r a g m e n t of t h e painting t h a t h a d
evidently slid down from t h e wall a n d was found n e x t t o it. The head we see
on t h e f r a g m e n t is t u r n e d to t h e l e f t . I t is t h a t of a m a n w i t h bald head f r a m e d
with grey hair, bearded and w i t h moustache, and w i t h an ear-ring. His face
a b o v e t h e sharply outlined collar of his robe is calm and concentrated, his
longish eyes turned upwards (Eig. 11).
The dontor (and m a y be all t h e donators on this painting) was p r o b a b l y
shown looking a t t h e composition with t h e B u d d h a , which was placed on t h e
s a m e wall on the left (east) of t h e f i r s t painting and, evidently, on a higher level,
A peace of plaster with a f r a g m e n t of t h e composition depicting the B u d d h a
and monks was found face down on a layer of the debris which had accumulated
on t h e floor of t h e aiwan to t h e e a s t of t h e entrance. T h e p a i n t i n g was cleared
a n d fixed on t h e spot and t r a n s p o r t e d t o Moscow to t h e W C N I L K R , where its
restoration was completed and t h e p a i n t i n g was prepared for an exposition in
t h e S t a t e Hermitage Museum.
The restored f r a g m e n t (Figs. 12ab) reveals figures of t h e B u d d h a sitting
u n d e r a tree with rich folliage (on t h e right) and three m o n k s standing each
u n d e r a tree (on t h e left). I t is highly probable t h a t it is only a p a r t of the scene
which might have included t h r e e o t h e r monks placed symmetrically on t h e
B u d d h a ' s right. The scene is identical with the Miran p a i n t i n g (Eastern Turkis-
tan) 2 5 b u t is executed in a style closer to the Indian m a n n e r . I t is peculiar t h a t
in t h e K a r a tepe wall painting t h e B u d d h a has not merely a nimbus b u t a halo
a r o u n d t h e body, which is p r o b a b l y to be traced not to t h e I n d i a n , b u t to t h e
local, Bactrian, artistic tradition (see further).
23
See, e.g. J . K O S E N F I E L D : T h e Dynastic Arts of the K u s h a n s . Berkeley 1967.
Figs 20, 21, 2 8 - 3 9 , 98, 98a, 104, 108, 109, 112.
24
See., e.g. J . K O S E N F I E L D : T h e Dynastic Arts. Pl. I I , 19 27, 29 38 ; Pl. I l l ,
5 0 - 6 0 ; Pl. IV, 61, 63, 6 5 - 6 6 ; P l . V, 8 4 - 9 1 , 9 3 - 9 7 , 9 9 - 1 0 2 ; Pl. VI, 1 0 3 - 1 2 3 ;
P l . V I I , 1 2 4 - 1 3 5 , 143, Pl. V I I I , 1 5 5 - 1 5 6 , 158159, 1 6 3 - 1 6 4 ; Pl. I X . 1 6 7 - 1 8 1 ,
185, 187; Pl. X , 188196, 199, 203, 205 208; Pl. X I , 209 - 2 2 9 ; Pl. X I I , 230 246
(imagos of gods and rulers).
25
M. B U S S A G L I : Le peinture de l'Asie Centrale. Genve 1963. 23 ; B. R O W L A N D :
The Art of Central Asia. N. Y., 1974. 33.

118 . S T A V I S K I Y
T h e head of t h e B u d d h a is depicted in f r o n t view (as distinct from t h a t of

t h e d o n t o r mentioned above), a n d t h e heads of t h e m o n k s t u r n e d slightly
t o w a r d s him. The trees are stylized. I n their foliage, above t h e monks, flowers
a n d ovaries may be discerned, and above t h e B u d d h a circles, evidently ripe
fruit.
T h e background of t h e wall paintings in t h e southern ai w a n of the com-
plex courtyard was blue, t h e faces of t h e personages flesh-coloured, the hair
of t h e B u d d h a b l a c k , his nimbus a n d t h e halo white, t h e robes multi-
coloured. The trees w i t h black foliage a n d golden t r u n k s were d o t t e d with red
a n d w h i t e ovaries a n d f r u i t .
All the wall p a i n t i n g s discovered in t h e course of e x c a v a t i o n s on K a r a tepe
were m a d e on stucco (alabaster) priming laid over clay plaster which covered
s a n d s t o n e and m u d - b r i c k walls of t h e premises. This t e c h n i q u e is known to be
used in the most a n c i e n t wall painting f o u n d in Central Asia, in Pessedjik-tepe
( S o u t h e r n T u r k m e n i s t a n ) d a t e d b y t h e 5 t h millennium B.C. 26 The K a r a tepe
wall paintings with t h e i r elaborate t e c h n i q u e are a p r o d u c t of a long history of
t h e development of p a i n t i n g , which b y t h e n already h a d ancient traditions.
Undoubtedly, w h a t we have here are not t h e first s t e p s of t h e local, Bac-
t r i a n school of p a i n t i n g b u t m a t u r e works of a r t .
SCULPTURE
J u d g i n g b y t h e a b u n d a n c e of f r a g m e n t s among t h e finds, stone and

s t u c c o (alabaster) s c u l p t u r e was common a t K a r a t e p e . A h a n d of a stucco
s t a t u e of the B u d d h a or a B o d h i s a t t v a and other f r a g m e n t s derived from t h e
t e m p l e courtyard of t h e A complex gives us ground to believe t h a t big Buddhist
s t a t u e s were placed in t h e grand niches of t h e aiwans, a n d , probably, in t h e
cells. T h e f r a g m e n t s of stucco relief with floral design f o u n d near the central
niche in the D complex c o u r t y a r d gives us good evidence t h a t carved stucco
w a s a n element of t h e decoration a t K a r a tepe.
B u t stone s c u l p t u r e prevails a m o n g t h e finds of this kind. Among t h e m
a r e column bases, blocks f r o m t h e facing of t h e big a n d t h e votive stupas and
also t w o stone capitals of pilasters reminding those of t h e B u d d h i s t platform of
statues near Surkh Khotal 2 7 as well as t h e pilasters adorning t h e terraces of the
K a n i s h k a temple. 28 T h e l a t t e r conclusion is p r o m p t e d b y t h e finds of a lower
p a r t of a pilaster a n d a f r a g m e n t of its base. 29 The capitals a n d f r a g m e n t of t h e
26
See Jl. . : VI -1 . . .
. . . . 3, 1977,
84, 93, 96.
27
See., e.g. D. S C H L U M B E R G E R : The Excavations at Surkh K o t a l a n d the Problem
of Hellenism in B a c t r i a a n d India. Proceedings of the British Academy XLVII. 1962.
P l . X X I I . a, X X I V . a, b.
28
D. S C H L U M B E R G E R : The E x c a v a t i o n s at Surkh K o t a l . Pl. X V I I I , b.
29
See -. 11. M. 1969, 159 a n d Pl. 37, a, 6.

pilaster were not found in situ, therefore their original disposition is still to
seek. Most likely t h e y were part of t h e lateral facing of the terrace upon which
courtyards, caves and other structures of the southern hilltop were s i t u a t e d .
One of the capitals (Fig. 14) was discovered in 1964 in the brickwork
walling t h e entrance to the cave sell of the P - I I temple, t h e other (Fig. 15) in
1976 in t h e filling of the D complex c o u r t y a r d . Both belonged to one type, w i t h
two belts of carved design. The lower belt represented a composite (i.e., Corin-
thianized) capital with volutes and a h u m a n half-figure against the background
of acanthus leaves : on the first capital a man in f r o n t view with a trefoil in
his right hand, on the second a girl in profil supporting a flower garland.
On t h e upper belt three animals are depicted, two lying zebu-bulls, their heads
turned outwards flanking a beast of prey a tiger or lion on the first capital,
and a fantastic horned and winged griffon on the second.
Some of the fragments of stone reliefs brought o u t in the course of t h e
K a r a tepe excavations might have been those of similar capitals. B u t the great-
er p a r t of the fragments are to be connected rather with t h e reliefs decorating
the stupas, the bases of Buddhist s t a t u e s and daises where statues were placed
(e.g. t h e wooden platform in the P - I I cave temple cell), etc. 3 0
As revealed by finds on K a r a t e p e and other sites in Middle and Central
Asia (Dalverzin tepe in Southern Uzbekistan, 3 1 monuments of Eastern Turkis-
tan 3 2 etc. 3 3 , the technique of preparation of alabaster sculptures was elaborate
enough. Clay pig, evidently on a wooden carcass was wound round with cloth
and covered with alabaster, which was then modelled or moulded and p a i n t e d .
The cloth, the imprints of which are easily discernible on the inner surface of
a number of fragments 3 4 formed a conjunctive tissue, so to say, between clay
a n d alabaster t h a t combines badly with each other. Inside fingers and toes of
the alabaster statues traces may be seen left by wooden sticks. Details (rosettes
and small balls adorning the garments, etc.) were modelled or moulded sepa-
rately and then fixed.
Stone sculpture of K a r a tepe, as well as stone architectural details (col-
u m n bases, capitals, fragments of t h e facing, etc.) are m a d e of white lime-stone,
t h e deposits of which are to be f o u n d on the right bank of the A m u - D a r y a ,
30
Ibid., 1 5 9 - 1 7 3 .
31
. . : -. (Ma-
gazine) 1970. No. 2, 62 ; the same a u t h o r . Les trsors de Dalverzine-tepo. Leningrad
1978. Figs 1 5 - 3 1 .
32
. M. :
. . V, . I, Petro-
grad 1918; A. STEIN: Serindia. Vol. IV, Oxford 1921, Pl. X L I X , C X X I X - C X X X I X ;
The same a u t h o r : Innermost Asia. Vol. I l l , Oxford 1928. Pl. IV, V I I I , I X , X I I ,
X I X , X X , L U I , LIV, L X X I I .
33
See . . : - 19611962 . -
, 1, M., 1964, 2 2 - 2 5 .
34
See -. j. PI. 25.

120 . STAVISKIY
p a r t i c u l a r l y on t h e H o d j a Gulsuar m o u n t a i n (to t h e northeast of Termez) where

a gold coin of K a n i s h k a 3 5 was found in 1904 in ancient pits. Special ptrographie
analysis of lime-stone samples has shown t h a t t h e lime-stone used for the K a r a
t e p e sculpture a n d architectural details m o s t likely came f r o m the H o d j a
G u l s u a r mines. 38
T h e alabaster a n d stone sculpture was brightly p a i n t e d and, sometimes,
in m a n y colours. T h e f i n d s of K a r a t e p e revealed traces of gilt laid over red
g r o u n d and red, blue a n d green paint against white background.
Mention should b e m a d e of white lime-stone chatras (umbrellas) adorning
t h e pole over the g r a n d s t u p a in t h e A complex northern c o u r t y a r d and t h e
v o t i v e s t u p a in t h e s a n c t u a r y between t h e A and complexes temple court-
y a r d s . T h e chatras were painted red a n d blue and gilded.
A n o t h e r B u d d h i s t a t t r i b u t e , along w i t h t h e wall painting, sculpture and
c h a t r a s , were u n d o u b t e d l y t h e discs (probably t h e lids of reliquaries) decorated
w i t h lotus flower reliefs. The discs (Fig. 15) were either m a d e of stone or of
b a k e d clay.
T h e sculpture also included t e r r a c o t t a s : a B u d d h i s t figurine, a relief
depicting the local B a c t r i a n mother-goddess, and other personages or figures of
a n i m a l s (for instance zebu-bulls reminding those on t h e stone capitals).
CERAMICS
T h e layer t h a t yielded the greater p a r t of ceramics, usually the most

n u m e r o u s among archaeological finds, was t h a t with the burials, in which more
t h a n 40 vessels, whole or f i t t e d together f r o m f r a g m e n t s h a v e been discovered.
Typical pieces of glazed or painted p o t t e r y f o u n d in f r a g m e n t s refer to t h e
1 0 t h 1 2 t h or later centuries. Finds t h a t m a y be d a t e d b y t h e time when t h e
K a r a t e p e Buddhist c e n t r e was functioning, are scant : oil lamps, several ves-
sels, which would be restored, and relatively few disconnected fragments.
I t m a y , evidently, b e explained by t h e n a t u r e of the m o n u m e n t : ceramics is
seldom found in religious or dwelling premises which were looked a f t e r , swept
a n d otherwise cleared o u t .
A m o n g t h e ceramic objects discovered on t h e floors of t h e caves and
g r o u n d premises of t h e K a r a tepe B u d d h i s t complexes oil l a m p s prevailed.
These were simple saucers with a small d e n t closer to t h e rim for a fuse (Fig. 16).
35
M. E. : ,
1917 1927 . , . II. 1928, 286.
36
The material w a s usually referred to in literature as marl lime-stone, according
to t h e definition given b y t h e geologist N. H. NIKOLAYEV who inspected the famous
A i r t a m stone relief representing three musicians (see M. E. :
. . , . I.
1933. 16). The erroneousness of this conclusion was revealed by E . N. AGAYEVA, mem-
ber of our expedition, in 1977.

Some of the decorative clay p o t t e r y may, with an a d e q u a t e degree of accuracy,

also be referred t o t h a t early period : vessels with s t u c k - o n moulded representa-
tions of a h u m a n head, bowls with lion heads or j u g s w i t h handles shaped like
birds or animals (e.g. monkey, Fig. 17). The vessels were m a d e of yellow or grey
clay, over which red, brownish-crimson or grey slip was laid. The polish e i t h e r
covered the entire surface of the vessel or formed a r c h e s stripes or rhombic n e t -
work. Some vessels were decorated with s t a m p e d p a t t e r n placed a t intervals.
S t a m p s common a t t h e time bore representations of a m a n standing u p r i g h t ,
smaller ones with t h e so-called Foot of the B u d d h a a n d also circles, rosettes,
herringbone, etc. A t t h e b o t t o m p a r t of t h e vessels trickles of slip are sometimes
t o be seen.
Ceramics f r o m t h e burials (Fig. 18) followed t h e traditions of t h e preced-
ing period. R e d or crimson slip bowls (tagora), one or two handle jugs, basins
and saucers were adorned with polished waves, arches and rhombs or s t a m p e d
continuous small flower p a t t e r n . Irregular flows of slip were, evidently, also used
as a decorative element. Vessels with long curved s p o u t s (tea-pots) m a y belong
t o a new t y p e of p o t t e r y .
The d a t a , based on t h e s t u d y of the K a r a t e p e ceramics and its comparison
with the p o t t e r y f r o m o t h e r sites of Northern B a c t r i a m a d e by N . S. Sycheva,
member of our expedition, gives sufficient ground t o believe t h a t Termez of t h e
K u s h a n and p o s t - K u s h a n periods had its own, local, centre of ceramic produc-
tion. 3 7
COINS
The K a r a t e p e excavations, since the time t h e y s t a r t e d , have b r o u g h t o u t

65 coins (one coin in 1936 by V. N . Kesayev, w h e n he was uncovering t h e
manhole into cave 2, five coins in 1937 by E . G. P c h e l i n a , and the rest, 59,
in 1961 1977 by our expedition). Thirty-five of t h e m has been defined. Chron-
ologically t h e y m a y be divided into two distinct groups. The first one, con-
nected with t h e B u d d h i s t complexes, consists of K u s h a n coins, the second de-
rived from t h e layers with t h e burials, includes Kushano-Sasanian, post-
K u s h a n (Chionite ?) and early H e p h t h a l i t e coinage. All of them are copper,
except one early H e p h t h a l i t e coin which is m a d e of silver.
Of the 15 K u s h a n coins, 13 have been defined w i t h greater chronological
accuracy i38 1 as belonging t o t h e coinage of Vima Kadphises (the definition
is conditional for t h e image on t h e coin is almost effaced) ; 3 t o t h a t of
K a n i s h k a ; 3 of H u v i s h k a , 5 of Vasudeva, a n d 1 an imitation of t h e
coinage of Vasudeva. I t should be noted t h a t two of t h e coins of H u v i s h k a h a v e
37
See H. . : -. - IV. M. 1975. 8 8 - 148.
38
Definitions given by M. E . MASSON (1937 finds), E . V. ZEIMAL (1961 71 finds)
a n d . I . VAHLBERG ( 1 9 7 2 7 7 f i n d s ) .
A r ta Antiqua Academiae Scientiarum Hungaricae 28, 1980

122 . STAVISKIY
been f o u n d inside a m u d brick, and one coin of Vasudeva in a crack in a lime-

stone pilaster base, which is i m p o r t a n t for d a t i n g reconstructions of the K a r a
tepe complexes.
Of t h e 20 coins of t h e second group 1 was defined as belonging to the
coinage of the Sasanian K u s h a n s h a h Hormizd 1,4 to the coinage of Varahran
I (one of them suppositionally) ; 4 to t h e coinage of Varahran I I (three of
t h e m suppositionally) ; 5 as Kushano-Sasanian coins (a more accurate dating
could n o t be gained) ;39 5 coins (Fig. 19), with t h e bust of an unidentified ruler
on t h e averse and t r i b a l m a r k (tamga) on t h e reverse, 40 might be minted by
a local, post-Kushan (probably Chionite) ruler, and 1 coin (Fig. 20), early
H e p h t h a l i t e , referred to t h e 6th century A.D. 4 1 Most of the Kushano-Sasanian
coins (11 out of 14) h a v e been derived f r o m t h e layer with t h e burials in the
c e n t r a l chamber of t h e P - I temple in t h e A complex, while t h e Chionite ( ?)
coins c o m e from the same layer in t h e P - I I temple of the complex and in a
hiding-place in the D complex. I t may be t h a t the coin of Varahran I I ( ?) found
in t h e ground cleared o u t of the complex is also connected w i t h t h e burials
in t h e P - I I temple, a n d t h e Kushano-Sasanian coin, discovered in t h e rubble
in t h e A complex courtyard southern aiwan, with the burials in t h e P - I tem-
ple. T h e early H e p h t h a l i t e coin found in cave 3 of the complex, along with
a n o t h e r two small coins (Kushano-Sasanian or Chionite ?). W h e t h e r it comes
f r o m t h e burials of t h e 6 t h century or was brought here by some later visitor is
still t o be cleared.
INSCRIPTIONS
A m o n g the m a j o r finds t h a t immediately attracted t h e a t t e n t i o n of

scholars t o the K a r a t e p e site were u n d o u b t e d l y the inscriptions of t h e Kushan
a n d post-Kushan priode. The former, m a d e in black or Indian ink, are to be
f o u n d on pottery, and t h e latter were scratched on cave walls a n d entrance
niches, probably upon t h e decline of t h e B u d d h i s t complexes b u t before the
premises were used as b u r i a l grounds. T h e graffiti left by Moslems visiting the
place in t h e 10th12th centuries form a group of their own. T h e y have been
s t u d i e d b y S. B. Pevzner, specialist in Arabic languages. 42
Of t h e inscriptions on pottery we are in possession of only t w o complete
t e x t s one on a jug t h a t was preserved almost intact, and t h e other on the
f r a g m e n t s of a vessel. 43 T h e rest have been found in fragments.
39
D e f i n e d b y V . G . LTJKONIN a n d E . V . ZEIMAL.
40
D e f i n e d b y . I . VAINBERG a n d E . V . ZEIMAL.
41
D e f i n e d b y . I . VAINBERG.
42
See . . : -. -. II.
1975. 8 2 - 8 7 .
43
See -. IV. 1975. Fig. 11 13, and -. III. M. 1972. Pl.
XXIII.

The K a r a tepe inscriptions have been studied by the Soviet Indologists

T. V. Grek and V. V. Vertogradova and the specialist in Iranian languages V. A.
Livshits and also by Prof. J . H a r m a t t a from Hungary and Prof. H . H u m b a c h
from West Germany. 44
Most of the inscriptions were written in Indian Kharoshthi and Brhm,
some in the Kushan script (the language of such inscriptions is commonly
known as Bactrian). There were also bilingual texts in Brhm and Kushan
sfcripts. Three fragments revealed an unknown script which G. Pussman con-
ventionally called Kambodian. 45 In the opinion of V. A. Livshits it might be
the script of the Sacas. 46
Considerable part of the K a r a tepe inscriptions on pottery was made by
or in the name of donators. 47 In the bilingual (Brhm and Kushan scripts) in-
scription collected from fragments and on a number of fragments the dontor
mentioned is Buddhasira dharmakathika (the Buddha preaching a good
deed or creed in the Bactrian variant). 48 Saiighapala 49 of the Kharoshthi
inscription found in fragments might be the name of another dontor. Some
inscriptions are of didactic nature. 5 0 Peculiar is the mention in a number of in-
scriptions of the Mahasanghika Buddhist school. 51
The K a r a tepe graffiti of the post-Kushan period were discovered at the
entrance to the P - I I temple its corridors (southern, eastern and northern)
and its northern entrance niche in the complex courtyard. These inscriptions
have been published and discussed in a number of articles. 52 The bilingual brief
44
See -. I. 1964. 62 81; -. II. M. 1969. 32 - 39, 47 76;
-. III. M. 1972. 114-117. 118-121, 122-123; -. IV. 1975. 47 52,
6269, 7081. See also J . HARMATTA : The Bactrian Inscriptions a t Kara Tepe. Cen-
tral Asia in the Kushan Period. Vol. I . Moscow 1974. 328.
45
See. G. FUSSMAN : Documents epigraphiques Kouchans. B E F E O t. L X I , 1974.
22 34 ; G. FUSSMANM. LE BERRE : Monuments bouddhiques de la Region de Kaboul.
MDFA, t . X X I I , P., 1976. 94.
46
. A. : . (ed. I. T. KRUGLI-
KOVA) 1976. 165166, note 14.
47
The donatory nature of the inscriptions on pottery was first pointed out to by
J . HARMATTA, who also offerred the most complete interpretation : To the (Buddhist )
community of the four cardinal points (with the consent of the monks (or tutors) Ma[ha]-
sanghika (monk's gift) from Sahghapala. See . :
-. -. II. . 1969. 34 35.
48
See . . . . : -. -. III.
1972. 118120; . . : -.
Ibid. 114116. Cf. . : -.
3739. For arguments against the interpretation of the inscription as a bilingual see X.
: -. -. IV. 1975.
62-69.
49
5
. : . . ., 3435.
! . . : -. -
. IV. 1975. 70 81.
51
Ibid., 81 ; . : . . ., 34.
62
. . : -. -. II.
1969. 40 - 46; . . : -. Ibid.
47 81 (particularly 5052, 76 81); J . HARMATTA: The Bactrian Wall Inscriptions
f r o m Kara Tepe. Ibid., 82125; the same a u t h o r : The Bactrian Inscriptions at K a r a

124 . STAVISKIY
inscription scratched on t h e cornice of t h e vaulted e n t r a n c e niche of cave 3 in

t h e complex 5 3 has also been published. T h e g r a f f i t i in K h a r o s h t h i and K u s h a n
scripts found in 1977 in t h e e n t r a n c e p a r t of t h e P-V cave of t h e D complex are
being prepared for publication.
As distinct f r o m t h e inscriptions on pottery, t h e I n d i a n g r a f f i t i are n o t
n u m e r o u s : 3 in K h a r o s h t h i and 2 in B r h m . According t o T. V. Grek, in t h e
inscriptions in Brhm mention is m a d e of Mahesvara and dharma. 5 4 The na-
t u r e of t h e K h a r o s h t h i inscription is n o t y e t clear.
Two inscriptions in Middle Persian, 5 5 a bilingual inscription (in K u s h a n
a n d Middle Persian scripts) 5 0 and over a dozen inscriptions in K u s h a n script 5 7
(Fig. 21ab) consist, primarily, of t h e names of visitors and, in some cases t h e
d a t e s of their visits. J . H a r m a t t a believes t h a t among t h e n a m e s scratched on
t h e walls of the P - I I t e m p l e there also were those of t h e Sasanian K u s h a n -
shahs. 5 8 V. A. Livshits, however, disagrees with him. 59 Deciphering of some of
t h e d a t e s has been d i s p u t e d . I t seems t h a t some of t h e d a t e s m a y be accepted,
such as those given in a Middle P e r s i a n inscription 60 or 61, 60 and in two
K u s h a n inscriptions f r o m t h e P - I I t e m p l e 97 61 and 35. 02
T o conclude, I would like to note t h a t there is no u n a n i m o u s opinion as
t o t h e reading and i n t e r p r e t a t i o n of t h e K a r a tepe g r a f f i t i which are fragmen-
t a r y a n d o f t e n effaced, a n d t h a t therefore one should display m a x i m u m caution
for conclusion from t h e d a t a acquired by their analysis.
Tepe. Central Asia in t h e K u s h a n Period, V. I, 1974. 328333. . . :

-. -. III, M. 1972. 117; X. -
: -- . Ibid., 122 128; H . HUMBACH: K a r a Tepe-
Tochi-Surkh Kotal. Mnchener Studien zur Sprachwissenschaft. 28 (1970) 4360 ; . .
: -. -. IV. M '975.
47 60; . N. MUKHERJEE : The Epigraphic Evidence from K a r a Tepe and the So-
called Old Saka E r a . E a s t a n d West 21 (1971) 69 73.
63
See -. IV. 1975. 8 a n d Pl. 10.
64
. . : . 117.
65
. . : . 40 46.
56
See note 63.
57
Here according to V. A. LIVSHITS (see . . : -
, 53 60), who believes there were 11 inscriptions in t h e P - I I temple.
P r o b a b l y , there were more.
58
J . HARMATTA: T h e Bactrian Wall-Inscriptions (Nos. 17, 26, 49, 50, 62, 61),
94 95, 100, 111 112, 114115, 122124; t h e same a u t h o r : The Bactrian Inscrip-
tions. 329, 332333.
59
. . : . 54, 55, 57, 58, 59, 60.
60
. . : . 4146.
81
J . HARMATTA: T h e Bactrian Wall-Inscriptions (No. 20), 9697 X, -
: - - , 128; . :
, 57. J . HARMATTA a n d V. A. LIVSHITS hold that n u m b e r 3 m i g h t have preceded
t h e d a t e which is then to be read as 397.
82
J . HARMATTA: T h e Bactrian Wall-Inscriptions (No. 36), 104108; X. -
: - , 128; . . : -
. . ., 59. J . HARMATTA believes t h a t in this case t h e first n u m b e r was omit-
ted a n d t h e inscription is to be dated b y 435. H . HUMBACH and V. A. LIVSHITS disagree
with him on this point.

GENERAL HISTORICAL SURVEY
Scholars who were the first to get interested in K a r a tepe d a t e d its caves
by t h e pre-Moslem period. The excavations of 1937 allowed E . G. Pchelina t o
m a k e more a c c u r a t e d a t i n g by referring t h e K a r a tepe B u d d h i s t m o n u m e n t s t o
t h e t i m e of t h e K u s h a n s . The 196177 excavations confirmed Pchelina's con-
clusions. The f a c t t h a t the K a r a tepe B u d d h i s t religious centre emerged a n d
functioned in K u s h a n times is evidenced b y the finds m a d e by our e x p e d i t i o n
K u s h a n coins, ceramics, inscriptions on pottery, a n d o b j e c t s d ' a r t . Details of
t h e K a r a tepe history, however, are n o t t o be easily ascertained.
The emergence of t h e centre is m o s t probably t o be connected with t h e
reign of t h e famous K u s h a n emperor K a n i s h k a . The only pre-Kanishka coin
f o u n d in 1961, t h a t of Vima Kadphises (thus defined b y its dimensions a n d
weight for t h e images were effaced) 6 3 is n o t enough to disprove this hypothesis,
while there are already three coins of K a n i s h k a a m o n g t h e finds. T h e e a r l y
ceramics of K a r a tepe draws analogies in t h e p o t t e r y connected with t h e h e y d a y
of the K u s h a n E m p i r e , a n d not with t h e period of its f o u n d a t i o n . The K u s h a n
script, based on t h e Greek alphabet, according to n u m i s m a t i c d a t a , was c r e a t e d
a t the time of K a n i s h k a , therefore even t h e earliest inscription in K u s h a n script
f o u n d cannot be referred to a p r e - K a n i s h k a period. Inscriptions analogous t o
t h e I n d i a n inscriptions on K a r a tepe p o t t e r y do not refer (as f a r as I can judge)
t o an earlier period either. According t o B u d d h i s t t r a d i t i o n s K a n i s h k a encour-
aged t h e spread of Buddhism and B u d d h i s t communities beyond the borders of
t h e I n d i a n sub-continent. Thus the c o n j e c t u r e as to t h e emergence of a B u d -
dhist religious centre on t h e outskirts of a big city of K u s h a n BactriaTermez
a t t h e time of K a n i s h k a is borne out b o t h b y archaeological d a t a and t h e B u d -
dhist t r a d i t i o n .
On t h e evidence of the finds t h r e e coins of H u v i s h k a and four coins of
Vasudeva (one of t h e m , as has already been mentioned, discovered in a c r a c k
in t h e base of a pilaster which was found in situ, p u t t i e d into a wall so t h a t only
one of its lower edges j u t t e d out) a n d also the p a l a e o g r a p h y of t h e cursive
a n d half-cursive inscriptions in t h e K u s h a n script a n d t h e K h a r o s h t h ! a n d
B r h m l inscriptions of pottery, we m a y well believe t h a t t h e Buddhist religious
centre existed on K a r a tepe all t h r o u g h t h e period of h e y d a y of t h e K u s h a n
Empire, which covered also the reign of K a n i s h k a ' s successors. I t tallies f u l l y
w i t h t h e d a t a on t h e cultural development of t h e K u s h a n E m p i r e . According t o
t h e general point of view, H u v i s h k a continued his famous predecessor's policy of
religious tolerance. I t is with his name, for instance, t h a t t h e Buddhist v i h a r a
63
See. E. . . . : - 1961 1971 .
-. II. 1972. 101 (. I).

126 . STAVISKIY
in J a m a l p u r is connected. Judging by t h e inscriptions it existed in Vasudeva's

time too, at least as far as year 77 of t h e Kanishka era. 64
I t is not so easy, however, to determine the cause a n d the time of t h e
desolation of the B u d d h i s t complexes. T h e d a t a collected so far are insufficient
for drawing any definite conclusions as to whether it was destroyed by enemies
or j u s t forsaken in good time by its i n h a b i t a n t s . One might be t e m p t e d to ex-
plain t h e destruction or desolation of K a r a tepe by the conquest of Termez b y
t h e Sasanian army. Of t h e appearance here of Sasanian troops we have good
evidence now t h a t Middle Persian g r a f f i t i have been discovered in the P - I I
temple. Especially since one of them was interpreted as a t e x t written by a dibir,
i.e., scribe or secretary a t the chancellery of the Sasanian ruler or military
leader. 6 5 I t is not unreasonable to suppose t h a t the mud-brick fire altar built
in t h e central (grand) niche in the aiwan of the complex western courtyard
is connected with t h e presence of Sasanian soldiers here. F o r t h e fact t h a t t h e
region of Termez was under Sasanian rule we also have evidence in Kushano-
Sasanian coins from t h e layer of burials. Nevertheless it would still be an over-
simplification to s t a t e t h a t the desolation of the complexes was the result of an
invasion or an exodus in anticipation of an invasion. Archaeological d a t a r a t h e r
p r o m p t s t h e conclusion t h a t the fire a l t a r and the Middle Persian inscriptions
appeared on the grounds after the complex and its P - I I temple were aban-
doned. However, how much time elapsed f r o m the moment t h e complex was
destroyed or abandoned till Sasanian troops appeared here a few days or a
n u m b e r of years is unknown.
I t seems t h a t t h e time of the desolation of the K a r a tepe Buddhist com-
plexes might be determined by the dates given in the g r a f f i t i . B u t , unfortu-
n a t e l y , in this case, too, it remains a m a t t e r for conjecture.
According to H . Humbach, the dates given in the graffiti inscriptions in
t h e K u s h a n script 35 and 97 might be t h e years of the Kushano-Sasanian
era of t h e famous inscriptions from t h e Tochi Valley (the beginning of t h e era
refers t o 232 A.D.). If so, the dates of t h e P - I I temple inscriptions m a y be
deciphered as 267/268 and 329/330 A.D. 6 6 If, however, we a d o p t the interpreta-
tion given by J . H a r m a t t a , the dates 97 a n d 35 are to be read as 397 and 435
of t h e so-called Old S a k a era which s t a r t e d in 67 B.C. and should, therefore, cor-
respond to 330 and 368 A.D. 67 Besides, we cannot be certain t h a t the era of t h e
K a r a tepe inscriptions in the K u s h a n script is the same as t h a t of the Tochi
inscriptions or of t h e post-Kushan inscriptions of t h e territory of Ancient
64
For inscriptions from the Vihara of H u v i s h k a see L. LDERS : Mathura Inscrip-
tions. Gttingen 1961. 2 9 - 6 3 , 5 7 - 1 0 5 .
65
See . . : . 41, 43 44.
66
X. : - - . 128.
67
J . HARMATTA: The Bactrian Wall-Inscriptions. 95 97, 104108, 122 125.

K A R A T E P E I N OLD TERMEZ 127
I n d i a . All this makes it impossible to determine the d a t e s of inscriptions in

K u s h a n script with sufficient accuracy.
Different i n t e r p r e t a t i o n s are also given of the d a t e 60 or 61 in t h e Middle
P e r s i a n inscription. The late W . B . Henning thought it t o be a d a t e of t h e royal
Sasanian era which corresponded to 264/265 A.D., 68 while V. G. L u k o n i n be-
lieves t h a t it is t h e 61st year of t h e reign of Shapur I I , i.e., 369/370 A.D. 69 I n t h e
f i r s t instance t h e d a t e of the Middle Persian inscription is close to those of t h e
inscriptions in t h e K u s h a n script offered by H . H u m b a c h , in t h e second t o
t h e d a t e s given b y J . H a r m a t t a . This alone shows t h a t coincidences of this order
m i g h t be fortuitous.
W e hope t h a t new discoveries will throw light on t h i s problem a n d help
t o establish the era of t h e inscriptions and, consequently, t h e exact year in
which t h e y were m a d e .
U n t i l then t h e inscriptions help us in answering one question only, t h a t
of how long t h e period of t h e complex desolation lasted, t h e period in t h e
course of which visitors scratched their names on the walls of this complex.
Since the K a r a tepe g r a f f i t i in K u s h a n script definitely belong to a single
e r a we m a y assert t h a t the i n t e r v a l in time between t h e t w o inscriptions k n o w n
t o us is either 62 years (if we accept H . H u m b a c h ' s i n t e r p r e t a t i o n of t h e d a t e s
35 a n d 97 as belonging to one century) or 38 (if we agree w i t h J . H a r m a t t a who
reads t h e m as 397 and 435). Correspondingly we m a y conclude t h a t t h e period
of desolation of t h e complex and, most likely, of t h e K a r a tepe B u d d h i s t
c e n t r e as a whole, covered n o t less t h a n 62 or a t least 38 years.
I t will be remembered t h a t a t t h e end of this period t h e abandoned p r e m -
ises were used for mass burials. T h e d a t e of the burials f o u n d in P - I t e m p l e
m a y be established on the basis of t h e copper coins of t h e K u s h a n o - S a s a n i a n
rulers Hormizd I, Varahran I a n d V a r a h r a n I I t h a t occurred in them.
The dating of the Kushano-Sasanian coins generally f l u c t u a t e s within t h e
limits of over 100 years, t h o u g h t h r e e new approaches t o t h e problem h a v e
been offered lately to replace t h e o u t d a t e d system of E . Herzfeld. 7 0 A. D . H .
B i v a r refers their coinage to t h e period ranging from t h e middle of the 3rd cen-
t u r y t o t h e late 4th century, 7 1 R . Gobi to 371 400, 72 a n d V. G. L u k o n i n
t o t h e 60s of t h e 4th century t h e 30s of the 5th century. 7 3 The divergence
68
See -. II. 46, note 24.
89
. . : , 41, 44, 46.
70
E . HERZFELD : Kushano-Sasanian Coins. Memoires of the Archaeological Survey
of I n d i a No. 38. Calcutta 1930.
71
A. D. BIVAR: The Kushano-Sasanian Coins Series. J o u r n a l of tho Numismatic.
Society of India 18 (1956) Pl. I . Cf. Central Asia ed. by C. H a m b l y , L. 1970. 61 52.
72
I i . GBL : Dokumente zur Geschichte der Iranischen H u i m e n in Bactrien u n d
Indien. Wiesbaden 1967. Bd. I. Ss. 2425 ; Bd. I I . Ss. 47 48.
73
. . : - . 18 (1967)
1633; the same a u t h o r : -
. 1971, 3, 3944; t h e same a u t h o r :
, . 124 151.
9* Acta Antiqua Academiae Scientiarum Hungaricae 28, 1 9 8 0

128 . STAVISKIY
is t h e least, however, in relation to the d a t i n g of the V a r a h r a n I reign. According

t o L u k o n i n this K u s h a n s h a h ruled f r o m 384 t o 388, and A. D . H . Bivar places
his reign 25 30 years earlier. If we believe Lukonin, K u s h a n s h a h Hormizd I
w a s t h e predecessor of V a r a h r a n l a n d ruled f r o m 381 t o 384. B i v a r , however,
refers t h e rule of H o r m i z d I to 277 286. I t m a y be noticed, however, t h a t
t h e coin of Hormizd I w a s found a t K a r a t e p e (in t h e l a y e r of burials in the
c e n t r a l p a r t of t h e P - I temple) together (in one complex) w i t h the coins of
V a r a h r a n I and V a r a h r a n I I , which speaks against the d a t i n g of this K u s h a n -
s h a h ' s rule to such a n early period. V a r a h r a n I I ruled, according to Bivar,
a r o u n d the year 360, a n d according t o L u k o n i n from 389 even u p to 440. . I .
Vainberg, however, w h o agrees in principle w i t h the chronology given by Luko-
n i n convincingly p r o v e d t h a t Lukonin's d a t i n g of t h e V a r a h r a n I I coins is
d i s p u t a b l e . In her opinion, t h e coins were m i n t e d either b y V a r a h r a n I or his
predecessors. At a n y rate, she concluded, we have no sufficient ground to
believe t h a t the K u s h a n o - S a s a n i a n coins were minted a f t e r 389.74 This con-
clusion seems reasonable enough. I n a n y case t h e coins of H o r m i z d I, V a r a h r a n
I a n d V a r a h r a n I I f o r m a chronologically compact group a n d m a y be referred
t o t h e 50s60s or 80s of t h e 4th c e n t u r y . T h e burials in t h e P - I temple of t h e
A complex are d a t e d accordingly.
The p o s t - K u s h a n coins with t h e half figure of a ruler on t h e averse and
his t a m g a on the reverse f o u n d in the burials (or in connection with them) in t h e
complex ( P - I I temple) and the D complex hiding-place, were, evidently,
m i n t e d a f t e r the b r e a k - u p of the Sasanian domination in B a c t r i a by a local
(Chionite ?) ruler.
Chronological a n d geographical ties of these coins w i t h t h e coinage of t h e
Goboziko group, which, according to Vainberg, were in circulation in N o r t h e r n
T o k h a r i s t a n in t h e period between 389 a n d 440, 75 is as y e t n o t ascertained.
As the d a t a given above shows, it is still impossible t o reconstruct with
a d e q u a t e coherency a n d in detail the history of the B u d d h i s t religious centre of
K a r a tepe. The general outlines of its history, however, m a y be presented in t h e
f o r m of the following t a b l e (see p. 129).
K A R A TEPE AND T H E PROBLEM

OF BUDDHISM IN KUSHAN TERME Z
I n the 7th c e n t u r y , Termez and its environs, according to H s a n Tsang,

b o a s t e d ten B u d d h i s t monasteries and o t h e r m o n u m e n t s a n d u p to a thousand
monks. 7 6 One should n o t , probably, give full credit t o t h e words of t h e great
74
. . : IVV . -.
III. M. 1972. 134-135.
Ibid., 1 3 0 - 1 3 4 , 137, 1 5 0 - 1 5 1 (Pl. 1).
7|>
S. BE AL : Si-Yu-Ki. Buddhist Records of the Western World Translated from
t h e Chinese of Hiuen Tsiang. L. 1906. Vol. I. 39.

KAKA T E P E I N OLD TERMEZ 129
Table I
General Historical Outlines of t h e Buddhist Religious Centre on K a r a tepe
The rise of the Buddhist

complexes the time of Kanishka ?
Their functioning, reconstructions

and repairs the time of Huvishka and Vasudeva
Destruction or the beginning of
desolation the time of Vasudeva around 260 or 370 A. D.
Burials in P-I temple the time of Kushanshahs Hormizd I, Varahran I
and Varahran II 60 60 or 80 of the
4th cen. A. D.
Length of the period of desolation
of the P-II temple not less than 38 or 62 years 1
Burials in the P-IX temple and the time of an identified (Chionite?) ruler . .
the D complex after the 80s of the 4th cen. A. D.
Penetration of visitors through
gaps and breaks . . from the 9th10th cen. (6th in the complex?)
B u d d h i s t pilgrim, for he personally never visited Termez. There is no d o u b t ,

however, t h a t there were large B u d d h i s t communities a n d numerous B u d d h i s t
m o n u m e n t s in Termez in the early (pre-Moslem) Middle Ages as in Balh, Qun-
d u z and other big cities of T o k h a r i s t a n . 7 7
U n f o r t u n a t e l y , we have no w r i t t e n evidence as to t h e number of B u d d h i s t
m o n u m e n t s or t h e size of monastic communities in B a c t r i a - T o k h a r i s t a n of an
earlier, K u s h a n , period. The n a m e of Termez in p a r t i c u l a r crops up only once in
connection with t h e history of B u d d h i s m in a colophone of a B u d d h i s t
t r e a t i s e of the K u s h a n period t r a n s l a t e d into T i b e t a n , which gives t h e n a m e of
t h e a u t h o r as D h a r m a m i t r a , native o f T a r m i t a [ = T e r m e z ] on t h e b a n k of t h e
V a k s h u [ = Oxus = A m u Darya]. 7 8
Archaeological d a t a acquired by Soviet researchers so far reveals t h a t
B u d d h i s m in K u s h a n Termez e n j o y e d t h e same s t a t u s as in t h e period described
b y Hsiian Tsang.
I t is to be recalled t h a t a p a r t f r o m K a r a tepe on t h e outskirts of t h e Old
Termez site (Fig. 26), three more B u d d h i s t m o n u m e n t s h a d been discovered
b e y o n d the outer city wall n o r t h a n d e a s t of it. Not m u c h remained of t w o of
t h e m o n u m e n t s . On t h e site of one of t h e m , o n the south-east of t h e city, ruins
of a g r a n d s t u p a have been discovered, which is now known as t h e Z u r m a l a
77
I share t h e opinion of m a n y o t h e r scholars that Tokharistan is what ancient
Bactria came to be called since t h e 4th century A. D. Thus, theso are two names of one
a n d tho same region.
78
. H. : . 1963. 6.122; the same
a u t h o r : . . -
1964. . 65, 141; H. W. BAILEY: Indo-Iranica. I I I . 5. T a r m i t a , BSOAS 13 (1950)
400-403.
9 Acta Antiqua Academiae Scientiarum Hungaricae 28, 1980

130 . STAVISKIY
Tower. 7 9 On the other site, close to the north-eastern corner of t h e city, under-
ground premises with s t e p s leading up h a v e been unearthed. 8 0
T h e stupa was a g i a n t structure of u p t o 16 metres in height. I t s cylinder-
shaped body, 15 metres in diameter, was crowned with a cupola and set on a
rectangular dais of 22 x 16 metres. While differing from t h e two K a r a tepe
s t u p a s in appearance, t h e Zurmala Tower was similar to t h e m in construction
t e c h n i q u e : it was b u i l t of m u d brick and p a k h s a a n d faced w i t h lime and baked
brick (the latter, p r o b a b l y , used for the sake of economy considering the size
of t h e structure). The studies made by G. A. Pugachenkova a n d Z. H . Hakimov
leave no doubt as to i t s being built in t h e K u s h a n period.
T h e dating was confirmed by the finds of pottery and Buddhist inscrip-
tions of the Kushan period discovered in t h e underground premises. The finds
also bear witness to t h e Buddhist n a t u r e of t h e ground structures connected
with these premises, of which, unfortunately, nothing came to us.
W e can only m a k e conjectures as to the composition of the Buddhist
complex of which t h e Zurmala Tower and t h e underground premises were a
p a r t . B u t it is highly probable t h a t next to t h e Zurmala Tower there was a sub-
u r b a n monastery similar to the third B u d d h i s t monument on t h e outskirts of
K u s h a n Termez.
This third m o n u m e n t , known as Faiaz tepe site, was situated north of
Termez a t a distance of about one kilometre from K a r a tepe. 8 1 I t is in a better
s t a t e of preservation t h a n the other two, and even t h a n K a r a tepe. This com-
plex (a typical s u b u r b a n monastery) consisted of a rectangular building, 110
m e t r e s in length a n d 35 metres in width spreading from northwest to southeast
a n d a s t u p a to the n o r t h e a s t of it t h a t might initially, be surrounded by walls.
T h e monastery building was divided into three functional p a r t s approximately
equal in size, each w i t h an inner courtyard.
T h e northwestern p a r t , evidently, consisted mainly of dwelling quarters
giving access to the inner courtyard which h a d an aiwan along one of its walls.
This should have been t h e monastery proper, which above t h e monks' hostel
included other premises.
79
. . : . 1967.
. 3, 257261 and bibliography.
80
Data on t h e 1973 excavations has n o t yet been published. I h a v e had occasion
to visit them due to t h e kindness of L. I. ALBAUM who is carrying out the digs. For
brief information on t h e basement and t h e inscriptions found there see M. . -
- : . -
, 1974 . 1. 117.
81
The excavations of Faiaz-tepe were carried out from 1968 to 1976 by L. I .
ALBAUM. Data has not y e t been published. F o r brief information see : Jl. . :
- ( 1968 1972 .). -
. . 1974. 53 - 5 8 ; t h e same author : . 1975. 90 93.
I t a k e this opportunity to thank L. I. ALBAUM and deputy director of the Termez
M u s e u m V. A. KOZLOVSKIY for allowing m e to visit t h e Faiaz tepe site and familiarize
myself with the d a t a .

The south-eastern p a r t adjoining a large c o u r t y a r d housed store rooms,

kitchens with ovens t o cook food (e.g. f l a t cakes) and, probably, refectories.
The central p a r t was, evidently, a place of worship. I t s c o u r t y a r d , was
similar to t h e temple c o u r t y a r d s of K a r a tepe : aiwans along its perimeter w i t h
wooden columns resting on limestone bases and a lowered unroofed p i t c h in t h e
middle. W h a t brings it still closer to the K a r a tepe A and complexes t e m p l e
courtyards was the e n t r a n c e to a s a n c t u a r y in t h e southern aiwan, and also t h e
wall paintings uncovered on the walls of t h e aiwan and inside t h e s a n c t u a r y .
They remind those of K a r a tepe b u t are in a b e t t e r s t a t e of preservation. Here,
too, t h e paintings present t h e B u d d h a and a group of donators : their faces
are depicted in profile and their bodies and feet in a f r o n t a l position, 8 2 which
confirms t h e assumption we have m a d e in connection with t h e K a r a tepe mur-
als. Clay, alabaster and lime-stone sculptures found here are also b e t t e r pre-
served. W h a t was preserved t h a t were t h e alabaster head of t h e B u d d h a or
B o d d h i s a t v a and a r e m a r k a b l e lime relief showing t h e B u d d h a under a tree and
two monks on his right and left. 8 3 B o t h pieces of s c u l p t u r e are executed in t h e
best traditions of t h e G a n d h a r a n art.
The Faiaz tepe site have also yielded f r a g m e n t s of p o t t e r y with I n d i a n
inscriptions in K h a r o s h t h i and Brhm, as well as inscriptions m a d e in K u s h a n
script (most of them are n o t yet published).
The comparison of t h e Faiaz tepe and K a r a tepe sites reveals, along with
t h e similarity of their temple courtyards, certain distinctions in t h e s t r u c t u r e of
these basically synchronous monuments. The former was erected according t o
a single and distinct plan on an open f l a t terrain, while t h e l a t t e r was probably
built in stages, as separate complexes, on a high hill seen from afar. Faiaz t e p e
was, therefore, a one-storey compact structure, while t h e B u d d h i s t complexes
of K a r a tepe, its ground and cave premises, were scattered over the slopes of t h e
hill, on different levels, sometimes one directly above t h e other. However t h e
basic difference between t h e m lay elsewhere, i.e., in their purpose, which, in t h e
final analysis, determined also the choice of place.
T h e builders of t h e Faiaz tepe B u d d h i s t monastery followed a traditional
architectural p a t t e r n and created a complex t h a t answered perfectly the needs
of a monastic c o m m u n i t y completed with dwelling and household premises,
mess rooms and a place of worship. The planning is precise and easily divined.
T h e K a r a tepe complexes do not offer us either t h e precision of the plan-
ning nor complete architectural p a t t e r n . The only typological distinction we
can p o i n t to so far is a t e m p l e courtyard with its aiwans and t h e lowered central
p a r t . As to temple premises, their planning was each t i m e to be determined
anew (see cave 2, temples P - I and P - I I ) . Much in t h e s t r u c t u r e and purpose of
82
Jl. . : . 90 and Pl. L I .
'3 Ibid. 90 and Pis. 31, 32.

132 . STAVISKIY
t h e K a r a tepe religious centre is still t o seek b y f u r t h e r excavations. Thus, we do

n o t y e t know w h e t h e r t h e complexes h a d m o n k hostels, refectories a n d other
premises common in monastic B u d d h i s t communities, a n d if so, whether each
complex had his own premises of this kind, or t h e y were shared b y t h e whole
community.
Of one thing, however, we m a y speak with c e r t a i n t y a large n u m b e r of
B u d d h i s t temples a n d sanctuaries was concentrated on K a r a t e p e t o be used n o t
b y a single monastic group b u t , most probably, b y a numerous B u d d h i s t com-
m u n i t y of t h e town of Termez. F r e q u e n t visits to its ruins and inscriptions left
b y t h e visitors might t e s t i f y to its p o p u l a r i t y as a religious centre. I t should be
s a ; d t h a t this p o p u l a r i t y , and also its proximity t o t h e t o w n now give m u c h
t r o u b l e to its s t u d e n t s : t h e a r t m o n u m e n t s of t h e neighbouring Faiaz tepe
preserved b e t t e r b y far t h a n those of K a r a tepe.
Nevertheless, ruined as it is b y time, and even more b y people, t h e K a r a
t e p e site together w i t h other B u d d h i s t m o n u m e n t s of K u s h a n Termez bears
v i v i d l y witness t o t h e spread of B u d d h i s m a n d also t o t h e w e a l t h of t h e B u d -
d h i s t community t h a t was, evidently, numerous a n d enjoyed m u c h a u t h o r i t y
i n these parts in t h e K u s h a n period.
The n a t u r e of this c o m m u n i t y m a y be established on t h e basis of the in-
scriptions on p o t t e r y of K a r a t e p e . Already a t t h e initial stages of s t u d y , J . H a r -
m a t t a assumed t h a t t h e inscriptions m a d e in K h a r o s h t h I script were to be con-
n e c t e d with t h e school of M a h a s a n g h i k a , and those in B r h m with t h e
S a r v a s t i v a d a school. 84 T h e assumption was disputed on t h e grounds t h a t t h e
s c r i p t alone could n o t be a conclusive evidence in referring an inscription to a
d e f i n i t e school of B u d d h i s m . I t was noted, in particular, t h a t an inscription in
K h a r o s h t h I belonging to a S a r v a s t i v a d a school h a d been f o u n d in H a d d a
(Southern Afghanistan). 8 5 However, H a r m a t t a ' s assumption was confirmed b y
f u r t h e r studies. A m o n g t h e newly-found inscriptions on p o t t e r y there was one
which contained t h e word mahasanghika ( H a r m a t t a h a d only p a r t of t h e
w o r d : . . . sanghika). 86 Besides, a complete inscription in B r h m on a vessel
discovered in t h e ground s t r u c t u r e above t h e P - I I t e m p l e a n d a number of
f r a g m e n t i z e d inscriptions of this order, according t o V. V. Vertogradova, bear
evidence of a dispute between t h e followers of the early H i n a y a n i c A r h a t ideal
a n d adepts of t h e M a h a y a n i c B o d h i s a t t v a ideal which was a t t h a t t i m e infil-
t r a t i n g the H i n a y a n i c school. 87 P r e s u m a b l y different trends existed and Maha-
y a n i c ideas were c u r r e n t in t h e K u s h a n period on K a r a t e p e and in t h e Termez
B u d d h i s t c o m m u n i t y , as in other B u d d h i s t centres for t h a t m a t t e r .
84
. : -.
-, II. 1969. 3 2 - 3 9 .
85
G. FUSSMAN : Une inscription Kharosthi H a d d a . B E F E O 56 (1969) 59.
86
. : -. 35.
87
. . : -. -
. IV. 1975. 708 b

The K a r a tepe inscriptions in which mention is m a d e of d h a r m a k a t h i k a

Buddhasira, give grounds t o assume t h a t this w o r t h y donator-monk m i g h t
h a v e been like D h a r m a m i t r a of t h e T i b e t a n colophon, a learned interpreter of
B u d d h i s t t e x t s . D a t a on B u d d h i s t m o n k s from B a c t r i a indulging in t r a n s l a t i o n
in China in t h e 2nd4th centuries 8 8 makes it all t h e more probable t h a t in
B u d d h i s t communities of K u s h a n B a c t r i a , in Termez c o m m u n i t y in p a r t i c u l a r ,
t e x t s were t r a n s l a t e d from S a n s k r i t i n t o t h e local, B a c t r i a n language. I t is n o t
unlikely t h a t Buddhaira, whose bilingual inscriptions bespeak the knowledge
of Indian, as well as B a c t r i a n languages, might have been one of such t r a n s -
lators.
The K a r a tepe site, its decoration and the finds m a d e during t h e excava-
tions reveal t h e syncretistic n a t u r e of t h e a r t of the K u s h a n Empire, and also
of Buddhism of t h e K u s h a n Termez (and of the entire B a c t r i a ) .
Stupas a n d cave premises (A a n d complexes) were derived from I n d i a ,
along with B u d d h i s m , whereas t h e planning of the cave temples is alien t o t h e
ancient Indian t r a d i t i o n and is r a t h e r t o be traced to t h e religious a r c h i t e c t u r e
of t h e Middle E a s t , where similar f i r e temples could be f o u n d consisting of a cell
a n d t h e corridors encircling it. 89 As t o t h e temple c o u r t y a r d s of K a r a tepe, t h e y
go b a c k to t h e R h o d o s - t y p e E a s t Hellenistic courtyards t h a t became c o m m o n
in B a c t r i a and t h e neighbouring P a r t h i a already in t h e 3 r d 2 n d centuries B.C.
Vivid illustrations are t h e large c o u r t y a r d of the Ai K h a n o u m palace complex 9 0
a n d t h e original courtyard of t h e S q u a r e House in the P a r t h i a n Nica. 9 1
The K a r a t e p e wall paintings are equally syncretistic (as distinct f r o m t h e
B u d d h i s t s c u l p t u r e of K u s h a n Termez, which, as is evidenced by t h e f i n d s ,
followed the Indian, or, to be more exact, Gandharan, t r a d i t i o n ) . The images of
t h e B u d d h a and t h e monks are executed in Indian m a n n e r , while the d o n a t o r s
are presented in t h e style of t h e K u s h a n dynastic art, which spread not f r o m t h e
I n d i a n subcontinent, like B u d d h i s m , b u t from Bactrian territory and m a y be
examplified by t h e clay sculptures of Khalchayan 9 2 and s e p a r a t e earlier finds. 9 3
88
The latest brief d a t a based mainly on the works by P . CH. BAGCHI, I n d i a a n d
Central Asia, Calcutta, 1955 a n d E . ZRCHER, The Buddhist conquest of China, Leiden,
1966, m a y be found in t h e book by . . and . . : -,
1971. 114.
89
Numerous examples are contained in the fundamental work : K. SCHIPPMANN :
Die iranischen Feuerheiligtmer. . N. Y. 1971.
90
P . BERNARD : A K h a n o u m on t h e Oxus : a Hellenistic City in Central Asia.
Proceedings of t h e British Academy 53 (1967) 7880. See also : Fouilles d'A K h a n o u m .
I . R a p p o r t prliminaire publie sous la direction de P . Bernard, P . , 1973.
91
See . . : , 1967. Pl. on p. 3 5 ;
. . : . . 1974, Fig. 37.
92
See . . : . 1971; cf. . -
1966 by the same author.
93
See . . : : .
1977, 162, 224; M. . :
() 1950 1951 .). 37. .-. 1953. 286; . . -
: - , -
, . 8., , 1956, 2 2 3 - 2 2 4 .

134 . STAVISKIY
Moreover, such details in t h e representation of t h e B u d d h a as t h e nimbus a n d

t h e halo around t h e e n t i r e figure can h a r d l y be traced t o I n d i a n tradition.
A . Grnwedel and A. F o u c h e r noted in their t i m e t h a t such a t t r i b u t e s were
e x t r e m e l y rare in t h e a r t of G a n d h a r a . A. Grnwedel assumed t h a t the nimbus
a n d especially the halo in I n d i a n sculpture were, evidently, derived from paint-
ing. 9 4 T h e K a r a tepe wall paintings t h a t m a y be dated to t h e t i m e of Kanishka,
o n whose coins the i m a g e of the B u d d h a with the nimbus a n d the halo first
occurs, confirm the a s s u m p t i o n of the f a m o u s German scholar. I t is appropriate
t o ask here was it n o t in B a c t r i a t h a t this iconographie t y p e first appeared ?
I t might be interesting to note t h a t B u d d h i s t temples similar to those of
K a r a tepe in their p l a n n i n g , as well as representation of t h e B u d d h a with the
n i m b u s and the halo a r o u n d t h e whole b o d y become c o m m o n in Eastern Tur-
k e s t a n and then in t h e F a r E a s t in t h e p o s t - K u s h a n period, which is probably
t h e result of the i m p a c t of t h e syncretistic Buddhism of K u s h a n Bactria.
T h e syncretism of culture and a r t creative assimilation of different
c u l t u r a l and artistic t r a d i t i o n s , which, embodied in the work of a r t , present an
organic combination of various artistic elements m a y be witnessed in t h e
K a r a tepe architectural decoration well. A vivid e x a m p l e is the two-belt
c a p i t a l s of the decorative pilasters. 95 T h e reliefs placed one above the other
b r i n g t h e m closer t o t h e R o m a n and early B y z a n t i n e capitals, and distinguish
t h e m from G a n d h a r a n c a p i t a l s . T h e half-figure against t h e a c a n t h u s background
m a y be traced to t h e classic tradition, t h e trefoil on t h e f i r s t capital and t h e
g a r l a n d on the second h a v e analogies in t h e a r t of G a n d h a r a , while the slanted
eyes of the girl holding t h e garland m a y be a local touch, a reflection of aesthetic
n o t i o n s of the n o m a d conquerors of B a c t r i a . The composition of t h e upper belt
( t h e central animal in f r o n t view and t h e two others f l a n k i n g it with heads
t u r n e d outwards) was k n o w n in ancient I n d i a n art. H o w e v e r , capitals with
s c u l p t u r e d animal figures, protomes, m a y be t r a c e d to ancient Persian architec-
t u r e . A t the same t i m e , t h e motif of a b e a s t of prey assailing a bull was wide-
s p r e a d not only in t h e Ancient E a s t b u t also in the world of t h e stoppes. F r o m
t h e r e also comes t h e image of a horned a n d winged griffon, depicted in the centre
of t h e upper belt of one of t h e capitals.
Of lesser artistic value, b u t no less peculiar is a f r a g m e n t of a ceramic ves-
sel with a handle in t h e s h a p e of t h e f i g u r e of a monkey. 9 6 Zoomorphic handles,
j u d g i n g by the area of t h e finds of this kind were t r a d i t i o n a l with the peoples
94
See A. GRNWEDEL : Buddhistische K u n s t in Indien. . Lpz., 1920. S. 8 3 ;
A FOUCHER: L'Art grco-bouddhique. T. I I , 1. P., 1917. 370.
95
For details see : . J . STAVISKY : T h e Capitals of Ancient Bactria. East and
W e s t , vol. 23, No. 3 4, R o m a 1973. 265 277 ; See also . . :
. 234 236. N e i t h e r of the works include the data on t h e second Kara-tepe
capital.
96
See -. III. 59 and Pl. X V I I ; . . : .
236 237.

of the E a s t European steppes, K a z a k h s t a n and Northern Central Asia. 97 I n t h e

Volga region zoomorphic handles are to be found as far back as the 2nd1st
centuries B.C. 98 Zoomorphic handles in K u s h a n Bactria are definitely a sign of
the influence of t h e notions of the steppe tribes. However, the K a r a tepe handle
differs from zoomorphic handles of the steppe world. The latter usually depicted
rams or other local animals with heads turned up to the mouth of the vessel,
which, it was believed, had to guard the contents of t h e vessel from evil spirits.
Represented on t h e K a r a tepe vessel is the monkey, a traditional image of
Indian Buddhist a r t . The head of t h e animal is t u r n e d away from the vessel,
which makes the figure rather a decorative element. This piece of handicrafts
of K u s h a n Termez is thus a vivid illustration of intermingling of different
artistic traditions the Indian and t h a t of the steppes meant probably t o
please the tastes of the local Buddhist community.
I t should be stressed t h a t handles analogous to t h a t of the K a r a t e p e
vessel 99 are not to be found either in the Eurasian steppes or in the Indian sub-
continent.
The present article does not lay claim to being an exhaustive analysis of
the finds made on K a r a tepe in the course of the 1937, and 1961 77 excavation
seasons, or a complete summary of the conclusions to be drawn on its studies.
My purpose has been to familiarize the reader with t h e m a j o r results of several
years of work on this large Buddhist religious centre of the K u s h a n period on
the bank of the Oxus.
Moscow.
97
For the recent d a t a on ceramic vessels with zoomorphic handles see : . A.
: - . 1968 (particularly 313). See also
an enlarged and amended edition of the work in German in Central Asiatic J o u r n a l
16 (1972) 2 4 4 - 2 8 8 ; 20 (1976) 47 73.
98
See M. . : .
1969, 2, 69.
99
An analogous find lias been m a d e in the vicinity of Termez a t t h e Shor-tepe
site. See . . : : -
. 1973 121, 123, Fig. 40.
A rta Antiqua Academiae Scientiarum Hungaricae 28, 1980

143 J . WOLSKI
LES SOURCES DE L'EPOQUE HELLNISTIQUE

ET PARTHE DE L'HISTOIRE D'IRAN
D I F F I C U L T S D E L E U R I N T E R P R T A T I O N E T P R O B L M E S DE L E U R
VALUATION
P o u r a b o r d e r les p r o b l m e s m o n t r s d a n s le t i t r e , il m e semble i n d i q u
d e m ' e n r e m e t t r e d e u x p r o p o s d ' i m p o r t a n c e e t q u i s o n t lis i n t i m e m e n t a v e c
le s u j e t en q u e s t i o n . E t t o u s les d e u x , ce q u ' i l m e p a r a t p r o p r e d e souligner, o n t
t exposs ici, B u d a p e s t , p e n d a n t les confrences consacres l'histoire d e
l'Asie Centrale p r i s l a m i q u e . C'est M. le P r o f e s s e u r G i g n o u x qui a v o q u le
rle e t l ' i m p o r t a n c e d e s sources orientales, u n i q u e m e n t destines recon-
s t r u i r e le cours d e l'histoire orientale. 1 E t a n t en p r i n c i p e d ' a c c o r d a v e c son
expos, je m e suis rserv, e t j ' y tiens encore a u j o u r d ' h u i , u n e exception d o n t
la validit n e m e s e m b l e p a s possible d ' t r e mise en d o u t e aussi l'avenir. I l
s'agit, bien e n t e n d u , de l ' p o q u e p a r t h e e t d e celle, c o u r t e qu'elle soit, q u i l a
prcde e t q u ' o n se p l a t d ' a p p e l e r hellnistique. 2 L a t c h e qui m ' a t confi,
e t c'est ici q u e je dois p r s e n t e r m e s r e m e r c i e m e n t s M. le Professeur J n o s
H a r m a t t a , r d a c t e u r en chef d e la Collection d e s sources sur l'histoire d e
l'Asie Centrale prislamique, m e pose d e n o u v e a u e t c e t t e fois p a s d u c t
thorique, gnral, m a i s p r a t i q u e , en v u e d e la ncessit d ' l a b o r e r le v o l u m e
d e u x de c e t t e Collection, d e v a n t le p r o b l m e d o n t il a t question plus h a u t .
Les sources d o n t n o u s disposons p o u r l ' p o q u e hellnistique et p a r t h e
s o n t , p o u r la p l u s g r a n d e p a r t i e , ou bien grecques ou bien latines. E n c o m p a r a i -
son avec le l a p s d e t e m p s qu'englobe l'histoire de l ' I r a n d e p u i s la m o r t d ' A l e -
x a n d r e le G r a n d j u s q u ' la c h u t e d e la m o n a r c h i e des Arsacides, donc d e 323
a v . n. e. 226 d e . ., d o n c cinq sicles e t demi, les sources archologiques,
d f a u t de fouilles a y a n t p o u r b u t de dgager les m o n u m e n t s de c e t t e p o q u e ,
ne p e u v e n t e n t r e r q u e d e faon restreinte en ligne d e c o m p t e p o u r r e c o n s t r u i r e
c e t t e p h a s e d e l'histoire d e l ' I r a n . 3 Cela n e v e u t p a s dire q u e nous ne disposions
q u e de sources crites grecques et latines. Certes, il y en a u n assez g r a n d n o m b r e ,
1
Pendant la confrence tenue Budapest en 1976.
2
Cf. J . WOLSKI : Points de vue sur les sources grco-latines de l'poque p a r t h e
dans : Prolegomena to the sources on the History of Preislamic Central Asia. Budapest
1976 (publi 1979). 17 26.
3
Cf. R . GHIRSHMAN : L'Iran des origines l'Islam. Paris 1951. A. G. KOSCHE-
LENKO : K u l t u r a Parfii. Moscou 1966, qui donnent l'tat do la question.

138 J. WOLSKI
p o u r n'numrer que les auteurs armniens d ' u n e valeur douteuse, 4 les chro-
n i q u e u r s chinois, mais d o n t les donnes difficiles parfois interprter p o r t e n t
seulement sur un secteur de l'histoire de la Bactriane. P o u r en finir, il f a u t se
r a p p o r t e r aux sources juives, la t r a d i t i o n indienne d o n t l'importance est
aussi prouver. P o u r fermer la liste globale, nous devons tenir compte d'in-
scriptions, semes long intervalle,sur t o u t e l'histoire de l'Iran, de papyrus, de
parchemins, de textes cuniformes p r o v e n a n t de la Babylonie. 5 Mais, vrai dire,
c e t t e varit des sources ne doit pas nous tromper sur leur importance et leur
rle d a n s la reconstruction de l'histoire de l'Iran l'poque hellnistique et
p a r t h e ; sans les a u t e u r s grecs et latins avec leur continuit d u rcit qui fait
d f a u t ailleurs, le processus historique d a n s cette rgion a p p a r a t r a i t comme
vide, comme dnu de faits, de liaisons intrinsques.
E t c'est pourquoi il nous f a u t nous poser de nouveau la question capitale :
quelle est la valeur des sources grco-latines, qu'est-ce qu nous pouvons en tirer
p o u r tre en t a t de rpondre la t c h e pose d e v a n t nous. Il va de soi, et il ne
nous semble pas ncessaire de le souligner, qu'elles reprsentent, e t elles le
d o i v e n t , bien sr, le point de vue macdonien ou romain. 6 Issues du milieu
grec et romain, ces sources, leurs a u t e u r s , se sont laisss conduire par leur
raison d ' E t a t , habitus, et qui s'en tonne, de traiter les peuples de l'Orient
c o m m e infrieurs en comparaison avec les leurs, comme barbares. Cette
a t t i t u d e , faonne e t maintenue le long des sicles, a influenc les tendances
exposes dans la science, a donn l'histoire p a r t h e un aspect partial, conu
d u p o i n t de vue des Sleucides d ' a b o r d , des Romains ensuite : les lments
p a r t h e s en f u r e n t presque exclus. U n f a c t e u r additionnel d o n t l'importance est
p o u r nous clair c'est le m a n q u e , du ct des auteurs grecs et latins, d'orientation
d a n s les institutions iraniennes, la mauvaise traduction et explications des
t e r m e s juridiques propres la socit iranienne, 7 phnomne d o n t l'apparition
r e m o n t e , du reste, l'poque des Achmnides.
4
C'est seulement, a u t a n t que je sache, B. P . LOZINSKI : The Original Homeland
of t h e P a r t h i a n s . 'S-Gravenhage 1959, qui s'est prononc en faveur, q u a n t sa valeur,
de cette couche des sources.
5
P o u r cette catgorie de sources, consulter N. C. DEBEVOISE : A Political History
of P a r t h i a . 2 ed. Chicago 1968. Les xecherches venir apporteront, certainement, la
solution du problme.
* De la nombreuse littrature relative cette attitude, on ne peut citer que les
positions les plus typiques comme J . DOBIAS : Les premiers rapports des Romains avec
les P a r t h e s et l'occupation de la Syrie. Archiv Orientlni 3(1931) 212 256, H . H . VON
OSTEN: Die Welt dor Perser. 3 ed. S t u t t g a r t 1956. 118 et passim, J . B. BURY : The
Hellenistic Age and the History of Civilisat ion d a n s : The Hellonistic Age. 2 ed. Cambridge
1968. 1 - 3 1 , qui, expressis verbis, traitent les P a r t h e s de barbares.
' C'est J . WOLSKI : L ' E t a t p a r t h e des Arsacides, Essai de reconstitution de leur
volution intrieure. Palaeologia 7 (1958/9) 91 98, qui a soulign cet aspect des sources
grco-latines. Id., avec une ample documentation, Le classi inferiori dlia popolazione
nel regno dei Parti dans : Storia sociale ed economica dell'et classica negli studi polacchi
contemporanei. Milano 1976. 55 61.

LES SOURCES DE L ' P O Q U E HELLNISTIQUE ET P A R T H E 139
Il est bien significatif que ce n ' e s t que ces derniers temps que la concep-
tion trs borne de l'histoire de l ' I r a n sous les Arsacides, d u complexe iranien
d a n s le cadre de l'histoire d'Alexandre le Grand et de ses successeurs, des Sleu-
cides, commence prendre une a u t r e allure. 8 Probablement, il f a u t en chercher
la cause d a n s le changement de l'apprciation du rle des peuples o r i e n t a u x ,
des cultures extra-europennes, opr a u j o u r d ' h u i non seulement d a n s le do-
maine p u r e m e n t scientifique, mais aussi dans la vie politique et conomique.
P o u r t a n t , ce revirement globalisant ne tient pas compte d ' u n e situation p a r t i -
culire d e m a n d a n t un t r a i t e m e n t dtaill comme c'est le cas avec l'histoire
de l ' I r a n l'poque hellnistique et p a r t h e surtout. P o u r faire ici u n p a s en
a v a n t , pour dbarasser le tissu de l'iiistoire iranienne des d f o r m a t i o n s accu-
mules le long des sicles, il f a u t t r o u v e r une approche approfondie, t o u j o u r s
conscient d ' u n d f a u t inn toutes les recherches relatives l'histoire d e c e t t e
poque, savoir des sources indignes. C'est ici qu'on se heurte un obstacle
capital qu'on doit surmonter d a n s les cadres des possibilits, bien e n t e n d u .
P o u r crer une atmosphre plus favorable la tche que nous nous s o m m e s
poss et qui d e m a n d e encore des t r a v a u x assidus, il nous f a u t reconsidrer
a v a n t t o u t une nouvelle mise au point qui s'tait forme d a n s l'aperu d u pro-
cessus historique. Au lieu d ' u n e discontinuit, parfois conue sous l'impression
des faits politiques, le plus facilement saisissables et, en consquence, s'impo-
sant avec le plus de force, on cherche sous cet aspect extrieur des f a c t e u r s
moins s u j e t s la t r a n s f o r m a t i o n , p l u s stables et c'est pourquoi plus i m p o r t a n t s ,
plus jirts nous faire une ide de la continuit des principaux c o u r a n t s d u
processus historique."
E t c'est ici que nous pouvons nous rapporter a u x rsultats des recherches
qui contribuent, sous maints aspects, tablir une a u t r e approche l'histoire
de l'poque qui a suivi la conqute de l ' I r a n par Alexandre le G r a n d . C'est
l'opinion bien sduisante et m a r q u e d ' u n nombre de rfrences de Pierre
Briant de l'chec du conqurant macdonien en vue de s'attirer la bienveillance
des Perses, des Iraniens en entier. 1 0 L a preuve de son m c o n t e n t e m e n t f u t
l'incendie de l'ensemble des palais Perspolis. Grce cette opinion trs
a t t r a y a n t e nous sommes en t a t de constater d ' u n refus de la p a r t des I r a n i e n s
8
P o u r une conception globale d u problme voir FR. ALTHEIM et R . STIEHL :
Geschichte Mittelasiens im A l t e r t u m . Berlin 1970.
9
C'est J . WOLSKI : L ' I r a n d a n s la politique des Sleucides, Festschrift J . H a r -
m a t t a ; A. A. A. S. H . 25 (1977, publi 1980) 1 7 25, qui a expos ce point d e v u e p o u r
l ' a d a p t e r l'histoire de l'Iran.
10
1'. BRIANT : Conqute territoriale et stratgie idologique : Alexandre le G r a n d
et l'idologie m o n a r c h i q u e aehmnide. D a n s : Actes du colloque i n t e r n a t i o n a l sur
l'idologie m o n a r c h i q u e dans l'antiquit, Zeszyty Naukowe Uniwersytetu .Tagiollonskiego.
P r a e e Historyczne 63 (Cahiors scientifiques de l'Universit Jagellone, T r a v a u x Histori-
ques 63). Cracovie 1980. 37 83. P . BRIANT, a p p u y sur une analyse des sources trs
p n t r a n t e , a m o n t r les m t h o d e s capables de nous rveler l'existence de la t r a d i t i o n
iranienne cache dans les sources grco-latines.
Acta Antiqua Aeademiae Srientiarum Hunguricne 28, 1080

140 J. WOLSKI
de se subordonner de bon gr, de la volont de rsister, d'entretenir l'esprit

d'indpendance vis--vis du conqurant. Hritiers d'un glorieux pass, les
Perses et les autres Iraniens, comme le montre la rsistance tmoigne par les
Bactriens et les Sogdiens, sans parler des peuples habitant l'arrire-pays de
l ' I r a n , ont fait cause commune contre l'envahisseur macdonien. Ce ne sont
pas les nouvelles sources capables de nous donner des dtails pour combler le
tissu historique mais ce sont les donnes qui nous font apparatre l'attitude des
Iraniens, certainement de l'aristocratie foncire, de son activit qui contraste
avec les opinions t a n t de fois suggres par les savants modernes de leur passi-
vit. Ces considrations sont mme de nous livrer la clef pour rsoudre maints
problmes dans l'avenir.
Si les propos de Briant mettent en lumire le ct idologique, trs ng-
lig dans les recherches, de la socit iranienne dans son comportement envers
les dynasties macdoniennes ce qui permet de voir d'un autre il le contenu
des sources grco-latines, ceux qui ont t prsentes, aussi ces derniers temps,
p a r Heinz Kreissig 11 donnent accs au milieu conomique de l'Iran l'poque
hellnistique, milieu continu aussi, sans doute, celles des Parthes. De ses
recherches on peut tirer la conviction que les nouveaux matres n'ont introduit
que peu de changement touchant la structure conomique des pays conquis en
Orient, donc aussi en Iran. En particulier, ils ont retenu les modes de produc-
tion hrits du pass. L a force des institutions faonnes depuis des sicles
sinon des millnaires, a conserv sa vigueur malgr les vicissitudes des temps
et a cr la base de la nouvelle structure politique en Iran issue de la conqute
de ce territoire par les Arsacides. Cependant, il faut s'en rendre compte, les
propositions de Briant et de Kreissig ne nous fournissent pas de dtails, mais
elles forment une nouvelle approche l'histoire de l'Iran dans la phase qui
nous occupe. Si l'on veut laborer un commentaire rpondant aux exigences des
temps actuels, il ne s u f f i t pas, comme il me semble, d'expliquer schement les
faits politiques, il f a u t les interprter et les appuyer, en analysant les sources
grco-latines, au moyen d'information tires d'autres couches du processus
historique. Bien sr, cela ne peut pas tre de rgle, pourtant sans en tenir
compte, sans en faire mention dans l'occurence, ce commentaire ne prsente-
rait aucun progrs, si important dans la science.
J e tiens exposer ce caractre d'tudes visant approfondir notre con-
naissance de l'poque en question parce que c'est grce cela que nous trouvons
une approche convenable nos propres tches. Le manque de sources iraniennes,
la division ingale des relations grco-latines, o seul Justin donne un rcit
continu des vnements, depuis les origines des Parthes jusqu'au rgne d'Au-
11
Quand il s'agit d u problme controvers d e l'influence exerce par le milieu
hellnique en Asie Antrieure dans le domaine socio-conomique, voir H . KREISSIG :
Gesellschaft und W i r t s c h a f t im Seleukidenreich. Berlin 1978, qui se prononce dcidment
c o n t r e l'opinion exagre de cette influence en e x p o s a n t le rle des facteurs traditionnels.

L E S SOURCES DE L ' P O Q U E H E L L N I S T I Q U E E T P A R T H E 141
guste, t o u t cela nous incite nous pencher sur le terrain jusqu'ici inexploit
mais propre nous d o n n e r des renseignements prcieux. Dgager le sens vri-
table des sources grco-latines et, ce qui est d ' i m p o r t a n c e , d'en tirer les infor-
mations provenant d u milieu parthe, cela, mon avis, se pose d e v a n t nous comme
t a n t d ' u n intrt m a j e u r , si nous voulons construire u n e base solide capable
de soutenir la reconstruction de l'histoire de l ' I r a n d o n t nous nous occupons. 1 2
U n e fois d j , et c'est ici, Budapest, que j'ai essay de m e t t r e en relief la
valeur de la t r a d i t i o n grco-latines pour l'histoire des Parthes. 1 3 Mon propos
s'est ramen, cette fois, vrifier, en vertu de l'analyse de quelques exemples
choisis, la valeur de S t r a b o n et de J u s t i n p o u r la h a u t e p o q u e de l'histoire de
l ' I r a n . L a critique d a n s bien des cas a p u rvler un h a u t degr de vracit
les informations fournies p a r ces auteurs. Cela ne v e u t pas dire qu'on n ' y
t r o u v e pas de d f a u t s , signals dj plus h a u t , c'est--dire d u m a n q u e de pr-
cision q u a n t il s'agit de termes juridiques, d'institutions politiques e t sociales
propres l ' I r a n .
E n t r o u v a n t cela conforme nos tches, il me semble juste d'largir notre
analyse en la s o u m e t t a n t des dmarches r p o n d a n t a u x nouveaux b u t s que
nous sommes proposs. Bien entendu, ce q u ' o n a dit plus h a u t sur une nouvelle
conception de l'histoire de l'Iran sert nous introduire d a n s un climat diffrent,
mal conu jusqu'ici e t a d o p t dans les recherches. Cependant, pour approfondir
notre connaissance de l'poque, pour gagner un lment additionel, a p t e dis-
siper les malentendus surgis propos de la barbarie des P a r t h e s , leur infriorit
culturelle en comparaison avec la supriorit suppose des Grco-Macdoniens
et des Romains, 1 4 il nous f a u t faire une chose de plus. L a critique textuelle des
sources ne suffit p a s nous satisfaire d a n s ce domaine. J e ne voudrais pas
m'en remettre mes recherches prcdentes relatives l'poque du H a u t -
E m p i r e parthe, donc de cette phase de son histoire qui finit, du point de vue
de R o m e , avec l'activit d'Auguste d o n t la politique consistait tablir un
modus vivendi avec les Parthes. 1 5 E t il me semble, d'accord avec le droulement
du processus historique chez les P a r t h e s , de mettre, e t de leur p a r t , la csure
e n t r e le H a u t - et le Bas-Empire en 20 av. n. . environ. C'est peu peu que
commence en ce temps-l de se dssiner le changement d a n s la s t r u c t u r e de la
r o y a u t parthe, arsacide expose de plus en plus a u x a t t a q u e s de l'aristocratie
12
II est bien clair que c'est do ce m a n q u e do sources iraniennes que provient,
dans les recherches, cet aspect proromain dans l'histoire dos relations romano-parthes.
Voir, p. ex., dernirement, M. G. ANGELI BERTINELLI : R o m a l'Oriente. R o m a 1979.
13
Cf. inon article cit plus haut, Points de vue sur les sources grco-latines de
l'poque parthe.
14
C'est ce problme que j'ai consacr m a communication au VI e Congrs Inter-
national d ' E t u d e s Classiques, Madrid 1974, sous le titre : Les Parthes et leur a t t i t u d e
envers le monde grco-romain. Bucurcti Paris 1976. 455 462.
15
J e tiens ici renvoyer mon tude I r a n und Rom, Versuch einer historischen
W e r t u n g der gegenseitigen Beziehungen. A N R W I I 9. 1. Berlin New York 1976. 195
214, o j'ai prsent l'tat de la question.

142 J . WOLSKI
foncire, des puissantes familles parthes, inspires par l'inimitie de la monar-

chie absolute considre comme oppressive, comme tyrannique. D ' a u t r e p a r t ,
en v e r t u du processus intrinsque, on voit s'accomplir au sein de la socit ira-
nienne un retour la tradition vieille-perse, aux traditions des Achmnides,
ensevelies pour quelque temps sous la couche superficielle d ' a p p o r t s hell-
niques. 1 6
E n comparaison avec l'poque du Haut-Empire, cette phase de l'histoire
p a r t h e , assombrie p a r les dfaites subies de la p a r t des Romains, par la triple
prise par eux de Ctsiphon, capitale parthe, est assez mal connue, si nous l'en-
visageons, bien entendu, du ct des Parthes. Contrairement l'poque du
H a u t - E m p i r e , nous ne disposons pas, pour l'poque du Bas-Empire, d ' u n
rcit continu, comparable celui de J u s t i n . Il nous f a u t donc nous contenter
d'informations exposant, du point de vue de Rome, les exploits des armes
romaines, des empereurs romains et o on ne tient compte que de faon trs
faible des conditions socio-conomiques en vigueur dans l ' E t a t des Arsacides.
E t c'est pourquoi, comme il me semble, on doit d ' a u t a n t plus mettre l'accent
sur la tradition grco-latine, en extraire des donnes accusant la provenance
p a r t h e . C'est mon avis l'unique vois qui puisse nous conduire vers une plus
g r a n d e confiance l'gard de la tradition romaine, nous inspirer plus de foi
d a n s sa vrit, sape, et qui s'en tonne, pour le I I I sicle de n. ., par la d-
couverte des Res gestae divi Saporis. 17 E t c'est par cela qu'on peut aller
l'encontre de la suggestion avance par le professuer Gignoux de donner la
priorit pour reconstruire l'histoire de l'Orient aux sources orientales, cette
fois parthes. Ce serait, du reste, demander t r o p cette t a p e des recherches,
de parvenir dj des rsultats plus importants, d'puiser le problme qui ne
vient que d'tre pos. Quand mme, il y a quelques points de repre assez signifi-
catifs dont l'impression nous prdispose suivre la voie une fois choisie.
Dans mes tudes mettre de la lumire sur le problme de l'idologie des
Arsacides, sur leur capacit, t a n t de fois mise en doute, de se laisser conduire
d a n s leur politique par des plans d'envergure, je me suis pench sur une phrase
de Tacite, Ann. V I 31, qui mentionne un vnement a y a n t t r a i t l'histoire
des relations romano-parthes. Il s'agit d'une lettre adresse p a r le roi p a r t h e
A r t a b a n I I l'empereur Tibre. 1 6 L a voici: missis qui reposcerunt ; simul
veteres Persarum ac Macedonum terminos, seque invasurum possessa primum
Cyro et post Alexandro per vaniloquentiam ac minas iaciebat. Sans a t t a r d e r
a u x expressions issues sans doute du milieu romain et tmoignant de la manire
16
Cf. J . YVOLSKI : Les Achmnides et les Arsacides, Contribution l'histoire
de la formation des traditions iraniennes. Syria 43 (1966) 65 89.
17
Pour une dernire mise au point, voir M. G. ANGELI BERTINELLI : R o m a >
l'Oriente. 103 ss., o on trouve cite la littrature la plus rcente relative cette question.
18
Cf. J . WOLSKI: Les Achmnides et les Arsacides. Syria 43 (1966) 65 89,
o c e t t e question a t analyse en dtail.
Acta Antiqua Academiae Scientiarum Hungaricae 28 ,1980

LES SOURCES DE L ' P O Q U E H E L L N I S T I Q U E E T P A R T H E 143
habituelle du ddain avec lequel les cercles impriaux traitaient les P a r t h e s ,

et d ' a u t a n t plus, sur le tmoignage trs convaincant de la pense politique des
Arsacides, se posant en successeurs des Achmnides, il f a u t tirer de cette p h r a s e
une conclusion valable pour nos recherches. C'est que, d a n s les archives romai-
nes, probablement de l'poque impriale, on a conserv un nombre de t e x t e s
officiels parthes, utiliss ensuite p a r les crivains romains dans les p a r t i e s per-
tinentes de leurs rcits relatifs a u x P a r t h e s . Cet exemple confirme l'existence
de tels documents ce qui ne semble tre une preuve de premier ordre, c o m p t e
tenu de ce peu d o n t nous disposons ; cependant, fortiori, les gains qu'ils
nous a p p o r t e n t , sont encore plus multiples. D ' a b o r d ils nous inspirent de la
confiance dans la vracit des sources romaines, grecques et latines, a p p u y e s ,
d a n s une mesure qui nous chappe, bien entendu, sur une q u a n t i t de docu-
ments, lettres, traits, rapports etc., t a n t la preuve de l'activit d i p l o m a t i q u e
du milieu arsacide. 1 9 S'il en est ainsi, et il n ' y pas de raison d'en douter d ' a p r s
l'exemple cit plus h a u t , tous ce d o c u m e n t s ont t transmis la postrit p a r
l'intermdiaire de l'historiographie romaine.
Si nous ne pouvions citer que la phrase de Tacite mentionne ici p o u r faire
valoir l'ide exprime ci-dessus, on n'en a p u tirer q u ' u n e conclusion limite,
sans pouvoir l'utiliser dans le b u t de crer une conception plus large, plus gn-
rale. Cependant, nous ne nous t r o u v o n s pas dans une situation exceptionelle,
au contraire, il y a d'autres cas p o r t a n t dans le mme sens. C'est l'occasion
de t r a i t e r le problme du Caucase et de la Mer Noire d a n s la politique de R o m e
et des P a r t h e s , que j'ai pu me rfrer une lettre d ' u n souverain p a r t h e d o n t
la teneur nous a aussi t transmise p a r Tacite. 2 0 Cette fois, le personnage prin-
cipal est le roi p a r t h e Vologse I qui, l'occasion de l'intronisation de Tiridate,
son frre, et du rglement de l'affaire armnienne a prcis dans u n e l e t t r e
adresse l'empereur Nron les fondements, les directives de la politique de
l ' E t a t p a r t h e face l'Armnie. L a t e n e u r de Tacite, Ann. XV 24, est la sui-
v a n t e : se (Vologesis) priora et totiens i a c t a t a super optinenda Armenia n u n c
omittere. Ce n'est pas ici la place d ' a n a l y s e en dtail le programme a r m n i e n
des Arsacides. Ce qui importe, c'est de m e t t r e en relief, cette fois encore, le
rle de la tradition parthe, des d o c u m e n t s parthes insrs dans la t r a m e d e la
n a r r a t i o n de l ' a u t e u r latin. L ' a n a l y s e d u texte de Tacite suggre et nous f a i t
19
Que l'activit de la cour de Ctsiphon, en vue de prsenter les plans politiques
des Arsacides sur les problmes relatifs aux relations romano-parthes, ne se limitait
pas a u x genres cits dans le texte, qu'elle utilisait d'autres moyens pour a t t e i n d r e les
b u t s proposs, nous en avons des preuves indirectes dans une phrase clbre de T. Live.
I X 18,6 : Levissimi ex Qraecis qui Parthorum quoque contra nomen Romanum gloriae favent.
C'est un grand dommage pour la science de voir toute cette littrature perdue qui, con-
serve, serait mme, certainement, de jeter un peu de la lumire sur les m t h o d e s de
la propagande partho.
20
Cf. J . WOLSKI : Caucase et Mer Noire entre Rome et les Parthes. Actes d n XVI"
Congrs International des Sciences Historiques. Bucarest 1980.

144 J. WOLSKI
croire l'existence, d u reste pas exclusivement chez Tacite, d ' u n e ample docu-
m e n t a t i o n p r o v e n a n t de cercles officiels p a r t h e s et conserve dans les archives
romaines, exploite p a r les historiens r o m a i n s dans les parties pertinentes de
leurs uvres.
E t en v e r t u d e ces exemples d o n t le n o m b r e p e u t certainement tre
a u g m e n t , se fortifie la conviction p o u r le contraire on devrait citer des
a r g u m e n t s d ' u n poids s u f f i s a n t de la h a u t e valeur des sources romaines
t o u c h a n t les relations romano-parthes. Cela d i t , on est loin d'englober d a n s
c e t t e opinion t o u t e l'histoire de l'Iran de l ' p o q u e p a r t h e qui de telle manire
m e n a c e r a i t de n ' t r e perue que sous l'angle romain. Qu-il y a encore beaucoup
d e domaines lacuneux, presque vides de f a i t s historiques, cela s'ensuit de l'exa-
m e n des sources l'histoire de l'Iran p r o p r e qui restent m u e t t e s et ne nous
d o n n e n t pas rponse beaucoup de questions. E t elles auraient t r a i t l'histoire
intrieure de l ' I r a n des P a r t h e s , aux querelles dynastiques, a u x sditions des
s a t r a p e s , aux c o n t a c t s t a n t t pacifiques, t a n t t belliqueux avec les voisins
p a r f o i s turbulents. 2 1
Ce qui nous c h a p p e , ce sont les causes, les voies, les milieux o s'est
o p r le grand processus intrieur et qui, p a s pas, a permis a u x Iraniens de
r e f a i r e leurs liens traditionnels avec l'hritage des Achmnides en f r a y a n t de
telle faon la renaissance grandiose de l ' p o q u e des Sasanides 22 . E n vue de
l'opinitret avec laquelle les Sasanides o n t perscut les souvenirs de l'poque
d e s Arsacides, pour en effacer la mmoire m m e au moyen d ' u n e telle manipu-
l a t i o n que de limiter la d u r e de leur rgne 200 ans, 2 3 il me semble indiqu de
soulever cette question pour nous rendre prsents les problmes qui restent
encore rsoudre p o u r la pense historique moderne. Les d f a u t s des sources
y s o n t clairs.
J ' a i cru ncessaire de faire l'examen de toutes ces questions abordes
seulement trs r a r e m e n t d a n s les recherches a f i n d'exposer mon parti pris, de
l ' a u t r e , pour m o n t r e r quelles lacunes d o i v e n t tre prises en considration et
d o i v e n t tre marques a v a n t que nous puissions trouver notre tche, de donner
u n commentaire complet des sources sur l'histoire de l'Asie Centrale p o r t a n t
s u r l'poque arsacide, comme puise.
Pour finir, on ne p e u t pas omettre de souligner le rle des considrations
c o m m e celles-ci sur n o t r e aperu des possibilits des P a r t h e s . L a structure int-
21
L'exemple le plus m a r q u a n t en est l'histoire des Juifs de Babylonie dont l'atti-
t u d e a parfois caus l ' E t a t p a r t h e de graves inconvnients. Cf. J . NEUSNER : A History
of t h e J e w s in Babylonia. I. The Parthian Period. Leiden 1965. 50 ss.
22
L'opinion d u rle actif de l'poque arsacide dans la transmission de l'hritage
des Achmnides aux Sasanides commence a u j o u r d ' h u i de plus en plus prendre de la
valeur dans la science. Cf., p. ex., G. WIDENGREN, Die Begriffe populorum ordoi) und
ram als Ausdrcke des Standesgliederung im Partherreich. Festschrift Walter Baetke.
W e i m a r 1960. 384. note 25.
23
Cf. J . WoLSKi : Arsakiden und Sasaniden. Beitrge zur Alten Geschichte und
d e r e n Nachleben. Berlin 1969. 315-322.
Acta Antiqua Aeademiae Srientiarum Hungarirae 28, 1080

L E S SOURCES D E L ' P O Q U E H E L L N I S T I Q U E E T P A R T H E 145
rieure de leur E t a t s'avre plus dveloppe qu'ou ne le souponnait jusqu'ici.

L ' a c t i v i t diplomatique de la chancellerie royale mene probablement en lan-
gue grecque, les directives de leur politique rvles d a n s les lettres des rois,
la stabilit dans l'excution des plans une fois conus et mis en action, t o u t cela
nous oblige d'enclure dans les rflexions, d o n t l'effet sera le commentaire, u n tel
aspect de l'poque p a r t h e qui rpond a u x ralits de l'histoire parthe.
Krakow.

153 E. A. DAVIDOVICH
THE FIRST HOARD OF TETRADRACHMAS OF

THE KUSNA HERAIOS
T h e f i r s t h o a r d of silver t e t r a d r a c h m a s of Heraios h a s been found in t h e

s o u t h e r n p a r t of T a j i k i s t a n in t h e t e r r i t o r y of t h e d i s t r i c t c e n t r e of V a k h s h
(district Vakhsh) a t t h e construction of a house. 1 The t y p e of t h e t e t r a d r a c h m a s
of Heraios has been described a l t h o u g h very briefly, in l i t e r a t u r e . H e r e it is
i m p o r t a n t t o stress t h o s e details and characteristics, which can b e essential for
further study.
Obverse. T h e b u s t of t h e ruler : f a c e with m o u s t a c h e (and with a s h o r t
b e a r d ?) in r i g h t profile, t h e shoulders b e i n g a l m o s t in f r o n t - v i e w . T h e d i a d e m
s u r r o u n d s t h e f o r e h e a d a t t h e root of t h e hair, t h e f r o n t a l s h o r t locks are b r u s h -
ed b a c k , while t h e r i g h t long locks are freely hanging d o w n , covering t h e ear.
T h e f a c e is s o m e t i m e s m o r e r o u n d , a l m o s t p u f f y , and s o m e t i m e s more t h i n a n d
e v e n vehement. 2 Stylistic differences can be observed in certain details.
T h e eye is represented sometimes in f r o n t - v i e w a n d s o m e t i m e s in profile.
T h e m o u s t a c h e is s o m e t i m e s q u i t e s t r a i g h t and s o m e t i m e s slightly pointed w i t h
s o m e w h a t d r o o p i n g or t u r n i n g u p ends. T h e locks are c u t s t r a i g h t a t their ends
f o r m i n g sometimes slight a n d s o m e t i m e s s h a r p waves. T h e r i g h t lock of t h e hair
t u r n e d back f r o m t h e f o r e h e a d f r e q u e n t l y falls freely on t h e d i a d e m , b u t some-
t i m e s i t fuses w i t h t h e long locks i n t o o n e line. T h e ends of t h e d i a d e m are q u i t e
s h o r t and curved, t h e y slightly v a r y in t h e length and s t e e p n e s s of t h e arch, b u t
' T h e original dimensions of the hoard are not known. The late N. N. ERSHOV,
a well-known Tajik specialist in ethnography has informed of the circumstances and
place of its find. He has already succeeded to collect six coins (N-377), which he handed
over to the Institute of History of the Academy of Sciences of the Tajik SSR (further
on IIT). One coin (N-341) was handed over to the I I T by the journalist V. . SITNIKOV.
And an additional coin was given to the Republican Museum of History and Culture of
the Tajik People (further on RMT, Inv. No. 1-2973) by the collective farmer H. IDRISOV.
To the eight coins of the hoard mentioned later in the article the following aerial numbers
have been given : Nos. 1 to 6, H T , N-377/1 6 ; No. 7, IIT, N-341 ; No. 8, RMT, 1-2973.
2
I t is possible that this is not simply a stylistic difference b u t a difference of age,
if the representation is a portrait. In this connection, however, it is important to stress
the great similarity of the coins of Heraios to tho sculptural representations of two
monuments of different times and distant from each other : by Khalchayan in Southern
Uzbekistan (G. A. PUGACHENKOVA : , V D I 1965 No. 1. 132134;
eadem : . Tashkent 1966. 188189; eadem : . Moscow
1971. 45 46) and by H a d d a (J'.HACKIN, L'uvre de la Dlgation Archologique
Franaise en Afghanistan (1922- 1932). I. Archologie bouddhique. Tokyo 1933. 11.
Fig. 12; R. ROWLAND: Zentralasien. Baden-Baden 1970. 27 28, 48).

148 E . A. DAVIDOVICH
t h e y always s t a n d o u t backwards, not hanging down. T h e head is sometimes

slightly lifted u p and thrown back, and sometimes it shows a more straight posi-
tion. The bust of t h e ruler is surrounded by a f r a m e consisting of longish beads
alternating with two vertical strokes each (resembling occasionally dots).
I n those cases, when this f r a m e has been preserved below, i t can be seen t h a t on
different coins t h e bust was cut in different heights. The clothing is always the
same, viz. underneath a shirt m a d e of soft textile with b r o a d round neckline,
keeping open t h e neck with a wrinkle under the chin ; above there is a c a f t a n
w i t h a roll-collar (or with border ?). The representation and the f r a m e are
arranged exactly in t h e centre of t h e flan only on coin No. 1, and in this case it
can be seen very clearly t h a t t h e size of the stamping-die was slightly smaller
t h a n the size of t h e flan. On t h e other coins of the hoard, t h e representation is
s h i f t e d sometimes slightly and sometimes considerably in some direction, so
t h a t p a r t of the f r a m e (below occasionally also an insignificant part of the bust)
is completely outside t h e flan, and if t h e representation a n d t h e frame were dis-
placed still entirely, then on t h e margin of the opposite side, there remains a
large free field on t h e flan.
Reverse. Mounted figure with quiver on horseback shown from the right
s i d e ; the quiver reaches down to t h e level of the foot. F r o m behind the goddess
N i k e flies towards t h e mounted figure ; her relative size, pose and position com-
p a r e d with the horse and the m o u n t e d figure, as well as t h e stylization of the
costume are somewhat different. T h e horse is pacing, its l e f t foreleg is lifted up.
I t s tail is tied round so t h a t sometimes five and sometimes six knots (represented
w i t h dots) are formed. The end of t h e tail is pointed. T h e horseman is sitting
freely, the reins in his left hand are slackened, his right h a n d bent in t h e
elbow reaches backwards, t h e torso is turned almost i n t o front view, and
t h e head is in profile. The mounted figure is obviously t h e same ruler as the one
represented on t h e obverse. In spite of the small size of t h e representation, the
characteristics of t h e coiffure and t h e face are definitely stressed, and the de-
tails of the attire are given even more accurately. The border (roll-collar or
l a p p e t ?) of the c a f t a n reaches down to the waist. The t y p e of t h e representation
is t h e same. The representation differs on all the eight coins of the hoard in t h e
details, in the m a n n e r and skill. T h e legend is arranged partly above t h e
m o u n t e d figure in an arched line a n d p a r t l y under and between the legs of the
horse in two straight lines. On t h e different coins the arched line of the legend
differs in the bending of the arch, in its base (the beginning and the end are
approximately in t h e same level, b u t more above or more below in comparison
t o t h e horseman), a n d in the size of t h e characters. The word between the legs
of t h e horse consists sometimes of five (coins Nos. 1 to 4 a n d 6 to 7) and some-
t i m e s of seven characters (coins Nos. 5 and 8), the a r r a n g e m e n t of which can be
d i f f e r e n t (see Table 1 ).

T H E F I R S T HOARD OF TETRA DRACHMAS OF ' H E R A I O S ' 149
Table 1
No. Hind legs of Front legs of

the horse the horse
/ ,
1 z AN A
2 z AN A
3 z A
L z [A]
5 I ZA 8 V
6
M AN A
7 z A
8 z* N * V
Reverse : characters arranged between the legs of t h e horse
T h e r e l a t i o n s h i p of t h e a x e s of t h e o b v e r s e a n d r e v e r s e s i d e s is d i f f e r e n t .
B o t h t h e o b v e r s e a n d r e v e r s e s i d e s of all t h e e i g h t c o i n s of t h e h o a r d j u d g e d
f r o m t h e d e t a i l s a n d m a n n e r of t h e r e p r e s e n t a t i o n , a s w e l l a s f r o m t h e p e c u l i a r i -
ties of t h e inscriptions were m i n t e d w i t h d i f f e r e n t dies.3
3
A . N. Z GRAF, author of the most thorough and many-sided research in t h e
coins of Heraios, on t h e basis of the obverse and the inscription between the legs of t h e
horse on the reverse side, divides t h e tetradrachmas into two groups (A. N. ZOGRAF :
. T a s h k e n t 1937. 15) : 1) low cut of the bust, straight carriage of t h e
head and accurate beard (on the reverse side the word between t h e legs of the horse

T h e legend on t h e t e t r a d r a c h m a s of Heraios has raised a great polemic

a n d i t cannot be regarded as definitively deciphered and understood even today.
I n t h e course of the accumulation of material it has become clear t h a t the legend
is t h e same on all t e t r a d r a c h m a s , b u t on some coins it differs in certain charac-
ters. Consequently, on certain coins there were errors and distortions in it t h a t
p e r m i t t e d of different restorations of its original, basic t e x t . This uncertainty
r e n d e r e d its deciphering and understanding even more difficult, and has
b r o u g h t about f u r t h e r hypotheses.
T h e legend consists of four words. A t present t h e original spelling (al-
t h o u g h there are errors in the execution), the reading (with t h e consideration of
t w o palographie details) 4 and the translation of the first word TYPAN-
NOYNTOY of the ruling do not raise any doubts. Among t h e t e t r a d r a c h m a s
of t h e hoard this word appears on coin No. 8 in correct form, a n d on coins Nos.
1 t o 7 with different errors. There exists, however, no convincing historical
i n t e r p r e t altion so far. According to t h e interesting comment by Kennedy, this
t i t l e reflects the vassal condition of its bearer. 5
T h e restoration of the original form of the second word HIAOY on the
t e t r a d r a c h m a s does not raise any earnestly founded r e f u t a t i o n , although its
second character is frequently fully missing or appears in t h e form of a short
s t r o k e upwards ; there are also errors. On coins Nos. 2 a n d 8 of the hoard the
spelling is exactly t h e same, viz. HIAOY, while on coin No. 6 there is a short
s t r o k e in place of t h e second character, on coin No. 4 it appears without the
second character, and on the other coins it contains errors. I t is i m p o r t a n t to
n o t e t h a t the first character on seven coins is an eta, on one it is an omicron
in q u a d r a t i c form (the origin of this a n d the other errors will be explained
below), and the mu does not occur as an initial character a t all (here we do not
discuss the spelling of this word on the obols).
T h e most accepted reading (Heraios) and interpretation (Greek proper
n a m e ) are, however, still disputed and doubted. 6
consists of seven characters) ; 2) high cut of t h e bust (almost a t t h e border of the shirt),
t h e neck bent ahead, t h e head of rather big size a n d slightly lifted up, the beard is not
so a c c u r a t e (on the reverse side the same word consists of five characters). The coins of
t h e h o a r d partly support this classification (see Pis. IIII) : Nos. 5 a n d 8 low cut of
t h e b u s t and this word on the reverse side consists of seven characters ; Nos. 1 3 high
cut of t h e bust and this word on the reverse side consists of five characters. The carriage
a n d t h e size of the head, apparently, are no accurate criteria for classification.
4
The third character ro usually has the form of an inclined stroke, and the omicron
is of q u a d r a t i c form, in t h e setting, however, it is always substituted for a round form.
5
ZOGRAF : . 11 12.
6
Y . G. VINOGRADOV kindly stated to m e t h a t the name 'Hgaio (or "Hgaio)
is a theophoiic name t o be derived from the n a m e Hera or from the m o n t h 'Hgalog of
c e r t a i n Doric and other calendars (Crete, Delphi, Byzantium, Bithynia, Chios, Zeleia,
see A. SAMUEL: Greek a n d R o m a n Chronology. Mnchen 1972. 74, 88, 117, 131, 175),
d o c u m e n t e d : 1) on t h e coin from K y m e a n d in several inscriptions, see W. PAPE
G. BENSELER : Wrterbuch der griechischen Eigennamen. Braunschweig 3 1884. 462 ;
2) in Chersonesus I O S P E I 2 , 359 ; E. I. SOLOMONIK :

T H E FIRST HOARD OF T E T R A DRACHMAS OF ' H E R A I O S ' 151
T h e t h i r d w o r d , a r r a n g e d u n d e r t h e f e e t of t h e h o r s e , w a s r e a d a s K u s -
n a . T h e r e a d i n g d o e s n o t r a i s e a n y d o u b t . O n t h e t e t r a d r a c h m a s , e.g., t h e b a s i c
r e p r e s e n t a t i o n is KOI I AN Y, b u t t h e t h i r d a n d f o u r t h c h a r a c t e r s s t a n d s o m e
t i m e s separated a n d sometimes t o g e t h e r , and have s o m e t i m e s a slightly a r c h e d
f o r m t h a t s h o u l d n o t b e r e g a r d e d a s e r r o r s , b u t a s o t h e r v a r i a n t s of d i f f e r e n t
m e a n i n g , s h o w i n g t h e a t t e m p t s t o r e n d e r t h e s o u n d s. T h e i n t e r p r e t a t i o n o f t h e
w o r d Kusna a s t h e n a m e of o n e o f t h e f i v e t r i b e s of t h e Y u e h - c h i h , 7 h o w e v e r ,
is n o t i n d i s p u t a b l e . A f t e r K u j u l a K a d p h i s e s , t h e K u s n a yabghu, h a s s u b d u e d
t h e o t h e r f o u r yabghu, his e m p i r e w a s c a l l e d K u s n a . A l l o t h e r p e o p l e s u s e d t h i s
n a m e , a n d o n l y t h e C h i n e s e H a n d y n a s t y called t h e m i n v a r i a b l y G r e a t Y u e h -
c h i h . T h e q u e s t i o n o n t h e o r i g i n o f t h e w o r d Kusna a n d o n i t s m e a n i n g b e f o r e
i t b e c a m e t h e n a m e of t h e d y n a s t y a n d t h e e m p i r e i n t h e i r o w n l a n g u a g e , a n d a s
s u c h it was accepted b y t h e o t h e r peoples, however, m u s t not be e x a m i n e d
i s o l a t e d , o u t s i d e t h e g e n e r a l c o n t e x t of i n f o r m a t i o n s o n t h e Y u e h - c h i h .
I n f a c t , i n t h e s o u r c e s t h e r e is n o d i r e c t r e f e r e n c e t o t h e f i v e t r i b e s ( c l a n s )
of t h e Y u e h - c h i h . I n t h e H a n - s h u a n d t h e H o u H a n - s h u f i v e h s i - h o u ( y a b g h u )
a r e m e n t i o n e d . T h e i n f o r m a t i o n s of these two sources on t h e f i v e hsi-hou a r e n o t
e n t i r e l y i d e n t i c a l . T h e y s e e m t o b e d i f f e r e n t in r e s p e c t of t h e i r c o n t e n t s , b u t
t h e y s t r e s s in t h e s a m e d e g r e e t h e n e c e s s i t y of t h e i n d e p e n d e n t e x a m i n a t i o n of
t w o q u e s t i o n s i n t h i s c o n n e c t i o n , viz. 1. o n t h e b e l o n g i n g of t h e f i v e t e r r i t o r i e s
, [II]. Kiev 1973. 146. No. 147 ; 3) in the Bosporus K B N , 166, 692, 1137, B,
2.22; V. F . GAJDUKEVICH: , . Moscow
1966. 70; 1 . 1 . TOLSTOJ: .
MoscowLeningrad 1953. 93, No. 149 ; 4) in Athens, see J . KIRCHNER : Prosopographia
Attica B. 1901. Nos. 6435, 6436 ; 5) in t h e island of Eresos, t y r a n t a t the time of Alexan-
der OGIS, 8.37. By these, in his words, all men tionings of this name, well spread u p to
t h e Roman period, are not being exhausted.
I t must he n o t e d that the interpretation of this word as a name, but in t h e spell-
ing MIAOY and MIAIOY, proposed in the initial phase of t h e study of these coins, was
convincingly r e f u t e d b y the n u m i s m a t i s t s on the basis of a more objective analysis of
the old and the newly a c c u m u l a t e d material. In spite of this, in certain recent mono-
g r a p h s devoted to the history a n d numismatics of the K u s n a s this spelling with the
first character mu a n d with the r e a d i n g Miaou has emerged again. Thus, BHASKAR
CHATTOPADHYAY (The Age of the K u s h n a s A Numismatic Study. Calcutta 1967.
11 16, 20), without any motivation a n d evidence, only in t h e beginning gives t h e n a m e
of this ruler as Miaus (or Heraus), a n d later in the whole excursus there a p p e a r s defi-
nitely the writing MIAOY and the reading Miaus. I n this t h e author relies u p o n the
tetradrachma of the British Museum, published by P. GARDNER (but on t h a t , judged
from t h e photograph, there appears n o t an M but exactly an H ; P . GARDNER himself
also read II, see P. GARDNER : Catalogue of Indian coins in t h e British Museum. Greek
a n d Scythic kings of Bactria and I n d i a . Chicago 1966. X L V I I , 116, Pl. X X I V , 7), and
also upon the obol of the Museum of I n d i a in Calcutta (on which the first character does
not at all resemble to an M). It is possible, that the acceptance of such a writing and
reading is explained by the circumstance t h a t the author is not acquainted with t h e whole
literature of the question. B. N. MUKHERJEE (Studies in K u s h n a genealogy and chronol-
ogy. Vol. I. The K u s h n a genealogy. Calcutta 1967. 92, note 6) asserts t h a t he has studied
t h e coins of the British Museum a n d he has convinced himself that on t h e m a t least,
one must read Miaos (MIAOY). I n his text wo find everywhere consistently Miaos.
' ZOGRAF : - 31 32.

or possessions with cities-capitals and on their place in the empire of the Great
Yueh-chih ; 2. on t h e origin of the names of t h e hsi-hou and t h e transformation
of t h e i r meaning. 8
Chang Ch'ien, who visited the Great Yueh-chih between 130 and 125 B.C.
(some scholars mention m o r e accurate, b u t different dates), describes the fol-
lowing situation : t h e G r e a t Yueh-chih conquered the territory of Ta-hsia
(south of the Amu D a r y a ) , b u t they settled down north of t h e river and led
t h e r e a nomadic way of life. I t is important t h a t in the description of the bound-
aries of the empire of t h e G r e a t Yueh-chih Ta-hsia is mentioned as the southern
boundary.
Five districts or possessions are not mentioned either in t h e territory of
n o m a d i c life north of t h e river or in Ta-hsia. T h e lack of reference is, of course,
no evidence. Chang Ch'ien, however, did visit Ta-hsia, and his detailed descrip-
tion (They have no g r e a t kings or heads, b u t everywhere in their walled cities
a n d settlements they installed small heads) 9 does not render likely the exis-
tence of five or less districts or possessions. T h e situation described by Chang
Ch'ien is called by us conditionally the f i r s t stage in the history of the Yueh-
chih a f t e r they conquered t h e territory on b o t h sides of the Amu Darya.
I n the Han-shu (completed about 80 A.D., the events are described up
to 25 A.D.) a completely different situation is depicted, viz. t h e residence of the
king of t h e Great Yueh-chih is in a city n o r t h of the river ; t h e southern bound-
a r y is no more Ta-hsia, b u t the district K i - p i n ; they themselves originally
developed as a nomad s t a t e ; there are five hsi-hou, which are subjected to the
G r e a t Yueh-chih. 1 0 T h u s m a n y things have already changed towards the joining
of t h e nomad conquerors in the local culture and economy (and, partly, very
likely in the settled life) a n d towards a more definite formation of the state of
t h e G r e a t Yueh-chih itself. This is shown by t h e residence in a city, b y the estab-
lishment of the boundaries of the state (the annexation of Ta-hsia originally
conquered only formally), as well as by some determination of t h e system of
s t a t e schematically fixed b y the existence of t h e king of t h e G r e a t Yueh-chih
a n d t h e five districts or possessions governed by subordinated hsi-hou (yabghu).
T h e development and d u r a t i o n of the existence of this situation is unknown,
8
We used the English translations of the most significant contemporary sinologists
PULLEYBLAMC and ZRCHER t h a t are in certain essential details different from the
R u s s i a n translation by BICHURIN, E . G. PULLEYBLANK : Chinese Evidence for the Date of
K a n i s k a , in Papers on t h e D a t e of Kaniska, s u b m i t t e d to the conference on the date of
K a n i s k a , L., 20 22 April 1960, ed. by . L. BASHAM. Leiden 1968; E . ZRCHER: The
Yeh-chih and Kaniska in t h e Chinese Sources (ibidem) ; N. J a . BICHURIN (IAKINF) :
, , I I . Mos-
cow Leningrad 1950.
9
Z R C H E R : op. cit. 360.
10
ZRCHER: op. cit., 3 6 4 ; A. F. P. HULSEW : China in Central Asia. The Early
S t a g e : 125 B. C . - A . D. 23. Leiden 1979. (Sinica Leidensia, Vol. XIV.) 121 123.

T H E F I R S T HOARD OF TETRADRACHMAS OF ' H E R A I O S ' 153
i t is only clear t h a t it existed earlier t h a n 25 A.D. 1 1 and it represents a new stage

in t h e history of t h e Yueh-chih t h a t is called by us t h e second stage.
Who were the yabghu, the rulers of these five districs or possessions, each
of whom, similarly to the head of t h e Great Yueh-chih, had his residence in his
own district ? The existing different viewpoints have to be divided into t w o
groups, viz. 1. yabghu no Yueh-chih ;12 2. yabghu Yueh-chih.
The t e x t of the Han-shu, t a k e n isolated, permits both assumptions.
The logical conclusion from the f i r s t one (i.e. if the yabghu is no Yueh-chih)
will be as follows : Kadphises I, yabghu of one of the five districts, was also no
Yueh-chih, b u t a local or pre-Yueh-chih dynast. Thus t h e K u s n a s t a t e estab-
lished by him was not Yueh-chih either. Such a conclusion, however, contradicts
to t h e sources of the later period and to the fact t h a t in t h e m the K u s n a empire
appears as Yueh-chih. The logical conclusion from t h e second assumption
(yabghu is Yueh-chih) will mean t h a t there existed a hierarchy among t h e
Great Yueh-chih, and the yabghu t h e heads of the five tribes (clans) were
in a certain dependence from t h e head of the Great Yueh-chih, i.e. from t h e
whole of their tribal union. In this case the t e x t of t h e Han-shu would m e a n
t h a t the five Yueh-chih yabghu received five districts or possessions in t h e ter-
ritory of the s t a t e of the Great Yueh-chih, but practically or formally they were
not independent. Very likely it can be presumed t h a t originally this was a
nomadic division t h e territories of which, in accordance with the adoption of
sedentary life and culture, gradually were transformed into more definite pos-
sessions. 13 Since the envoy Chang Ch'ien does not speak about the five nomadic
camps and possessions, and in Ta-hsia he even describes a different situation,
it can be presumed t h a t the quite accurate division of t h e nomadic camps a n d
even more of the possessions took place later, when it formed a significant com-
ponent of the second stage. W i t h other words, even in t h e short and general
testimonies of the two sources from different times in t h e report of Chang
Ch'ien and in the Han-shu undoubtedly i m p o r t a n t changes are reflected,
giving so to say t h e minimum a b o u t the two stages. There is no doubt t h a t
changes must have taken place also later and, therefore, their reflection can
also be expected in later sources.
If we look a t the next, third source the Hou Han-shu (in this source
the information about the Western Regions dates back to 125 A.D.) from such,
11
E . PUI.LEYBI.ANK (op. cit. 250254) has convincingly shown t h a t the passage
on t h e five hsi-hou appears to be a later insertion from t h e years 7475 A. D . a n d
reflects t h e situation of these years. T h e latter does not appear to bo compulsory ; t h e
insertion may be an additional information on the past history.
12
On the different assumptions a b o u t their origin in this case see ZRCHER :
op. cit., 365, note 1 ; G. HALOUN : Zur etsi-Frage. ZDMG, Bd. 91, 1937, 257, n o t e 7.
13
Undoubtedly, these nomad camps or possessions corresponded both to t h e
physico-geographical and (judged from t h e existence of capitals) t o t h e historical a n d
cultural rogions, what is important for the localization.
Arta Antiqua Academiae Scientiarum Hungaricae 28, 1980

not static viewpoint, then it appears t h a t it does not c o n t r a d i c t the Han-shu

and is no distortion (as it was thought b y certain researchers), b u t it simply
reflects another new s t a g e in the history of the Yueh-chih t h a t we call the third
stage.
The compilator of t h e Hou Han-shu himself stated t h a t he was interested
only in the situation f r o m the time of t h e foundation of t h e L a t e r H a n dynasty,
which is entirely d i f f e r e n t from the previous one. I t is not by accident t h a t t h e
whole early history of t h e Great Yueh-chih is limited with him to two informa-
tions, viz. they clashed with the Hsiung-nu and moved to Ta-hsia. The follow-
ing events stood in t h e sphere of direct interest of the compilator of the Hou
Han-shu : the yabghu K u s n a subdued t h e other four yabghu, declared himself
king a n d called his s t a t e Kusna. Till t h e time of the unification, therefore, t h e
five possessions headed by Yueh-chih yabghu were independent, there was no
hierarchy and subordination to the head of t h e Great Yueh-chih. Very likely it
is n o t by accident t h a t t h e unifier was not t h e head of the Yueh-chih either (in
this case he is not mentioned at all), b u t one of the five yabghu, and the form of
t h e unification was n o t the strengthening of the central power of the Great
Yueh-chih, but the v i c t o r y of one yabghu over the other four. This is the new
s i t u a t i o n described in t h e Hou Han-shu. I t s compilator, however, knows t h a t
these five possessions h a d existed since long, he even knows about how long
these five possessions existed before t h e unification, viz. over one hundred
years. However, it does not interest him a t all how these possessions were origi-
nally (at the time of their formation), because these nuances of the past are out-
side his chronological interest. Thus, s t a t i n g the fact of t h e antiquity of these
possessions, he writes a b o u t them as t h e y looked like before the unification.
This is why the following phrase is very i m p o r t a n t : the Yueh-chih divided their
c o u n t r y among five (or into five) hsi-hou (yabghu).11
Thus, in the H a n - s h u the districts governed by the yabghu (the origin oi
t h e yabghu is not defined a t all by this source) were subdued by the Great Yueh-
chih. I n the Hou H a n - s h u the country of t h e Great Yueh-chih is divided into
districts unconditionally governed by Yueh-chih yabghu, which became inde-
pendent at t h a t time, when one of t h e yabghu conquered the other four.
The early informant speaks about t h e whole (state of the Great Yueh-chih)
and a b o u t the parts (possessions of t h e yabghu) subordinated to it. For t h a t
chronological section of t i m e t h a t interests the late i n f o r m a n t , the parts be-
came so independent a n d the details of t h e p a s t of their former subordination
became so distant t h a t he turns over t h e situation near to his time (indepen-
14
The translations of this important phrase on t h e division of the country by
PULLE YB LANK (op. cit., 247 : They divided t h e country among five hsi-hou (Yabguj . . .
a n d by ZRCHER (op. cit. 367 : . . . and divided their country into the five hsi-hou
(yabgu) . . .) differ in t h e arrangement of t h e accents. PULLEYBLANK'S translation gives
more expedient to regard t h e five hsi-hou as tribal name (see below).

T H E F I R S T H O A R D OF T E T R A DRACHMAS O F ' H E R A I O S ' 155
dence of the possessions) into this past and speaks about the division of t h e
s t a t e among the five yabghu. Behind this differing information of the t w o
sources one can clearly see the tendency of development of the still whole,
nomadic s t a t e of t h e Great Yueh-chih, already adopting sedentary life and cul-
ture, with subordinated Yueh-chih yabghu (second stage), towards the dissolu-
tion of the unified empire into independent possessions of the yabghu (third
stage). 15
There is no need to give preference to one of the two sources (as is usually
done). They are not contradictory, but define the general tendency, the different
stages and boundaries of the history of the Yueh-chih, and the territories con-
quered by them on both sides of the Amu Darya. In this long period the devel-
opment advanced in the direction of the separation and strengthening of t h e
independence of the live districts or possessions of the Yueh-chih yabghu,
reflected in the formulation of the Hou Han-shu and brought about t h a t t h e
centre of the new unification became one of the possessions.
In our opinion the most complicated question is not the seeming m u t u a l
exclusion of the testimonies of the Han-shu and the Hou Han-shu regarding t h e
five hsi-hou, b u t the origin of their designations. Are t h e y the designations of
districts, of some pre-Yueh-chih dynasties or of tribes (clans) of the Yueh-chih
themselves ? In fact these three viewpoints are raised in literature. The decision
of this question is outside our competence, b u t it has no conceptual significance
for the present paper either. Here it is more i m p o r t a n t to reconstruct the f u r t h e r
transformation of these designations t a k i n g into consideration any variant of
their original meaning. The key for the understanding of the semantic trans-
formation is the designation Kusna. Since Kadphises I was a Kusna yabghu
and after the conquest of the other four yabghu his s t a t e was called Kusna,con-
sequently the meaning of the dynastic designation was borne by this word in t h e
stage of the five independent possessions and also a f t e r their unification. Thus,
if the five designations according to their origin are pre-Yueh-chih, 16 then a f t e r
the dissolution of the nomadic camps and the formation of possessions they be-
came the dynastic designations of the five Yueh-chih yabghu. This would be one
of the examples for the territorial designation of t h e dynasties. If, however,
this was t h e designation of the five tribes (clans) of t h e Yueh-chih, then a re-
versed process can be presumed, viz. t h e extension of t h e tribal designations
to the districts or possessions received. B u t a p a r t from t h e dependence from t h e
original meaning, the result will be the same also in accordance with the order
15
The fact t h a t the Hou Han-shu mentions the capital of the Great Yueh-chili
south of the river, does not a t all contradict to t h e conclusion about the dissolution of t h e
country into independent possessions. The existence of a capital is no evidence a t all of
t h e strength and significance of the central power. Cf. Chang Ch'ien : in Ta-hsia t h e r e
is no supreme head a t all, but the capital is designated.
16
Directly designations of districts, or (what is less likely) designations of pre-
Yueh-chih dynasties extended to the districts.

of t h e process, viz. t h e melting of t h e designation of the districts and t h e dy-

nastic designation yabghu in one of t h e stages. Therefore, t h e conclusion ap-
p a r e n t l y will not raise any objection t h a t t h e designations of t h e five hsi-hou
in t h e second stage (transformation of t h e nomadic camps f i r s t into vassal
possessions) and in t h e third stage (development of their independence) till t h e
unification by Kadphises I can be regarded as dynastic-territorial designations
independently of t h e ancient origin of t h e words themselves.
T h e proposed interpretation of t h e reports of the three sources as reflec-
tions of three stages in the history of t h e Yueh-chih enables us to see the f i r s t
word T VIANN V N TOE of the ruling of the inscription on the coins of
Heraios in a new light and to define t h e place of these t e t r a d r a c h m a s from t h e
viewpoint of periodization and relative chronology. The vassal character of
this t i t l e noted by K e n n e d y (see below) f i t s well into the framework of t h e sec-
ond s t a g e (i.e. the situation according t o t h e Han-shu) and seems to be a very
good Greek translation or equivalent of t h e title hsi-hou (yabghu) for t h a t long
period, when in the second stage t h e yabghu were in a different and changing
phase of subordination to the head of t h e Great Yueh-chih. Such a relative
chronology in combination with all t h e other d a t a can also be useful for t h e
absolute chronology.
T h e fourth word between the legs of the horse raised a serious polemic in
all aspects and this polemic has not come t o an end. Of the proposed variants of
reconstruction of this word we mention t h e three most i m p o r t a n t ones, viz.
1) SAKA (without t a k i n g into consideration the f i f t h character beta or without
a convincing interpretation of it) ; 2) S AN AB SANABOY ; 3 ) SAKAPOY.
The f i r s t variant can be regarded as discarded. The study of a more significant
material has shown t h a t this word consists sometimes of seven characters (in
genitive case, as t h e whole inscription), and sometimes of five characters
SAN AB, i.e. a form consisting of four characters (i.e. SAKA) is excluded.
Cunningham proposed the reconstruction of the spelling as SAN AB
SANABOY on the basis of six coins of his collection (they are k e p t a t present
in t h e British Museum). Bataille accepted an analogous spelling on the basis of
eight t e t r a d r a c h m a s (seven in the B r i t i s h Museum and one in t h e P a r i s Cabinet
of Medals), and A. N. Zograf supported such a reading with t h e publication of
f u r t h e r coins. 17
T h e third variant of spelling a n d reading was proposed b y R . M. Ghirsh-
man 1 8 on t h e basis of a f u r t h e r coin, which he himself called unique. I n this line
of t h e legend there are seven characters. I t differs from the original spelling of
17
A. CUNNINGHAM : Coins of the Indo-Scythian king Miaiis or Heraus. NC, 1888,
Ser. 3. Vol. V I I I , Sec. o f f p t . p. 10 ; G. BATAILLE : Notes sur la numismatique des Kou-
shans. Arethuso 1928. 19 ff. ; ZOGRAF : . 15 16.
18
R . GHIRSHMAN : Begram, recherches archologiques et historiques sur les
Koushans. Le Caire 1946. 109112.

T H E F I R S T H O A R D O F T E T R A DRACHMAS OF ' H E R A I O S ' 157
the second variant essentially only in t h e f i f t h character, where in place of t h e

beta there is some other character, which was proposed by R . M. Ghirshman to
be read as ro. I t really has in the upper p a r t an inflexion permitting t h e reading
ro, b u t below there is a horizontal twirl (the origin of which in the words of
R . M. Ghirshman himself is not to be clarified), which is not at all necessary
for the character ro. Among the published and not published t e t r a d r a c h m a s of
Heraios known to us there is no other one with an analogous peculiarity of t h e
f i f t h character in this word consisting of seven letters. I t appears to us t h a t this
is rather the result of the technical deficiency of the die, and due to abrasion
and cracking in the lower part of t h e f i f t h character. At the same time R . M.
Ghirshman proposed exactly the character ro to be regarded as original, t h e
character beta on all the other coins as a later error of the engravers of t h e dies,
and the strange form (the unnecessary horizontal twirl below) of the original
character ro as a cause of their error. To support this conclusion he mentions
still two arguments, viz.-. 1) the f i f t h character on the t e t r a d r a c h m a of t h e P a r i s
Cabinet of Medals is written so t h a t Bataille hesitated between the reading ro
and beta 2) t h e cursive nature of the character beta itself renders such a reading
doubtful, because one ought to expect a lapidary, capital form.
The doubts of Bataille have been resolved already by Zograf. Regarding
the cursive form of the letter beta, however, we should note the followings. I t is
very i m p o r t a n t t h a t this character has n o t a t all the same shape on the d i f f e r e n t
coins. I t is sufficient to examine the coins of the published hoard (Pis. I t o I I I )
to be convinced about this fact. On coins Nos. 1, 2 and 8 the cursiveness is
clearly expressed, the shape of the characters is round, this character is some-
what higher (especially on coin No. 8) t h a n the other ones in this word. T h e
form of the letter beta on coins Nos. 3 a n d 6, on the other hand, can h a r d l y be
considered cursive, it can be regarded nearly (No. 6) or fully (No. 3) as l a p i d a r y .
The shape of t h e beta on the t e t r a d r a c h m a s of Heraios Nos. 3 and 6 does n o t
differ from t h e form of this character on m a n y Graeco-Bactrian and K u s n a
coins, e.g. in t h e word BABIAEQB. Moreover, examining this subject of t h e coin
hoard, we have discovered on a series of coins of the Nameless king19 t h a t the,
character beta in the first word of the inscription has exactly such acursive form
as on the t e t r a d r a c h m a s Nos. 1 and 2 of Heraios. In the same word we see t h e
clearly cursive form of the letter beta on t h e Bactrian bronze coins minted a f t e r
the model of the tetradrachmas of Heliocles. Thus, the cursive form of t h e f i f t h
character in t h e word discussed on certain tetradrachmas of Heraios in itself
does not furnish any baeis to doubt t h a t this is exactly the letter beta, a n d no
error or deformation of the character ro.
As regards the third letter of this word, which was read by the a u t h o r s of
the first v a r i a n t () and the t h i r d v a r i a n t (BAKAPOY) of its reconstruc-
19
TIT, N-294. Found in the district of Hissar, Tajik SSR.

158 E. A. D A V l D O V I C H
tion n o t N (nu), b u t (kappa), this reading, besides conceptual and intellectual

considerations (which we do not discuss now), was based on the following
palaeographic d a t a . T h e third character on t h e tetradrachmas has two forms,
viz. N and . Already Gardner read the (as the Russian ) as the basic,
original form, and such a form was f r e q u e n t l y assumed by the kappa on t h e
latest Bactrian and P a r t h i a n coins. 20 Now, when a sufficiently large material
has been accumulated, this conclusion, f r o m the palaeographic point of view,
seems to be entirely fictitious. Even if we confine ourselves only to certain
t e t r a d r a c h m a s of t h e hoard discussed, it becomes clear t h a t there is no reason
to regard the in t h e word discussed as original and to read it as x (kappa).
F o r the sake of a comfortable comparison, we included the inscriptions
on t h e coins of t h e hoard in Table 2, a n d gave a general numeration to all
characters. 2 1
L e t us s t a r t with the statement t h a t t h e character and the alternation
N are typical n o t only of the word discussed (in which according to t h e
general numbering on Table 2 this is character 28 : N on coins Nos. 1, 2, 6 and
8 ; on coins Nos. 3, 4 and 7). The s a m e t w o forms appear in the first word
(characters 5 and 6 : M o n coins Nos. 1 , 2 , 5 , 6 , 7, 8 ; A o n coin No. 3 ; only
on coin No. 4), the reading of which does n o t raise any doubt, because the origi-
nal form is here exactly A (nu), and the is t h e reflected image of the nu. B o t h
forms appear also in t h e third word (character 23 : A o n coins Nos. 1, 2, 6 , 7 , 8 ;
on coin No. 3, and also on coin No. 5 in character 20), in which the original
form is undoubtedly also the character nu, i.e. N. All these examples already
themselves render t h e palaeographic f o u n d a t i o n of the reading of character 28
as (kappa) d o u b t f u l . B u t it is even more i m p o r t a n t t h a t in the inscription of
the t e t r a d r a c h m a s there appears the character (character 18) as the f i r s t
character of the third word. And in each case, when it has been preserved well
on t h e coins of the hoard (coins Nos. 1, 3, 6, see Pis. I II), it is a quite clearly
and correctly written character (kappa). And if on the t e t r a d r a c h m a s of
Heraios the character (kappa) has still preserved its correct writing, t h e n
there is no reason to d o u b t t h a t character 28 in its source is actually A (nu)
sometimes reflection-like deformed into . W i t h other words, for the readings
YAKA and YAKAPOY neither the character (kappa) nor the character P (ro)
furnishes any palaeographic basis. Consequently, the v a r i a n t YANAB
YANABOY remains most authentic.
20
GARDNER : Catalogue . . . , X L V I I .
21
The blank square on Table 2 indicates t h a t t h e corresponding character is missing
in t h e inscription. The fully erased characters a r e marked with three parallel strokes.
In those cases, when t h e inscriptions have been preserved very badly and it cannot, bo
absolutely established, whether certain characters were erased or they did not exist,
then a n interrogation m a r k is put into the s q u a r e on the figure. On Table 2 we gave
the main characteristics of the writing a n d position of the characters, but it is n o t
absolutely authentic for the study of those changes t h a t took place on the exterior of
t h e characters on account of the wear and tear of the dies or of other special reasons.
Acta Antiqua Aeademiae Scicntiarum Hungarirae 28, 1980

Table 2
1 2 3 4 5 6 7 8 9 1011 12 1314 151617 1819202122232425 26272829303132
1 TV/ANNUVDT H AEV 1 1 KNE V SANAS

2 * AU R "Q f Y >NA
3 T Y f AN EYETEST . KtHiAHCT E A HA
,
4 / .V ADY *
5 TV//)NNnvnVH ADV '1 1 1 ' EAZAEf
6 TY/AWNnvnrnr H'AEY KQCANOV ANAS
7 TV / ANNE Y DT Y UV * ANO4 ZT>
8 TV/ANNOYNTOE KI*EY NNY ^NEV
Errors in the inscriptions on the tetradrachmas of the hoard
The question of t h e meaning of this word has not been decided and in its
essence this circumstance has not been examined by specialists. The idea of
Zograf seems to be very useful (casually mentioned and rejected at once by him)
t h a t this is a proper n a m e of Iranian origin. 22 This idea was energetically sup-
ported by S. P . Tolstov, 2 3 convincingly rejecting all other explanations. The ab-
sence of this word on t h e oboli of Heraios confusing Zograf is not a serious
obstacle, because the inscriptions of coins even of the same denominations and
even more of different denominations were frequently d i f f e r e n t and did not
always contain proper names (cf. the coins of the Nameless king). At t h e same
time, it has to be noted in advance t h a t t h e word Sanab as a proper n a m e by
itself does not decide at all the problem of the second word, HIAOY, of the
inscription. The existence of two names on the t e t r a d r a c h m a s (the Greek He-
raios and the Iranian Sanab) should not be excluded mechanically, although
t h e s t u d y of HIAOY as a constituent p a r t of the titulature 2 4 is also very useful.
In the restored inscription of the t e t r a d r a c h m a s of Heraios we accept
not t h e traditional order of t h e words, inasmuch we do not regard the word
below, under the feet of the horse as the last one, b u t the word Sanab between
t h e legs of the horse.
1. TYIANNO YNTOE HIAOY KOIIANOY SAN ABO Y

Of the ruling ( = yabghu) HIAOY K u s n a Sanab.
2. TYIANNOYNTOS HIAOY KOIIANOY SANAB,

Of the ruling ( = yabghu) HIAOY K u s n a , Sanab.
As has been already noted, there are m a n y errors in t h e inscription of the

t e t r a d r a c h m a s of Heraios, and the first task of research was, of course, to
break through these errors, to establish the original writing and the original
t e x t . At t h a t the errors themselves were not the m a t t e r of special examination.
T h e peculiarities of these errors and their possible regularity were not analysed.
They were mechanically placed to the account of the gradual deformation of
t h e inscription. Sometimes they were implicitly and sometimes explicitly
regarded as the sign of the illiteracy of t h e engravers of t h e dies, and sometimes
also as unskilful forgeries.
A more careful examination of these errors permits their classification,
and the classification suggests an entirely diffrent assumption regarding the
reasons and origin of these errors. Here we consider only t h e inscriptions of coins
of t h e hoard (see Table 2 a n d Pis. I to I I I ) .
On the basis of t h e material of t h e hoard, these errors can be divided into
five types, which are numbered with R o m a n numerals.
22
ZOGRAF : . 3 0 3 1 .
23
S. P . TOLSTOV : . V D I 1939. No. 2. 117.
24
Ibidem.

T H E F I R S T HOARD OF T E T R A DRACHMAS OF ' H E R A I O S ' 161
I. Reflection-like reversed writing of the characters

a) N (see above)
b) \ /1 in each word (characters 4, 15, 22, 29).
I I . Substitution of the necessary characters and groups of characters with

other characters and groups of characters
a) in place of N (character 9 on coins Nos. 1 to 7)
b) H in place of N (character I I on coin No. 5)
c) Y in place of T (character 10 on coin No. 5)
d) in place of H (character 13 on coin No. 5)
e) Y in place of \ (character 15 on coin 7)
f) in place of II (characters 20 and 21 on coin No. 7)
g) in place of group KOII (characters 18 to 21) on coin No. 8 NNY
I I I . Shifting of characters and groups of characters 25

a) on coin No. 3 AI in place of IA (characters 14 and 15)
b) on coin No. 5 AH in place of (characters 20 to 23)
c) on coin No. 8 OB in place of BO (characters 30 a n d 31)
d) on coin No. 5 characters 11 to 13 (parts of two words) are also the result
of shifting : . . . . . . in place of . . . . . .
IV. Omission of characters

a) (character 11 on coin No. 1)
b) N (character 6 on coin No. 3)
c) I full and abbreviated (character 14 on coins Nos. 1, 4, 5 and 7)
V. Inversions by 90, 180 and other divergences from the correct posi-
tion
a) Z in place of N (character 29 on coin No. 5)
b) in place of the character A (character 4 on coin No. 3, character 15 on
coin No. 1), possibly it is the inversion of Y by 180
c) character 4 on coin No. 6 alpha, but the impression remains t h a t its
upper p a r t forms here the left end of t h e base, i.e. it is inverted by about 120
to 130. An approximate position is possible also in the case of character 27 on
coin No. 4.
" T h e published coins give a very interesting variant of translocation of an already

erroneous combination of characters, not represented in the hoard. Characters 8 10 in
the first word should be YNT, a very frequent error (with the replacement of character
9 : in place of N, see coins Nos. 1 7 of t h e hoard) is YOT. On t h e published coins in
this place (characters 8 10) we find TOY instead of YOT. (See CUNNINGHAM : Coins . . .,
10, No. 4, Pl. I l l , 4 ; R . B. WHITEHEAD : Notes on the Indo-Groeks, NC 1940. Ser. 5,
No. 7g, 120, No. 17) ; ZOGRAF: , 34 (in the rendering of character 8 with
ZOGRAF we find Y, and it should be T, seo t h e photograph).

d) example of divergence from t h e correct position is character 27 (alpha)

on coins Nos. 2 a n d 3.
The first word is written correctly only on coin No. 8 ; the second word
on coins Nos. 2 a n d 8 (if we regard t h e writing HIAOY as correct) ; the t h i r d
a n d fourth words on coins Nos. 1, 2 and 6. B u t there is not even one among t h e
eight coins of t h e h o a r d , on which the inscription would not contain some t y p e s
of errors.
I t has already been mentioned t h a t t h e reverse sides of all the eight coins
were minted with different dies, viz. t h e details of representation of the horse-
m a n and the goddess Nike are different ; t h e bending of t h e arched part of t h e
inscription and its arrangement in comparison to t h e representation are dif-
f e r e n t ; there is a difference in the length and arrangement of the lower word
u n d e r the feet of t h e horse ; and t h e arrangement and size of the characters of
t h e fourth word between the legs of t h e horse are also different. Now we h a v e
to a d d to this also another very i m p o r t a n t characteristic, viz. the stock of errors
and the types of errors are individual on each coin ! We e n u m e r a t e these individ-
u a l stocks of errors (omitting the errors of t y p e I).
Coin No. 1 : t y p e I I (character 9) -(- t y p e IV (characters 11 and 14) +
t y p e V (character 15).
Coin No. 2 : t y p e I I (character 9).
Coin No. 3 : t y p e I I (character 9) -f- t y p e I I I (characters 14 and 15) +
t y p e IV (character 6) -f- type V (character 4).
Coin No. 4 : t y p e I I (character 9) -f- t y p e IV (character 14).
Coin No. 5 : t y p e I I (characters 9 a n d 10) -f- t y p e I I I (characters 20 to 23
a n d characters 11 to 13) + type IV (character 14) -f- t y p e V (character 28). 26
Coin No. 6 : t y p e I I (character 9).
Coin No. 7 : t y p e I I (characters 9, 15, 20 and 21) -f- t y p e IV (character 14).
Coin No. 8 : t y p e I I (characters 18 to 21) type I I I (characters 30 to 31).
I t is i m p o r t a n t to stress t h a t n o t only the inscriptions showed stocks of
errors of different types, but even t h e single characters or groups of characters
contained errors of several types. For example, an incorrect character was t a k e n
( t y p e II), and in addition it was also inversed (type V) ; or an incorrect group
of characters came into being as a result of the substitution of one character
(type II), and t h e n t h e y were also s h i f t e d here and there.
B u t w h a t is t h e origin of these errors ?
26
On the coin published by CUNNINGHAM, who reads t h e inscriptions not entirely
correctly, judged f r o m t h e photograph, t h e r e is a n entirely analogous, composite set of
errors (character 31 resembles to I, because it is damaged by t h e representation of t h e
foot of the horse). CUNNINGHAM'S coin h a s been preserved badly, but still it seems to be
m o r e likely t h a t it a n d coin No. 6 of t h e hoard were minted with different dies. Only
t h e t y p e of the set of errors is identical (see CUNNINGHAM : Coins . . . , 10, No. 5 PI.
I I I , 6).

Of course, every new die for t h e m i n t i n g of coins was prepared a f t e r a

model a n d , consequently, the engraver of t h e die could r e p e a t t h e certain ac-
cidental errors of t h e model, b u t he could also m a k e some s e p a r a t e errors. T h u s ,
t h e errors of the f i r s t t y p e can easily be explained with negligence on t h e p a r t
of t h e engraver of t h e die. Certain especially frequently repeating s u b s t i t u t i o n s
of certain characters could be regarded as inherited, i.e. p a r t of t h e errors of
t y p e I I (e.g. in place of N, character 9), etc. B u t , obviously, t h e errors of t y p e s
I I t o V as a whole and t h e individuality of t h e stocks of errors in each inscription
will be understood, if we presume t h a t on the dies the inscriptions were n o t
engraved individually, b u t they were minted with punches of t h e certain
characters.
T h e punching technique 2 7 simplified and accelerated t h e p r e p a r a t i o n of
t h e inscriptions b u t , a t t h e same time, it also furnished a considerable possibility
of errors exactly of types I I to V. Simplicity and a d v a n t a g e of t h e p u n c h i n g
technique are obvious. The character m u s t have had t h e same form on t h e
punch as later on t h e coin, i.e. t h e c h a r a c t e r on the punch did not have a mir-
rored b u t a normal representation t h a t psychologically facilitated t h e work of
the engraver. B u t , of course, this is n o t t h e main thing. The main thing was t h a t
one and t h e same punch could be used m a n y times. The character engraved on
it once, later on was repeated on m a n y dies, thus accelerating and facilitating
their preparation.
T h e engraver of t h e dies had a set of punches. And just like in t h e m o d e r n
hand-operated typographic set, he m a d e t h e same errors. The model was lying
before t h e engraver j u s t like before t h e setter. He took t h e correct c h a r a c t e r ,
b u t inverted the punch, and thus an error of type V was m a d e . He correctly
took 2 3 characters, which followed one a f t e r the other, b u t he arranged t h e
punches incorrectly, and t h u s an error of t y p e I I I arose. H e did not t a k e t h e
same character, and t h u s an error of t y p e I I came into being, etc. And t h e r e
were no proof-readings. Thus an individual set of errors of different types c a m e
into being. Only t h e errors of type I were errors in the strict sense of t h e word.
These are errors m a d e a t the time of p r e p a r a t i o n of the punch. The rest of t h e
errors, with s t a n d a r d forms of t h e c h a r a c t e r s themselves, were errors of t h e
setter.
This explanation is also supported by two observations. Firstly, on cer-
tain coins t h e inscriptions are arranged s t h a t certain characters practically
run into t h e representation and, t o some e x t e n t , even d a m a g e it. This can
clearly be seen a t t h e contact of the c h a r a c t e r s with the representation of Nike,
of t h e head of the horseman, as well as w i t h t h e representation of t h e head a n d
feet of t h e horse. We mention a few concrete examples (see Pis. I I I I ) . Coin
27
About tho application of the p u n c h i n g technique in ancient times see Wrter-
buch der Mnzkunde, od. by F . SCHRTTER. . 1970. 538.

N o . 2 : characters 7 a n d 13 ; coin No. 3 : characters 5, 7, 10, 11, 13, 14 a n d 29 ;

coin No. 4 : characters 6 to 8 ; coin No. 5 : characters 26, 30 and 31 (character
30 below does not come out from behind t h e representation of the hoof) ; coin
N o . 6 : character 15 ; etc. On coin No. 2 characters 29 a n d 30 ran into each
o t h e r . A t the e n g r a v i n g directly on t h e die t h e engraver could easily have avoid-
ed these deficiencies with a slight s h i f t i n g a n d reducing.
Secondly, a t t h e individual engraving on t h e die, in t h e case of a set of
punches, the engraver ought to have m a d e a marking of t h e arch on t h e die f o r
t h e first two words a n d a marking of a s t r a i g h t line for t h e t h i r d word under t h e
f e e t of the horse. If he h a d engraved t h e inscription direct on the die, then t h i s
m a r k i n g would h a v e been an accurate basis for him a n d t h e n he would not h a v e
d a m a g e d it. I n t h e punching technique it is more complicated to observe such
a marking. This consideration also speaks in favour of t h e punching technique,
because on certain coins this marking is considerably damaged. In this con-
ception coins Nos. 6 a n d 8 are especially characteristic, viz. the third word
u n d e r the feet of t h e horse on coin No. 6 is arranged in an undulant form, while
on coin No. 8 it is a r r a n g e d arched. There are smaller deficiencies of the marking
also on other coins. 28
The writing of certain characters in t h e arched p a r t of the inscription a t -
t r a c t s a t t e n t i o n . F o r example, t h e characters H and N are sometimes
broadening u p w a r d s t h a t corresponds t o t h e arch and therefore, in the e a r l y
phase, explicably, t h e s e characters were either engraved direct on the die or on
t h e punch (marked o u t for the arched inscription). B u t when the character
(coin No. 3, c h a r a c t e r 23) is broadening u p w a r d s in t h e linear third word u n d e r
t h e feet of t h e horse, t h e n this speaks in f a v o u r of t h e punching t e c h n i q u e
(a punch marked o u t for the arched inscription was t a k e n ) . Character /Vin t h e
arched inscription of t h e t e t r a d r a c h m a belonging to t h e Cabinet of Medals in
Paris 2 9 is even more convincing. This is c h a r a c t e r 5, it is broadening not u p w a r d s
b u t downwards ! I t is clear t h a t the p u n c h was t u r n e d r o u n d with the legs u p -
wards, at 180, j u s t like other characters (errors of t y p e V) were turned r o u n d .
Thus, all f a c t s a n d observations mentioned, and especially the presence of
errors of types I I t o V on the coins a n d t h e individuality of t h e set of errors of
these types on t h e d i f f e r e n t coins of t h e hoard, render t h e conclusion very prob-
able t h a t the inscriptions on the dies for t h e minting of t h e reverse of the t e t r a -
28
The dimensions of the characters a n d their height are interesting. Usually, t h e
height of the characters in the same inscription is almost identical. B u t on certain coins
it is very different even in the single words (see coins Nos. 4, 6 a n d others). Although
this can bo explained also with the inexperience of the engraver of the die, still it m u s t
b e noted that it is more complicated to keep t h e accurate height of the characters a t
t h e preparation of s e p a r a t e punches, t h a n in the direct engraving on the die if there is
punctuation. This alone does not speak in favour of the punching technique, but together
with the previous more weighty facts a n d conceptions it still receives some significance.
29
GHIRSHMAN : Begram . . . , Pl. X X I I I , 5.

T H E F I R S T H O A R D O F T E T R A DRACHMAS OF ' H E R A I O S ' 165
drachmas of Heraios were not engraved individually on each die, but s t a m p e d

on the dies with sets of punches. 3 0
From the palaeographic point of view, in the inscriptions of the coins of
t h e hoard the character beta is most interesting, because i t furnishes v a r i a n t s
from the cursive u p to the almost perfectly lapidary forms (see above). The o t h e r
characters of t h e coins show a r a t h e r standardized writing for this group of
coins known already from publications (Table 3). 31 The tetradrachmas of t h e
Table 3
letters 20-21
letter Kon
on Table 2
omikron
Table 2
y psi Io n
Greek
kappa
sigma
alpha
letters A
Z)
(V AZ A
-Q a
on the
coins of

A Y H N I T 1 II
the
hoard
A 8 /
((
A
Basic forms of the characters on t h e tetradrachmas of the hoard
?
hoard do not show any further or different transformations of the characters,

except, perhaps, t h e peculiar writing (with hook) of character 14 on coin N o .
2, which a t the first glance could s u p p o r t its reading as ro t h a t , on the o t h e r
hand, would support the reading of the whole word as Heraios. B u t one m u s t
carefully examine this hook (Pl. I, 2). Between character 13 and character 15
30
Among other things, we have examined in this context t h e obols of Euoratides
m i n t e d in Bactria a n d to be found in Southern Tajikistan. Two unpublished obols f r o m
t h e burial-ground of Tup-khana near Hissai' (excavations of 1961, burial 96 and excava-
tions of 1971, burial 3) convince us t h a t t h e inscriptions on the dies foi the reverse side
were also prepared exactly with punches.
31
The small variation (not considered on Table 3) is sometimes t h e result of incli-
nation to t h e left or to the right (especially in t h e arched inscriptions) a n d of some indi-
viduality of the writing, and sometimes t h e result of t h e wear a n d tear of the dies o r
punches. The Greek names of the last two characters and composite characters are n o t
given on Table 3 : t h e final reading of character 14 on the t e t r a d r a c h m a s has not b e e n
established, and characters 2021 served for t h e rendering of t h e sound

t h e r e is a too small space for character 14, it is so to say sticked to the alpha
(to character 15) and repeats its writing. I t seems t h a t this is t h e case when t h e
p u n c h with the character 15 (aljiha) was either simply dislodged or stamped
twice (after the first failure, when it did n o t appear fully), as a result of which
t h e left side of character 15, i.e. alpha in t h e die and thereafter also on the coin
appeared as practically duplicated forming t h e above mentioned hook. If this
is so, then on coin No. 2 character 14 is simply left out, w h a t is very typical of
this word on the tetradrachmas of Heraios. 32
I t is interesting to note the rendering of the two variants of the sound
s (characters 20 and 21), viz. with two s t r a i g h t or with two slightly bent, parallel
strokes. They are placed so close t h a t visually they are perceived as one charac-
ter. I n the literature on the obols of Heraios these forms are laid down, and
also two combinations uniting both writings l( and P(, where the first character
of this two-charactered sound in outlines is sometimes a clear ro. The assertion
of A. M. Mandelshtam. according to which on the obols of Heraios for t h e
rendering of the sound s the kappa was also used, 33 is a mere misunderstanding.
T h e whole inscription consisting of two lines is read by A. M. Mandelshtam as
follows : 34 [ ?] A [ ? ?]OY [KjOKANOY. T h e coin was found in the south of
T a j i k i s t a n and is kept in the IIT. We h a v e carefully examined it, and the in-
scription has the following form (Pl. I l l , 9) : HAjOY \K]OHAI(OY. Here in
b o t h words there appear errors of t y p e I I I , viz. shifting of characters. I n t h e
f i r s t word in place of HAJOY we m u s t p u t HI AO Y (to which in this case t h e
parallel bendings of t h e second and t h i r d characters correspond very well) ;
in t h e second word in place of [K]OHAI(OY we have to p u t [K]OI(AHOY.
Consequently, the sound s is rendered also here with one of the variants of t h e
usual combination of characters. And t h e nature of t h e errors itself in the in-
scription of the obol permits the assumption t h a t the punching technique of
preparation of the inscriptions in the dies was used for the obols of Heraios
j u s t like for his tetradrachmas.
To show t h a t t h e coins have become important sources for political and
economic history, we must elucidate their own biography. As regards the in-
scriptions of the tetradrachmas of Heraios, there is still no convincing and
unambigous reading and translation accepted by everybody. The hoard of
t e t r a d r a c h m a s from Tajikistan does n o t furnish special additional materials for
i t . B u t the coins have supported the original variant of this inscription on t h e
32
Cf. coins Nos. 1, 4, , 7.
33
A. M. MANDELSHTAM : . EV, 14 (1961) 54.
34
On the basis of this reading of t h e f i r s t word and also of t h e writ ing of t h e same
word on the oboli published by ZOGRAF ( , 6), G. A. PUGACHENKOVA
presumed t h a t the obols a n d tho t e t r a d r a c h m a s belong to t h e minting of two different
rulers, viz. the obols to the successor or predecessor of Heraios (
13. ). The essence of t h e errors (translocation of t h e characters) both on the t e t r a d r a c h m a s
a n d (see below) on t h e obols, to be traced back to the punching technique of preparation
of t h e inscriptions in t h e die, remove the assumption of PUGACHENKOVA.

t e t r a d r a c h m a s , t h e y have rendered possible the classification of the errors a n d

different readings, and on the basis of this classfication also the e x p l a n a t i o n of
t h e specific feature, reasons and origin of t h e errors to be traced back t o t h e
punching technique of the preparation of the inscriptions of dies.
I n the biography of the coins of Heraios there is still a series of white
spots, like t h e chronology, genealogy, a n d the place of minting.
At present t h e place of the m i n t where t h e coins of Heraios were s t r u c k ,
viz. Bactria, 3 5 raises t h e slightest disputes. The conclusion is based on t h e com-
parison of three circumstances, viz. t h e absence of K h a r o s t h i in t h e inscrip-
tions, the essentially Attic s t a n d a r d of weight, provenance of t h e coin finds.
The a t t e m p t s to take t h e mint of Heraios o u t of the boundaries of B a c t r i a is
not supported b y a n y facts. 3 6
I t must be t a k e n with reservation t h a t t h e coins m a d e of precious m e t a l s
could be t a k e n f a r o u t of the boundaries of those places, where t h e y were m i n t -
ed. The isolated and sporadical finds of silver coins of high s t a n d a r d do n o t a t
all outline the boundaries of the districts issuing them a n d t h e y require g r e a t
carefulness. Finds of coins of Heraios, e.g. are now recorded from T a x i l a t o
Tashkent. 3 7 Therefore, only a mass registration of compactly situated prove-
nances of silver coins of Heraios could be regarded as a reliable source. Till
recent times, however, t h e researchers h a v e had information only on isolated
and territorially scattered finds. Their n u m b e r can now be considerably increased,
viz. there are already three finds in t h e territory of Southern T a j i k i s t a n .
We have already mentioned the obol of Heraios from t h e site of K e i - K o b a d -
shah. 3 8 F o u r obols of Heraios have been found in 1972 b y the archaeologist
B . A. Litvinskiy a t t h e excavations of t h e necropolis (two specimens in t h e
funeral s t r u c t u r e I I ) a n d in the gorodishche (two specimens, slope) of S h a h t e p e
near the b a n k s of river P y a n j (district Shaartuz). 3 9 And, finally, t h e published
hoard of t e t r a d r a c h m a s . Thus, exactly N o r t h e r n B a c t r i a (city of Termez 4 0 a n d
33
There are two viewpoints about t h e question, what territory was called in
ancient times Bactria, viz. only t h a t situated south of the A m u Darya or also n o r t h
of this river, i.e. comprising Southern Uzbekistan and Southern Tajikistan? H e r e a n d
later we accept the second variant.
38
The territory south of the H i n d u K u s h has already been convincingly excluded.
There are, however, a t t e m p t s to take t h e coins of Heraios beyond t h e boundaries of
Bactria north and north-oast of it. S. P . TOLSTOV proposed, even as a hypothesis, e.g.
first Chorasmia or K u s n i y a on the Zerafshan (VDI, 1939. No. 2. 119), then Sogdiana
or Bactria ( . Moscow 1948. 181) and again K u s n i y a on the Zerafshan
( . MoscowLeningrad 1948. 149). These hy-
potheses are not based on facts, the constituent p a r t of the conception is t h a t Heraios
is a vassal of the king of K'ang-chii. I t was convincingly pointed o u t by PUGACHENKOVA
t h a t t h e location of the mint of the coins of Heraios to Kusniya on t h e Zerafshan is
unfounded ( . 131, 134135).
37
I t is true, t h e find in Tashkent was received with dubitation (PUGACHENKOVA :
. 131).
38
MANDELSHTAM : . 53 54.
39
These obols are the subject of special publication.
40
ZOGRAF: . 3435.
Artii Antiqua Aeademiae Scientiarum Hungaricae 28, 1980

168 E. A. DAVlDOVICH
three sites in Southern Tajikistan) has already given four compactly situated
occurrences of coins of Heraios. 41 T h e t o t a l number of coins from these four
places is now fourteen (five obols and nine tetradrachmas), b u t there were more
of them, because t h e hoard of t e t r a d r a c h m a s from Tajikistan has not yet been
collected fully.
This, of course, does not give a firm basis to assert t h a t the mint issuing
t h e coins of Heraios was exactly in N o r t h e r n Bactria. B u t it shakes the r a t h e r
wide-spread and earnestly not founded point of view (to which, by the way,
also Zograf adhered) t h a t the coins of Heraios were minted in Southern Bac-
t r i a (in Northern Afghanistan). If we do n o t w a n t to overestimate the facts in
f a v o u r of t h e conception, then it can quite objectively be s t a t e d only t h a t t h e
new occurrences of coins prove beyond a n y d o u b t t h a t t h e coins of Heraios
were really minted in Bactria. By themselves they do not answer two i m p o r t a n t
questions, viz. whether there was one m i n t or several mints ; accordingly,
were the coins of Heraios minted in N o r t h e r n Bactria, or b o t h in Northern and
in Southern B a c t r i a ? To answer these questions the registration of the finds of
coins alone is, apparently, insufficient, exactly because t h e coins are m a d e of
silver N
The indisputability of the conclusion about the Bactrian (and preferably
N o r t h e r n Bactrian) origin of the coins of Heraios induces us to turn to t h e
question of the five hsi-hou, i.e. the five districts or possessions of the Yueh-chih
yabghu. The localization of these possessions has attracted t h e attention of t h e
most i m p o r t a n t scholars several times a n d it has raised a series of hypotheses.
T h e K u s n a hsi-hou, e.g., was located in different places from the valley of t h e
K a s h k a D a r y a to B a d a k h s h a n and K a f i r i s t a n , 4 3 b u t outside the boundaries of
t h a t s t a t e of t h e Great Yueh-chih, which was established in the second stage
(according to our chronology) and occupied t h e territory of Bactria (north and
s o u t h of t h e A m u Darya). The d a t a on t h e place of minting of the coins of
Heraios do not leave any doubt t h a t t h e K u s n a hsi-hou was situated exactly
in t h e territory of Bactria. The earlier a t t e m p t s to locate his settlement could
41
One of t h e obols of Heraios from t h e collection of t h e late historian a n d region
researcher A. E . MAJI is kept now in the R M T (1-1884/2). I t has not been published, a n d
t h e site of this find is not known.
42
The d a t a of t h e archaeological excavations have led PUGACHENKOVA to the
conclusion t h a t t h e m a i n possession of t h e K u s n a s before t h e foundation of t h e empire
was exactly N o r t h e r n Bactria : Under t h e rule of Heraios K h a l c h a y a n obviously
p l a y e d t h e role, if n o t of t h e capital, then of t h e main residence of the king, as soon as
a sculptural composition of the building of t h e palace was devoted to the glorification
of t h e clan of Heraios ( . 132134).
43
A s u m m a r y of d a t a see A. M. MANDELSHTAM : --
X . . ., -
, 53 (1957) 6264. The basis for the location a n d t h e method is t h e
phonetic similarity a n d t h e d a t a of a later source (Pei-shih). I n connection with this
MANDELSHTAM (not taking into consideration even t h e work of ZOGBAF and the numis-
m a t i c data) manifested a quite sober doubt regarding the full identity of the d a t a of Pei-
shih and the earlier sources, and especially concerning the location of the Kusnas.

T H E FIRST HOARD OF TETRADRACHMAS OF ' H E R A I O S ' 169
not yet rely on t h e d a t a of the coins, and the general historical situation as well
as t h e chronology of the Yueh-chih before t h e establishment of the K u s n a
empire (as these are described on the basis of three sources, viz. Chang Ch'ien,
t h e Han-shu and t h e Hou Han-shu, see above pp. 153 to 155) were not considered
by them. I t has to be added t h a t the four other hsi-hou should also be looked
for and located in those boundaries of the s t a t e of t h e Great Yueh-chih as t h e y
were in the second stage (when the hsi-hou were still vassal) and continued to
exist in the third stage, because the expansion started already under Kadphises I,
i.e. a f t e r the unification of the five possessions. 44
One of the most complicated and definitively not y e t solved questions is
t h e time of minting of the tetradrachmas of Heraios. A f t e r the most f u n d a -
mental research of Zograf (in 1937) which, on t h e basis of a comparative anal-
ysis of a large n u m b e r of relevant d a t a , d a t e d t h e coins of Heraios to the mid-
dle or even to the second quarter of the 1. century B.C., and mentioned also a
series of arguments against the earlier and later d a t e proposed before him
(within the limits f r o m 128 B.C. to 100 A.D.), and found followers in this ques-
tion, in the works of several authors there appeared again t h e tendency to d a t e
these coins to a later period. This is, however, etxactly a tendency, because t h e
method and conclusions of Zograf are not refuted, new methods and facts are
not proposed, b u t t h e new dates appear, as a rule, in t h e form of appendices
to the conceptions. 45
The first hoard of the tetradrachmas of Heraios itself does not solve t h e
question of their d a t i n g either. B u t certain observations m a d e in the course of
t h e study of the hoard can be very useful in this respect.
The form and execution of t h e t e t r a d r a c h m a s of Heraios deserves a t t e n -
tion. Zograf took this circumstance into consideration in t h e dating of the coins
of Heraios, viz. In the field of the execution the t e t r a d r a c h m a s of Heraios
with their solid circle with almost f l a t surface on both sides and even, plain edge
are far from the Bactrian tetradrachmas, just like all hellenistic coins of t h e 3.
and 2. centuries B.C. clearly showing bulge on the observe side and depression
on t h e reverse side with a sharp edge.46 Below, Zograf stresses once more t h a t
44
Cf. the correction in the Hou Han-shu in connection with t h e fifth hsi-hou,
mentioned in the H a n - s h u (ZRCHER: op. cit., 367 368), a n d also the other d a t a on
t h e beginning of the K u s n a conquests (ibidem, 367).
45
We mention a few excellent examples. R . M. GHIRSHMAN (1946) : last q u a r t e r
of t h e 1st century B. C. (Begram . . . , 112 ; it is true t h a t GHIRSHMAN did not h a v e a
possibility to use t h e work of ZOGRAF) ; I. M. DIAKONOV (1950) : 1st century A. D.,
from t h e turn of this century (M. M. DIAKONOV : ,
15. 1950. 1 7 4 1 7 5 ) ; BHASKAR CHATTOPADHYAY in 1967 ( T h e A g e of t h e K u s h -
nas . . . , 15) : last q u a r t e r of the 1st century . C. ; D. W. MACDOWALLN. G. WILSON
in 1970 (The References to the Kiisnas in the Periplus a n d F u r t h e r Numismatic E v i d e n c e
for I t s Date. NC 1970. Ser. 7. Vol. X . 227 foil.) : Heraios is identical with the Nameless
King, his coins wore m i n t e d in the period between the K u s n a kings Kadphises 1 a n d
Kadphises I I .
46
ZOGRAF : . 22, 24.
Artii Antiqua Aeademiae Scientiarum Hungaricae 28, 1980

t h e t e t r a d r a c h m a s of Heraios and those of the later Bactrian rulers sharply

d i f f e r in their execution. 4 7
I t is, however, f a r from being so. R . B. Whitehead, on the contrary,
h a s n o t e d t h a t t h e coin published by him is distinguished not by f l a t b u t by
concave reverse. 48 T h e same can be said also about six (of t h e eight) coins of the
h o a r d discussed here. The degree of bulging of the obverse side and the concav-
i t y of their reverse side are different, b u t t h e y exist, and the edges of all the six
coins are not even b u t exactly sharp. 4 9 T h u s , this essential characteristic the
execution of the t e t r a d r a c h m a s does n o t at all separate (sharply a t t h a t )
t h e hellenistic coins from the 3. and 2. centuries B.C. (including t h e Graeco-
B a c t r i a n ones) and t h e tetradrachmas of Heraios. On the contrary, the t e t r a -
d r a c h m a s of Heraios in their execution closely join the Graeco-Bactrian coins,
b u t these typical features of the l a t t e r ones so to say gradually fades away in
t h e m . This circumstance is essential for t h e absolute and relative d a t i n g of the
t e t r a d r a c h m a s of Heraios, especially in connection with t h e tendency to d a t e
t h e m to a later period already mentioned.
T h e standard, a n d weight of the t e t r a d r a c h m a s of Heraios are very im-
p o r t a n t . Seven coins of the hoard h a v e been tested, the s t a n d a r d of .six coins
was 800 (the testing was made in Moscow, in the GIM), and t h e s t a n d a r d of one
coin w a s 873 (the testing was made b y t h e coin tester of the Office of Standard-
izing Control in Tashkent). For the t i m e being we do not discuss a certain dif-
ference, we note only the main thing, viz. t h e tetradrachmas of Heraios have
been minted of silver of high quality, t h e silver in them is about or above 80%.
This is a very i m p o r t a n t essential fact against the background of t h e character-
istics of the development of minting in Bactria and in certain neighbouring
territories after the fall of the Graeco-Bactrian Kingdom and before the reform
of t h e K u s n a ruler Kadphises I I .
T h e weight of all the 8 coins of t h e h o a r d is high (Table 4), it gives a com-
p a c t group within t h e limits of 15.2 to 16 grams.
I t is not possible to determine t h e average weight of all these coins with
t h e graphic method (histogram). T h e average arithmetic weight of the eight
coins of the hoard is 15.56 g. This n u m b e r , in the given case, is convincing,
because the coins of t h e hoard do n o t show unexpected differences of weight,
a n d a s has already mentioned, they are quite compact in regard to weight.
47
Ibidem. 25.
48
WHITEHEAD : N o t e s . . . , 122.
49
Since among t h e tetradrachmas b o t h variants appear, viz. flat ones with
plain edges, and convex-concave ones with sharpened edges, from this point of view it
is absolutely unnecessary to examine t h e t e t r a d r a c h m a s of Heraios kept in other
m u s e u m s , especially in t h e rich collection of t h e British Museum. Certain coins of those
p u b l i s h e d by CUNNINGHAM (Coins . . . , P l . I I I . Nos. 34) arouse our interest : judged
f r o m t h e photographs, t h e sharpening of t h e edges and the convex-concave execution
is e x a c t l y the same on t h e m that, of course, h a v e to be controlled in t h e originals.
Acta Antigua Academiae Scientiarum Hungaricae 28, 1980

Table 4
W E I G H T AND STANDARD OF T H E TETRADRACHMAS OF T H E HOARD
Inv. Weight in grm. Standard Inv. Weight in grm. Standard
1 N-377/1 15.20 800 5 N-377/5 16.00 800

2 N-377/2 15.55 800 6 N-377/6 15.79 800
3 N-377/3 15.42 800 7 N-341 15.84
4 N-377/4 15.40 800 8 RMT-2973 15.30 873
The weight of t h e t e t r a d r a c h m a s of Heraios, as a source for their d a t i n g ,

was thoroughly studied by Zograf. Working with t h e weights of ten coins,
he received t h e following d a t a : t h e e x t r e m e values are 11.95 g. and 15.67 g.,
a n d the average weight is 13.93 g. S t a r t i n g out from t h e assumption t h a t t h e
changes of weight (the average weight, the a m p l i t u d e of fluctuations in t h e
weights, t h e q u a n t i t y of specimens with small weights below 14 g.), in several
eastern states took place synchronously, A. N. Zograf arrived at the conclusion
t h a t the t e t r a d r a c h m a s of Heraios most closely join with the t e t r a d r a c h m a s
of Orodes I (57 37. B.C.) and with t h e R o m a n m i n t i n g in Antioch (47 20.
B.C.). 5 "
Does the hoard change the conception about t h e weight of the t e t r a d r a c h -
mas of Heraios ? Definitely, although the question is n o t a t all simple and a t
present it cannot be examined in its whole range. First of all we must elucidate
t h e facts themselves of the high weight of all coins contained in t h e hoard, of
their compactness of weight, and accordingly their high average weight with-
in t h e hoard. This characteristic of t h e hoard can reflect t h e fortuity of t h e set
or the regularity of t h e m i n t of t h e coins and the circulation of money. Since t h e
general range of the f l u c t u a t i o n of weight of the t e t r a d r a c h m a s of Heraios is
very broad (now it is even broader, u p to 4.14 g. : f r o m 11.86 g. to 16.00 g.),
it can be presumed t h a t the owner of t h e hoard specially selected specimens of
high weight. T h e possibility, however, c a n n o t be excluded t h a t Heraios him-
self changed t h e s t a n d a r d of weight a n d also the average weight limits in t h e
minting of the t e t r a d r a c h m a s . This would mean t h a t in one stage the t e t r a -
d r a c h m a s were of high weight and the average weight limits were comparatively
small (the f l u c t u a t i o n s of weight were within the limits of 1 g. or slightly higher),
a n d in another stage they were of smaller weight and w i t h a very high range of
t h e fluctuation of weight (over 2 g.). If this is so, t h e n t h e method proposed by
Zograf for t h e s t u d y of t h e weight of t h e t e t r a d r a c h m a s of Heraios for their
d a t i n g is unacceptable, because it will be necessary to s t u d y and to compare not
50
ZOGRAF : , 21 23. At present one thinks t h a t it was Orodes
I who reigned in these years.

t h e general average weight of all t e t r a d r a c h m a s and not t h e general range of

t h e f l u c t u a t i o n s of t h e i r weights, b u t two a v e r a g e weights a n d two ranges of
t h e f l u c t u a t i o n s of weight.
N o w we know t h e weight of 23 t e t r a d r a c h m a s of Heraios. 51 The histo-
g r a m (Table 5) seems t o speak in f a v o u r of e x a c t l y two stages in t h e minting of
t h e t e t r a d r a c h m a s of Heraios. I n this case t h e hoard would r e f l e c t one of t h e
t w o stages of coin m i n t i n g and the circulation of money. F o r t h e t i m e being t h i s
is a working hypothesis. A possible m e t h o d of control is t h e collation of t h e
w e i g h t of the coins w i t h all of their other characteristics, viz. t h e testing of
s t a n d a r d of fineness, execution, and with t h e characteristic of v a r i a n t s , and also
w i t h t h e stylistic characteristics of t h e representation, t h e errors as well as with
t h e sets of types of errors in t h e inscriptions, etc. I t is also essential to com-
p a r e t h e weight of t h e t e t r a d r a c h m a s of Heraios with t h e weight of his obols,
a n d also with the weight of other coins m i n t e d and circulated in t h e t e r r i t o r y
of B a c t r i a after its c o n q u e s t by t h e Yueh-chih. This already exceeds the f r a m e
of t h e t a s k s of t h e p r e s e n t publication, b u t it is i m p o r t a n t to stress t h a t t h e as-
s u m p t i o n about two stages of weight in t h e minting and circulation of t h e
t e t r a d r a c h m a s of Heraios and t h e necessity to find a m e t h o d for t h e control
of t h i s hypothesis are p r o m p t e d exactly by t h e hoard discussed. A n d depending
on w h e t h e r there a p p e a r s a compactness of high weights (15.2 g. to 16.0 g.) of
t h e eight coins of t h e h o a r d and whether their high average weight (15.56 g.)
is t h e result of accident (selection by t h e owner) or t h e result of regularity
( m i n t i n g of coins a n d m i n t i n g policy), these d a t a of weights are different, b u t
in b o t h cases they r e p r e s e n t an additional source for the absolute and relative
d a t i n g of the coins of Heraios.
Another fact is also interesting, viz. all t h e eight t e t r a d r a c h m a s of t h e
h o a r d were minted w i t h different dies. The m a j o r i t y of the coins published ear-
lier as far as it can be judged f r o m t h e photographs were m i n t e d also w i t h
d i f f e r e n t dies. The a s s u m p t i o n can be expressed only a b o u t certain specimens
t h a t t h e y were minted w i t h t h e same dies. 52 T h e small number of t e t r a d r a c h m a s
of Heraios known a t present should n o t lead t o a mistake, viz. their minting
w a s very numerous, w h a t is shown exactly b y t h e significant number of dies,
used for the minting of such a small n u m b e r of coins known so f a r . This f a c t ,
51
The weight of nine t e t r a d r a c h m a s of t h e British Museum (1 specimen published
b y P . GARDNER ; 1 s p e c i m e n of H . N . WRIGHT a n d 7 s p e c i m e n s o f CUNNINGHAM) was
told m e by D. W. MACDOWALL, for what I am very grateful to him. I n t h e publications
only t h e weight of p a r t of t h e coins of t h e British Museum was mentioned. Certain d a t a
of weight have been t a k e n from t h e publications of WHITEHEAD a n d ZOGRAF, the weight
of t h e specimen of t h e M I U Z (N-16/4) and t h a t of the eight specimens of the hoard have
been added.
52
We are giving some examples. Tho reverse side of coin No. 5 of the hoard a n d
of coin No. 5 of CUNNINGHAM was minted with one die ; t h e reverse side of the coin
found in Termez (ZOGRAF : . . . , 34) a n d t h a t of coin No. 4 of CUNNINGHAM,
etc. were also minted with tho same die.

Table
W E I G H T OF 23 TETRADRACHMAS OF HERAIOS
(the black circles represent the coins of t h e hoard)
Weight Coins Weight Coins

in grm. in grm.
16.0 13.9
16.9 13.8
16.8 13.7
16.7 13.6
16.6 13.5
16.6 13.4
16.4 13.3
15.3 13.2
16.2 13.1
15.1 13.0
15.0 12.9
14.9 12.8
14.8 12.7
14.7 12.6
14.6 12.5
14.6 12.4
14.4 12.3
14.3 12.2
14.2 12.1
14.1 12.0
14.0 11.9
examined not isolated, b u t together with t h e weight and fineness of the silver
t e t r a d r a c h m a s of Heraos, a n d also with their already doubtless B a c t r i a n
origin, clearly show t h a t in the t e r r i t o r y of B a c t r i a , a t the t i m e when they were
minted, there was a sufficient q u a n t i t y of silver (in metal or in old coins).
T h e favourable condition of t h e silver reserves in B a c t r i a u n d e r Heraios and
t h e intensive minting of his coins are also f a c t s t h a t are very i m p o r t a n t for their
d a t i n g t a k i n g into consideration t h e financial reform of K a d p h i s e s I I .
I t is true t h a t Macdowall a n d Wilson maintain t h a t t h e selection of silver
byHeraios for t h e minting of coins does not have a significance for the d a t i n g .
I n their opinion t h e selection of coinage metal (copper, fine silver, billon) by
t h e rulers (examples are given) entirely depended on t h e accessibility of t h e
rich silver mines of Panjshir. 5 3 Then it should be explained, w h y did Kadphises
53
MACDOWALL WILSON : T h e R e f e r e n c e , 225 foil.

I I (whom they regard as t h e direct heir of Heraios, whom they identified with
the Nameless King), t o whom the mines of P a n j s h i r (if t h e y were functioning)
were definitely accessible, reject the minting of silver entirely and the subse-
q u e n t K u s n a kings did also not return t o t h e silver. T h e question of the selec-
tion of t h e coinage m e t a l should not be decided so unilaterally and simplified.
This selection depends on the interrelation of several economic, political,
psychological, and sometimes also on individual reasons and factors, which in
addition are concrete for each region. I n P a r t h i a , e.g. t h e minting of silver was
not discontinued. B a c t r i a , on the other hand, definitely experienced a local
crisis of silver coins, which ended with t h e reform of Kadphises I I (see below),
it is only important t o determine when it started and when it developed. The
silver of the Bactrian coins of Heraios is, therefore, a very i m p o r t a n t fact for
t h e d a t i n g of these coins and also for t h e development of t h e crisis of the silver
coins in Bactria. 5 4
T h e reform of Kadphises II was observed by t h e numismatists long ago.
E . V. Zeymal, who studied the K u s n a coins intensively and usefully, summed
u p a n d evaluated his observations and those of his predecessors regarding t h e
contents and reasons of its realization as follows : Essentially Vima Kadphises
laid down the beginnings of the minting of gold in B a c t r i a and in Northern
I n d i a . T h e changing over to the gold coin by Kadphises I I was usually explained
first of all with the exhaustion of the silver reserves in t h e country, w h a t mani-
fested itself in the Indo-Greek period in t h e deterioration of the quality of t h e
minting metal. The second reason is seen in the establishment of close connec-
tions by the Kusnas with the R o m a n Empire. There were also economic rela-
tions t h a t resulted in t h e influx of western gold to t h e country. Our attention
should also be turned t o t h e circumstance t h a t as from t h e 2. century B.C. t h e
territories incorporated in the K u s n a Empire thereafter, were distributed
a m o n g small governors. T h e uniting of these isolated possessions into a vast and
powerful state must h a v e influenced its economic possibilities favourably and,
apparently, it also b e c a m e one of the most i m p o r t a n t conditions of the appear-
ance of the minting of gold under Vima Kadphises. 55
One of the sides of this reform t h e minting of gold is studied now
thoroughly. We m u s t specially mention t h e meticulous observations of Mac-
dowall in two excellent articles of his (in 1960 and 1968). Pointing out t h a t t h e
idea itself of the m i n t i n g of gold by Vima Kadphises was possibly (1960) or
54
Apparently it is n o t important for MACDOWALL a n d WILSON, because they
accepted the conception t h a t Heraios is t h e Nameless King a n d he ruled between
Kadphises I and K a d p h i s e s I I . This conception, however, is n o t based on facts, a n d it
exactly contradicts t h e m . I t is sufficient t o compare t h e palaeography, the titulature,
a n d especially the iconography of the silver coins of Heraios with the bronze coins of
t h e Nameless King to be convinced t h a t such an identity is entirely impossible, they
are entirely different individuals in the direct and in t h e figurative sense.
65
E. V. ZEYMAL: , . -
, 3. . 9 (1967) 61.

T H E F I R S T H O A R D OF TETRA DRACHMAS O F ' H E R A I O S ' 175
even undoubtedly (1968) inspired by the R o m a n aureus. He, a t the s a m e time,

also showed t h e independence of t h e metrological side of t h e reform, a n d t h e
f u r t h e r course of t h e financial affairs. 5 6 Zeymal corectly stressed t h a t t h e unifi-
cation of the isolated possessions into t h e v a s t s t a t e of t h e Great K u s n a s
created favourable conditions for t h e realization of this side of the reform.
The systematical minting of gold coins, however, in itself does not d e m a n d
t h e discontinuance of t h e minting of silver. The idea a b o u t t h e exhaustion of t h e
silver reserves in t h e country is useful, but it does n o t elucidate everything.
The exhaustion of t h e silver reserves would have led t o a serious change of t h e
r a t i o of gold to t h e silver, which m u s t have led in some stage (especially u n d e r
t h e conditions of development of international trade) t o a mass influx of t h e
silver metal and silver coins from those regions, where t h e y were minted. T h i s
could have eased t h e shortage of silver and could have smoothed out t h e r a t i o ,
i.e. it could have prepared the objective conditions for t h e returing to t h e m i n t -
ing of silver coins. B u t t h e realization of such a possibility depended on o t h e r
reasons.
The history of financial affairs showed such examples, when the s t a t e on
account of various reasons could not mobilize silver m e t a l for the minting of
silver coins, both in t h e conditions of bad and w h a t is especially i m p o r t a n t
in t h e conditions of a favourable s t a t e of the silver reserves. Among t h e reasons
disturbing the mobilization of m e t a l for the minting of coins the psychological
reason should n o t be disregarded either, viz. if t h e m i n t i n g of silver coins w a s
seriously discredited in t h e eyes of t h e population, t h e n t h e returning t o it be-
came more difficult t h a n the organization of minting f r o m another currency
metal.
The question a b o u t t h e reasons of the entire rejection of silver coinage in
t h e s t a t e of the G r e a t K u s n a s requires special s t u d y in which one should n o t
disregard either of t h e two assumptions, viz. t h e exhaustion of t h e silver
reserves and t h e impossibility of its mobilization for t h e minting of coins.
The interrelation of both factors appears to be most likely, b u t their significance
is different in t h e different chronological periods of t i m e under the conditions
56
D. W. MACDOWAIL : Tho Weight Standards of tho Gold and Copper Coinage
of t h e Kushana D y n a s t y from WimaKadphises to Vasudeva. J N S I 22 (1960) 63 foil.; idem:
Numismatic Evidence for t he Date of Kanika. Papers on tho d a t e of Kaniska [Australian
National university centre of oriental studies. Oriental monograph Series. Vol. I V ] .
Leiden 1968. 136, 143 foil. On the basis of the study of the moan weight of tho K u s n a
denarii, MACDOWALL inferred their weight standard, viz.: a b o u t 8g. (1960. 64) or accurate-
ly 8 S (1968. 144). [The histograms and the allowance for t h e mean erasion, as far
as t h e coins were p u t into circulation, suggest the assumption t h a t the weight s t a n d a r d
of the denarii of Kadphises I I was even higher than 8 g. E . D.]. Tho Roman aurei h a d
such a weight only in the years 19 to 12 B. C., what is too early for Vima Kadphises.
The pre-reform (years 56 to 64 A. D.) aurei of Nero weigh by 0.4 g. less than the denarii
of Kadphises I I . Besides this, the K u s n a kings preserved t h e weight standard, b u t
they gradually reduced t h e standard of fineness, while t h e R o m a n emperors, on t h e
contrary, preserved it b u t reduced the weight of the gold coins.

of t h e changing political (phases of the attack of the Yueh-chih ; stage of the

five vassal possessions ; stage of the five independent possessions ; the state of
the Great Kusnas, the expansion of the boundaries) and economic situation
(definite development of the commercial production demanding the increase of
the q u a n t i t y of coins).
The reform of Kadphises I I is a qualitative turn prepared by the charac-
teristics of the development and change of the political and economic situation
in general and of the coinage and circulation of money in particular, in the
course of a long time after the fall of Graeco-Bactria.
The dating of the coins of Heraios for the time being was examined es-
sentially without this broad context and without the analysis of all reasons and
conditions of the implementation of the reform of Kadphises I I . The chronolog-
ical place of the silver coins of Heraios between Graeco-Bactria (the coinage
system of which was based on silver) and the reform of Kadphises I I (who
organized the coinage system based on gold) cannot be determined in the frame-
work of the publication of the hoard of tetradrachmas and only on the basis of
t h e coins of the hoard, we propose to devote a special work to this question.
And the publication of the first hoard of tetradrachmas of Heraios is similar
to t h e edition of a new source containing useful material, partly repeating some-
thing already known, partly telling something new, and partly playing the role
of an impulse in the perception or re-interpretation of certain facts laid down
earlier. All these d a t a can be summed up as follows.
1. From the palaeographic point of view the hoard has not furnished any-
thing principally new, supporting the stability and single-stage character of
the transformation of the Greek characters on the tetradrachmas and the dif-
ference, observed already long ago, between the obols and the tetradrachmas of
Heraios in this respect. At the same time, the coins of the hoard furnished dif-
ferent variants of the character beta, the comparison of which with other occur-
rences (from the Graeco-Bactrian coins to those of the Great Kusnas) has
definitively proved t h a t on the tetradrachmas of Heraios this letter can be
considered beta.
2. The comparative analysis of the errors in the inscriptions has stated
t h a t they are not the result of deformation on account of mechanic imitation,
and the increasing number and different character of the errors do not arise
from the misunderstanding of the content of the inscriptions, b u t the result of
some kind of negligence of the setter, in the punching technique of the prepa-
ration, although the set itself of the punches with characters has remained
standardized for the tetradrachmas. These observations, of course, could also
be made by other authors earlier, but it appears that they were suggested by
the hoard.
3. The hoard convincingly supports t h a t the execution of the tetradrach-
mas of Heraios occupies an intermediate place, many specimens still have this

T H E F I R S T H O A R D O F T E T R A DRACHMAS O F ' H E R A I O S ' 177
bulging-concave form and sharp edge important for t h e dating. At t h e same

time the conception about the t e t r a d r a c h m a s of Heraios as flat coins with an
even cut of t h e edge, i.e. with their execution diverging from the Graeco-
B a c t r i a n tetradrachmas, can be regarded as removed.
4. Eight coins of the hoard have been minted with different dies. This has
suggested the conclusion (supported also by the coins published earlier) t h a t t h e
issue of the tetradrachmas of Heraios was no accidental and short episode,
b u t i t was a r a t h e r long and a b u n d a n t emission.
5. The hoard has entirely changed the conception a b o u t the weight of t h e
t e t r a d r a c h m a s and has suggested the very probable assumption a b o u t two
stages of weight. I t has also become clear t h a t in t h e s t u d y of the weight of
t e t r a d r a c h m a s it is methodically incorrect to work with t h e arithmetic average
of all t e t r a d r a c h m a s as their average weight, viz. if t h e assumption a b o u t two
stages of weight proves to be true, then the t e t r a d r a c h m a s of each stage will
have their own average weight.
6. I t is also doubtless t h a t t h e tetradrachmas of Heraios covering a quite
large sphere of trade, had no piece by piece circulation, since the f l u c t u a t i o n
in weight were very high, and between the pieces of t h e smallest weight a n d
those of the highest weight the fluctuations were over 4 g. Such t e t r a d r a c h m a s
differed from each other by a whole drachma.
7. The test of fineness of t h e coins of the hoard has shown t h a t t h e y were
minted from a quite high quality silver.
8. The hoard has fixed a new occurrence of the coins of Heraios in N o r t h -
ern Bactria. Together with the d a t a known earlier and also with the new d a t a
published here for the first time, we have now information about four compactly
situated provenances of the coins of the Kusna Heraios in the territory of
Northern Bactria. This permits to speak about Northern Bactria not only as
a place of circulation of the coins of Heraios, b u t as a possible (even if not
necessarily single) place of their production.
9. The analysis of the Chinese sources and the three-staged periodization
of the history of the Yueh-chih (from the conquest of t h e territory of B a c t r i a
to t h e foundation of the K u s n a state) in comparison with t h e first word t h e
vassal title TYIANNOYNTOE in the inscription of t h e t e t r a d r a c h m a s (an
equivalent of the vassal yabghu) permits the assumption t h a t this issue belongs
to t h e second stage.
Moscow.

ILYA GERSHEVITCH
THE COLOPHON OF THE

NOKONZOK INSCRIPTION*
No. 1. M : 9. euo
, -

10. mg!
mg 2
No. 2. M : 9. And this well and dromos by Khirgman was built, together
with Kozgask's son Burzmihr the Hastilugnian, honouring-
the-memory-of t h e Lord of the Marches Nokonzok by com-
mand of the Emperor.
10. And by suopavo (it) was written, together with Burzmihr's
son Mihrman mgj together with Mihrman mg 2
No. 3. : 9. po aao oqôpioqo xiqi t xoyapxi novqo vaOxiAoyaveLyo

voxovtxt xaqaAqayyt paqrjyi mo yoarjo ipqopavo A NAME m g j
ANOTHER NAME m g 4
No. 3. : 9. And the well b y Burzmihr was built, t h e son of Kozgak, t h e

Hastilugnian, honouring-the-memory-of t h e Lord of t h e
Marches Nokonzok by command of the E m p e r o r A NAME mg 3
A N O T H E R N A M E mg 4
No. 4. A : mg x apioqapavo mg 2
No. 5. Dilberjin : sio pavo xostyrio Auayo
Above under No. 1 you see t h e colophon of the Nokonzok inscription as

i t stands on t h e monolith, the final of t h e three versions of t h e inscription, called
M for monolith.
* (Paper delivered at the International Seminar on pre-Islamic Sources for the

History of Central Asia on Tuesday, 30 September 1980, at 9.00 a. m., in Budapest.)
12* Acta Antiqua Aeademiae Scientiarum Hungarirae 28, 1980

180 I. GEBSHEVITCH
Below it, as No. 2, you see m y English translation of t h e two paragraphs,

9 a n d 10, of which t h e colophon consists in the M version. The translation is
largely based on the improvements in t h e understanding of t h e syntax of this
passage which Nicholas Sims-Williams published in IF, 1973, 96. B u t I differ
f r o m him in t h a t I h a v e reverted to Henning's interpretation of aaxiXoyavaetyt
as a nisbe, and in t h e interpretation of t h e last seven words of 9. My translation
of these represents, as i t were, a contamination of Henning with Sims-Williams,
p r o m p t e d by considerations which I explained in a paper given a t the Inter-
national Seminar of K u s h a n Studies held in K a b u l in November 1978, and in
a more detailed version of t h a t paper due shortly to appear in print in Volume
2, 1979, pp. 55 73 of Afghan Studies, the journal of t h e British Academy's
Society for Afghan Studies.
Below t h a t translation you see, as No. 3, the colophon as it stood a t t h e
end of version B, whose blocks, you will remember, were found, like the blocks
of version A, immured at random as p a r t of the lining of the walls of the well's
dromos. Version ends with 9. T h e two names which respectively precede im-
mediately monogram v 3 and monogram 4 here abbreviated m g , and mg 4 , are too
poorly preserved for identification, b u t they were certainly different from any
n a m e appearing in t h e M version.
If the colophon of is shorter t h a n t h a t of M, t h e one of A, the earliest
a n d original version of the inscription, is even shorter. I t consists solely of w h a t
above is given as No. 4, namely of t h e same a/Liiv)ooa/iavo sandwiched between
t h e s a m e mg 4 and mg 2 as terminate t h e colophon of the monolith M, No. 1 above.
This is all I have to say b y w a y of introduction, except t h a t in the se-
quence Euoyavo, the second word of 10 in M, I have always seen a proper name,
I y u m a n , a view which Sims-Williams has accepted.
Of course I y u m a n , or I y u m a n , is not exactly a common name in I r a n i a n
onomastics, in fact until recently one knew of it only here. Indeed, since in 9
of M EUO occurs by itself as second word, in the clear meaning of this, f r o m
O l r a n . iyam, it would be perfectly legitimate to t r e a t suoyavo in 10 as con-
sisting of two words, EUO this plus yavo. The only reason for not doing so, is
t h a t syntactically we need, before t h e past participle voiyxo was written,
a word expressing t h e agent of writing whom n a t u r a l l y one expects to be de-
noted by a proper n a m e .
T h a t we need such an animate agent placed immediately before voiyxo,
is rendered a b u n d a n t l y clear by t h e f o u r t h and f i f t h words of 9 of (see No. 3
above), which together mean by Burzmihr was built.
Therefore the only acceptable alternative to t r a n s l a t i n g euoyavo voiyxo
in 10 of M as by I y u m a n was written, would be and this (euo) was written
b y Man, or by Man. I t is not a very a t t r a c t i v e alternative.
There would be no more to be said on the subject, h a d not the fragment of
a B a c t r i a n inscription fromDilberjin published in 1979by Livshits and Kruglik-

T H E COLOPHON OF T H E NOKONZOK I N S C R I P T I O N 181
ova on pp. 108112 of the second volume of (Ed. I . T.

Kruglikova, Moscow), brought to our notice a second occurrence of euoyavo,
this time spelled with one iota only (see above, No. 5). The sequence, which
Livshits divides into two words, is followed there by another sequence which he
again divides into two words. The l a t t e r he interprets as a hyphenated proper
name -, duly querying it with two question marks. I appre-
ciate the query, b u t see no reason for quarrelling with this interpretation, espe-
cially as the context is in any case quite obscure.
However, - is in itself so strange a name, t h a t Livshits
understandably refuses to regard t h e preceding eioyavo as yet another s t r a n g e
name asyndetically prefixed to it. H e accordingly suggests t h a t in the Dilberjin
inscription yavo, as two words, means this building, and he therefore n a t u -
rally wants Euo yavo in 10 of M to have the same meaning, in which case
voiyro would in t h a t paragraph mean not was written b u t was inscribed.
So far, so good. W h a t is less good are the consequential a d j u s t m e n t s t o
which Livshits finds himself inexorably driven within t h e ambit of M's colo-
phon. He of course still needs an agent for voiyro, b u t having deprived himself
of euoyavo as the first-named of two agents, he takes recourse to finding t h a t
the verb has as one a n d sole-named agent the following Mihrman. To t h e pre-
position ayo preceding t h e latter he thereafter has no option but to a t t r i b u t e
a meaning equivalent to English by in agential sense.
This is the first step on slippery ground, of the chain-reaction which Liv-
shits's dismissal of euoyavo as agent releases. The second is that, since to him
in 10 of M votyxo ayo yivgayavo now has to mean was inscribed by Mihrman,
Livshits is bound to w a n t xtgo ayo ogCoytvgo in 9 of M to mean was built by
Burzmihr, despite t h e f a c t t h a t in 9 of this very meaning is expressed, as one
would expect, by ogZoyiogo xtgt.
The next and final step in this consequential chain-reaction is for Livshits
inevitably the inference t h a t the sequence ytgyoyavo, which precedes xtgo in
9 of M, is not a proper name denoting the agent of xtgo, but a term for y e t
another structure, asyndetically aligned with the preceding euo yo aao oo
, so t h a t his translation becomes and this well and dromos, terrace
(ytgyoyavo), was built by Burzmihr.
This is where I balk, notwithstanding my boundless admiration for Liv-
shits, a truly great Iranianist, and friend of mine in addition. And if ytgyoyavo
refuses to be anything b u t the logical subject of XLQO j u s t as in 9 of ogCoytogo
is even for Livshits t h e logical subject of XLQL, what else can we do b u t retrace
all the steps of his slippery chain-reaction until we are back a t euoyavo in 10
of M having to remain, a f t e r all, the animate logical subject of votyro.
We are back a t euoyavo as a n i m a t e agent, b u t even I can no longer say
t h a t this makes me h a p p y . For naturally I do accept Livshits's premise, t h e
unlikelihood t h a t so strange and rare a name as I y u m a n , or rather the bearer

182 I. GERSHEVITCH
of i t , should occur in both inscriptions, a t Dilberjin even followed asyndetically

b y a n even stranger n a m e .
Caught in this onomastic fix, my mind happened to s t r a y t o the difference
between Dr Livshits's a n d my own f i r s t names. I t then struck me t h a t mine
disposes of an a d v a n t a g e which his could n o t possibly have offered him on this
occasion. For reasons no d o u b t egocentric, t h e name I bear has always induced
m e t o t a k e an interest in words beginning w i t h I L . And so, before long, it dawn-
ed on me t h a t semantically, and partly even etymologically, English ilk fits the
eioyavo situation like a glove the hand. A t Dilberjin eioyavo immediately pre-
cedes a proper name, in other words, qualifies a person. In t h e Nokonzok colo-
p h o n ' s M version the same euoyavo ought to be a person, t h e logical subject of
voiyxo, yet should not be, as Livshits rightly says, a personal name.
Well, English ilk means the same, a n d consists, as Germanicists have
long shown, of the pronominal element i, present also in B a c t r i a n et to from
iyam, plus a reduced form of what in English is like, in Gothic leik. Livshits
w a n t e d to find in t h e yavo of etto yavo a building structure, etymologically
therefore a cognate of dmna- house. B u t Bactrian yavo m a y just as well be-
long t o the Iranian base mn- to resemble, so t h a t , just as English ilk, yavo
would derive its meaning the same from this-like.
I t would follow t h a t a t Dilberjin we m u s t translate the same -
, who presumably was mentioned earlier on in the t e x t of which only a
f r a g m e n t has survived. A n d in the M colophon of the Nokonzok inscription we
m u s t translate and b y t h e same was written.
This would not be without interest to t h e understanding of the Nokonzok
inscription as a whole, whether or not you accept my new interpretation of
w h a t happened to t h e well of Baghlan, in t h e article about to a p p e a r in Afghan
Studies. For, the person referred to as the same in 10, can only be the m a n
called Khirgman in 9. A n d if Sims-Williams is right, as I think he is, in defining
t h e monogram mg 2 as t h a t of Mihrman, the monogram m g j as t h a t of the man
he believed bore the n a m e I y u m a n , then we learn at last w h a t hitherto we had
no reason for even suspecting, namely t h a t Khirgman figures already, represent-
ed so t o speak ideographically, by no more t h a n mg 1 , in the brevissimus colophon
of t h e A version (No. 4 above).
T h e gain accruing to us from this recognition, is a little more t h a n I had
t i m e or space to explain in translation-note i on p.66 of m y article about to
a p p e a r . Let me devote to it the rest of this paper.
1 8, i.e. the original Nokonzok t e x t constant in all three versions, to
w h i c h a different colophon was added in each of the three, n a m e only two per-
sons, Kaniska, too f a m o u s for his patronymic to be s t a t e d , a n d Nokonzok,
f a m o u s yes, but not f a m o u s enough for his patronymic to be omitted.
9 of M, i.e. the first paragraph of t h a t version's longissimus colophon (see
No. 1), introduces a third person, Burzmihr, stating his patronymic. He had

T H E COLOPHON OF T H E NOKONZOK INSCRIPTION 183
already appeared, complete with patronymic, in 9 of (see No. 3), which

paragraph by itself constitutes the whole of the longior colophon, intermediate
between t h e brevissimus and the longissimus not only in length, b u t also in time.
10 of M, i.e. t h e second p a r a g r a p h of the longissimus, which includes as
coda the brevissimus, supplies, just in f r o n t of the latter, the patronymic absent
from the brevissimus itself, of Mihrman. This patronymic we should never have
known h a d version A alone reached our ken. B u t we m a y be quite certain t h a t
local readers contemporary with the setting up of A, even if familiar with dozens
of Mihrmans, would have known exactly which Mihrman the brevissimus
designates, because mg 2 supplied as it were his patronymic ideographically.
The rule which emerges applied to Nokonzok, Burzmihr, and Mihrman,
is t h a t for identification name and patronymic must be given. B u t the custom
of not s t a t i n g the patronymic of a person whose name was to be immediately
followed by his monogram not only transpires from the brevissimus's pioqapavo
mg2 (see No. 4), b u t is assured by t h e longior's coda (see No. 3). The two names
preceding respectively t h a t coda's m g j and r r ^ cannot be read with any con-
fidence (see Asia Major, X I I , 1966, 94f.), b u t both certainly differ from all t h e
names mentioned in A and M and t h e rest of B, while yet neither bears its
bearer's patronymic. I t is clear, therefore, t h a t local readers extracted t h e two
patronymics which to us are beyond reach, from the monograms mg 3 and mg 4 .
Accordingly in the brevissimus of A, where not only the patronymic of
Khirgman, b u t even his name, fails t o appear, both had to be extracted by local
readers contemporary with the setting up of t h a t version, from no more t h a n
Khirgman's monogram m g r We m a y infer t h a t either K h i r g m a n was so well-
known t h a t the mere sight of his monogram sufficed for identification not only
of his patronymic b u t also of his name, or t h a t , although it did not really suffice,
K h i r g m a n thought t h a t it would. F o r quite likely the author of t h e brevissimus
was Khirgman himself, the boss of A's other engraver Mihrman.
Now all we need to note in addition is t h a t in the longior and in the longis-
simus Nokonzok's patronymic no longer appears, i.e. its absence from both colo-
phons is due merely to awareness of its having been s t a t e d a t the first mention
of Nokonzok's name, in 5 of the inscription. With this in mind we shall under-
s t a n d why of all t h e dramatis personae (apart from Kaniska) of M only Khirg-
man is named without patronymic : t h e author of the longissimus (who again
was quite likely K h i r g m a n himself, t h e author of the longior presumably being
Burzmihr) was aware t h a t the longissimus would have as coda the brevissimus
he had composed years earlier, whose initial mg4 sufficed, a t least in his opin-
ion, to inform readers of his patronymic. He therefore refrained, in 9 (see No.
1), from adding it to his name. B u t in 10 we may expect t h a t he would have
added it, seeing t h a t in t h a t p a r a g r a p h immediately before the brevissimus he
did add Mihrman's patronymic to his underling's name. He would have added
it also to his own name. B u t on reaching 10 it occurred to him for stylistic
Acta Antiqua Aeademiae Scientiarum Hungaricue 28, 1980

184 I . G E R S H E V I T C H : T H E COLOPHON OF T H E NOKONZOK I N S C R I P T I O N
variation to write, instead of his name, Uk. A n d to Uks even Bactrians were
evidently not in t h e h a b i t of attaching patronymics.
These are not aimless considerations. F o r whenever one differs from illus-
trious predecessors in t h e solution of problems, one m u s t t r y to understand, and
r e f u t e , not only t h e reasons they adduced for their solution, b u t also reasons
t h a t may have influenced them without their deeming it necessary to adduce
t h e m , if they are such t h a t they appear tacitly to support t h e solution with
which one disagrees. W e may well ask w h y Livshits, despite the egCoytogo
XIQI of No. 3 above almost forcing one to t a k e yigyoyavo in No. 1 for a personal
n a m e , preferred to consider yigyoyavo a building structure as Henning had done,
who however of our N o . 3 knew as yet nothing. I t seems to me obvious, although
Livshits does not s a y so, t h a t he did this o u t of reluctance to regard as a n a m e
a word which, unlike all the other names appearing in M (the sole Kaniska's
excepted), is unaccompanied by a patronymic defining it as a personal name.
The word also has as yet no etymology supporting onomastic interpreta-
t i o n . Let me end b y hazarding a shot into this dark : *rixtaka-manah- ( > *xir-
dag > *xir(d)g) the relaxed-minded, the even-tempered, with reference to the
meaning of German gelassen.
Cambridge.

J. HARMATTA
KIDARA AND THE KIDARITE HUNS IN KASMR
I t was recognized b y earlier research long ago t h a t precise historical d a t e s

can only be found in K a l h a n a ' s R j a t a r a n g i n i f r o m 813 A.D. onwards. This d a t e
(viz. t h e L a u k i k a year 3889 which began on the 7th March, 813 A.D.) is given
b y K a l h a p a for C i p p a t a - J a y p i d a ' s d e a t h . Before this e v e n t he can only offer
figures indicating t h e length of the reigns of individual rulers. 1 I t was also seen
already long ago t h a t t h e chronology of the whole period preceding t h e d e a t h
of C i p p a t a - J a y p i d a was based on calculated dates, computed by K a l h a n a on
the basis of some chronological and historical theories. 2
I n spite of the constructed character of t h e chronology for t h e earlier
periods of K a s m i r i a n history, one m u s t not regard t h e n a r r a t i v e of K a l h a n a as
a mere invention, being w i t h o u t a n y historical a u t h e n t i c i t y . This s t a t e m e n t
m a y be valid f i r s t of all for t h e earlier K r k o t a kings who ruled prior t o t h e
d e a t h of C i p p a t a - J a y p i d a . Here we find only a small error of 25 years. 3 E v e n
the d a t e for t h e rise of t h e d y n a s t y , deducted from t h e length of t h e reigns of
the earlier K r k o t a kings, can be supported by historical reasons. 4 I t follows
t h a t in general t h e historicity of K a l h a n a ' s n a r r a t i v e f r o m the beginnings of
t h e V H t h century A.D. onwards cannot be called in d o u b t .
B u t it was also pointed out by M. A. Stein t h a t K a l h a n a mentions some
well-known historical personalities even before t h e K r k o t a dynasty. 5 Thus, he
borrowed from t h e work of Chavillkara the names of f i v e kings of which a t
least four, viz. Asoka, H u s k a , J u s k a , K a n i s k a , are also known f r o m other
sources. T h e description of these kings given by K a l h a n a well corresponds t o
t h e evidence of other sources and as regards the t h r e e K u s n a kings he offers
even some details, u n k n o w n elsewhere, which point to a well-informed historical
work. Such details are t h e following :
1
Cf.M. A. STEIN : K a l h a n a ' s Rjatarangini. 2 Delhi PatnaVaranasi 1961. Vol. I.
58 foil, with earlier literature.
2
Cf. R j . I. 4856. M. A. STEIN : op. cit. Vol. I. 59 foil. (Introduction), notes to I .
4 8 - 5 0 , 53, 54 Vol. I. 1 0 - 1 2 (Translation).
3
C f . A. CUNNINGHAM: The Ancient Geography of India. I. London 1871. 9 1 ;
G. BHLER : Detailed R e p o r t of a Tour in Search of Sanskrit MSS. Made in Kamir,
R a j p u t a n a , and Central India. Bombay 1877. 43, 55 ; M. A. STEIN : op. cit. 67 foil.
4
J . HARMATTA: L a t e Bactrian Inscriptions. Acta A n t . H u n g . 17 (1969) 404.
3
M. A. STEIN : op. cit. 64 foil., 74 foil.

186 J. HARMATTA
1. The form of t h e names H u s k a and J u s k a .

2. The chronological interval between t h e three K u s p a kings and t h e
N i r v a n a of Buddha which is presumed to be 150 years.
3. The simultaneous reign of t h e three K u s n a kings which seems to be
supposed by K a l h a n a .
E a c h of these details m a y arouse a keen interest.
1. The forms Huska and Juska used instead of Huviska, Huveska, Oosko
a n d Vsiska, Vajheska, Bazsko were striking from the beginnings and Juska be-
c a m e identified with Vsiska only later. 6 I n fact, the possibility of these forms
can be proved from t h e view-point of historical phonology. T h e form Kaniska
obviously prevents us from ascribing the development of Huviska, Vsiska into
Huska, Juska to Gndhrl P r k r i t in which t h e outcome of Kaniska was *Ka-
nik(k)a. Accordingly, t h e rise of these striking forms can more probably be pre-
s u m e d in some E a s t e r n Iranian dialect, possibly in the language of the K u s n a s
themselves. Iu L a t e Bactrian or in some other Eastern I r a n i a n Language like
Yidgha-Munji, the form Huviska could develop into * Huska and then into
Huska. Thus, this form can easily be explained from the view-point of historical
phonology. As concerns Vajheska (to be read obviously *Vazeska), on the basis
of Munji WUZW- > WZW- 'to extinguish', Yidgha vzn-, Munji vzn- < O l r .
*ava-zn-, cf. Saka vaysn- 'to recognize', 7 we can presume a development
Vazeska > *Vzeska > *Vzuska, and even > *Zuska. The forms *Vzuska or
*Zuska could only be adopted by Sanskrit in t h e spoiling Juska. Contrary to t h e
f o r m s Oosko and Bazsko used in Bactrian, perhaps we have to do with some
local, Eastern I r a n i a n variants or even with the genuine K u s n a development
of these names here.
2. The striking s t a t e m e n t about t h e short chronological interval (150
years) between the Nirvria of B u d d h a and t h e three K u s n a kings has an
e x a c t parallel in t h e Khotanese t e x t P 2787-1-195 according to which Candra
K a n a i s k a was born when one hundred years elapsed since the deva B u d d h a
passed in parinirvna. 8 B o t h the Saka t e x t and K a l h a n a ' s source might have
preserved a genuine K u s n a tradition which wanted to shorten the long (more
t h a n six hundred years) interval between the Nirvna of B u d d h a and the three
K u s n a kings considerably.
3. Since the discovery of t h e inscription from r, it became clear t h a t
K a n i s k a I I and H u v i s k a (I ?) must have reigned partly simultaneously. In t h e
light of this fact it seems not to be excluded either t h a t K a n i s k a or/and Huviska
were already co-rulers of Vsiska. I n this case the Kasmirian tradition might
6
Cf. recently M. BUSSAGLI : The Problem of Kaniska as Seen by the Art Historian.
P a p e r s on the Date of Kaniska. Leiden 1968. 42 and L. PETECH : Kashmiri and Tibetan
Materials on the Date of Kaniska. Ibidem 244.
' Cf. G. MORGENSTIERNE : Indo-Iranian Frontier Languages. I I . Oslo 1938.
260, 264.
8
H . W. BAILEY : Visa Samgrina. AM NS 11 (1965) 107.

KIDARA AND T H E K I D A R I T E HUNS I N KAMIR 187
have correctly preserved the memory of the three K u s a kings as ruling simul-
taneously.
Thus, the historical value of t h e Buddhistic tradition, preserved in Chavil-
lkara's work a n d used by Kalhana, seems to be beyond a n y doubt. Moreover,
earlier research also recognized long ago t h a t t h e R j a t a r a n g i n preserved a
valuable tradition of the Hephthalite kings Mihirakula, Toramna, Khingila,
and L a h k h a n a , too. 9 On the basis of linguistic and historical arguments it seems
also to be very likely t h a t a t least one part of the list of kings from Mihirakula
until the Western T u r k conquest of Gandhara in the twenties of the VTIth
century A.D. can be regarded as authentic from historical view-point. 1 0
Now, we arrive a t an obvious contradiction here. I t is well-known f r o m
various sources t h a t the rule of the Kusijas (and L a t e Kusn.as) was followed
by t h a t of the K i d a r i t e Huns who were again succeeded by the Hephthalites.
This historical succession is valid also for the territory of Kasmr as it is clearly
proved by the coinage of the Hephthalite dynasty of Kasmr. The type of t h e
coins, minted by Toramna, Pravarasena, Narendrditya a n d Gokarna, exactly
reproduces the gold coins of K i d a r a , which again follow t h e type of the coins
struck by the L a t e K u s n a kings of North-Western India. 1 1
The Hephthalite kings of Kasmr had maintained even the legend Kidara
on t h e reverse of their coins. This f a c t can only be explained by the assumption
t h a t the Hephthalite dynasty of Kasmr was the immediate successor of t h e
Kidarite H u n kings who ruled there. As it is proved by the legend of their coins,
the Hephthalite kings of Kasmir wanted to be regarded as the legitimate de-
scendants and heirs of Kidara a n d this claim is only comprehensible if K i d a r a
also ruled over Kasmir.
I t follows f r o m all these considerations t h a t K i d a r a , t h e H u n king o u g h t
to have occurred in t h e enumerations of the Kasmirian kings. I t was impossible,
however, to discover Kidara or K i d a r i t e H u n kings in K a l h a n a ' s R j a t a r a n g i n
so far. In the m e a n t i m e the contradiction between historical probability a n d
Kalhapa's narrative became even sharper. One succeeded to identify some coins
of Hiranyakula who reigned in Kasmr according to the Rjataraiigipi before
Vasukula, the immediate predecessor of Mihirakula. 12 T h e coins of Hiranyakula
imitated even more exactly the coins struck by K i d a r a . Thus, they represented
a missing link between the coinage of Torama a n d t h a t of Kidara and proved
a t least to some e x t e n t the historicity of the list of kings who ruled before
9
A. CUNNINGHAM : Coins of Mediaoval India. London 1894. 25 foil. ; M. A. STEIN :
op. cit. 64 foil., 78 foil.
19
J . HARMATTA: op. cit. 100 foil.
11
A. CUNNINGHAM: Coins of Mediaeval India. London 1894. 42 foil., Pl. I I I .
For the coins of K i d a r a mentioned here cf. ; A. CUNNINGHAM : L a t e r Indo-Scythians.
Scytho-Sassanians a n d Little Kushans. London 1893. Pl. VI, 3, 10, 14.
12
A. CUNNINGHAM: Later Indo-Scythians. Ephthalites, or White Huns. London
1894. 98, 114, Pl. V I I I . 910.

188 J. HARMATTA
Mihirakula according to t h e R j a t a r a n g i n . I t became clear t h a t these kings

m u s t h a v e been K i d a r i t e H u n rulers because they reigned before the first
H e p h t h a l i t e king, Mihirakula, in K a m r . B u t the name K i d a r a seemed to be
missing there. Accordingly, the whole problem came to a deadlock.
A t this state of affairs unexpected help arrived on behalf of linguistics.
S t u d y i n g the Kidarite H u n and H e p h t h a l i t e names, I discovered t h a t the name
K i d a r a did not sthnd isolated among t h e names of the epoch. I t occurs four
t i m e s in the Kharosthi documents f r o m Eastern Turkestan, each times as a
honorific title or surname. The evidence is t h e following :
1 Doc. 118, C, 2 kirydari sugitasa muli I

2 Doc. 103 R e v . 6 kirytra sugitasa I
3 Doc. 146, B, 4a daavita kirytra mutreya
4 Doc. 242. Obv. , 1 kirytra sugitasa pake I
T h e honorific character of the title becomes clear from its use by t h e officer
daavita mutreya. I t s meaning may be 'honoured, hero, v a l i a n t ' or the like.
T h e n I found a b u n d a n t evidence for the word in K h o t a n e s e Saka texts.
T h e evidence for it was collected and discussed by H. W. Bailey in a detailed
m a n n e r . 1 3 The variants of t h e word are t h e following : kaidari, kidari, kaidara,
kidarr (locative pl.), kinarqna (genitive pl.), kimnara, kedhara, kedara, kirydari
a n d even kaidliar re ' K i d a r a king'. T h e variants kinara- a n d kimnara and even
more t h e occurrence of t h e word as t h e equivalent of Sanskrit Kirynara-, Kin-
narain t h e Khotanese Saka S u d h a n a - A v a d n a prove beyond a n y d o u b t t h a t
S a k a kaidari, kidara, kimdari etc. represent an adoption in K h o t a n e s e Saka of
I n d i a n Kirynara-, Kinnara. Taking into consideration t h a t t h e change -ryn-\-nn-
> -ryd-\-d- cannot be pointed out in Gndhri Prakrit, b u t it is well-known
in Sanskrit loan-words of Khotanese S a k a (Saka sandvta < Sanskrit sanni-
pta-, Saka vavanna-jvavaryda < Sanskrit upapanna-)u, we m u s t regard kim-
dari, kirytra in the N i y a documents as borrowed from K h o t a n e s e Saka.
A t this point we can return to K a l h a n a ' s R j a t a r a n g i n . If we recognized
t h a t t h e name Kidara was an I r a n i a n adoption and development of Indian
Kimnara-, Kinnara- probably meaning 'hero, valiant' then it becomes clear a t
once t h a t the king N a r a who is mentioned in Book I 197 of the R j a t a r a n g i n
i n t h e following way :
k i m n a r p a r a n m t h a kimnarairgtavikramah /
vibhsanasya p u t r o b h n n a r a n m nardhipah //
13
H . W . BAILEY : K a m a I I . BSOAS 10 (1941) 583, note 2. a n d Irano-Indica IV.
BSOAS 13 (1951) 921 foil.
14
H . W . BAILEY : R m a I I . BSOAS 10 (1941) 583, note 2.

KIDARA AND T H E KIDARITE HUNS IN KASMR 189
Then Vibhisana's son N a r a who also bore the n a m e of

K i m n a r a , and whose prowess was sung by the Kininaras, became king.
represents the Kidarite H u n king K i d a r a whose reign in Kasmr must have been
supposed already on the basis of historical considerations and numismatic
evidence b u t who could not be identified in the t e x t of the R j a t a r a n g i p
hitherto.
The identification of of king K i m n a r a / N a r a mentioned in the R j a t a r a n -
gini with the H u n king K i d a r a can also be corroborated by chronological argu-
ments. According to the n a r r a t i v e of K a l h a n a , after Huska, J u s k a , K a n i s k a six
kings reigned in Kasmr. The length of the reigns of the last five kings amounts
t o 190 years and 6 months b u t the duration of the reign of A b h i m a n y u I is not
indicated. Adopting 134 A.D. for t h e beginning of the K a n i s k a E r a and 195A.D.
for t h e end of the reign of H u v i s k a a n d adding to this figure the 190 1/2 years,
we arrive at the d a t e 385 A.D. to which we have still to add the regnal years of
Abhimanyu I, reckoning with an average reign of 15 25 years. T h u s we obtain
approximately 400 410 A.D. for t h e beginning of the reign of K i d a r a inKasmir
which exactly coincides with the lower time limit proposed by K . Enoki for the
I n d i a n conquests of Kidara recently. 1 5
Taking into consideration t h a t by the coinage of H i r a n y a k u l a an immedi-
a t e historical link between K i d a r a and the Hephthalite kings of Kasmr is
established, we can identify the K i d a r i t e kings of Kasmr with K i m n a r a and his
successors. Thus, it becomes clear t h a t the continuous historicity of the nar-
r a t i v e of Kalhana in the R j a t a r a n g i n i is not restricted to t h e period beginning
with 813 A.D. as it was presumed earlier but it can be extended up to the
beginning of Kuana rule in K a s m r a b o u t the middle of the I l n d century A.D.
Budapest.
15
K . ENOKI : On the Date of t h e Kidarites. Memoirs of t h e Research Departs
m e n t of t h e Toyo Bunko 28 (1970) 20.
Acta Antiqua Academiae Scientiarum llvnqaricae 28, 1980

PH. G I G N O U X
TITRES ET FONCTIONS RELIGIEUSES SASANIDES

D'APRES LES SOURCES SYRIAQUES
HAGIOGRAPHIQUES
Si l'on compare les sources arabes aux sources syriaques, n o t a m m e n t

hagiographiques dans le b u t de les utiliser pour l'histoire de l'administration
sasanide 1 on ne peut que constater combien ces dernires sont suprieures
aux premires, par la qualit de l'information plus prcise et gnralement
plus crdible. Cette supriorit pourrait tre due au fait que l'on n ' a p a s
assez not que beaucoup de personnages mis en scne dans les Actes des
martyrs perses2 ne sont pas seulement des chrtiens a y a n t vcu en Iran mais
des convertis du mazdisme, venus des hautes couches sociales et ayant exerc
des fonctions importantes au service d u Roi des Rois. P o u r prendre quelques
exemples, datables du 6 me sicle, le Catholicos Mar Ab tait n dans la reli-
gion mazdenne, il tait mme arzbed,3 m o t qu'il f a u t corriger en argbed, mon
avis, puisqu'il devint attach au secrtariat de l'mrgar, qui le dnoncera
comme chrtien. 4 De mme, Mar Yazdpanh, est prsent comme plus s a v a n t
que tous les mages nombreux et instruits qui vivaient Su, prs de K a r k d e
Ldan, et il tait chef et juge de son village. 5 E t encore, Mar Giwargis a v a n t
de se convertir au Christianisme, s'appelait Mihr-Mh-Gusnasp (plutt que
1
J ' a i tudie n o t a m m e n t la fonction de marzbn, surtout partir des sources
arabes, d a n s un article paratre dans les Jerusalem Studies in Arabie and Islam. J ' a i
aussi prsent au Colloque de Jrusalem (juin 1980) une communication Pour une esquisse
des fonctions religieuses sous les Sasanidos, qui fait le point des sourcos iraniennes.
2
Cf. P. BEDJAN: Acta martyrum et sanctorum, 7 vol., Paris 18901897. Ce sont
les vol. I I et IV qui contiennent les actes des m a r t y r s porses. J e cite cette dition-l,
et non celle d'AssEMANi : Acta sanctorum martyrum orientalium et occidentalium, I,
] 748, difficilement accessible.
3
BEDJAN : Histoire de Mar-Jabalaha, de trois autres patriarches, d'un prtre et de
de deux laques, nestoriens, Paris 1895, p. 210 : syr. 'rzbd.
4
BEDJAN : . c. p. 216 216. Le t i t r e est crit : hmrgrd muis la fin du m o t -grd
n'est sans doute d q u ' une confusion banale avec les m o t s iraniens se terminant p a r
-gird (cf. Yazdgird : yzdgrd). Le h initial pourrait indiquer que l'aspire tait encore
prononce ; les deux graphies avec ' et h coexistent en t o u t cas, sur les sceaux et les
bulles, comme on peut le voir dans mon Catalogue des sceaux, cames et bulles sasanides,
I I . Les sceaux et bulles inscrits, Paris 1978, p p . 141 et 144.
S
BEDJAN : Histoire . . . p. 396 : les deux titres sont donns en syriaque : r'
wdyn'. Il semble d'ailleurs que pour dsigner le juge, ce soit le plus souvent le t e r m e
syr. qui est utilis.
Acta Antiaa Academiae Scientiarum Hungaricae 28, 1980

192 PH. GIGNOUX
Mihrm-Gusnasp, comme on l'appelle en gnral), il portait donc un nom

mazden comme t a n t d'autres martyrs, et le pre de son pre tait de ligne
royale et gouverneur d'une petite ville, tandis que le pre de sa mre tait
mohd? Plus anciennement, dans les t e x t e s se rapportant au 4 me sicle, 011
trouverait aussi des exemples : le pre du m a r t y r Pusay avait pous une femme
perse de la ville de Bispr (Veh-Sbuhr) lorsqu'il f u t dport dans cette ville
a p r s avoir t fait prisonnier. 8
On peut penser ds lors que les rdacteurs des rcits de ces martyres
devaient tre assez bien informs d u contexte historique, administratif et
religieux, dans lequel vcurent ces chrtiens, et leurs donnes, sous cet aspect-
l, 11'ont pas t, il me semble, s u f f i s a m m e n t tudies, en dpit des t r a v a u x
rcents que l'on doit au Pre J.-M. Fiey ou G. Wiessner. Ce dernier, dans son
livre Zur Mrtyrerberlieferurig aus der Christenverfolgung Schajpurs II9, a
me
centr sa recherche sur les textes se r a p p o r t a n t au 4 s. (mais rdigs au
5 me s.), et relevant les termes utiliss en syriaque pour la hirarchie religieuse
mazdenne, il dplore leur confusion et considre que les mots diffrents
employs pour dsigner une seule et m m e personne seraient ds au fait que
les rdacteurs ne faisaient pas une distinction nette entre les deux grandes
classes de prtres mazdens, les mydn e t les mbadn.10 E t il ajoute que cette
confusion est comparable celle de la littrature grecque (o il n'est question
que de magos) et arabe (o l'on ne parle que de mbad).11 Cette interprtation
m e semble trop pessimiste, et pour juger de la valeur des termes administratifs
transmis en syriaque, je pense qu'il est prfrable de commencer par tudier
les donnes plus tardives, celles du 67me sicles, qui doivent normalement
se rvler plus prcises et plus sres, t o u t comme les sources arabes sont beau-
6
Le nom est en effet crit sous plusieurs graphies : cf. BEDJAN : Histoire p. 436 :
myhrmgwSnsp mais p. 664, 1. 6 et 12, le h est bien not : myhrmhgwSnsp. *Mihrm n'est
p a s usit dans l'onomastique sasanide, telle q u e je l'ai dcrite d a n s Pad nm i yazdn,
P a r i s 1979, p p . 3 6 1 0 0 .
7
BEDJAN : Histoire, p. 436. Son pre m m e tait stndr (syi. 'wstndr' Nisibe.
J . - M . FIEY n'a malheureusement pas compris ce passage, dans son livre Nisibe, mtropole
syriaque orientale et ses suffragants des origines nos jours, Louvain 1977, p. 274, a d d e n d a
n 1, et p. 11 de l'introduction [CSCO vol. 388, Subsidia t. 54]. Ma liste n'est videm-
m e n t pas exhaustive, car on pourrait citer encore le nom de K a r d a g , qui, a v a n t de se
convertir, tait un zoroastrien aux fonctions importantes, qui cumulait les titres de
m a r z b n de l'Adiabne et ptlcS' d'sr (srestn). Ce deuxime titre pourrait bien
reprsenter le pehl. bthSy des inscriptions, car le p initial peut s'expliquer, soit comme
u n e consquence de la rgle d'archasation de l'criture (qui veut qu'en pehl., un son
6 est souvent crit p), soit comme le signe d ' u n e tymologie pressentie d u m o t commen-
a n t p a r pali-. Il faut noter qu'il existe d e u x graphies du m o t en syr. : ptkS' BEDJAN :
AMS I V , 2 2 1 , e t p(hS\ c f . B E D J A N : AM S I I , 4 4 5 , 4 6 3 , 4 6 4 , 4 8 7 .
8
B E D J A N : MS I I , 208.
9
Gttingen, 1967 [Abhandlungen der Akademie der Wissenschaften in Gttingen,
Philologisch-Hist. Klasse, 3. Folge, n 67]. Ce livre est malheureusement difficile utiliser,
car il n'est pourvu ni d ' I n d e x ni de bibliographie.
10
WIESSNEK: o.e. p . 171.
11
WIESSNER : o.e. p . 171 n o t e 3.

T U R K S E T FONCTIONS R E L I G I E U S E S SASANIDES 193
coup plus crdibles pour la priode des Xusr que pour les d b u t s de l'poque
sasanide, et seulement ensuite examiner les sources plus prcoces du 4 me
sicle, que l'on pourra alors juger d'un regard moins svre et en tout cas, dans
une perspective diachronique.
J ' a i affirm, au d b u t de mon introduction, la prcision des sources syria-
ques. J e voudrais en donner quelques exemples. Dans un passage de la vie de
Mar Giwargis, martyr d u 6 me s., est cit un mot moyen-perse en syriaque qui
n ' a pas t reconnu jusqu'ici, m a connaissance, et qui est a b o n d a m e n t attest
dans les inscriptions du 3 me sicle, de Kirdlr Narseh. Le passage
w'tylih Ih 'yqr' bdrg' hd' dmtqry' pdHr
qui se traduit : Et il lui f u t donn une dignit dans ce rang qui est appel
padixSar Giwargis qui tait serviteur la table du Roi.
Mme dans une phrase aussi courte, le mot padixSar semble avoir un sens
assez prcis, et en t o u t cas diffrent de celui que possde 'yqr'.l f a u t donc voir,
partir des occurences en pehlevi, que je voudrais citer ici in extenso, quel
peut tre le sens exact du mot, qui a p p a r a t dans huit contextes diffrents :
a) K K Z 4/KSM 5 / K N r a 10 : 13
'Pm g'sy W pthSly 'pltly 'BYDWN...
Et il leva mon rang et mon padixSar
b) K K Z 6/KSM 8 / K N R m 16 :
'Pm wlhl'nc MLK'n MLK' PWN 'gl'dyhy W pthSly YHSNN. ..

Alors Vahrm roi des rois me maintint dans m a noblesse et mon
padixSar
c) K K Z 8/KSM 11/KNRm 2 2 - 2 3 :
ZKm BYN Stry 'pltly g'sy W ptMly 'BYDWN 'Pm g'sy W pthSly
ZY wclk'n YHBWNt...
Celui-ci ( = Vahrm II) leva mon rang et mon padixSar, et il me
donna le rang et le padixSar des Grands
d) K K Z 9/KSM 13/KNRm 28 ? :
12
BEDJAN : Histoire, p . 437, 1. 5 6.
13
J'utilise les sigles, utiliss dans mon Glossaire des inscriptions petitevies et parthes,
London 1972, et adopts par la plupart des iranistos : K K Z = inscr. do Kirdir la
K a ' a b a de Zoroastre ; KSM, = inscr. du mme S a r Mahad ; K N B r a = inscr. du mme
Naq-i R u s t a m ; NPi = inscr. do Paikuli.
13 Arta Antiqua Academiae Scientiarum Hungaricae 2S, 1980

194 P H . GIGNOUX
W mgwGBB' BYN Ury LB' ptMly YHWWNt...

E t les mogmardu f u r e n t d a n s le p a y s d ' u n g r a n d padixar
e) K S M 2 5 / K M R m 5 0 ( l a c u n a i r e ) :
HT LB' 'gl'dyhy W pth[S]ly QDM...

Si u n e g r a n d e n o b l e s s e e t u n g r a n d padixSar sur...
f ) N P i 9 :15
W GDH W Ury W NPSH W ptMly ZY nyd'k'n MN yzd'n

MKBL[WN...]
E t ( q u ' i l ) 1 6 a c c e p t e d e s d i e u x l e xvarrah, l'empire, son propre t r n e

e t le padixSar des anctres. ..
g) N P i 26 :
[W pt]Mly BR' LMYTNt. . .

E t il f u t d c h u ( ? ) d e s o n padixSar...
h) N P i 33 :
HT 'yw b'iy $hp[whry] . . . [p]sd'lyk k[r]ty 'Pn psd'lykn pthly

YHBWN ...
Si u n e f o i s S b u h r (roi d e s r o i s ? ) (le) f i t psdrlg, a l o r s il (lui) d o n n a
le padixSar d e s psdrlg ...
A i n s i , d ' a p r s c e s d i f f r e n t s p a s s a g e s , les r o i s o n t l e u r padixSar, les

G r a n d s ( k m z u r g n ) a u s s i , le g r a n d m a g e K i r d r g a l e m e n t ; e t m m e le s a i n t
14
Le mot mogmard pourrait bien tre u n titre dans la hirarchie clricale, bien
q u e l'usage qui en est fait dans les incriptions de Kirdr ne soit p a s trs clair. Le syr.
m e semble traduire ce m o t d a n s l'expression gbr' mgibS' que j'ai releve dans BEDJAN :
AMS I I , 51. Mais surtout, nous avons sans doute d'autres titres, forms de la m m e
manire, et dcelables d a n s des noms propres comme Azd-mard, qui se lit sur un sceau
d u Muse do Copenhague (cf. M.-L. BTJHL : A Hundred Masterpieces from the Ancient
Near East, in the National Museum of Denmark, 1974, p. 63, n 51. J e lis: 'c'tmlt' Y
mgw Y hwt'dbwt'n), ou comme Xvady-mard, que j'ai relev sur deux empreintes de
Qasr-i Abu Nasr : n 135 (36.30.249, ou D. 176, selon la publication de R . N. FBYE :
Saeanian Remains from Qasr-i Abu Nasr, Cambridge 1973, p . 62b), o est inscrit:
hwt'dmlt' Y mgw Y pws'n; et n 192 (36.30.118, ou D. 248, FBYE p. 63b) o je lis:
hwt'dmlt' Y mgw F [ ' n j . Or je considre ces noms comme de possibles anciens
t i t r e s , parce que l'inversion des membres d u compos no se fait p a s dans de tels noms
propres, comme on le constate par exemple d a n s Mard-bd (cf. m o n Catalogue de la B. N.
p . 146a, cit en note 4) et Bd-mard (Catalogue p. 143b).
15
Selon la nouvelle dition de H . HUMBACH et PEODS O. SKJAEEVC : The Sassanian
Inscription of Paikuli, P a r t 2, Synoptic Tables, Wiesbaden 1980. J e pense toutefois
q u ' e n N P i 9, on peut et doit restaurer u n W entre les mots 'LS' et pthSly.
16
II s'agit, d'aprs le contexte, du roi Narseh.

T I T R E S ET FONCTIONS R E L I G I E U S E S SASANIDES 195
Giwargis qui servait la table du roi. L'tymologie propose depuis longtemps

ne nous a p p r e n d pas grand'chose, et t o u t le monde t r a d u i t par honneur,
dignit. 17 Mais d'aprs nos textes, y compris le passage en syriaque, il s'agit
manifestement d'une certaine qualit ou qualification qui ne va pas de soi,
qui est accorde, peut-tre en vertu des mrites, et qui est proprement mazd-
enne. Si le m o t est vraiment hrit du parthe, 1 8 on pourrait penser qu'il dsigne
non un titre mais une qualit spciale, que les contextes ne nous p e r m e t t e n t
pas de prciser davantage, et qui existait d j sous les Arsacides.
Il est intressant de noter que le m o t est aussi attest en moyen-perse
manichen. D a n s un texte publi p a r W . Sundermann, 1 9 on lit en effet ceci :
'w(g) 'st'ySn 'wd pdxsr dyynd (')w$ d'nynd lew ley 'st
'wS (pd) wzrgyy(h) 'wd ' g r ' y y h gwg'y (hyn)d ...
Quel que soit le sens ici, on sera f r a p p tout de mme de constater l'al-
liance de l'expression padixSar ddan avec le mot agrdyh, comme dans plu-
sieurs passages des inscriptions.
D a n s les actes des martyrs perses, de nombreuses allusions sont faites
la religion mazdenne, au rituel et aux coutumes que les rdacteurs connais-
sent bien. Il n'est pas question ici de relever l'ensemble de ces donnes, mais
pour en montrer la prcision, je voudrais prendre seulement un exemple. D a n s
la vie de Mar Giwargis, 20 il est question d ' u n mage qui venant vers le saint,
sans savoir sans doute qu'il s'tait converti, selon l'habitude mazdenne
lui donne le vz au moment du pain (i.e. du repas), comme il est c r i t :
wyhb Ih wsq' b'dn' dlhm' ...
Le syr. wsq' reprsente le pehl. vzag de toute vidence. Or, l'expression

vz ddan, bien rendue ici en syr., exprime prcisment l'change de la prire
a v a n t le repas entre deux personnes. Certes, le terme courant pour dsigner
l'ensemble de cet acte rituel est vz griftan,21 mais il ne convenait pas d a n s ce
contexte.
Ces quelques exemples nous o n t m o n t r la validit des sources syriaques.
On p o u r r a donc faire confiance, avec la prudence requise, ce qu'elles nous
17
Cf. M. BACK: Die Sassanidischen Staatsinschriften, 1978, P. 248 [ = A c t a
Iranica 18] : <*pati-Sfara- Ehre, Ehrung.
18
BACK : o.e. p. 249, qui cite cela comme une suggestion d'O. SZEME RNYI.
19
Mtelpersische und parthische kosmogonische und Parabeltexte der Manichaer,
Berlin 1973, p. 102, ligne 1986/88. J e n ' t u d i e p a s ici les autres occurences du m o t , q u ' o n
peut relever d a n s les Milteliranische Manichaica, hrsgg. ANDREAS-HENNING, et q u ' o n
a t r a d u i t p a r honorer.
29
BEDJAN : Histoire p . 440.
21
Cf. M. BOYCE et E. KOTWAL : Zoroastrian b j a n d drn, d a n s BSOAS 24 (1971)
pp. 5673 et 298 313, n o t a m m e n t p. 58 et 60.

196 PH. GIGN0X
r a p p o r t e n t au sujet de l'administration religieuse l'poque sasanide. Toute-

fois, Wiessner n'est pas de cet avis, et reste pessimiste q u a n t cette validit.
I l considre que l'emploi de plusieurs m o t s pour dsigner une seule et m m e
fonction manifeste, de la p a r t des rdacteurs, une diffrence non claire entre les
d e u x grandes classes de prtres mazdens. 2 2
J e pense qu'il f a u t distinguer entre les textes du 45 me s. auxquels
s'est limite l'tude de Wiessner , et ceux du 6me sicle, car ils nous fournis-
sent une documentation assez diffrente, et on pourra ainsi voir se dessiner
u n e certaine volution, condition, encore une fois, de faire confiance a u x
textes.
Premire constatation : il n'est nulle p a r t question d ' u n mbadn mbad
au 4-5 m e s., alors que ce titre est continuellement a t t e s t au 6 me s., et ce
n ' e s t sans doute pas u n hasard. On ne p e u t videmment croire, avec Christensen,
sur la foi de Tabari, 2 3 que ce titre existait dj sous Ardachir I er . J ' a i m o n t r
ailleurs, 24 que cette fonction n'a pu commencer au 3 me s., car le mage Kirdr
l ' a u r a i t assurment exerce, et nous n'en avons aucune trace dans toute l'pi-
graphie iranienne. Si l'on en croit les sources syriaques, ce serait donc une
fonction relativement tardive. D'ailleurs, peu de temps aprs la parution de
l'ouvrage de Christensen, S. Wikander estimait que la dignit de Grand Mage,
ou de mbadn mbad, n ' a v a i t commenc qu'au 5eme sicle. 25
U n texte intressant de la vie de Mar Aba, qui f u t patriarche de 540
552, nous prsente u n ensemble de dignitaires, lors de la reconnaissance offi-
cielle de sa mort :
Et ensuite le R o i des rois et le Mbadn mbad envoyrent le Mbad
e t le J u g e (dayyana) et les autres mages (magus) avec eux pour voir si c'tait
bien le saint ou n o n . . . 2e
Le Mbadn m b a d est souvent accompagn des magus, ce terme dsi-
g n a n t la fois les prtres de rang infrieur et plus gnralement les prtres quels
qu'ils soient, puisque l'expression rabb magu sert dsigner simplement le
m b a d n mbad, comme l'indiquent d e u x passages de la vie de Mar Aba. 27
Cette quivalence est m m e exprime, de la bouche de Mar Aba, qui dclare :
22
O.e. p . 1 7 1 .
23
L'Iran sous les Sassanides, p. 119.
24
D a n s mon article sur le marzbn, cit d j note 1.
25
Feuerpriester in Kleinasien und Iran, L u n d 1946, p . 182. G. WIDENGBEN :
Les religions de l'Iran, Paris 1968, p. 292, ne se prononce pas sur une d a t a t i o n , et J .
DUCHESNE-GUILLEMLN : La religion de l'Iran ancien, p. 294, estime que ce fut peut-
tre sous Vahrm V . . ..
26
BED JAN : Histoire, p. 271. La fonction de juge, que je n'tudierai pas ici, devrait
correspondre au ddvar pehlevi. Le dayyana est distinct d u Mbad, mais d a n s d ' a u t r e s
passages (Histoire, p. 229), il est identifi a u rabb maguS. Le juge, comme le rad (cf.
plus loin), sont des titres qui peuvent tre donns u n mbad.
27
Cf. BEDJAN : Histoire, p. 232 et 233 : hydyn mwbdn mwbd wklliwn mgibS' et
l'expression quivalente, p. 230 et 233 : qm rd ' whbrivhy.

T I T R E S E T FONCTIONS R E L I G I E U S E S SASANIDES 197
w'mr Imwbdn mwbd w'p 'nt d'ytyk rb mgiv'

Et il dit au mbadn mbad : Toi aussi, qui es chef des mages. . . 28
Comme Ta not Wiessner, 29 le m m e personnage est souvent dsign p a r

des titres diffrents. P a r exemple, le rS mauhpat (chef de mbads) est appel
aussi simplement mauhpata (mbad) ou mme magua rabba (grand mage)
ou seulement maguSa. Mais toutes ces appellations p e u v e n t convenir u n
mage suprme, et ne prouvent pas ncessairement que le rdacteur est m a l
inform. Encore aujourd'hui, lorsque Ton parle du P a p e , on p e u t l'appeler
aussi pontife suprme, ou vque (de Rome), ou pasteur, ou mme prtre,
puisque c'est sa premire dignit.
Le Mbadn mbad est appel aussi rSa d' magu (chef de mages).
D a n s la vie de Mar Yazd-panh, les d e u x titres sont explicitement prsents
comme identiques : 30
mwbdn mwbd d'ytwhy rlT dmgw'
Prcisment, d a n s ce mme passage, nous trouvons une liste hirarchique

fort intressante, o sont numrs :
1 le Mbadn mbad, qui est ra d e magus ;
2 u n grand Mbad ( m a u h p a t a rabba) ;
3 le Mbad du Beth A r a m a y ( = Sleucie-Ctsiphon) ;
le t e x t e est le suivant :
wmtyw Ih Imdynt' pyrwzSbwr mwbdn mwbyd d'ytwhy rS' dmgwS'

wmwhpt' hd rb' wmwhpt' dbyt 'rmy'
Et ils arrivrent la ville de Prz-Sbuhr, le m b a d n m b a d
qui est le chef des mages, u n Grand mbad et le m b a d d u B e t h
Aramay. 3 1
Les donnes gographiques me paraissent confirmer la valeur d u t e x t e :

Prz-Sbuhr (ou Anhr) 3 2 est situe sur l'Euphrate, et c'tait le dernier vch
d u Beth A r a m a y en remontant l ' E u p h r a t e . Elle devait donc faire partie ausei
de la province civile d'srestn, o pouvait fort bien se rendre le Mbad
responsable de cette province, puisque les divisions ecclsiastiques nestoriennes
calques au d b u t sur les divisions administratives sasanides o n t d sans d o u t e
le rester par la suite.
28
BEDJAN : Histoire, p. 232. Cf. de mme, p. 226.
29
O.e. p. 170.
30
31
32
Comme l'a m o n t r A. MABIOQ : Classica et Orientalin, Paris 1965, p p . 9497.

198 PH. GIGNOUX
Ainsi, au-dessous du Mbadn mbad, il y a un grand mbad, et c'est

u n titre que l'on rencontre souvent et qui est n e t t e m e n t distinct du Mbad
qui a la charge de telle ou telle province, comme cela est mentionn dans les
textes. E n voici quelques exemples :
hydyn qm mwhpt' dbyt 'fmy'

Alors il se leva le mbad du B e t h A r a m a y . . . >>33
w'yty l'dwrbygn wlbk qdm mwhpt' d'tr' hw
. . . E t il e m p o r t a [ces crits] en d u r b d a g n et les plaa d e v a n t
le m b a d de cet endroit. 34
'lm 'nwn Imwhpt' dbhw zwn' lyt hw' bhdyb dnwbl 'nwn Iwt mlk'
Il les livra au mbad qui, en ce temps-l, avait le pouvoir en Adiabne,
pour qu'il les conduise auprs d u R o i . . . 35
Comme le m b a d n mbad, le grand m b a d se tient manifestement la

cour, devant le roi, lorsqu'il s'agit p a r exemple d'interroger un chrtien :
hydyn qm mwhpt' rb' qdm mlk'. ..

Alors il se leva, le grand m b a d d e v a n t le roi. . . 3fl
Mais il se dplace aussi, en dehors de la cour, comme en tmoigne ce

passage du m a r t y r e de Yazd-panh :
'zlw w'mrw Ih Imwhpt' rb' d'yt hw' b'tr' hw mtlth wmn dm' mwhpt'
hw Sdr wqryhy Iqdmwhy
Ils allrent parler contre lui au grand m b a d qui se trouvait en ce
lieu et ds que ce mbad (les) e t entendus, il l'envoya (chercher) et
le convoqua d e v a n t lui.37
Il n ' y a pas s'y mprendre ici : il s'agit bien du grand mbad et non
p a s d u mbad de la province. Que ce grand m b a d avait un pouvoir sur t o u t le
p a y s , le partageant sans doute avec le m b a d n mbad, me semble indiqu
p a r u n passage de la vie de Yazd-panh, o le roi promet au saint, s'il aban-
d o n n e le Christianisme, de le faire grand m b a d sur t o u t le pays : 3 8
n'bdk mwhpt' rb' bklh 'r"...
33
BEDJAN : Histoire p. 233, idem p. 300.
34
35
B E D J A N : AM S I V , p . 136.
36
37
33

Il me semble donc que les grands mbads pouvaient former une sorte de
collge autour d u Mbadn mbad, l'image de ce qui existait alors d a n s
l'Eglise nestorienne. Nous savons en effet qu'il y avait trois degrs dans sa
hirarchie : le Patriarche, les mtropolites ayant juridiction sur une rgion
trs tendue, et les vques 39 d o n t l'autorit ne s'exerce que sur une seule
ville. 40 Selon Labourt, le mtropolitain est en quelque sorte, le dlgu rgional
du Catholicos qui le n o m m e . . . Il p e u t intervenir dans les contestations qui
surviennent entre les vques. . . Mais pour les causes plus graves, il doit en
rfrer au patriarche. Car il ne possde pas, comme lui, la juridiction ordinaire
d a n s toute l'tendue de sa p r o v i n c e . . . Sa primaut est s u r t o u t une p r i m a u t
d'honneur. 4 1
Ainsi, de mme que le mtropolite a un pouvoir intermdiaire entre le
patriarche et l'vque, et moins territorial, car l'vque est matre absolu
d a n s les limites de son diocse, 42 de mme il semble que le grand Mbad soit
aussi une sorte d'intermdiaire entre le Mbadn mbad et le mbad provin-
cial. Cette hirarchie mazdenne est encore reflte dans u n passage de la vie
de Yazd-panh, o le grand mbad propose au saint de le nommer m b a d (il
tait seulement, a v a n t sa conversion, rSa et dayyana, comme on l'a vu), s'il
revient au mazdisme :
W 'n dly tttpys wtSm' Ihw mdm d'mr 'n' Ik thw' mwhpt' btr dyly
b'tr' hn'
Maintenant, si tu te laisses convaincre et coutes ce que je te dis,
tu seras mbad aprs moi en cette rgion.. . 43
Bien entendu, les titres o n t chang, au cours des sicles : le Patriarche

s'est d'abord appel archevque, puis grand mtropolitain, et partir de
DdiS' (421 456), l'vque de Sleucie se nomme presque exclusivement
Catholicos, puis patriarche.
Si l'on en croit le tmoignage d ' u n sceau trs intressant de la Bibl.
Nationale de Paris, 44 l'vque a y a n t autorit sur le H u l v n et le Balsagn por-
t a i t le titre de vuzurg Catholikos.ib Ce devait tre l'quivalent d ' u n mtropolite.
On peut donc supposer que le grand m b a d portait le t i t r e en pehl. de *vuzurg
mgbed, malheureusement non a t t e s t jusqu'ici. Par contre, les bulles sasanides
39
Cf. J . LABOURT : Le Christianisme dans l'empire perse sous la dynastie sassanide,
Paris 1904, p. 326 sv.
40
LABOURT : o.e. p. 332 : Il ne doit y avoir qu'un seul vque dans chaque ville.
41
L A B O U R T : o.e. p p . 3 3 1 3 3 2 .
42
L A B O U R T : o.e. p . 3 3 3 .
43
B E D J A N : Histoire P . 398.
44
Cf. m o n Catalogue, p. 64, n 7.5. Le professeur J . HARMATTA avait le premier
dchiffr l'essentiel de l'inscription, et je le remercie encore p o u r cette i m p o r t a n t e
dcouverte.
45
Pehl. : wcwlk <&> 'tlykws.

200 PH. GIGNOUX
nous o n t livr en de nombreux exemplaires, des empreintes de sceaux de m-

b a d s des villes,48 que je place au troisime degr de la hirarchie, qui avaient
d o n c sceller des documents, ce que ne faisaient peut-tre pas les grands m-
bads.
L a situation de la hirarchie mazdenne que j'ai t e n t de dcrire ici, est
celle que nous connaissons partir des textes syriaques l'poque des Xosr.
J ' a i indiqu plus h a u t que le mbadn m b a d n'tait p a s attest au 45 me
s. E n effet, cette poque, on parle seulement du rSa de mauhpat, le chef
des mbads, qui, au-dessus de tous les mbads, a d logiquement prcder le
t i t r e de mbad des mbads. 47 Mais la situation au 6 me s., mieux documente,
e t plus proche de nous, semble plus claire. C'est pourquoi nous nous limitons
ici cette priode-l, c o m p t a n t prsenter ailleurs un t a b l e a u de la situation
p l u s ancienne.
Le rad
Il est une fonction encore mal connue, celle qui est exerce par le Rad
et sur laquelle je voudrais apporter le tmoignage des t e x t e s syriaques.
Ce titre a p p a r a t p o u r t a n t maintes reprises dans les actes des martyrs,
c o m m e dans la l i t t r a t u r e pehlevie tardive, mais H o f f m a n n dans ses Auszge
semble l'avoir presque ignor, en l'interprtant comme u n nom propre Rd,
n o m qui n'est pas a t t e s t dans l'onomastique sasanide. 48
Il est vident q u e la graphie en syriaque est ambigu, puisqu'on peut
s u p p u t e r comme voyelle mdiane un a long ou bref, mais c'est le titre de rad
( < a v . ratu-) qui est gnralement attest.
E n effet, dans le m a r t y r e de Saint Pethion, H o f f m a n n cite un personnage
qui s'appellerait t a n t t Naihormazd t a n t t Naihormazdrdh.M Bien plus, il
semble q u ' H o f f m a n n identifie dans ce rcit trois personnes diffrentes, car
ct de ces deux noms, il fait du dernier nom compos, d e u x personnages s'ap-
p e l a n t Naihormazd e t Rdh !50 Or il est clair qu'il faut comprendre dans ce mot
compos Naihormazd le rad !
D e mme, d a n s l'histoire de Mar Giwargis, H o f f m a n n interprte le syr.
rd c o m m e le nom d ' u n mage, alors qu'il s'agit manifestement d ' u n e controverse
sur le feu entre le saint e t un Rad?1
46
P a r ex., d a n s le s e u l Catalogue de la B . N., o n t r o u v e les m b a d s de H u s r A s d -
O h r m a z d ( = Ctsiphon), s. 1.9 ; d'AbarSahr (Nipr), b. 1.1a, 1.3a, 1.4a, 1.5a; d ' E r n -
X v a r r a h - S b u h r (Suse), b . 1.4a, 4 . 8 a ; de Mdie, b . 7.8a.
47
C f . B E D J A N : AM S I I , 158.
48
G. HOFFMANN : Airszge aus Syrischen Akten Persischer Mrtyrer, Leipzig 1880,
p p . 6465. JUSTI : Iranisches Namenbuch, p . 256b, ne cite ce n o m que sur la foi
d'Hoffmann.
49
O.e. pp. 6 4 - 6 5 .
50
O.e. p. 68.
81
O.e. p. 109.

D'aprs les sources syriaques, la fonction de rad n'est pas une fonction
indpendante, mais elle tait exerce par le mbad d ' u n e province :
hydyn qm rd wmwh.pt' dbyt 'fmy'...

Alors il se leva, le r a d et mbad du Beth A r a m a y . . . 52
Le rad est attach une province :
w'Slmwhy Ird d'dwrbygn dSmh hw' d'dyn

Et ils le livrrent au r a d d'durbdagn d o n t le nom tait Ddn. 5 3
Dans la discussion de Mar Giwargis avec un rad, 5 4 celui-ci est appel r a d

(ligne 8), mais plus loin (ligne 14) maguSa, tandis que c'est de nouveau le r a d
qui est dsign ligne 16,de mme q u ' la page suivante, ligne 1. Mais cela ne doit
pas nous tonner, car comme nous l'avons vu plus h a u t , propos du m b a d n
m b a d , il est normal qu'un m e m b r e lev de la hirarchie ait t d ' a b o r d
mage, t o u t comme un vque ou u n mtropolite a v a i t d'abord t consacr
prtre.
L a fonction de rad tait distincte de celle d'ainbed (ou ynbed) que j'ai
tudie ailleurs, 55 car on lit d a n s l'histoire de Mar Aba :5
wmn btr d'Slmh rd I'ynbd. ..
U n peu plus loin, 57 le rdacteur utilise le pluriel (n'hdwn-yhy) p o u r

indiquer qu'il s'agit de deux fonctionnaires diffrents :
wSdrwhy Ird wl'ynbd dn'hdwnyhy ksy'yt

Et ils l'envoyrent au R a d et l'Ainbed p o u r qu'ils s'emparent de
lui en secret.
Que le rad ait t, en t o u t cas, un fonctionnaire i m p o r t a n t dans la hirar-

chie, cela est a t t e s t aussi d a n s les textes pehlevis. Ainsi, dans la procdure
d'ordalie, dcrite dans plusieurs chapitres du Mdayn hazr ddestn, c'est
62
BEDJAN : Histoire p. 235.
53
BEDJAN : Histoire p. 239. C'est u n excellent nom, bien attest (cf. Pad nam,
yazdn, 1979, p . 67), que n ' a p a s reconnu BEDJAN, qui a voulu l'interprter comme
dd-dn ! Do mme, O. BRAUN, Ausgewhlte Akten Persischer Mrtyrer, 1915, p. 204,
a cru bon de transcrire le nom D d [ d ] n . Certes, le nom de Dn-dd existe : BEDJAN :
Histoire p. 234.
64
55
Dans m o n art. Pour une esquisse des fonctions religieuses sous los Sasanides,
cit en note 1 : j ' y tudie les sources iraniennes ainsi que les passages des actes des
martyrs.
56
57

202 P H . GIGNOUX
sous l'autorit d'un r a d et non d'un seul m b a d que le jugement pouvait tre
a r r t . L a condamnation ne pouvait tre prononce par les seuls ddvarn
ou mbadn. 5 8
Cette tude, consacre quelques dignitaires mazdens, voudrait en
m m e temps manifester quelques aspects de l'intgration de la chrtient
nestorienne l'intrieur de l'empire sasanide. Le jugement de J.-M. Fiey,
a f f i r m a n t la non-assimilation des chrtiens en Iran, doit tre rectifi : dans
l'introduction son livre sur Nisibe, 59 il dclare de manire premptoire :
L'glise de Perse p u t bien s'tendre, . . . mais elle n'en devint jamais
p o u r a u t a n t iranienne, ni par la langue de sa liturgie, ou de ses crits spiri-
tuels, ni par sa hirarchie, le plus souvent importe, ni p a r ses racines hagio-
graphiques, ni s u r t o u t p a r sa mentalit. Ce jugement peu nuanc sert appa-
r e m m e n t d'argument pour expliquer l'effondrement rapide de la chrtient
d e v a n t l'Islam. Or, m m e si les premiers chrtiens d ' I r a n o n t t des prison-
niers d'importation grecque, ils ne le sont pas rests p e n d a n t quatre sicles !
P a r ailleurs, il ne f a u t p a s seulement distinguer, comme le fait Wiessner, entre
des chrtiens de langue grecque, de langue aramo-syriaque ou de langue moyen-
perse, mais il faut aussi poser le problme du bilinguisme. 60
Dans son excellente monographie sur l'glise nestorienne, toujours utile,
le Cal Tisserant notait d j que le Catholicos Ma'n, vers 420, avait traduit en
persan des ouvrages syriaques. 6 1 D'autres mtropolites du P r s , Rv-Ardasir,
composrent des hymnes et cantiques en persan. Le code juridique d ' I s ' b x t
f u t rdig en persan et t r a d u i t plus tard en syriaque. Certes, le syriaque est rest
la langue liturgique, mais les nestoriens o n t t certainement bilingues, utili-
s a n t le syriaque et le moyen-perse, tout comme aujourd'hui ceux qui vivent au
K u r d i s t a n , parlent la fois le syriaque moderne et le kurde. 62 U n passage de la
vie de Mar Aba tmoigne clairement de ce bilinguisme :
68
Cf. A. PERIKHANIAN : Ordalija i k l j a t v a v sudoproizvodstve doislamskogo
i r a n a , dans Peredneaziatskij Sbornik I I I , Istorija i filologija stran drevnego vostoka,
Moskva 1979, pp. 182192.
59
Nisibe, mtropole syriaque orientale et ses suffrayants des origines nos jours,
1977, p . 9.
90
G. WIESSNER, Zur Auseinandersetzung zwischen Christentum u n d Zoroastrismus
in I r a n , dans ZDMG S u p p l e m e n t a I, X V I I . Deutscher Orientalistentag, Wiesbaden
1969, p . 411. J e n'ai p a s compris l'intrt de cet article, qui compare u n rcit tardif, et
dclar inauthentique, savoir le roman de Qardag au Krnmag Ardasr, pour
d m o n t r e r le changement des deux ples du dualisme zoroastrien (bien et mal) opr
d a n s le christianisme iranien, dfini comme u n e sorte de dualisme invers. J e ne vois
p a s comment l'on peut v r a i m e n t diffrencier deux grandes religions partir d'lments
mythico-lgendaires c o m m u n s (qui signifient seulement que les r d a c t e u r s syriaques
o n t t influencs, ce qui est normal, par la littrature pique iranienne), ni je ne puis
croire que le christianisme nestorien ait j a m a i s t u n e forme de dualisme, ce que l'auteur
semble admettre.
61
DTC (dictionnaire de thologie catholique), fasc. X C I - X C I I , 1930, p. 264.
62
Comme je l'ai l o n g u e m e n t constat d u r a n t u n sjour en I r a k .

T I T R E S E T FONCTIONS R E L I G I E U S E S SASANIDES 203
wmn btr d'mr hlyn klhyn bUri swryy' hpJc brdrS w'mr ' blSn'
prsy' 'yJc dnSm'wn 'p prsy' wnd'wn mdm d'mr. ..
Et aprs qu'il et dit toutes ces choses en langue syriaque, il les rpta
en langue perse afin que les Perses aussi l'entendent et connaissent
ce qu'il avait dit. 63
J e vois aussi une autre preuve de ce bilinguisme d a n s les sceaux chrtiens.

E n effet, dans un article paratre, 6 4 j'ai runi une trentaine de sceaux que
j'identifie comme chrtiens essentiellement partir du contenu de leurs in-
scriptions. Or celles-ci sont soit en syriaque soit en pehlevi, soit mme pour l'une
d ' e n t r e d'elles dans les deux langues !65 C'est assez dire combien les nestoriens
n'avaient pas de peine utiliser l'une ou l'autre langue, mme dans des docu-
ments officiels comme les sceaux celui du grand Catholicos cit ci-dessus est
grav en criture et langue pehlevie ! ce qu'ils n ' a u r a i e n t certainement pas
fait s'ils taient rests l'cart de toute influence mazdenne et iranienne 66
et ferms sur leur propre communaut. L'expansion nestorienne vers l'est
partir du 6 me et 7me sicle tmoigne au contraire de l'ouverture des chrti-
ens au monde dans lequel ils vivaient, t o u t comme les actes des martyrs. C'est
pourquoi ceux-ci, comme j'ai t e n t de le montrer, nous informent la fois sur
la structure de l'glise mazdenne et sur la bonne insertion des nestoriens dans
l'Iran sasanide.
Paris.
63
64
Sceaux chrtiens d'poque sasanide, dans Iranica Antiqua 1980 (en hommage
la m m o i r e d e R . GHIRSHMAN).
85
C'est u n sceau du Brit. Museum n 120 260 ( = BIVAR NA 1), o l'on peut lire
deux mots : pehl. p'ky et syr. 4h', Dieu saint !.
68
Cette influence mazdonne se manifeste aussi dans les inscriptions de sceaux,
puisque les chrtiens ont utilis des formules zoroastriennes, m a i s en les a d a p t a n t et les
rendant conformes leur propre foi.

211 A. D. H. BIVAR
QUESTIONS OF INTERPRETATION IN THE

INSCRIPTIONS OF THE SASANIAN SEALS
The engraved gems of the Sasanian period in I r a n , with their s h o r t P a h l a v i

inscriptions, have a t t r a c t e d interest in recent years for t h e special light they can
throw on the history, culture, and religion of t h a t civilization. In m y present
paper, for t h e o p p o r t u n i t y to present which I express t h a n k s to the H u n g a r i a n
Academy of Sciences, 1 I shall be dealing w i t h some fairly limited problems of
interpretation which concern these inscriptions.
One of t h e most familar terms in t h e Pavlavi seal-inscriptions is the word
rst, either found alone, or in phrases. This has dictionary meanings true,
straight, direct, 2 b u t for the present discussion the opposed shades of mean-
ing are, on t h e one hand, correct, and, on the other, righteous. We shall come
in a moment to the associated abstract noun rstih (l'styhy) righteousness.
One line of interpretation, therefore, for the effective meaning of the former
term on a Sasanian seal is to regard it as a mere mark of authentication, signify-
ing t h a t a document is correct. Equivalent in the Arabic chancery usage is t h e
use of a phrase such as sahha dhlika that is correct as a mark of authentica-
tion for official documents in the states of the Maghrib. 3 I had been inclined to
this view until I saw Professor Gignoux's eloquent article in the Ydegr Nme-
ye Purdvud, which makes strong case for the interpretation of phrases contain-
ing this term as the expression of moral sentiments :4
Il rsulte, de ce qui prcde, que p a r t o u t o rdstjrsth n'est pas employ

seul, une exception prs (et encore douteuse : le dernier exemple), il
possde une connotation morale. Ds lors, q u a n d il est seul, pourquoi
' T h i s paper was read at the conference on the theme Problmes d'interprtation
des sources sur l'histoire de l'Asie Centrale prislamique hold a t Budapest on September
30th to 2nd October 1980 under the auspices of the Hungarian Academy of Sciences
(with the assistance of the Union Acadmique Internationale and the CIPSH). I am
grateful also to tho Council of the Corpus Inscriptionum Iranicarum, who assisted my
attendance.
2
D. N. MACKENZIE : A concise Pahlavi dictionary. London 1971. 71.
3
S. M. STERN : Fatimid decrees. London 1964. 139 40.
4
PH. GIGNOUX : Los formules des sceaux sassanides et la signification de rst
et rsth. Farhang-e Iran Zamin, X X I , 1354/1976 ( = Ydegr Nme-ye Purdvud), 49.

206 A. D. H . BIVAR
perdrait-il cette connotation ? Il reste en t o u t cas le prouver, si l'en v e u t

tablir qu'il servait authentifier un document
Of course, in one very obvious sense, every seal is designed for the authen-
t i c a t i o n of documents ; a n d in a s t a t e where the administration was largely in
t h e h a n d s of the priesthood, the distinction between a religious a n d an executive
act m a y not always h a v e been clear-cut. A seal recently published by J u d i t h
Lerner, 5 illustrating t h e Christian, or Jewish, theme of the sacrifice of Isaac,
bears t h e Pahlavi inscription ZNH mwdly r'st This seal is correct. Such a t e x t
is d i f f i c u l t to understand in other t h a n a practical sense t h a t an earlier seal
of t h e same owner had been lost, a n d only its replacement was now considered
valid. I t is true t h a t t h e iconography makes clear t h a t this was not a seal used
by t h e Zoroastrian hierarchy.
Y e t its engraving is sufficiently typical of Sasanian glyptic, and it is dif-
ficult t o doubt t h a t t h e word rst would have been understood as on other seals.
T h i s example, therefore, tends to some degree to swing back t h e balance, a n d
to leave the precise meaning of rst once more open to debate. Where we h a v e
knowledge only of t h e seals, and not of t h e documents they authenticated, i t is
difficult to grasp their message clearly ; y e t perhaps the fault is ours for seeking
to u n d e r s t a n d them too literally ; between an ethical exhortation and a s t a m p
of a u t h e n t i c i t y there m a y have been less difference t h a n we are inclined to sup-
pose.
W h e r e the formula involves t h e a b s t r a c t noun, rstih righteousness, t h e
ethical force is of course more obvious. A t t h e same time, the term had a long
life in t h e Iranian chanceries, and the words rsti, rsti appear in the sign man-
ual of Timur. 6 A different nuance, however, is conveyed by t h e s t a t e m e n t of
al-Mas'di (Murj, I I , 204), t h a t t h e revenue seal of Khosrau I Anoshirvan was
engraved with the word al- 'adle justice an Arabic translation of rstih. I t has
to be considered, therefore, whether the Sasanian seals engraved with this word
are likely to have been employed exclusively in the revenue d e p a r t m e n t .
Curiously enough, one of t h e more explicit phrases quoted b y Professor Gi-
gnoux, 7 Vstyhy TB ZY SB W PWN bhty which he variously translates La bon
d r o i t u r e des choses (se fait) par le sort or Bonne exactitude des biens (marchan-
dises) en partage, could, though the implications in either case are far from
clear t o me, be interpreted as an admonition to share the burdens of t a x a t i o n
e q u i t a b l y . However, though its editor appears to entertain t h e possibility of
5
JUDITH A. LERNEB : Christian seals of the Sasanian period. Leiden a n d I s t a n b u l
1977. 42, no. 17. [Subsequently, P . GIGNOUX:'6'; sasanides de diverses collections pri-
ves, L e u v e n 1982, 19 h a s justifiably shown t h a t t h e complete reading is ZNH mwdly
r'styn T h i s seal belongs t o R a s t e n " which eliminates this example f r o m t h e discussion.]
6
JAHNGR QA'IM-MAQM : M u h r h , t u g h r h u tawq'h-yi pdishhn-i I r n
az I l k h n i y n t payn-i Q j r i y a . Barrasih-yi trkh, IV, 3 4, 1348/1969 70. 128
7
GIGNOUX : 49.
Acta Antiqua Aeademiae Scientiarum Hungarirae 28, 1980

QUESTIONS OF I N T E R P R E T A T I O N 207
a fiscal purpose for this seal, I still feel, as I did some years ago, t h a t an ethical
connotation for rstlh provides t h e best solution a t present.
A formula of t h e Sasanian seals of which the implications can be puzzling
is t h a t which appears in its basic form as Iwb'n HZYTN (ruvn vn) see t h e
spirit. The ideogram IIZYTN, representing the past participle, or the impera-
tive, of the verb to see, might of course appear on seals in several connections,
literal as well as spiritual. At the present day, all those of us who travel a r e
familiar with visa stamps. The word -ruvn, as the n a m e of the ruler Anoshirvn
plainly testifies, might form an element of personal names or honorifics. On t h e
other hand it is difficult to understand in w h a t sense anything so invisible as
the spirit can be spoken of as to be seen. I find it difficult, nevertheless, to follow
Brunner (Sasanian stamp seals in the Metropolitan Museum of Art, 141) t h a t
Iwb'n IIZYTN represents a legal formula signifying witnessed. At the same
time, he suitably calls attention to almost the first instance of such a formula
to receive scholarly discussion, the epigraphic seal a t Berlin which had also been
noted by Shaked, reading : 8
tny ZY KBYR k'mk Iwb'n HZYTN
which with the other commentators I t a k e to mean Body of m a n y desires, con-

sider the spirit. I assume this is an ethical admonition, expressing the well-
known Zoroastrian value-judgment which, dualistically, constrasts the physical
unfavourably with the spiritual being. The admonition (not improbably a quo-
t a t i o n from a religious text), is I suppose addressed not to the reader of the docu-
ment, b u t directly to the owner of the seal, to remind him of the sovereign vir-
tues of austerity. Consequently, when we come to interpret such shorter for-
mulae as thmdyny Iwb'n HZYTN, I understand the implications a little dif-
ferently, not only from Brunner, b u t also indeed from Gignoux, 9 whose transla-
tion Regarde l'me de Tahm-den suggests t h a t he considers the admonition
addressed to the reader, and the spirit in question to be specifically t h a t of t h e
owner of the seal himself, so t h a t the whole legend has (as he suggests) a f l a v o u r
of commemoration. In my view, this would constitute, for the Zoroastrian,
an a t t i t u d e of self-glorification quite a t variance with t h e sentiments appro-
priate to t h a t f a i t h . R a t h e r , I should prefer, the admonition was a p e r m a n e n t
reminder to the t i t u l a r y of the seal to give precedence to spiritual over physical
values, and by no means a message addressed by him to the reader of t h e in-
scription.
8
SHAOUL SHAKED : Jewish and Christian seals of tho Sasanian poriod, in MYRIAM
ROSEN-AYAEON (ed.), Studies in memory of Oaston Wiet, 29 ; citing P . HORN a n d G.
STEINDORF, Saeeanidische Siegelsteine, Berlin 1891. 39, no. 2179.
P . 60.
14* Acta Antigua Academiae Scientiarum Hungaricae 28, 1980

208 A. D. H . BIVAR
These are useful preliminaries before turning to one of t h e most trouble-

some a n d puzzling seals which use a f o r m u l a of this type, the British Museum's
seal B B 1 , which has inspired an unusual number of conflicting interpretations.
T h o u g h there is no disagreement a m o n g scholars as to the reading of the letters,
a n d indeed of the words, the point of commencement, the word division, a n d
the p u n c t u a t i o n invite varying solutions. My first a t t e m p t 1 0 HZYTN LY
pr'whlmzdy ZY hwhymlwb'n Seen b y me, Farr-Ohrmizd, son of Huhmruvn,
m a d e t h e purport of t h e inscription t h a t of a simple visa-stamp, a n d of course
sidesteps t h e general question of Iwb'n. Shaked soon afterwards sent me his very
ingenious rendering 1 1 mzd ZY xwxymlwb'n XZYTN LY pr'wxl My frawahr
( = eternal spiritual being) has seen t h e reward of a man possessed of a soul of
good character. Professor Gignoux r e t u r n e d to the fray 1 2 w i t h pl'whlmzdy ZY
hwhym Iwb'n HZYTN LY Farr-Ohrmazd, l'me bienveillante, regarde moi !
H e was right to correct m y orthographic imprecision in failing to record t h a t
the second character was written w i t h t h e usual -I- for -r- ; y e t on the general
issue, if m y view is correct, i t was n o t t h e purpose of these inscriptions to call
a t t e n t i o n to the owners of the stones ; r a t h e r they were an ethical admonition
addressed to t h a t person. Consequently, Iwb'n HZYTN will be, as in other
examples, a closely linked phrase, a n d t h e general interpretation has again to be
revised. My first reading, in which I understood the pronoun LY (man) as t h e
second word of the sentence, was p a r t l y influenced by Professor Boyce's teach-
ing, t h a t the pronoun normally comes second in the Middle Persian sentence.
On t h e other hand, there is a gap in t h e legend before LY, of which the opening
character is larger t h a n others in the inscription, suggesting t h a t reading should
commence a t this point.
Recently I have seen how this p a r a d o x can be resolved. If we assume t h a t
t h e word mwdly (muhr) seal is understood, b u t not expressed, as the first word
of t h e sentence, then we can interpret t h e whole as (mwdly) LY, pl'whlmzdy ZY
hwhym: Iwb'n HZYTN, a text of perfectly conventional p a t t e r n . However,
problems m a y not y e t be a t an end. Are we to understand hwhym benevolent
as a proper name, or just as an epithet of religious tone, similar in fact to hwmtry,
of which Gignoux has shown t h a t it has, a t least in some connections, merely
an a d j e c t i v a l force ? Similar hesitation arises over another piece, Frye, Foroughi
Collection, LI, 178, which reads, yzdt-wlhl'n hwkly, which might be either an
invocation of the deity (Yazad Vahram hukar), or a personal n a m e with patro-
nymic. Of this group of epithets, hwpndy from the British Museum stone CT) 1
seems definitely a proper name.
10
Catalogue of the Western Asiatic Seals in the British Museum. Stamp seals.
II : The Sassanian dynasty. London 1909. 55.
11
SHARED : 30.
12
PH. GIONOUX : Cachots sasanides du British Museum, in Acta Iranica, Textes
et Mmoires, V, Varia : :h. 133.
Acta Antiqua Academiae Scieruiarum rira e 28, 1980

I t u r n now to w h a t seemed to me in my innocence of 15 years ago a v e r y

simple text, the British Museum stone AA2, 13 which reads mnctry ZY 'rtMtr-
Snwmy, and which I took to mean Manuchihr who (is) Ardashir's Joy t h e
latter not unusual t y p e of Sasanian honorific a t t e s t e d also in the Shpr KZ
inscription. Discussion with colleagues soon evoked t h e view t h a t this was f a r
from certain, and in his survey of my readings, to which we owe so m a n y valu-
able ameliorations, 1 4 Gignoux denied t h a t mnctry could spell the name in epi-
graphic Pahlavi of which the Avestan form is Manusira. For this seal he ac-
cordingly proposed an avowedly vague interpretation De la descendance de
Ardaxsr-n5m. I h a d been recalling, however, the well-known coins of the pre-
Sasanian princes of Fars, 1 5 a familiar variety of which, attested in London b y a
fortunately well-preserved specimen, (Pl. I) reads clearly mnctry MLK' BRH
mnctry MLK K i n g Manuchihr who (is) son of K i n g Manuchihr. There can b e
no d o u b t a t all t h a t t h e repeated word must represent t h e proper name of f a t h e r
and son. The etymological problems arising from these d a t a I have no choice
b u t to leave to expert etymologists.
The points I have made so far constitute merely a sifting of previously-
published discussions, b u t it is only right I should wind u p m y article with some
new material. This includes two readings which t h r o w a chink of light on t h e
history of the Sasanian fire-temples. Actual seals employed by the fire-temples
are rare, and I can quote o f f h a n d only those excavated a t dr-Gushnasp, a n d
perhaps the P a r i s stone 3.4, on which I commented in a review. 18 Yet d e s p i t e
dissenting views, there is a considerable case t h a t t h e names of fires, and various
compounds of these, occur frequently as the names of individuals. If t h a t view
were well founded we could hope to learn from these attestations of fire-temples
whose existence was otherwise unrecorded. Recently I was sent by an owner
living in a small English village a good enlarged photograph of a 5th c e n t u r y
A.D. Sasanian p o r t r a i t seal with a semi-cursive inscription interesting in this
regard (Pl. II). I am grateful to Mrs. L. Beck for permission to publish this spe-
cimen.
'twr'rtMtr Y 'twrplh'ndt
dr-Ardashir drfarroxn-dd
I t is t e m p t i n g to see in t h e names appearing on t h e specimen allusions t o

the names of Sasanian fire-temples. Gignoux has recently re-opened the discus-
13
Catalogue of the Western Asiatic Seals in the British Museum. Stamp seals. I I .
The Sassanian dynasty. 44.
"GIGNOUX: Cachets sassanides. 135.
15
Catalogue of Greek coins in the British Museum ; Arabia, Mesopotamia, Persia,
etc., b y G. F . HILL. London 1922.
18
JRAS, 1980. 94, anont PH. GIGNOUX: Catalogue des sceaux, camees et bulles
sasanides de la Bibliothque Nationale et du Muse du Louvre : I I . Les sceaux et les bulles
inscrits. Paris 1978. 25.

210 A. D. H. B I V A R
sion of Sasanian personal names of this t y p e , adopting a r a t h e r formalized

criterion to determine which examples indeed allude to t h e names of temples,
a n d which are merely theophoric compounds incorporating t h e element dr- ;
t h o u g h in fact cult-institution and personal deity may not have been rigorously
d i f f e r e n t i a t e d by t h e Zoroastrian mind. On the one hand he concedes :17
Certes, dans des noms comme d r - F a r r b a y et dr-Gusnasp si com-

munment attests, on ne peut rien voir d ' a u t r e que les noms des deux
grands feux devenues noms de personnes . . .
Elsewhere, he shows a tendency to prefer wherever possible the interpretation

of such compounds as des noms . . . constitus par deux noms de dieux. How-
ever, one of the examples he quotes in this connection, dr-Anahid, being ob-
viously the name of t h e celebrated fire-temple of Istakhr, might here inspire
in u s a note of caution. Surely the n a m e of t h e provincial fire-temple of F a r s
would be more in the mind of the Zoroastrian naming his offspring t h a n would
a r o u t i n e juxtaposition of the names of t h e two deities. When t h e names of so
m a n y other unattested fire-temples of I r a n are unknown to us, it is i m p o r t a n t
t o b e w a r e of undue dogmatism. There could be more allusions to particular fire-
t e m p l e s than is at first obvious, and a critical judgment, t a k i n g note of the
r a p i d l y expanding evidence, is required in every case.
I n the case of our example above, dr-Ardashir can hardly be regarded
as o t h e r than the n a m e of the dynastic founder's fire. This probability must
n a t u r a l l y affect our understanding of t h e second onomastic element on the seal.
T h e son having been n a m e d from a sacred fire, so too could have been the father,
a n element of parallelism within the f a m i l y being naturally to be expected, if
n o t invariable. E v i d e n t l y dr-farroxn-dd could be analysed in several ways.
MacKenzie 1 8 cites F a r r o b a g (plwbg) besides, presumably, plnbg as a name
of t h e priestly fire of Fars. Gignoux reads 1 9 plhwbg'tw(r) on a bulla from Qasr-i
A b N a s r , which seems to be a variant rendering of the same n a m e . However,
he does not list this reading in his t u d e typologique, 20 perhaps because it
seemed antagonistic to his general theory. These forms are somewhat parallel
t o d r f a r r o x n , a compound not a t t e s t e d hitherto. One could conceive of a fire
t e m p l e designated dr-Farroxn-dd Fire founded (or created) b y a person
n a m e d Farroxn, n a m e of a post-Sasanian King of Tabaristn a t a date of
course considerably later t h a n t h a t of our example. On the other hand, it is
17
PH. GIGNOUX : Les noms propres en moyen-perse epigraphique : tude typo-
logique, in Pad, nam i Yazdan (Universit de la Sorbomie Nouvelle, T r a v a u x de l'Institut
d ' E t u d e s Iraniennes, 9). 76.
18
Concise Pahlavi dictionary. 32.
18
PH. GIGNOUX : Los bulles sasanides de Qasr-i Abu Nasr in PH. GIGNOUX and
A . TAFAZZOLI (ed.), Memorial Jean de Menasce. Louvain 1974. 179, No. 214.
20
GIGNOUX : Les n o m s propres . . .

fairly obvious t h a t t h e element -dd in Sasanian proper names, as in many other

religiously inspired names of the ancient world, refers the child's conception t o
t h e answering of prayer directed to particular deity, or performed in a partic-
ular temple. On this fairly evident hypothesis, drfarroxn-dd might well
be t h e n a m e of child whose conception was a t t r i b u t e d to prayers performed a t
a temple named dr-Farroxn. And so I t a k e it to be in this particular case.
F a r r o x , or Farrox-bag could be understood as t h e designation of a deity, per-
h a p s identical with F a r r b a y . With regard to the n a m e of the Tabaristn ruler
F a r r o x n , it could well be t h a t he was named from a celebrated fire-temple,
t h e dr-Farroxn of our inscription, a n a m e presumably well known locally,
and thus shortened merely to F a r r o x n . If so, a n d the temple the same as on
our seal, it m a y well have been situated in Tabaristn, a theory which the style
of t h e engraving m a y do something to support.
My second example which again seems to m a k e allusion to a fire-temple
is from Frye's Foroughi album, 2 1 being the striking impression on Pl.L, 171. ( = P 1 .
I l l here). The inscription, not to my knowledge read previously, appears to r u n :
'thfpzrdrp Y l's(tywMmzdy: 'p\s'tn 'L y]
I t seems quite logical to resolve the first of these names as Ataxs-bzar-

d a r b ; taking bzar (a term still current in modern Persian, as Mr N. Chegini
reminded me) as t h e designation of the famous mercury gilding typical of t h e
Sasanian silverware. Darb, a by-form of dar door also current in modern P e r -
sian (the relation of which to its Arabic homonym, perhaps a loan from M P ,
must remain unresolved) is readily understood as door, giving the overall
meaning Fire-temple of the gilded door. This concept is perfectly probable.
A t t h e present day I know one religious building of I r a n with gilded doors,
t h e Madrasa a t I s f a h a n . I n the Sasanian heyday of silver-gilt, such decoration
would have been typical. I n this case a t least, there seems no difficulty in con-
ceiving the owner of t h e seal as having been named from a fire-temple, and no
possibility here of a theophoric re-interpretation of the name. The second n a m e
presents a difficulty. Of course, l's is a P a h l a v i spelling of Persian rh road,
yet though possible, road of Ohrmizd is not a very convincing personal n a m e .
On t h e whole it seems best to assume the omission of a character in the engrav-
ing, a n d emend to Rst-Ohrmizd The righteous (deed) of Ohrmizd whether
deity or ruler of t h a t n a m e be here in question. And rast, of course, was the t e r m
with which I began this paper.
London.
21
R . N. FRYE : Sasanian seals in the collection of Mohsen Foroughi. (Corpus I n -
scriptionum Iranioarum P t . I l l , Vol. vi, Portfolio II). (A reading based on a different
emendation is now proposed by GIGNOUX and GYSELEN: Sceaux sasanides de diverses col-
lections prives, 142, b u t shows no obvious superiority.)

. CZEGLDY
ZUR GESCHICHTE DER HEPHTHALITEN
Die Erforschung der Geschichte der Hephthaliten h a t whrend der zwei

letzten J a h r z e h n t e zu wichtigen Teilergebnissen gefhrt, die heutzutage die
Revision einiger Grundfragen der hephthalitischen Geschichte ermglichen.
Ich mchte diesmal von den letzteren, zwei miteinander zusammenhngende
Probleme ins Auge fassen : einerseits die F r a g e der N a m e n und der Stammes-
komponente der Hephthaliten, andererseits die Geographie der hephthaliti-
schen Einwanderung nach Ost-Tocharistan. Obwohl ich beide Problemen-
komplexe bereits vor 25 J a h r e n in einem ungarisch herausgegebenen Aufsatz 1
behandelt habe, ist es nicht meine Absicht meine frhere Arbeit hier zusam-
menzufassen. Vielmehr mchte ich Rechenschaft geben von meinen neueren
Untersuchungen, die ich auf Grund von neuen Quellenstudien, einschlielich
der chinesischen Berichte angestellt habe.
Die Beurteilung der ltesten Periode der hephthalitischen Geschichte ist
eng mit der Frage ihrer verschiedenen Benennungen verbunden. Der N a m e
heftal scheint weder altaischen, noch iranischen Ursprungs zu sein : diese ety-
mologische Frage ist aber vom Gesichtspunkte unserer Problemstellung weniger
aufschlureich Aus den Quellen geht nmlich eindeutig hervor, da heftal
ursprnglich nur N a m e einer zeitgebundenen Dynastie u m ungefhr 460 560
unserer Zeitrechnung war und erst nach 460 auf das Volk der Hephthaliten
bertragen wurde. Der A u f t r i t t der hephthalitischen D y n a s t i e ist also ein ter-
minus post quem f r die Authentizitt der Erwhnungen der Hephthaliten.
Mit anderen Worten : diejenigen Quellen, die die H e p h t h a l i t e n vor 460 erwh-
nen, antizipieren diesen Namen, was unter Umstnden auch f r die Beurteilung
ihrer Quellenwertes von Belang sein kann. Noch wichtiger ist aber, da wir f r
die Zeitalter der Erscheinung der Hephthaliten und fr ihre frhere Geschichte,
a n s t a t t des Namens heftal andere Benennungen zu erwarten haben. U n d das
ist in der T a t der Fall : die Hephthaliten erscheinen in den chinesischen Quellen
vor 460 u. Z. unter dem Namen Hua. In den Quellen des Mittel- und N a h e n
Ostens werden sie als hyaona, xyn, oder %on, beziehunsweise hon-V und huna
1
H e f t a l i t k , hunok, avarok, onogurok. Magyar Nyelv [Ungarische Sprache] 50
(1954) 1 4 2 - 1 5 1 .
Acta Antiqua Academiae 'cientiarum Hunguricae 28, 1980

214 . CZEGLDY
e r w h n t , whrend sie bei den Byzantinern Hunnen, oder Weie Hunnen

g e n a n n t werden.
Was nun das chinesische hua b e t r i f f t , sind die besten Kenner der chine-
sischen Lautgeschichte, u n t e r ihnen P . Pelliot, 2 B. Karlgren, 3 L. Ligeti 4 u n d
neuerdings auch E . G. Pulleyblank 5 der Meinung, da das chinesische Graphem
Hua im Mittelchinesischen als *ywt ausgesprochen wurde. Seine L a u t f o r m
entwickelt sich aber in derjenigen chinesischen Mundart, in der die meisten
Sprachreste der nrdlichen B a r b a r e n Innerasiens im V I I I X . J h . bermittelt
worden sind, ins iuad, oder hua, eventuell auch ' u a d , oder 'ua. Das y- ver-
s t u m m t e also in in dieser M u n d a r t stufenweise. In der F o r m mit modifiziertem
i gibt letzteres ein s t i m m h a f t e s h zurck. Dieser alte L a u t wurde zuerst von
P . Pelliot angenommen. Pulleyblank k a m neuerdings zum gleichen R e s u l t a t .
D a s stimmlose h h a t Ligeti auf Grund von sino-tibetischen Entsprechungen
vorgeschl agen.
I n meinem ungarischen Aufsatz ging ich von diesen wichtigen linguisti-
schen Erkenntnissen aus, u n d n a h m an, d a wir den alten N a m e n der H e p h t h a -
liten mit groer Wahrscheinlichkeit in *uar zu sehen haben, umsomehr, da auch
sonst mittelchinesisches fremdsprachiges r vertritt. Einer ganz anderen Mei-
n u n g ist dagegen K a z u o Enoki, 6 der in einem ausgezeichnet dokumentierten
A u f s a t z neue Lsungen der hier errterten Fragen vorlegt. E r zweifelt darin
a u c h meine in 1954 verffentlichten Ausfhrungen an. Enoki, ein hervorragen-
der Kenner der chinesischen Quellen betrachtet die letzteren nach den Prinzi-
pien der modernen Quellenkritik und demonstriert auch in diesem Falle ber-
zeugend, da wir ohne einen kritischen Vergleich der chinesischen Kompilatio-
n e n , welche denselben Bericht in verschiedener Fassung darbieten, k a u m zu
einem befriedigenden R e s u l t a t gelangen knnen. Seine bersetzungen und
K o m m e n t a r e enthalten viel wertvolles Material zur Frage. Auch seine Ansich-
t e n auf dem Gebiete der chinesischen Lautgeschichte t r a c h t e t er sachgem zu
fundieren. So verlt er sich im Falle des Namens Hua auf die kompetente
Rekonstruktion B. Karlgrens, indem er die zeitgenssische Lautform dieses
N a m e n s als ywt rekonstruiert, das er auf eine mit Asteriskus versehene Form
*Ghwr zurckfhrt. *Ghwr, seiner Meinung nach, ist nichts anderes als der
a l t e Name der nordwestafganistanischen Provinz, der spter, in den islamischen
Quellen o f t als r belegt ist. Mit diesen Errterungen v e r k n p f t schlielich
2
T'oung P a o 16 (1915) 9, cf. J o u r n a l Asiatique 1921/1, 139 sqq.
3
B. KARLGREN: Grammata Serica Recensa. 133134, No 486/E ; 9192,
N o 302/m.
4
I n einein V o r t r a g , g e h a l t e n 13 O k t . 1945 an der Ungarischen Akademie der
Wissenschaften.
5
Asia Major 9 (1962) 258.
' KAZUO ENOKI : On t h e N a t i o n a l i t y of the E p h t h a l i t e s . Memoirs of the Toyo
B u n k o 18 (1959) 1 - 5 8 .

ZUR G E S C H I C H T E D E R HEPHTHALITEN 215
Enoki seinen wichtigsten neuen Deutungsversuch, indem er die H e p h t h a l i t e n

f r ein in Ostiran autochtones Volkselement erklrt.
E s fllt allerdings sofort a u f , d a Enoki in diesem Falle gar nicht sagt,
in welcher Sprache die Entwicklung *Ohwr > r s t a t t g e f u n d e n h a t . Diesbe-
zgliche Angaben fehlen in unseren lteren Quellen und auch den Muslimen war
die Sprache vom unzugnglichen r, wie auch sein heidnisches Volk im allge-
meinen, f a s t gnzlich unbekannt. E n o k i nimmt eine m i t Asteriskus versehene
ursprngliche Form von r als *Ohwr offenbar deshalb an, weil sie dem mit-
telchinesischen Lautwert (*ywt) von Hua am nchsten stehen wrde.
Weitere Schwierigkeiten h u f e n sich, wenn wir die geographischen Err-
terungen Prof. Enoki's nher ins Auge fassen. Die wichtigste Beschreibung des
Landes Hua finden wir wie schon lngst bekannt in dem 54. K a p . des
Liang-u aus der ersten H l f t e des 6. Jahrhunderts. Hier finden wir eine Reihe
von wichtigen Angaben auch bezglich der Nachbarlnder von Hua. Der W e r t
der Berichterstattung des Liang-su wird aber dadurch beeintrchtigt, da die
Verfasser dieses Jahrbuches, H o f b e a m t e r der sdlichen Liang-Dynastie, mit
der Lage der einzelnen Lnder des fernen Nordwestens, im Verhltnis zu einan-
der, nicht ganz im Klaren waren. So unterlief ihnen ein Fehler im Falle des
Landes Hua und der Provinz Balch, indem sie behapteten, da sich das L a n d
I l u a auf secsh Tagesmrsche westlich von der Provinz befinde, whrend weniger
spter, im selben Kapitel b e h a u p t e t wird, da der westliche Nachbar der
Provinz Balch Po-s, das heit Persien, wre. Es ist jedoch glcklicherweise gar
nicht schwierig festzustellen, welche von den zwei einander ausschlieenden
Berichten zu akzeptieren sei. Alle historischen und geographischen Quellen
nmlich, welche die Geschichte des persisch-hephthalitischen Krieges behan-
deln, lassen keinen Zweifel d a r b e r bestehen, da das eigentliche Zentrum des
Hephthalitenlandes sich immer stlich und nicht westlich von der Provinz
Balch befand. Deshalb erscheint es einigermaen befremdend, da E n o k i die
andere Angabe (Hua westlich von Balch) fr richtig erklrt, indem er Hua,
als soeben gesagt, mit der Provinz Gr identifiziert, obwohl letztere sich sd-
sdwestlich und auf einer groen E n t f e r n u n g von Balch befand.
Umsomehr entscheidend fllt in die Waagschale, d a sich weitere wich-
tige Angaben bezglich des Landes H u a , d. h. *Uar in den chinesischen Quellen
einerseits und muslimischen historischen und geographischen Quellen anderer-
seits befinden, welche die Anwesenheit des Volkes Hua-*Uar stlich von der
Provinz Balch, im eigentlichen Tocharistan bezeugen. U m hier nur die zwei
wichtigsten zu nennen : H : a n Tsang, der berhmte chinesische Pilger, durch-
kreuzte whrend seiner Reise nach Indien um 630 auch Tocharistan. E r ver-
weilte damals auch in der hephthalitischen H a u p t s t a d t . Letztere befand sich,
gem seiner Aufzeichnungen, sdlich vom Oxus, in der Gegend des jetzigen
K u n d u z . Der N a m e der H a u p t s t a d t lautet bei Hian Tsang Huo f g , also wieder
einmal ein chinesisches Wort, dessen mittelchinesische F o r m ywt lautet, also
Acta Antiqua Academiae Scientiarum Hunquricue 28, 1980

216 . CZEGLDY
identisch mit der des chinesischen Volksnamens Hua ist. Die natrliche Annah-
me, da wir auch in diesem Fall mit dem Hephthalitennamen *Uar, als Name
der hephthalitischen H a u p t s t a d t zu t u n haben, wird erhrtet durch Birni's
Angabe, wo gesagt ist, da die H a u p t s t a d t der Hephthaliten in alten Zeiten
den Namen War-waliz trug. Mit Rcksicht auf Hian Tsang's War erscheint
n u n sehr wahrscheinlich, da es sich hier um einen zusammengesetzten Namen
handelt, dessen erste H l f t e den Hephthalitennamen *Uar enthlt, whrend
die zweite wie es lange her vermutet wurde ein Wort m i t der Bedeutung
'Burg, Stadt' sein kann. 7 Wir sollen dabei die Tatsache nicht vergessen, da
*Uar sich im eigentlichen, stlichen Tocharistan befand, also auf altem irani-
schem Boden, wo neben der Sprache der Eroberer auch noch das mittelbaktri-
sche Idiom gesprochen wurde. Ich bin daher zur berzeugung gekommen, da
wir genug Grund dazu besitzen, zwischen diesem wallz und dem mittelbaktri-
schen malizo einen Zusammenhang zu suchen, und diese Annahme scheint
auch vom linguistischem Stndpunkt aus plausibel zu sein. 8
Was nun die oben genannten anderen Bezeichnungen der Hephthaliten
b e t r i f f t , die mit dem Namen der Hunnen zusammenhngen (yyn, y un, hun),
ich habe mich in meiner frheren Arbeit mit den sprachlichen Eigentmlich-
keiten dieser Namensformen eingehend beschftigt. J e t z t bemerke ich nur, da
wir bezglich des Endnasals in der chinesischen Form hiung- (im Namen hiung-
nu, archaisch-chinesisch yiwong-nuo) besser unterrichtet sind als frher.
J . Hamilton und E . G . Pulleyblank haben nmlich schwerwiegende Argumente
vorgetragen fr die Beobachtung, da der chinesische Nasal in manchen Trans-
kriptionen fremdsprachiger Streudenkmler den Lautwert -n, anstatt des ge-
whnlichen y besitzt. Das bedeutet, da wir die archaische Form von hiung-nu
problemlos als yiun-nuo rekonstruieren knnen, welch letzteres mit dem mittel-
persischen yyn bis auf den im mittelpersischen schon nicht mehr bezeugten,
7
Prof. HARMATTA, d e m ich diesen etymologischen G e d a n k e n v o r t r u g , h a t t e die
G t e d a z u zu b e m e r k e n , d a so eine E r k l r u n g o h n e weiteres a k z e p t a b e l ist, aber u n t e r
d e r Bedingung, d a w i r v o m E l e m e n t lizo d e s b a k t r i s c h e n W o r t e s ausgehen.
8
D a s Z u s a m m e n l e b e n der H e p h t h a l i t e n u n d der Qarluqen in Ost-Tocharistan
g e h t besonders d e u t l i c h a u s einer von ENOKI n i c h t a n g e f h r t e n Stelle des T'ang-hui-yao
(ed. Shanghai 1935. 1315), a u s d e m J a h r e 649. 10. Monat 3. Tag) hervor, wo von der
a d m i n i s t r a t i v e n O r g a n i s a t i o n der W e s t l n d e r die R e d e ist, welche die Chinesen n a c h
i h r e n berwltigenden Siegen ber die T r k e n p l a n t e n . Hier wird a u c h eine P r f e k t u r
(), n a m e n s Ko-lo ( a b g e k r z t von Ko-lo-lu ~ Qarluq) a u f g e z h l t fr zwei H o r d e n ,
die Q a r l u q und die H e p h t h a l i t e n eingerichtet. W i c h t i g in diesem B e r i c h t e ist, da hier
eine Verwechslung, wie i m Falle des arabischen G r a p h e m s h.l.h (mgliche Lesungen :
halluh = qarluq oder halag = qalac) ausgeschlossen ist.
Die Yabgus (bzw. Sade) von T o c h a r i s t a n u n d von den ihm angrenzenden Gebieten
g e h r t e n zum w e s t t r k i s c h e n Zweig der h e r r s c h e n d e n A-schi-na D y n a s t i e . Einige von
i h n e n trugen den N a m e n P'u-lo oder Ko-lo P'u-lo. I m Brczi Gza E m l k k n y v [Fest-
s c h r i f t f r Gza Brczi. B u d a p e s t 1963. 60] h a b e ich versucht zu beweisen, da die mittel-
chinesische F o r m dieser N a m e n : b'uok-l, k-l b'uok-l eine g e n a u e Wiedergabe des
t r k i s c h e n Buyra, Qara Buyra darstellt. Glechzeitig habe ich die B e d e u t u n g dieser
A n g a b e n f r die F r a g e des Ursprungs der B u g r a oder Qara-Chanidischen Dynastie
ge wertet.
Acta Antiqua Aeademiae ScierUiarum Hungaricae 28, 1980

ZUR G E S C H I C H T E D E R H E P H T H A L I T E N 217
also wohl auch schon im Original entfallenen Endung -uo identisch ist. Die
Hephthaliten besassen also, wie auch die Awaren, neben den Uar-Stmmen,
auch starke hunnische Stammeskomponente. Der gemeinsame Ursprung der
verschiedenen Namensformen der Hunnen (hiung-nu, yyn, y un, hun) t r i t t im
Falle der hephthalitischen Hunnen am deutlichsten hervor.
Budapest.

225 M. GRIGNASCHI
LA CHUTE DE L'EMPIRE HEPHTHALITE DANS

LES SOURCES BYZANTINES ET PERSES ET
LE PROBLME DES AVAR
Les sources perses sur la c h u t e d e l ' E m p i r e H e p h t h a l i t e o n t t t u d i e s

en d t a i l p a r M. W i d e n g r e n 1 et n o u s r e p r e n o n s a u j o u r d ' h u i cet a r g u m e n t p a r c e
q u e nous c r o y o n s que, si l'on t i e n t c o m p t e des r a p p o r t s e n t r e les t e x t e s a u x -
quels le s a v a n t sudois a e m p r u n t les d i f f r e n t e s riwyah se r a p p o r t a n t
cet v n e m e n t , il est possible d e se f o r m e r u n e ide p l u s claire d u rcit origi-
naire d u R o d y n a m a h sassanide. T o u t c o m m e M. W i d e n g r e n , n o u s n e n o u s
a r r t e r o n s p a s d a n s c e t t e c o m m u n i c a t i o n sur le p r o b l m e de savoir s'il a e x i s t
un seul R o d y n m a h sassanide ou si, d j l ' p o q u e perse, il y a v a i t p l u s i e u r s
versions d e ce livre. 2 Nous dirons s e u l e m e n t q u e si l'on prfre croire q u ' i l a
exist u n seul IJodynmah, ainsi q u e nous s o m m e s enclin le p e n s e r , il
f a u t alors a d m e t t r e que d j ce livre perse c o n t e n a i t p a r f o i s des versions d i f f -
rentes d u m m e v n e m e n t l ' i n s t a r d e ce q u e n o u s r e n c o n t r o n s si s o u v e n t
d a n s les o u v r a g e s arabes. 3 Il t a i t d ' a i l l e u r s ncessaire de r e p r e n d r e le p r o -
b l m e des sources perses sur la c h u t e d e s H e p h t h a l i t e s parce que t r e n t e a n s
d j se s o n t couls depuis la p a r u t i o n d e l'article d e M. Widengren e t q u e les
progrs d a n s la connaissance d e la n u m i s m a t i q u e des E t a t s f o n d s p a r les
n o m a d e s d a n s l ' I r a n Oriental et l ' A f g h a n i s t a n , ceux q u e M. Gobi a a p p e l les
Huns iraniens 4 e t notre d c o u v e r t e des f r a g m e n t s d ' u n K r n m a g - i - A n s i r -
vvn r e m o n t a n t l ' p o q u e sassanide d a n s l'histoire d ' I b n M i s k a w a y h 5 p e r -
m e t t e n t a u j o u r d ' h u i de s o u m e t t r e la version pehlvie u n e critique p l u s serre.
Q u a n t a u x sources grecques, analyses t a n t de fois p a r les s a v a n t s , d e r n i r e -
m e n t en p a r t i c u l i e r p a r M. H a u s s i g e t M. Czegldy, 6 elles m r i t e n t elles-aussi
1
Orientalia Suecana I. p. 69 95.
2
W I D E N G R E N : 1. c . p . 8 5 .
3
Ainsi par ex. Eutychius raconte tout d'abord que Bahrm s'tait fait confier
par la noblosso porse le mandat de dsigner les successeur de Yazdagird I et que, fort
de ce mandat, il se proclama lui-mme Shenh. Ensuite, la fin du chapitre, Eutychius
rsume en quelques mots la riwyah l ien connue qui racontait que Bahrm Gor a v a i t
vcu la cour d'al-Nu'mn b. Munjjir et que ce fut avec l'appui des forces arabes de l a
principaut do R i r a qu'il obtint le trne.
* R . GBL : Die iranischen Hunnon. Wiesbaden 1965.
5
J . A. 1966 p. 1 6 - 4 5 .
8
H . W. HAUSSIG : Theophylakts' E x k u r s ber die skythischen Vlker. Byzan-
tion 23 (1953), Die Quellen ber die zontralasiatische Herkunft der europischen Awaren.
CAJ 2 (1956) et Zur Lsung der Awaronfrage. Byzantinoslavica t. X X X I V 1973 fasc. 2.
Acta Antiqua Acudemiae Scientiarum Hungrii 28, 1980

220 . GRIGNASCHI
d ' t r e reprises en e x a m e n car, une fois tablie avec certitude la d a t e de l'effond-

r e m e n t de l'Empire Hephthalite, elles apparaissent sous une lumire nouvelle.
Ce f u t Stein qui, le premier, m o n t r a comment les sources grecques nous
livrent un terminus a n t e quem pour la f i n de l'Empire H e p h t h a l i t e : 561.
Cependant Stein envisageait la possibilit qu'Anirwn se f t tourn contre
son voisin de l ' E s t immdiatement aprs la trve stipule avec l'Empire d'Ori-
e n t en 557'. De son ct dj Specht a v a i t rappel l'attention sur une indica-
tion de L'histoire des Tcheou qui, son avis, nous donnait le terminus post
quem de l'effondrement de l ' E t a t H e p h t h a l i t e : en 558 une dernire ambas-
sade des J - t h a s'tait rendue la cour des Tcheou. 8 E n dpit de cette indica-
tion, on a parfois prfr admettre qu'en fait, au moment o ces ambassadeurs
arrivrent la cour des Tcheou, l ' E t a t hephthalite n'existait plus et qu'il
s ' t a i t croul sous les coups des armes t u r q u e s et perses dj a la fin de 558."
A u j o u r d ' h u i le Krnmag-i- Ansirwn nous apprend qu'au cours de l'anne
28 de son rgne, qui commena le 5-VII-559, Ansirwn renona conduire
personnellement une grande enqute sur les conditions des paysans cause de
la guerre qui tait en cours contre les ennemis de l'Iran et de la ncessit de se
rendre sur le thtre des oprations. 10 E n t e n a n t compte de la dateen 559 la
trve entre Ctsiphon et Constantinople continuait et, d a n s les premires
annes qui suivirent le partage de l ' E m p i r e Hephthalite, les relations entre
Ansirwn et Istemi taient encore amicalesil ne peut tre question que des
hostilits sur les frontires nord-est de l ' I r a n . Donc la chute de l'Empire Heph-
thalite d a t e soit de la f i n de 559, soit de 560. Ajoutons que les fragments de
Mnandre p e r m e t t e n t de prciser deux a u t r e s points importants : l ' a t t a q u e
principal contre les Hephthalites f u t dirig p a r le Qagan t u r c N.Ciov/.og11,
le Silgib ou Singib des sources perses e t islamiques, 12 sans doute un titre
dcern p a r l a Cour Chinoise Istemi Qagan, le Ereptaxayav de Thophylacte,
le Exepe yayv de Nicphore Xanthopulos 1 3 . Le Roi hephthalite tait rest
K . CZEGLDY : H e f t a l i t k , hunok, avarok, onogurok, MNy 50 (1954) = I V I X szzadi

npmozgalmak a steppen (par la suite abrg en Npmozgalmak) ; N o m d npek vn-
dorlsa Napkelettl N a p n y u g a t i g . Budapest 1969 (par la suite abrg en Nomd npek.
N e connaissant p a s le hongrois, nous nous basons sur l'analyse de ces deux t r a v a u x d u
savant magyar que M. A . MOHAY (Priskos* F r a g m e n t ber die Wanderung der Steppen-
vlker) a donne d a n s les Studies in the Sources on the History of pre-Islamic Central
Asia (Budapest 1979) et n o u s en reproduisons les indications.
7
Mnandre P r o t e c t o r De leg. R o m a n o r u m frag. 11. STEIN: Histoire du Bas E m p i r e
p . 518 . 1 Cependant STEIN pensait que la campagne d'Anirwn contre les Hephthali-
tes a v a i t eu lieu i m m d i a t e m e n t aprs la trve signe avec Byzance en 557.
8
J . A. Octobre 1883 p . 345.
9
Npmozgalmak p . 11 (MOHAY: 1. e. p . 133).
10
Episode V I I 1. c. p . 20.
11
Excerpta de Sententiis d. Mai p. 354.
12
MARKWART : Historische Glossen zu den alttrkischen Inschriften, W Z K M X I I
p. 184185. Voir aussi HAUSSIG: Byz p. 332 et suiv. Cependant d a n s son article Die
H e r k u n f t der Awarischen Hunnen, o il propose p o u r Sif.siovo; la reconstruction
Silig-gebu, Haussig semble tablir une distinction entre ce yabgu et Istemi.
13
Histria Ecoles. X V I I I 30. MORAVCSIK : Byz.-Turc. I I p. 291.

LA C H U T E D E L ' E M P I R E H E P H T H A L I T E 221
sur la dfensive en suivant le conseil frauduleux d ' u n grand personnage de sa

cour KaxovXqio ,u qu'il avait offens et qui ensuite passa l'Iran. De plus,
Mnandre nous apprend que ZXfovXo reut la nouvelle de la fuite des
"Aagoi, . -d. les OVO et Xovvvi, les pseudo-Avares de Thophylacte, les OVCLQ-
ycovxat de Tovoavi alors que la guerre contre les Hephthalites tait encore
en cours. 15 Or, la date de l'arrive des ambassadeurs des pseudo-Avar Con-
stantinople est bien connue. Leurs ambassadeurs arrivrent en 558 et le gros
de la horde se trouvait cette anne-l dans le pays des Alains et, en croire
Mnandre, elle ne se dplaait pas dans une grande hte. 1 6 Il faudra en conclure
que la premire expdition desT'ou-kiue contre les Hephthalites fut dirige en
555 par Sse-kin, le fils de T'ou-men connu sous le n o m de Mou-han K h a n ,
ainsi que l'affirment les sources chinoises, 17 qu'une seconde campagne f u t con-
duite parEiXtfiovXo^ (Istemi) dj a v a n t 558, l'anne de l'arrive des pseudo-
Avar dans l'Alanie, donc en 556 ou 557 au plus tard, que ce f u t ce moment-l
qu'Istemi apprit la nouvelle de leur fuite et que probablement dj en 556
Istemi tait matre de la Sogdiane car, d'aprs a t - T a b a r le mariage entre
Ansirwn et la fille d'Istemi avait prcd d'un an l'ouverture des oprations
militaires contre le R o y a u m e Hephthalite ; en t e n a n t compte des renseigne-
ments des sources chinoises on comprendra que ce mariage, ou le contrat de
ce mariage 1 8 avait prcd d'un an la seconde campagne turque. 1 8
La chronologie de ces vnements revt plus d'importance encore pour
le problme si obscur des Avar et des pseudo-Avar que pour l'histoire des
Hephthalites eux-mmes. E n effet elle suffit exclure la possibilit que l'mi-
gration des pseudo-Avar en Europe ait t la consquence directe de la chute
de l ' E m p i r e Hephthalite. 2 0 Aussi, quand Thophylacte affirme que les Avar
arrivs en Europe en 558 taient en ralit des tribus U a r et Xunni e t que
Mnandre cite le qagan Tourksanthos qui les appelle Uarxnitai, nous n'avons
plus le droit d'interprter ses paroles dans le sens qu'il s'agissait d'une confd-
ration de tribus War (les Hoa de l'histoire des Liang) e t de tribus Chionites
(les Hyon des textes pahlvs provenant du Tohristn, 2 1 voire des steppes
14
De lgat. Gentium frag. 18 (Ed de Boor p. 450).
15
Exoerpta de Sententiis d. Mai p. 354.
18
Mnandre Exc. de leg. gentium frag. 4 E d . DE BOOR p. 442. Voir aussi PIGU-
LEVSKAJA : Sirijskije Istoniki p. 139 (Joh. Ephes.).
17
ST. JULIEN : Documents sur les Tou-kiue Orientaux. J . A. V. srie vol 3 P. 335.
18
D'aprs Eutychius, le Q a g a n envoya sa fille Anirwn seulement aprs la
victoire sur les Hephthalites.
18
At-Tabar 1,2 p. 899.
10
P a r contre HAUSSIG (Byz. 1. c. p. 370) parle d ' u n kausaler Zusammenhang
entre l'effondrement des Hephthalites et la migration des A v a r . De son ct MAROUART :
p. 73 rattachait l'arrive des Avar en Europe la dfaite des Jeou-jan. Nous
prfrons songer la chevauche de Mou-han travers la Sibrie en 555.
21
K . CZEGLDY : N p m o z g a l m a k p . 2 4 (MOHAY : 1. c . p . 132). R a p p e l o n s q u e ,
pour expliquer la diffrence entre la description dos Hephthalites par Maniach, qui les
dfinissaient des h a b i t a n t s des villes, et les renseignements dos sources chinoises qui
reprsentent les Y-tha comme des nomades, HAUSSIG propose son tour (Byz-Sl. p.

222 M. GRIGN ASCHI
n o r d d'Abrahr, 2 2 d e u x rgions qui, selon t o u t e vraisemblance, en 557 n'avaient

t p a s encore touches p a r la guerre. E t d'ailleurs, mme si plusieurs minents
sinologues ont considr que l'ancienne prononciation de Hoa t a i t War et
q u e l'analogie entre W a r a n de la Gographie Armnienne e t W a r a a n , la
ville des H u n s d a n s le Caucase ne s a u r a i t tre ignore, la f o r m e War est
loin d ' t r e certaine. E n o k i prfrait lire ywat (Ghr) 23 e t M. H a r m a t t a Qal,
Xwal et y retrouver le nom des Qalc ou Qilgi.24 E n t o u t cas, mme
si n o u s ne croyons p a s que H o a puisse tre la ville de Ghr sur le H a r r d ,
ainsi que le c o n j e c t u r a i t Enoki, les sources chinoises p e r m e t t e n t de localiser
le p a y s de Hoa, plus p r o b a b l e m e n t des Hoa 2 5 sudouest de Balkh. 2 6 Ainsi
l ' h y p o t h s e de M. Czegldy qui propose d'expliquer la distinction faite par
Silziboulos entre les A v a r (. . d. les Uarxnitai) et les H e p h t h a l i t e s en
a d m e t t a n t que les premiers t a i e n t les t r i b u s nomades d u Tohristn Ori-
e n t a l e t les seconds les h a b i t a n t s des villes de la Bactriane 2 7 n'aide pas
e x p l i q u e r pourquoi les U a r x n i t a i a u r a i e n t quitt leur p a y s d j a v a n t le
d n o u e m e n t de la guerre t u r c o - h e p h t h a l i t e ni comment ils auraient tra-
vers la Sogdiane d j t u r q u e . Ou alors force serait-il d ' a d m e t t r e avec M.
H a u s s i g , qui localise les pseudo-Avar d a n s la rgion d ' A b r s a h r , que ces tribus
t r a v e r s r e n t t o u t l ' I r a n e t passrent le Caucase. Mais le passage d'Evagrius,
q u e M. Haussig i n v o q u e l'appui de sa thse, 2 8 ne parle n u l l e m e n t d ' u n pas-
sage d u Caucase. Il d i t seulement que les A v a r taient l'un de ces peuples qui
a v a i e n t les chariots p o u r demeure e t h a b i t a i e n t les plaines a u del (vno) d u
Caucase. 2 9 Nous a j o u t e r o n s qu'on ne possde en ralit a u c u n e preuve certaine
q u e des tribus a y a n t f a i t p a r t i e de la confdration h e p h t h a l i t e se soient ta-
blies ensuite dans le Caucase. L a m e n t i o n des Kamicik H e p V a l k dans l'His-
175/6) une distinction assez semblable. De plus, Haussig voulait claircir par cette mme
distinction le rapport e n t r e les nom de Y - t h a et d'Abdel. Les premiers (*Yabdel) auraient
t les tribus nomades do ce peuple h a b i t a n t l'est du Pamir, les seconds (Abdel,
Hephthalites) les h a b i t a n t s de la Bactriane. E t le premier a t t a q u e t u r c se serait dirig
c o n t r e les Y-tha/Abdel (1. c. et Byz. p. 325.) Nous devons faire n o t e r que l'auteur d u
Pien-i-tien dans sa Remarque sur Mou-ho khan affirme que ce dernier avait avanc
j u s q u ' 10.000 li de la Mer Occidentale (la Mer Caspienne) et a u sud jusqu'au nord du
G r a n d Dsert. On en conclura que Mou-ho a v a i t vaincu les Y - t h a aprs une chevau-
che travers la Sibrie, car a u t r e m e n t l'historien chinois aurait parl d'une campagne
en direction du Tsong-ling. (ST. JULIEN J . A. V srie vol. I l l p. 323 et suiv.).
22
HAUSSIG : B y z . Slav. p. 178 et note 46.
23
Memoirs of t h e Toyo Bunko 1958 On t h e Nationality of t h e Ephthalites, p. 5.
24
A . MOHAY : 1. c. p . 142 n o t e 70.
25
La traduction du Liang-chou d'ENOKI (1. . p. 1) The c o u n t r y (kouo) of H u a
is a n o t h e r branch of t h e Ch'-shh (Tourfan) nous cause de la perplexit et, t o u t en
i g n o r a n t le chinois, nous croyons qu'il f a u d r a prfrer la traduction plus libre de SPECHT,
Le r o y a u m e des H o a . Ce peuple est u n e branche particulire des Tche-sse. D'ailleurs
ENOKI a traduit lui-aussi quelques lignes plus loin kouo par community.
20
Liang-chou 1. 54 traduit par ENOKI 1. c. p . 3.
27
Npmozgalmak p . 10 (MOHAY 1. c. p. 133.)
28
Byz-Sl p. 182.
29
J . BIDEZ (Londres 1898) p. 196 : "E&vo Exv&txv o't "Aagoi xwv a;oiwv,
xwv vuq rv Kavxaaov t nxetva neta vefiopvv ol rov yetxviwvxa Tovoxov; naaaai
netpevyxe, ctei xaxw Tig avzwv inenv&eaav, im zov BanoQov cpixovxo.
Acta Antiqua Academiae Scientiarum Htingaricae 28, 1980

LA CHUTE D E L ' E M P I R E H E P H T H A L I T E 223
toire des Agouan I ch. 27 de Moses Dasxuranci 3 0 indique seulement, si toute-

fois on peut a j o u t e r foi la lgende de Mrope, que des barbares Kamiik
se trouvaient d a n s le Causace au IV sicle et au V sicle. On en conclura que les
Kamiik faisaient partie de la confdration des Ovvvoi Xevxoi et que des tribus
de cette horde s'taient arrts dans les Caucase, o les H u n s Blancs sont si-
gnals en 395, 31 tandis que d'autres Kamiik se fixrent plus tard dans le To-
hristn. 3 2 On sait que les H u n s Blancs, les Sveta Huna arrivrent jusque dans
l ' I n d e . Quant aux War, les H u n s de Waraan, dont la prsence au Caucase est
a t t e s t e seulement au V I I sicle, ils peuvent appartenir soit la horde BaQ-
arjXx qui lia son nom aux plaines au nord du Daghestn, 3 3 peut-tre une conf-
dration des W a r (BQ) et des ZXoi (les Cl?), 34 soit plus probablement la
horde des Uar et X u n n i (on ne saurait oublier que le Shib-as-sarr portait le
titre de Awar), 35 soit enfin avoir t, tout comme les Kamiik, une tribu des
H u n s Blancs qui s'tait fixe dans le Caucase alors que leurs congnres trans-
migraient vers le sud. E n tout cas, rien ne permet de penser qu'ils venaient du
Tohristn, pas m m e la polyandrie, car ce que Movsis Dasxuranci raconte 3 8
n'implique pas que chez ces H u n s la famille tait fonde sur un systme poly-
andrique rgl jusque dans les dtails comme il en t a i t le cas chez les Heph-
thalites, mais seulement que les murs de ces H u n s taient fort relchs si bien
que les frres avaient parfois une femme en commun q u i t t e possder d ' a u t r e s
concubines.
E n faveur d ' u n e origine hephthalite des Avar d ' E u r o p e on a parfois
avanc que les noms de Uarxnitai, de Uar et Xunni, d'Avares et de H e p h t h a -
lites dsignaient en ralit le mme peuple 37 Altheim croyait mme avoir
dcouvert dans les sources syriaques la preuve dfinitive du fait que les Heph-
thalites incorpors dans l'Empire Sassanide aprs l'croulement de leur E t a t
taient connus sous le nom d'Avar. 3 8 Dans l'Histoire Ecclsiastique de J o h a n -
nes d'Ephesos ( I I I partie, livre V I ch. VII), il est question de deux mille vierges
30
Passage mis en vedette p a r MAROTTART : Wehrot u n d Arang p. 98 et K o m a n e n
p. 34.
31
Voir la Vie de Pierre d ' I b r e Trad. RAUBE p. 6 et suiv., MARKWART E r a n a h r
p. 95 et les tudes de HONIGMANN sur Pierre d'Ibrie et sa Vie.
32
Hwarezm, Mafth-al-'ulm p. 119 ; Al-Muqaddas 283(8).
33
Voir G. MORAVCSIK : Byzantinoturcica. I I . 1958. p. 89.
34
Mnandre E x c e r p t a de legatis d. B O O R p. 4 4 3 / 9 ; M A R K W A R T
. . 1. 15 (1911) . 17, HAUSSIG (Byz. P. 368) prfre
songer une confdration de Var et de Sir (ZfjQe). Cependant la lettre grecque 9 rond
assez souvent le son t u r c et persan dj (HAUSSIG : CAJ 2. p. 15).
35
Voir MARQUART : K o m a n e n p. 74. R a p p e l o n s que d a n s cette page MARQUART
envisageait une hypothse, sur laquelle il faudra probablement revenir. E n tenant compte
d u fait que les H u n s appelaient lo n i p e r Var, MARQUART envisageait la possibilit
q u ' l'origine ce terme ait p u constituer un nom topographique et non pas un nom
ethnique.
36
Moses Dasxuranci (trad. PATKANIAN) p. 194 :
;
.
3
' . CZEGLDY : N o m d n p e k p . 95 ( A . MOHAY: 135).
38
Geschichte der H u n n e n . . p. 26 28.
15* Acta Antiqua Aeademiae Scientiarum Hungaricae 28, 1980

224 M. GRIGN ASCHI
chrtiennes faites prisonnires par Ansirwn Dr en 575 que le Shenh

d c i d a de remettre aux barbares vivant l'intrieur de son Empire et qui
sont appels Abars. 39 Parvenues sur les rives d ' u n grand fleuve, une dis-
t a n c e de cinq tapes de ces barbares, les vierges choisirent de s ' y noyer plutt
que d ' t r e dshonores par des payens. Mais les Avar de cet pisode n'taient
p a s les Hephthalites de l ' I r a n Oriental, ainsi que le supposait Altheim. E n effet
T h o p h y l a c t e nous a p p r e n d incidemment q u ' u n e colonie d ' "Aagoi se trouvait
c i n q tapes de Kigxijatov (Qarqsiyyah) 4 0 la ville situe sur la confluence de
l ' E u p h r a t e et du K h a r b u r . E t ds lors il f a u d r a penser non pas une colonie
h e p h t h a l i t e mais u n e tribu avar du Caucase qui, l'instar des tribus sabir,
a v a i t o b t e n u la permission d'entrer dans l ' I r a n condition de servir dans l'ar-
me sassanide. 4 1
I l existe enfin un dernier argument mis en lumire par Shiratori qui per-
m e t d'carter dfinitivement l'hypothse d ' a p r s laquelle les Avar d'Europe
v e n a i e n t du Turkestan Occidental. Ils portaient la queue qui suscita t a n t de
curiosit chez les Byzantins, une coiffure inusite, ainsi que le notait le savant
j a p o n a i s chez les H u n s arrivs en Europe a v a n t les Avar, chez les Huns Ira-
niens e t chez les Yue-pan. 4 2
Mais c'est prcisment quand on a cart cette hypothse et c'est le
seul r s u l t a t certain, nous ne nous le cachons pas, que les textes notre dispo-
sition permettent d ' a t t e i n d r e qu'on se heurte la difficult la plus grave :
le problme des rapports entre les vrais Avar, les pseudo-Avar et les Jeou-
j a n . P o u r s'orienter au milieu des donnes contradictoires de nos sources et
des solutions qu'on a proposes,il importe t o u t d'abord de distinguer soigneuse-
m e n t les diffrents aspects de notre problme.
I) Les vrais Avar de Thophylacte et les Apar de l'inscription de Klte-
gin 4 3 peuvent-ils tre identifis avec les J e o u - j a n ?
I I ) A-t-on droit d ' a d m e t t r e que cette horde connue des Chinois seule-
m e n t sous le nom de J e o u - j a n dans les Annales du Nord, de Joui-joui d a n s les
A n n a l e s du Sud voire p a r un jeu de mots pjoratif de J o a n - J o a n portait aussi
le n o m d'Abar/Apar? Ou alors Thophylacte a-t-il donn aux Jeou-jan le nom
d ' A v a r par une mprise ?
I I I ) Les Avar de Priscus qui expulsrent les Sabir de leurs siges taient-
ils les Jeou-jan ou les Apar (si on prfre distinguer ces deux peuples) ou enfin
les pseudo-Avar de Thophylacte ?
39
Ms. Vatican.
40
Thophylacte L. I I I ch. 10.
41
Dans l'pisode I X d u Krnmag-i-Ansirwn (1. c. p. 40) il est question de
aiba'atu-asnfin mina-l-'atrk, donc de q u a t r e peuples hunniques diffrents, de la
r g i o n des Bazar qui avaient d e m a n d asyle la Perse.
42
The Queu a m o n g t h e peoples of N o r t h Asia. I. The Queu of t h e Hsiung-Nu
(Toyo B u n k o I. p. 6364).
43
Inscription de Kltegin ct est ligne 4 (ORKUN : Eski T r k Yazitlari. Vol.
I. 31).

LA CHUTE DE L ' E M P I R E H E P H T H A L I T E 225
IV) Quelle confiance peut-on faire la distinction tablie par ce seul

historien byzantin entre les vrais Avar e t les pseudo-Avar ?
Sur aucun de ces points l ' u n a n i m i t ne s'est faite p a r m i les savants qui assez
souvent ont modifi leurs opinions au f u r et mesure qu'ils approfondissaient
l ' t u d e des textes 1 4 , ce qui prouve assez combien les documents notre
disposition sont insuffisants et incertains. Ainsi, aprs plusieurs hsitations,
Chavannes a fini pour considrer que les vrais Avar de Thophylacte t a i e n t
les J e o u - j a n et non pas les Apar, mais il n'a su proposer aucune explication
pour la diffrence des noms sous lesquels cette horde a t connue des Chinois
et des Byzantins. 4 5 M. Haussig et M. Czegldy acceptent eux-aussi que les
vrais Avar de Thophylacte t a i e n t les J e o u j a n , mais ils divergent sur le
second point. M. Haussig qui, au cours de toutes ses tudes a maintenu que les
Avar et lesAbades sources chinoises t a i e n t en ralit un peuple du Turkes-
tan Occidental, 4 6 a propos dernirement une explication de l'emploi fait p a r
Thophylacte d u nom d'Avar pour les J e o u - j a n . L'crivain byzantin a u r a i t
suivi une mode des Sogdiens qui a u r a i e n t donn a u x J e o u - j a n le nom d u
peuple que les Chinois appelaient Wou-houan, des idogrammes sous lesquels
se cacherait la prononciation Avar. E t ils leur auraient donn ce nom, car
ils habitaient les anciens territoires des Wou-houan, u n peuple que, d ' a p r s
M. Hau3sig, les Sogdiens connaissaient bien puisqu'il t a i t arriv dans le T u r -
kestan Occidental avec les Chion que le s a v a n t allemand considre comme les
H u n s de Lieou-Ts'ong. 4 7 Il s'agit d ' u n e hypothse ingnieuse mais qui notre
avis a t t r i b u e a u x Sogdiens des connaissances gographiques et historiques
que difficilement ils possdaient. E t d'ailleurs les anciennes siges des Wou-
houan taient englobes dans l ' E m p i r e J e o u - j a n mais elles ne constituaient
pas son centre.
De son ct, M. Czegldy qui, d a n s son premier travail, considrait
l'identification des Avar (Ouar) avec les Jeou-jan comme invraisemblable, 4 8
a cru ensuite pouvoir infrer des Annales chinoises que ses a u t e u r s employaient
exclusivement le nom de Jeou-jan p o u r la partie de cette horde qui t a i t reste
d a n s la Mongolie e t les noms de J e o u - j a n , de Hiong-nou e t de H o a p o u r les
44
Ce qui, avouons-le, nous est arriv notre tour.
43
Documents sur les Tou-kiue Occidentaux p. 309.
46
D'aprs HAUSSIG le nom d'Abrahr driverait des A v a r qui auraient h a b i t
cette rgion pondant deux sicles. (Byz-Slav. p. 193). E t ces A v a r seraient les Wou-
houan des Chinois. Nous avouons ne voir d a n s le nom d'Abrahr que le terme pahlv
d'apar (suprieur, septentrional), une interprtation que HAUSSIG a lui-aussi envisage
mais qu'il a carte cause de la thse de MARQUART qui voulait voir dans ce nom l'ethni-
que d ' u n peuple. Toutefois MARQUART songeait-il aux Aparnoi, une explication en
tout cas plus vraisemblable que celle de HAUSSIG. Quant aux indications de Thophylacte
qui situait la ville de Taugast sur les confins de l'Inde et du ct de la Bactriano, des
indications que HAUSSIG a mises en vedette, un dtail suffit lui seul leur enlever
toute v a l e u r : le souverain do Taugast p o r t a i t le titre de Tasan et Thophylacte
traduit le fils do Dieu, le fils du Ciel.
" Byz-Slav. p. 189 et 191.
48
N p m o z g a l m a k p . 10 n . 5 (MOHAY : p . 133).

226 M. G R I G N A S C H I
t r i b u s qui avaient transmigr dans le T u r k e s t a n Occidental. 4 9 E t , ds lors, M.

Czegldy conjecture que la horde J e o u - j a n se composait de tribus J e o u - j a n ,
Hiong-nou et H o a (War). Nous ne reviendrons pas sur nos doutes q u ' o n
puisse reconnatre d a n s H o a une ancienne prononciation W a r . Nous ferons
p l u t t remarquer que, d'aprs les Annales chinoises, la horde Jeou-jan a c q u i t
d e l'importance p e i n e en 394, q u a n d elle migra des m o n t a g n e s qui s p a r e n t
le Chen-si de la Mongolie Intrieure, au nord d u Grand Dsert. 5 0 P a r consquent,
on ne saurait considrer ni les Chions d ' A m m i e n Marcelin ni m m e les H o a qui,
d ' a p r s le T ' o u n g - t i e n livre 193, a v a i e n t fond leur E t a t 80 90 ans a v a n t
456 5 1 comme des b r a n c h e s de J e o u - j a n . E t d'ailleurs les renseignements des
sources chinoises p o u r l'histoire de l'Asie Centrale au cours d u IV sicle e t d u
V sicle sont a r b i t r a i r e s et sujettes caution, ainsi que l ' a tabli Enoki. 5 2
E t nous ajouterons q u e l'indication de T'ou-yeou qui p e r m e t de penser que les
H o a taient arrivs d a n s le Tohristn d j d a n s la I I moiti d u IV sicle, r e n d
encore plus obscur le problme des Sabir. E n effet, si on p r t e foi cet a u t e u r ,
les Avar qui, d ' a p r s Priscus, peu a v a n t 463 53 ont dlog les Sabir de leurs
siges sur le bas S y r - D a r i a , ou plus p r o b a b l e m e n t des plaines qui se prolongent
a u nord du Balkash d e l'Irtych Noir au b a s Syr-Daria, 5 4 n ' a u r a i e n t pu tre en
a u c u n cas les t r i b u s H o a / W a r en marche vers le T u r k e s t a n Occidental. E t il
e s t aussi difficile d ' i d e n t i f i e r les A v a r de Priscus avec les Jeou-jan, car les
Annales Chinoises n e connaissent pas u n e expansion de cette horde vers l'Occi-
d e n t et m e n t i o n n e n t u n i q u e m e n t une c a m p a g n e de D a d a r (Ta-t'an) contre les
Y u e - p a n de la rgion d e l'Ili a v a n t 429 e t u n e campagne contre les Tseu-K'iu d u
49
N o m d n p e k p . 59 62 (MOHAY : p . 1 3 4 1 3 5 ) .
50
PARKER : A t h o u s a n d years of t h e T a t a r s . Book I I I . L o n d o n 1924.
51
Voir ENOKI : On t h e Nationality of t h e Ephthalites p. 2 et note 3. Le prof.
SABBATINI-ROSSI n o u s a confirm que le t e x t e chinois affirme bien que l'migration
d e s Y-tha avait eu lieu 80 90 ans a v a n t le rgne de Wen-tch'ing-ti (452 465). Il est
v r a i que, ainsi que le r e m a r q u a i t ENOKI, nous ne savons pas quelle source T'ou-yeou
a e m p r u n t ce renseignement. Mais il est probable qu'il l'avait t r o u v dans la rdaction
originale du Wei-chou.
52
Voir ENOKI : 1. c.
53
Nous ne n o u s cachons pas que le t e x t e de Priscus ne contient aucune indication
p e r m e t t a n t d'tablir avec certitude l'poque de la transmigration des SaAr. Dans son
article Die H e r k u n f t der europischen Awaren p. 21 HAUSSIG a mme considr le
passage de Priscus c o m m e le premier tmoignage de la prsence des Avar dans le Caucase.
Cependant de la lecture des passages des Excerpta de legationibus et de Suda on gagne
l'impression trs forte q u e vers 463 les Saragour et leurs allis venaient seulement d'arri-
ver dans la Ciscaucasie et qu'ils ne devaient p a s avoir abandonn leur ancienne patrie
depuis longtemps.
54
Tour comme CZEGLDY (Nomd npek p. 92, MOHAY: p. 136) nous songeons
la zone du bas S y r - D a r i a cause d u nom de l'ancienne ville des Onogour conserv
p a r Tbophylacto : B a k a t h , un nom qui fait songer invitablement au sogdien. Mais,
d a n s ce cas, [il est certain que Thophylacto a obtenu ce renseignement d'informateurs
sogdiens qui peuvent avoir modifi la forme originale turque. Voir les remarques d e
MOHAY : p. 138. C e p e n d a n t c'est seulement d a n s une rgion expose aux influences
sogdiennes qu'on p e u t concevoir que les nomades turcs aient commenc habiter des
villes.

T o u r f a n en 460. 55 Il est vrai que ces sources a t t r i b u e n t aussi Ch-louen d ' a v o i r

soumis la horde des Kao-kiu et Cheloun d'avoir annexs des tribus hiong-nou
se t r o u v a n t nord-ouest de son E m p i r e , qui devait ainsi atteindre le h a u t
I r t y c h . E t de plus en 429, en 443, en 449 e t en 459 les E m p e r e u r s des Yuen-Wei
T'o-pa Fu-li et T ' o - p a Siun conduisirent des expditions victorieuses contre
les J e o u - j a n . E n 429 Fu-li dpassa m m e les territoires des J e o u - j a n et a t t a q u a
les Kao-kiu de l'Alta. Aussi ne serait-il pas impossible q u ' la suite de ces
campagnes des Yuen-Wei, la horde jeou-jan se soit dplace vers le nord-ouest
et se soit jete sur les Sabir quitte r e n t r e r ensuite d a n s ses siges au n o r d d u
Grand Dsert. 5 6 Mais nous n'en possdons q u ' u n seul indice 57 et qui est p a r sur-
crot incertain : la mention de l ' E t a t de A-fu-t'ai-han d a n s le Pe-chih ch.
98 et d a n s le Wei-chou ch. 102, E t a t qui se trouvait entre B o k h a r a et K h w -
rizm. M a r q u a r t pensait y retrouver la mention d'un Grand K h a n des A-fou
(ancienne prononciation About), des Avar. 5 8 Mais M. P a u l D a f f i n nous a com-
muniqu que, de rgle, les historiens chinois ne traduisaient pas une p a r t i e
d ' u n titre tranger. Aussi pourrait-on songer tout au jilus et M. D a f f i n
est sceptique mme sur ce point - u n t a r h a n des A-fou. 5 9 E t ds lors on
p o u r r a i t imaginer q u ' a p r s la victoire sur les Sabir vers 460 quelques t r i b u s
a v a r avaient avanc j u s q u ' a u x confins d u Khwrizm. 6 0 Mais, dans ce cas, il
f a u d r a i t se d e m a n d e r pourquoi les Chinois auraient parl d ' u n tarhan des
About ? Etait-il parce qu'ils n ' a v a i e n t p a s reconnu dans ce petit r o y a u m e du
55
Voir PARKER: A thousand years of t h e Tatars 1. e. et GROUSSET : L ' E m p i r e
dee Steppes p. 104.
58
Nous devons signaler une difficult supplmentaire. Priscus raconte que trois
peuples s'taient enfuits devant les Sabir : les Saragouroi, les Ourgoi et les Onogouroi.
Or, avec la seule exception de HAUSSIG (Voir MORAVCSIK : Byz-Ttirc. I I p. 227) tous
les savants ont corrig la forme Ourgoi de nos mss. en Ogouroi, une correction justi-
fie par la leon du ps. Zacharias R h t o r 'w.g.r. D'autre p a r t , il est vident que les
Sabir a t t a q u s par les Avar devaient se trouver nord, nord-est des Yue-pan qui occu-
paient la valle de l'Ili. Mais la toponymie de la Sibrie Occidentale (nous renvoyons
aux t r a v a u x de P a t k a n i a n ) prouve assez que des tribus sabir ont continu habiter cette
rgion p e n d a n t de longs sicles encore. E t , de mme, Thophylacte signale au VI sicle
une grande confdration ogr vivant sur le Til Noii, notre avis le K a r a I r t y c h .
11 faudra probablement en conclure que seules les tribus mridionales des Sabir et des
Ogr participrent a u x transmigrations signales par Priscus.
67
E n effet on ne saurait tirer aucune conclusion de la prsence des A ar d a n s la
liste des peuples au nord du Caucase d a n s l'appendice du ps. Zacharias Rhetor,
l'auteur de cette liste s'est inspir de Priseus. (Voir K. CZEGLDY : Studia Turcica.
Budapest 1971. Psctido- Zacharias R h e t o r on t h e nomads p. 141.)
58
Das Reich Zabul p. 264, A Catalogua of the Provincial Capitals of Eranahr p. 43.
69
Ainsi, dans la liste des gouvernements du protectorat de Ngan-si (CHAVANNES
Doc. p. 69) on rencontre une expression analogue : le gouvernement du Ta'i-Han tabli
dans la ville de Houo-lou des Hephthalites et ce ta'i-han n'tait certainement p a s
un Grand K h a n mais un simple Tarhan.
60
Cette ville do A-fou-t'ai-han ne figurait pas, selon toute vraisemblance, d a n s
le rapport sur les Pays d'Occident de Toung-Youan, ainsi que le croyait MARQUART.
Toung-Youan, qui visita l'Occident vers 437, avait des ides fort confuses sur la rgion
entre les deux mers et au sud des eaux croupis et ce fut le compilateur du Wei-ehou
qui a j o u t a la description de plusieurs p a y s do l'Occident qui ne figurait pas dans le rap-
port de Toung-Youan. (Voir ENOKI : Sogdiane a n d the Hsiung-nu. Central Asiatic J o u r .
1955 p. 55 et suiv.)

228 M. GRIGN ASCHI
K h w r i z m qui n ' e n t r e t e n a i t pas des r a p p o r t s avec la Chine, de consanguins

d e s tribus auxquelles ils donnaient le nom de J e o u - j a n ?
Ainsi on revient une fois de plus au p o i n t le plus difficile de notre pro-
b l m e , celui de savoir si on a le droit d ' a d m e t t r e que les J e o u - j a n portaient
aussi le nom d ' A p a r / A b a r . M. Haussig1 cite u n e inscription du Chen-si a p p a r -
t e n a n t aux J e o u - j a n , d a n s laquelle ils s'intitulaient le R g n e des Ta (Grands)
J o u - j o u (ou Nou-nou). De son ct Pelliot croyait possible de reconnatre
d a n s le nom de J e o u - j a n / J o u i - j o u i le t e r m e de jojin, u n t e r m e qui d a n s les
l a n g u e s altaques signifie l'tranger. D a n s le turc hqniyyah yat dsigne
l ' t r a n g e r entr au service d ' u n e tribu ou d ' u n chef militaire. Ds lors on se
d e m a n d e r a s'il n'existe quelque lien e n t r e ce sens du t e r m e de jojin et ce que
n o u s racontent les annalistes chinois de l'adoption p a r cette horde du patro-
n y m e Ukuru Toba p o u r affermir leurs liaisons avec les Yuen-Wei. 6 2 Seul
Blochet, rappelons-le, a cru pouvoir dcouvrir dans la f o r m e Joui-joui des
annalistes chinois d u Sud une ancienne prononciation ou-in qui pourrait
r e n d r e le nom awir. 63 C'est une explication qui, p o u r a u t a n t que nous le
sachions, n'a pas t retenue p a r les sinologues et qui p a r surcrot pourrait
difficilement expliquer la forme apar d u t u r c et la forme obri du slave.
Vraisemblablement ce f u t seulement en grec et en syriaque 6 4 que la lettre b
ou p d u nom Abar se t r a n s f o r m a en v (Avar).
D a n s ces conditions, il est comprhensible que d ' a u t r e s savants aient
r e j e t l'identification des A v a r avec J e o u - j a n . E n particulier P a r k e r proposait
d e voir dans les pseudo-Avares de T h o p h y l a c t e les Y u e - p a n , les tribus t u r -
q u e s ennemies des J e o u - j a n qui vivaient d a n s la rgion de l'Ili. Il pensait que
l'idogramme yue ait pu rendre une a initiale et que l'idogramme pan
soit une forme abrge de pan-lo, la transcription chinoise d u terme sanscrit
d e vara. D a n s son article Ueberde W u - s u n S t a m m in Centralasien, 6 3 Shi-
r a t o r i n'cartait p a s la possibilit que les idogrammes dsignant les Y u e - p a n
a i e n t pu tre prononces l'poque ancienne yeut-bat et correspondre
A b a r et cela quoiqu'il n ' a c c e p t t pas l ' a r g u m e n t principal a p p o r t par P a r k e r
l ' a p p u i de sa thse : savoir que les Y u e - p a n ne sont plus mentionns par les
historiens chinois aprs l'migration des A v a r en Europe. 6 6 E t Shiratori n'ac-
c e p t a i t pas cet a r g u m e n t car il croyait r e t r o u v e r les Yue-pan dans les Yen-mien,
d ' a p r s les Annales des Soei une branche des Tie-le sur le Balkash. 6 7 Cependant
61
Byz. p. 354 et suiv et note 293.
62
The Toba of t h e northern wilderness (PARKER 1. .).
63
E . BLOCHET: Le p a y s de T c h a t a et les Hephthalites R S O 1925.
64
K. CZEGLDY : Ps. Zacharias Rhetor p. 137.
65
Keleti Szemle I I I p. 133.
66
Asiatic Quarterly Review Vol. X I I I p. 355.
67
CHAVAJSTNES (Doc. p. 311) nouiissait des doutes propos de cette identification.
D'ailleurs dans la liste des gouvernements d u P r o t e c t o r a t de Ngan-si (CHAV. DOC. p. 71)
o n rencontre le gouvernement de l'arrondissement des Yue-pan. Il faudra en conclure
q u e des tribus y u e - p a n s'taient transfres d a n s la Tohristn Oriental dans la rgion
d e Kiu-lan (Kourn).

plus t a r d ce f u t le m m e Shiratori qui a p p o r t a la p r e u v e dfinitive que les

Yue-pan n'avaient rien en c o m m u n avec les Avar d ' E u r o p e qui portaient la
queue : ces tribus t u r q u e s de coupaient et l a v a i e n t les cheveux. 8 8
De son ct, M. Gumiljv se refusait catgoriquement d ' a d m e t t r e que les
vrais Avar de Thophylacte aient t les Jeou-jan et il les identifiait avec les
A-pa, qu'il considrait comme une tribu proto-sibrienne et que les annales
chinoises signalent la fin d u V I sicle dans la Dzoungarie. 8 9 E n 585 elles en-
registrent un a t t a q u e des A-pa contre le camp royal d'A-po-khan, l'un des
K h a n s des T'oukiue Orientaux qui disputait le pouvoir Cha-po-lio e t u n e
intervention des troupes chinoises. 70 A ce moment-l leurs siges ne p o u v a i e n t
tre q u ' a u nord du Grand Dsert. Elles les m e n t i o n n e n t une dernire fois
l'occasion de l'insurrection des tribus T-ie-le, Se-kie, Fou-likiu, Sie-sa et P o u -
kou contre Ta-T'eou 7 1 et, ainsi que le faisait valoir Chavannes, d ' a p r s le
T'ang-chou ch. 62, T'ie-le, Se-kie, Hoen et Pou-kou a p p a r t e n a i e n t aux Tls
une horde de la Sibrie. Si les A-pa se trouvaient d j au V sicle dans la rgion
o les historiens chinois les signalent la fin du V I sicle, ils pourraient corres-
pondre a u x Avar de Priscus, ainsi que le conjecturait M. Gumiljv. Mais on
se d e m a n d e r a si cette tribu, selon les sources chinoises d ' u n e importance secon-
daire, a u r a i t t capable de vaincre la puissante confdration sabir 7 2 . E t nous
croyons t o u t aussi difficile d'accepter les arguments a p p o r t s p a r M. Gumiljv
pour considrer les vrais Avar de Thophylacte et les J e o u - j a n comme d e u x
peuples diffrents et p o u r rfrer a u x A-pa ce que l'historien byzantin r a c o n t e
des vrais Avar, savoir qu'ils avaient t vaincus p a r les Turcs et qu'ils
s'taient rfugis en p a r t i e Taugast, c. . d. auprs des Si-Wei, et en p a r t i e
chez les Moukri, d ' a p r s M. Gumiljv un peuple sien-pi v i v a n t sur les rives de
88
T h e Queu among t h e Peoples of North Asia, I. The Queu of the Hsiung-nu
(Memoirs of the Toyo Bunko I, 1923, p. 03).
89
Malheureusement nous ne connaissons les thses de GUMILJV qu' t r a v e r s
son article IV. . VDI 1969 . 129 et ses notes ajoutes la
d'ARTAMANOV (Moskva 1962), en particulier la note de p. 106107.
70
J . A. V srie Vol. I l l p. 494.
71
Soei-chou ch. L I (CHAV. DOC. p. 50). E t le texte continue : Ces tribus deman-
drent de venir se soumettre la Chine. Tout le peuple de Ta-t'eou s'tant dispers,
(Ta-t'eou) s'enfuit dans l'ouest chez les T'ou-kou-hoen. Il e n rsulte de ce rcit q u '
ce m o m e n t Tardou se trouvait dans les rgions orientales de l ' E m p i r e Tou-kiue et qu'on
ne saurait considrer ce passage, ainsi que le faisait HAUSSIG (Byz. p. 330), comme une
preuve que les A-pa habitaient dans le Turkestan Occidental.
72
E n reprenant une thse de DARK, GUMILJV croyait que les paroles de Priscus
q u ' u n peuple habitant prs de l'Ocan avait mis en mouvement les Avar, se rapportaient
aux H e p t h t a l i t e s qui vivaient prs de la Mer Caspienne, une mer quo les Anciens, m m e
aprs Ptolme, continuaient trs souvent considrer comme un golfe de l'Ocan.
Cette hypothse f u t envisage un certain moment aussi p a r HAUSSIG (CAJ 2 p. 34).
Cependant, l'poque de l'expulsion des Sabir de leurs siges, expulsion antrieure en
t o u t cas 463, la Sogdiane taient encore aux mains des H u n s Kidarites, les Cl habi-
taient les bords orientaux do la mer Caspienne (Tabari I 884) et les H o a venaient peine
de constituer un E t a t . Ce f u t seulement dans la dernire dcade d u V sicle qu'ils a t t a q u -
rent les Turcs Kao-kiu prs d'Ouroumchi (ENOKI : On the nationality of the E p h t h a f i t e s
1. . p. 25).

230 M. GRIGN ASCHI
Tili 7 3 . P a r contre les J e o u - j a n , fait noter le s a v a n t russe, a v a i e n t cherch asyle

chez les Pei-Ts'i. Or, cet a r g u m e n t n ' a r r i v e pas nous convaincre car il est
e x a c t que A-na-lo-chen, le successeur de A-nakouei le dernier qagan Jeou-jan,
s ' t a i t rtir dans le territoire des Pei-Ts'i et non pas chez les Si-Wei. Mais
e n s u i t e les dbris des J e o u - j a n cherchrent un refuge a u p r s de l'Empereur
W e i Thai-tsou.' 4 E t de plus, il est certain que dans les langues de l'Asie Cent-
r a l e le nom de T a b g a dsignait t o u t e s les anciennes rgions des T'o-pa et
p a r t a n t mme le r o y a u m e des Pei-Ts'i. Mais un dernier a r g u m e n t de M. Gumil-
j v mrite d'tre e x a m i n de plus prs car il pourrait tre dcisif. A son avis,
les A p a r qui, d'aprs l'inscription rige p a r Bilge Qagan, envoyrent en 552
d e s ambassadeurs a u x funrailles de B o u m i n (T'ou-men) Q a g a n ne pouvaient
t r e en aucun cas les J e o u - j a n qui c e t t e d a t e continuaient encore leur rsis-
t a n c e contre les Tou-kiue 7 5 et qui ne constituaient plus une h o r d e indpendante.
Toutefois, a v a n t d ' a c c e p t e r cet a r g u m e n t , il f a u t t o u t d ' a b o r d se demander
quelle confiance on p e u t faire a u x renseignements de Bilge Qagan pour les
v n e m e n t s des annes 552 555. Or, il est certain que Bilge Qagan qui, en
732, f i t lever la stle en honneur de Kltegin, n'avait que des ides fort vagues
propos de l'histoire d u premier E m p i r e Tou-kiue. Il s a v a i t t o u t juste que
B o u m i n (recte T'ou-men) Qagan et I s t e m i Qagan a v a i e n t fond un E m p i r e
a l l a n t de la fort de Q a d i r q a n au Ternir Qapig (les P o r t e s de P e r dans la
Sogdiane) et il imaginait que tous les peuples puissants s ' t a i e n t fait reprsen-
t e r a u x funrailles de d e u x fondateurs de l'Empire turc, ainsi qu'il en avait
t le cas l'occasion des crmonies f u n b r e s en honneur de Kltegin. Mais
il est bien certain que les R o m a i n s d ' O r i e n t (Purum) n ' e n v o y r e n t pas de dl-
g u s a u x funrailles de T ' o u - m e n d o n t ils ignoraient j u s q u ' l'existence. E t il
e s t t o u t aussi certain q u ' e n 552 et en 576 les Tpt (T'ou-fan) d u Tibet Cen-
t r a l n'avaient pas encore constitu un E m p i r e et qu'ils n ' e n t r e t e n a i e n t pas de
r a p p o r t s avec les Tou-kiue. E t de plus Bilge Qagan parle u n i q u e m e n t des
T a b g a c , les Chinois, sans faire distinction entre les Si-Wei, les Pe-Ts'i et les
Pei-Tcheou. P r o b a b l e m e n t il imaginait q u ' l'poque des f o n d a t e u r s de l ' E m -
p i r e Turc, la Chine t a i t unie sous u n e seule dynastie ainsi qu'il en tait le
cas sous les Tang. D a n s ces conditions, on a le droit de conclure que les Apar de
l'inscription de K l t e g i n , les vrais Avar de Thophylacte qui taient renom-
73
Nous ignorons malheureusement les arguments que GUMILJV possdait pour
a v a n c e r cette identification et qu'il allait exposer d a n s un article qui n o u s a t inaccessible
Tri iseznuvix naroda Sibirii. E n gnral on s'accorde les voir dans Moukri un
p e u p l e de la Core (MORAVCSIK : Byz-Turc. I I p. 193) et on explique ce nom per le t e r m e
sanscrit do Mukuri, en tibtain Muglig, la Core. Mais on so d e m a n d e r a comment les
T u r c s auraient adopt un t e r m e du sanscrit p o u r dsigner la Core u n e poque laquelle
ils n'avaient q u ' u n e connaissance superficielle d u bouddhisme. On prfrera l'opinion
d e HENNING (Sogdica 7) qui proposait de corriger Moukri en Biikli, d a n s les inscriptions
d e l'Orchon los peuples des forts de l'est-.
74
S. JULIEN : J . A. V srie vol. 3 p . 325, E x t r a i t du Pien-i-tien.
75
Sous le c o m m a n d e m e n t de Teng-cho-tseu l'oncle de A-nakouei (S. JULIEN:
J . A. V srie vol. 3 p. 330).

LA CHUTE DE L ' E M P I R E HEPHTHALITE 231
mes chez tous les h a b i t a n t s de la Scythie pour leur valeur militaire, d e v a i e n t

tre un peuple f o r t puissant, d o n t le souvenir se conservait encore chez les
Turcs et que ds lors ce peuple ne p o u v a i t tre que les Jeou-jan. 7 8 E t ce qui est
en t o u t cas certain, c'est que Thophylacte a donn a u x J e o u - j a n le nom d ' A v a r
parce que les a m b a s s a d e u r s turcs eux-mmes leur a v a i e n t donn ce nom. Il ne
leur a pas donn p a r c e qu'il l'avait t r o u v dans l'histoire de Priscus. E n e f f e t
il nous raconte aussi q u ' u n e partie des Avar s'tait rfugie chez les Moukri,
un peuple qui lui t a i t certainement t o u t fait inconnu et q u ' a u c u n e a u t r e
source grecque ne mentionne.
Or, une fois t a b l i ces deux points, savoir que les pseudo-Avar de
Thophylacte n ' a v a i e n t rien en c o m m u n avec les H e p h t h a l i t e s et que ses
vrais Avar t a i e n t les Jeou-jan, on p e u t esprer de s'approcher de la solution
du problme de l'origine des Ouar et X o u n n i . Il ne f a i t p a s de doutes que l'ex-
cursus de T h o p h y l a c t e sur les peuples scythes fourmille d'erreurs en d p i t
des renseignements d'origine t u r q u e qu'il possdait. Ainsi Thophylacte semble
ne s'tre mme p a s aperu que ses renseignements propos des premires
campagnes t u r q u e s se rapportaient l'poque de la fondation de l ' E m p i r e
tou-kiue et non p a s au rgne du Qagan Tardou, qui a v a i t envoy l ' E m p e r e u r
Maurice la lettre, d o n t il nous a donn le dbut. D e mme, il prtend que la
premire campagne des Turcs avait t dirige contre les Abdel-Hephthalites,
alors que les Tou-kiue devinrent indpendants seulement avec la victoire de
T'ou-men sur les Jeou-jan. 7 7 P a r surcrot, il prtend que l'expdition t u r q u e
contre la confdration des Ogr, d o n t faisaient p a r t i e les tribus Ouar e t
X o u n n i avait eu lieu aprs les victoires sur les Abdel e t les Avar, t a n d i s
que nous savons q u ' u n e partie de ces deux tribus t a i t arrive dans l'Alanie
d j en 558 a v a n t la dfaite finale des Hephthalites. E t il fait peu de d o u t e s
que l'insurrection de Touroum a p p a r t i e n t p a r contre l'poque de Tardou car
d a n s cet excursus elle reprsente le seul fait nouveau que le Qagan p o u v a i t
dsirer de c o m m u n i q u e r Maurice. E t il est aussi certain que les KoxQyrjQot qui
76
Cependant voir ENOKI : On the d a t e of the Kidarites 2. Mem. of the Toyo-
Bunko X X V I I I , 1970, p. 26 27. Mais nous ne saurions a d m e t t r e qu'on puisse a t t r i b u e r
au terme de Apar d a n s une inscription t u r q u e du V I I I sicle la valeur d ' A p a r k h s h a t r
dans l'inscription de Chapour I. De son ct, HAUSSIG a rappel l'attention sur le fait
qu'on rencontre une mention des Apar d a n s u n texte t u r c d u Tourfan (Byz-Sl. p . 188
n. 120).
77
CHAVANNES (Doc. p. 260) pensait qu'on puisse retrouver dans les Abdel les
Tls b a t t u s p a r les Tou-kiue alors qu'ils taient encore les vassaux des Jeou-jan en 546.
Cependant il existerait une explication plus simple pour l'ordre chronologique des cam-
pagnes turques chez Thophylacte. Mou-ho khan conquit la Sibrie et avana j u s q u e
dans la Sogdiane a v a n t d'exiger de l'Empereur Tha-tsou des Si-Wei la livraison des
Jeou-jan rfugis Tch'ang-ngan (f'ion-i-tien 1. c.). Aussi les Tou-kiue auraient-ils p u
considrer ce massacre comme le couronnement de leur guerre contre les Jeou-jan qui,
dans ce cas, se serait close aprs leur premire victoire sur les Hephthalites. Mais il sera
plus prudent d ' a d m e t t r e que Thophylacte no possdait que des relations imprcises
sur les dclarations faites par les ambassadeurs turcs en 698, trente ans avant la compo-
sition do son livre.
Acta Antiqua Acudemiae ScicrUiarum Uungaricae 28, 1980

d ' a p r s Thophylacte appartenaient eux-aussi aux Ouar et Xounni, taient

en ralit les KovxQyovQot de Procope, les KovTQyovQoi/KovTgyrjQoi de Mnandre
arrivs en Europe a v a n t les pseudo-Avar. 7 8 Mais en dpit de toutes ces erreurs
on ne saurait croyons-nous ignorer l'assertion de Thophylacte que les Ogr
e t les Ouar et X o u n n i habitaient sur le Til Noir, une fleuve qui ne peut tre
la Volga puisqu'en 555 les armes de Mouho-khan avaient a t t e i n t la Sogdiane
m a i s non pas la Mer Caspienne et que p a r consquent cette horde devait se
t r o u v e r quelque p a r t dans la Sibrie mridionale, probablement sur le I r t y c h
Noir. 7 9 Mnandre affirme explicitement que les Avar taient parvenus dans
l'Alanie aprs de longues prgrinations, 8 0 Evagrius qu'ils avaient eu souffrir
de leurs voisins, les Turcs. 81 E t on ne saurait non plus ngliger l'affirmation
catgorique de Thophylacte que les A v a r d'Europe taient des pseudo-Avar
puisque nous apprenons de Mnandre que les Turcs les considraient comme
d e s Ouarxnitai, 82 alors que, nous l'avons constat, les vrais Avar, les
J e o u - j a n s'appelaient en turc Apar. Certes, il est permis de penser que Tho-
p h y l a c t e a donn une forme plus d r a m a t i q u e ses renseignements de source
t u r q u e . A n'en point douter, il a invent les hros ponymes U a r et Xounni.
Mais une invention de ce genre s'expliquerait mal si les pseudo-Avar ne se com-
posaient pas rellement de deux tribus : les Ouar et les Xounni. E t ce propos
on ne saurait oublier qu'aprs l'invasion avar il existait d a n s le Caucase la
ville de Waraan et que nos mss. grecs contiennent la variante Xeounni qui,
ainsi que le faisait noter Chavannes, rappelle la tribu des Hoen, faisant partie
a u V I I sicle de la confdration ougoure. 8 3 On a aussi le droit de douter que
ce f u t seulement la suite d'une mprise des Sabir, des Onogour et des Barsilt
q u e les Ouar et X o u n n i assumrent le nom d'Avar et cela lors mme qu'il
ne serait pas inconcevable que des populations de la Ciscaucasie aient pu se
t r o m p e r sur l'identit d'un peuple de la Sibrie Occidentale. Mais en fait il
n ' e s t pas rare, ainsi que l'a signal Shiratori, que des tribus de la Mongolie et
de la Mandchourie aient adopt au cours de leur histoire le nom d'une tribu
p l u s puissante. P a r ex. au moment d u triomphe des Sien-pi, des tribus hiong-
n o u en assumrent le nom. 84 Ds lors il ne serait non plus impossible que les
t r i b u s Ouarxnitai voire les tribus Ouar et Xounni aient adopt dj dans
la premire moiti du V sicle le nom de Abar, c. . d. le nom turc ou mongol
78
MORAVCSIK : B y z - T u r c I I p . 171 172.
79
Rappelons quo CHAVANNES (DOO. p. 251) songeait plutt a u Tola car, l'poque
d e s Soei, la confdration ougoure se t r o u v a i t sur le rive de ce fleuve. Cependant la tur-
cologie n'accepte plus l'identification des Ogr avec les Ougour.
80
Mnandre De leg. Gentium E d . DE BOOR p. 442 (frag. 4).
81
Voir note 29.
82
D e leg. R o m . f r a g . 4 3 (DE BOOR p . 205).
83
CHAVANNES : Doc. p. 88 n. E d . de B o n n p. 284 : oi rovrov ov ei'ov -
raroi fagxoi Ovg xai Xovwl vofidtiov.
84
SHIRATORI : The Queu among t h e Peoples of North Asia. I p. 25 26.

des Jeou-jan. E t , d a n s ce cas, les Avar de Priscus a u r a i e n t pu tre eux-aussi

des Ouar et des Xounni. E n f a v e u r de notre hypothse savoir que le Ouar-
X o u n n i avaient pris le nom de leurs suzerains les Apar, il n'existe q u ' u n indice
qui est loin d ' t r e dcisif. Maniach, l'embassadeur sogdien d ' I s t e m i Constan-
tinople en 568, donne lui-aussi le nom d ' A v a r a u x nomades qui s'taient rfugis
en E u r o p e et leurs congnres qui taient rests d a n s leurs anciens siges sous
l ' a u t o r i t turque 8 5 et le mme nom leur est donn p a r Silziboulos. 86 On p o u r r a i t
en conclure que p o u r dsigner ces populations les Turcs employaient t o u r
t o u r les nom de Ouarxnitai et de Apar/Avar e t que des liens quelconques
devaient exister e n t r e les Ouar-Xounni d u Til Noir et les J e o u - j a n de la Mongo-
lie puisque, t o u t comme ces derniers, les A v a r d ' E u r o p e portaient la queue,
une mode qu'ils n ' a v a i e n t certainement pas e m p r u n t e aux Tou-kiue, sous
lesquels ils avaient vcu un ou deux ans au plus.
Bien e n t e n d u nous ne nous cachons pas qu'on p o u r r a i t renverser n o t r e
argumentation e t voir dans l'emploi d u nom d ' A v a r p a r Maniach et p a r Silzi-
boulos et mme d a n s la queue des pseudo-Avar a u t a n t de raisons p o u r consi-
drer les Ouar e t X o u n n i comme des tribus jeou-jan. D u moins notre h y p o t h s e
p e r m e t t r a i t d'expliquer la prsence de quelques petites tribus A-pa d a n s la
Dzoungarie vers la fin du V I sicle, d a n s lesquelles les annalistes des Soei ne
reconnaissaient plus les J e o u - j a n et l'arrive de t r i b u s A v a r en E u r o p e sans
tre oblig de corriger nos sources qui se c o n f i r m e n t les unes les a u t r e s :
la relation d'origine t u r q u e sur la fuite des Apar/Avar en Chine et chez le Mouk-
ri, le rcit de l'anantissement de ce peuple d a n s les Annales chinoises e t la
dclaration e m p h a t i q u e de Thophylacte que les A v a r d ' E u r o p e t a i e n t des
pseudo-Avar a p p a r t e n a n t la confdration Ogr et qu'ils provenaient de la
rgion du Til Noir, une fleuve qui, en aucun cas, ne saurait tre identifi avec la
Volga. 87
Mais venons l'examen des sources perses tudies p a r M. Widengren. 8 8
Nous constatons t o u t d ' a b o r d un fait. Les riwyah les plus anciennes p a r m i
celles que M. Widengren a recueillies, se p a r t a g e n t entre trois traditions diff-
rentes :
I) U n e chronique de la moiti du I X sicle qui a utilis en p a r t i e le
f j o d y n m a h sassanide. At-Tabar a transcrit presque entirement c e t t e
histoire et nous en possdons une seconde version peu prs identique, celle
de l'Anonyme Sprenger. 8 9 A la c h u t e des H e p h t h a l i t e s cette chronique consacre
85
Mnandro De leg. gentium fr. 18 (Ed. DE BOOR p. 452). Nanmoins ce passage
ne peut pas tre considr commo dcisif car, au cours de la conversation avec Maniach,
c'est l'empereur J u s t i n I I qui, le premier, emploie le nom d ' A v a r pour les tribus qui
s'taient soustraites l'autorit t u r q u e et qui s'taient tablies dans la Pannoni.
86
Excerpta de Sontentiis (Ed. MAI p. 354 55).
87
Cependant voir CZEGLDY: Npmozgalmak p. 810 (MOHAY: p. 133).
88
Orientalia Suecana 1 p. 69 99.
89
B E O 1974 D a m a s ch. I I I et Appendices et I I I A B.

234 M. GRIGNASCHI
une courte notice insre correctement aprs le rcit de la premire guerre perso-
byzantine de l'poque de Justinien I, d'une campagne d'Ansirwn dans le
Caucase contre les H a z a r (en ralit contre les Sabir et les peuples du Caucase
allis de Byzance) et du blocus lgendaire d u dtroit d'Aden. Selon cette tra-
dition, aprs ces victoires, Ansirwn dcida de venger son aeul Proz. Dans
ce b u t , il pousa la fille du H q n des Turcs, le poussa attaquer les Hephtha-
lites et, en profitant de la victoire du H q n qui avait tu leur Roi, il occupa
Balh, la Transoxiane et la Ferghana. 9 0
I I ) L'pitom du Hodynmah conserv par Eutychius et I b n Qutay-
bah. R s'accorde le plus souvent avec la tradition prcdente. Dans notre cas,
il donne un texte presqu'identique celui d'at-Tabar. Cependant Eutychius
a j o u t e et, probablement cet ajout n'est pas rejeter, ainsi que le faisait M.
Widengren 9 1 que le mariage d'Ansirwn avec la fille du H q n f u t clbr
seulement aprs la victoire sur les Hephthalites.
I I I ) La rlaboration d'une traduction du Hodynmah vraisembla-
blement de celle d'Ibn-al-Muqaffa' par un crivain arabe des premires annes
du I X sicle que nous avons propos d'appeler le pseudo-Ibn-al-Muqaffa'. 9 2
Mais, ct de la traduction du Hodynmah cet auteur utilisait d'autres
sources appartenantes la Su'biyyah, un mouvement qui possdait assur-
ment de bons renseignements sur l'histoire de l'ancien Iran mais qui n'hsitait
pas les manipuler jJour mieux exalter la gloire des Sassanides. 93 Ce texte f u t
son tour remani plusieurs reprises au cours du mme sicle. At-Tabar a
puis ses extraits la version la plus ancienne de cet ouvrage, qui f u t utilise
aussi par al-Ya'qb. Ad-Dainawar en possdait une rdaction dj modifie,
tandis que, dans sa forme actuelle, la Nihyat-al-'arab f 'ahbri mulki 1-
F u r s wa-l-'Arab reprsente la dernire forme de ce livre. 94
90
N o t o n s que le t e x t e de c e t t e riwyah d a n s a t - T a b a r m a n q u e de clart, ce
q u i a i d e c o m p r e n d r e q u ' a l - M a s ' d ( M u r g - a d - d a h a b I I p. 203) et BATrni (ZOTENBERG
I I , 161 a i e n t a t t r i b u A n s i r w n d'avoir t u lui-mme le R o i des H e p h t h a l i t e s . Mais
n o u s convenons avec M. WIDENGREN qu'il f a u t t r a d u i r e le passage d ' a t - T a b a r et de
l ' A n o n y m e Sprenger : Et il (le R q n ) vint e u x et il t u a leur R o i et il e x t e r m i n a (et
n o n p a s il e m m e n a de force) les gens de sa m a i s o n . I l (Ansirwn) dpassa B a l h et la
T r a n s o x i a n e et il tablit ses a r m e s d a n s la F e r g h a n a .
91
L . c. p. 75. R a p p e l o n s q u ' E u t y c h i u s est i n d p e n d a n t d ' a t - T a b a r . Si l'on accepte
son indication, ce m a r i a g e a u r a i t t clbr en 560. Cette notice est d ' a u t a n t plus vrai-
s e m b l a b l e qu'il a u r a i t t d a n g e r e u x p o u r la fiance d u Shenh de t r a v e r s e r le R o y a u m e
H e p h t h a l i t e avec lequel les T u r c s taient en t a t de guerre depuis 555. Il est vrai que
sa c a r a v a n e a u r a i t p u e m p r u n t e r la r o u t e d u K h w r i z m qui, ainsi que nous allons le
c o n s t a t e r , tait l'poque u n r o y a u m e i n d p e n d a n t .
92
Voir notre t u d e La Nihyatu-1- ' a r a b f ' a h b r i - l - F u r s wal-'Arab et les Siyaru
Mulki-l-'Agam d u ps. Ibn-al-Muqaffa'. BEO t . X X V I D a m a s 1974 p . 115 ch. I I I
Le p s . Ibn-al-Muqaffa'.
93
Voir notre t u d e sur la rforme fiscale d'Anirwn d a n s les Quaderni dlia
A c c a d e m i a Nazionale dei LinceiConvegno sulla P e r s i a (Rome 1970) et Les Rgles d'Ar-
dar b . B b a k p o u r le g o u v e r n e m e n t d u R o y a u m e q u e nous a v o n s publies d a n s l'Islam
T e t k i k l e r i Enstitiisii Dergisi ( I s t a n b u l 1973).
91
La N i h y a t - a l - ' a r a b etc. B E O D a m a s 1969 et 1974 ch. I .
Acta Antiqua Aeademiae Seientiarum Hungarieae 23, 1980

Or, les riwyah que M. Widengren a appeles les lignes I, I I et IV, ap-
partiennent toutes la tradition du ps. Ibn-al Muqaffa' seule la ligne IV de
Firdaus, sur laquelle nous allons revenir, peut tre en partie indpendante
et elles relatent diffrents pisodes du rgne d'Ansrwn prsents dans une
succession arbitraire la diffrence de ce que faisaient les traditions I et I I .
I) Episode : la reconqute par Ansirwn des provinces que ses anctres
avaient perdues.
a) at-Tabar 1,2 p. 894 I. 1013. Dans ce passage le ps. Ibn-al-Muqaffa'
avait oubli de prciser que ces p a y s appartenaient a u x Hephthalites, vrai-
semblablement parce qu'il l'avait dj dit dans les paragraphes prcdents.
D'ailleurs il y fait explicitement allusion dans le troisime pisode.
b) al-Ya'qb (Vol I p. 186). Cependant ce dernier a modifi le passage
et fait vaguement allusion la conqute de pays qui, par le passe, n'avaient
pas appartenu l'Iran.
c) la Nihyah qui spcifie que les pays perdus taient sous la domina-
tion des Rois hephthalites.
d) ad-Dainawar qui fait prcder la listes des provinces recouvres par
l'avertissement : Il porta des armes dans le pays des Hephthalites et qui la
fait suivre par le rcit d'une expdition de Singib H q n en direction du Horas-
sn. Mais, ainsi que l'a fait remarquer M. Widengren 95 c'est seulement d a n s la
version de Mirhond, qui par ailleurs copie ad-Dainawar que la conqute des
pays des Hephthalites et la campagne de Singib sont considres comme con-
temporaines.
II) Episode : les mesures prises par Ansirwn contre les peuples pillards
des frontires (at-Tabar et Nihyah),
III) Episode : Les fortifications qu'Ansirwn f i t lever dans le Caucase
que Singib f u t impuissant forcer. A ce propos, pour donner plus de relief
au succs du Shensh, on raconte incidemment que Singib tait le souverain
qui avait vaincu et tu W. z. r. (Warz?), 9 6 le puissant Roi dee Hephthalites,
d t r u i t la plus grande partie de ses armes et conquis ses provinces l'excep-
tion de celles qu'Anirwn avait prcdemment occupes. 97 On notera que
l'apparition d'Istemi d e v a n t Derbent, sans aucun doute un fait historique, doit
appartenir aux annes 572 ou 573 aprs la rupture de la paix entre J u s t i n I I
et Ansirwn. Mais dans cette tradition la campagne d u Roi perse contre J u s -
tinien I est raconte aprs cet pisode.
On rencontre la mme succession des vnements d a n s les Ourur de
T a alibi (Ta alibi I et I I I de M. Widengren. E d . Zotenberg p. 610).
De son ct, Firdaus runit ces trois pisodes, parle vaguement de la
soumission de certains Rois trangers Kisr, d'un voyage du Shenh le
85
L. c. p. 77.
96
Voir WIDENGREN : 1. o. p. 73 note 4.
97
'ill rnft kna Kisr galaba 'alaylii minh.
Acta Antiqua Academiae Scientiarum Hungaricue 2S, 1980

236 M. GRIG.4ASCIII
long des frontires de son rgne, d ' u n e muraille btie sur les e a u x et des mesures
prises contre les peuples pillards sur t o u t e s les frontires d u R o y a u m e .
Mais, q u a n d on compare les versions que les historiens qui dpendent du
ps. I b n - a l - M u q a f f a ' d o n n e n t de ces pisodes, il a p p a r a t i m m d i a t e m e n t que
d j l ' a u t e u r de l a N i h y a h et, plus encore, les diteurs successifs de ce livre
o n t modifi le t e x t e original soit pour suivre la mode littraire de l'poque soit
p o u r enjoliver le t e x t e . N o u s nous bornerons deux exemples. Le ps. Ibn-al-
M u q a f f a ' commenait le chapitre du rgne d'Ansirwn p a r le t e x t e de la
l e t t r e o le Roi a n n o n a i t au padgospn de l ' A z e r b a d j n Wr-bn-Nahr-
g n son lvation au trne, u n e lettre que Nldeke considrait comme une pice
originale. Al-Ya 'qb y f a i t allusion. P a r contre, dans la Nihyah cette lettre
a t remplace p a r u n discours du trne farci de lieux p a r n t i q u e s , que nous
r e t r o u v o n s dans leShnmah d e F i r d a u s . D e m m e , a t - T a b a r raconte qu'An-
sirwn punit cruellement les Wrz et les Sl (Col). De ces derniers il n'pargna
que 80 hommes qu'il t a b l i t dans la ville de S a h r m - F r z . L ' a u t e u r de la
Nihyah a t r a n s f o r m ce massacre dans u n e victoire glorieuse du Shensh
qui a u r a i t fait construire l'intention des prisonniers u n e ville gale celles
qu'ils habitaient a u p a r a v a n t , ainsi qu'il le fera plus t a r d p o u r les dports
d'Antioche. 9 8 L a plus g r a n d e prudence s'impose, on le voit, q u a n d on fait
recours a u x riwyah des a u t e u r s les plus rcents. Toutefois ces derniers nous
p e r m e t t e n t de reconnatre que la version intgrale du ps. I b n - a l - M u q a f f a ' ou
l'dition primitive de la Nihyah contenaient au moins d e u x autres pisodes.
IV) Episode : L e mariage d'Ansirwn avec la fille d u H q n . D'aprs
a t - T a ' l i b , l'Anonyme d u Mugmal-at-tawrh et de F i r d a u s le H q n
a v a i t pris l'initiative d ' o f f r i r sa fille Ansirwn dans le b u t , ainsi qu'il en
ressort explicitement d u rcit de Firdaus, d ' v i t e r un choque avec l'arme perse
masse sur les frontires d u IJorassn. E t Ansirwn a v a i t runi cette arme
ds qu'il avait reu la nouvelle de l'crasement des forces hephthalites et de
l ' i n t e n t i o n du H q n , l'hritier d'Argasp, d'envahir l ' I r a n . Ad-Dainawar
d e v a i t avoir trouv lui-aussi dans la premire dition de la Nihyah l'indi-
cation que Singib se p r p a r a i t a t t a q u e r le IJorassn. E t Firdausi ajoute
qu'Ansirwn, t o u t en a c c e p t a n t la proposition du H q n , chargea son
conseiller Mihrnsitd de choisir entre les filles du souverain turc celle qui
a v a i t p o u r mre la H t n e t non pas u n e simple concubine. E t aprs avoir
r a c o n t comment Mihrnsitd d j o u a le s t r a t a g m e de la J J t n qui voulait
v i t e r de lui donner son e n f a n t et c o m m e n t le IJqn accda sa demande lors-
48
Cependant M. WIDESGREN (1. c. p. 77 n o t e 2) considrait que la tradition d'at-
T a h a r est fort lgendaire et que la Nihyah corrige d'une faon heureuse la notice
d ' a t - T a b a r . Quant nous, nous prouvons inoins de difficult croire un massacre
gnral q u ' l'assertion de la Nihyah que ces nomades possdaient u n e ville. D'ailleurs
n o u s avons dj donn u n oxemple significatif des modifications que les diteurs de la
Nihyah apportaient a u t e x t e du ps. Ibn-al-Muqaffa' dans notre tude Le ps. Ibn-
al-Muqaffa' (Damas 1974) p. 1 3 8 - 1 3 9 .

LA C H U T E D E L ' E M P I R E HEPHTHALITE 237
qu'un devin lui apprit qu'il allait devenir le grand-pre du futur Shensh,
Firdaus prtend que, pour s'assurer la reconnaissance d'Anirwn, le ffqn
vacua la Transoxiane et la Ferghana. Ainsi cette version du mariage d'Ansir-
wn avec la fille du f f q n s'oppose directement celle que nous ont conserve
les traditions I et I I (At-Tabar-Anonyme Sprenger ; Eutychius-Ibn Qutay-
bah) qui dclarent en toutes lettres que c'tait Anirwn qui avait demand la
main de la fille du f f q a n pour obtenir ensuite son aide dans la guerre contre
les Heplithalites. Cependant un heureux hasard nous apprend que, mme dans
la tradition originaire du ps. Ibn-al-Muqaffa', c'tait le Shenh qui avait
pris l'initiative de ce mariage. L'auteur de notre Nihyah a rlabor le
Roman de Bahrm Cbn pour lui donner un ton plus dramatique et il y a
insr de nouveaux pisodes, par ex. les pripties qui marqurent la fuite de
Parwz vers la Syrie aprs que son oncle Bindo l'avait sauv de la poursuite
des cavaliers de Bahrm C b n " et la prophtie d'un moine qui avait lu dans
le Livre de Daniel l'annonce du prochain triomphe des Arabes. Et, de mme,
l'auteur de la Nihyah a imagin que dans le conseil de guerre convoqu par
Hurmuz, que l'offensive du Roi des Turcs avait jet dans l'affolement, l'ancien
conseiller d'Ansirwn Mihransitn (recte Mihrnsitd), dsormais un veillard
dcrpit, avait fait le rcit de son voyage la cour du f f q n dfunt. Il s'y
tait rendu pour apporter une lettre d'Ansirwn qui demandait au f f q n de
lui accorder la main de sa fille. Mais il avait l'ordre de choisir entre ses filles
celle qui avait pour mre la f f t n . E t , tout comme dans le rcit de Firdaus,
mme dans cet pisode le ffqn et la f f t n ont recourt l'artifice de revtir
leur enfant d'une robe use afin de dtourner d'elle l'attention de l'envoy
perse. Plus tard ils donnent leur consentement ce mariage quand un devin
annonce que de cette union allait natre le f u t u r roi de l'Iran. Seulement, dans
cette version, le devin ajoute que ce Roi aura la chance d'tre servi par un
commandant glorieux, dont il donne la description, et que ce hros vaincra
les envahisseurs turcs de l'Iran. Nous donnons dans l'appendice cette page de
la Nihyah car elle suffit prouver que, tout comme l'auteur de notre
Nihyah arabe, l'diteur de la version de ce livre qui forma la base pour
l'histoire d'at-Ta'lib et pour le Shnmah en prose persane mis en vers par
Firdaus, devait avoir trouv lui-aussi dans son modle que c'tait Ansirwn
qui, le premier, avait propos au f f q n cette alliance. Mais cet diteur ren-
versa les rles afin d'exalter davantage la figure du Shensh. E t le fait que
cette modification a t reprise par Firdaus, at-Ta'lib et l'Anonyme du
Mugmal-at-tawrih laisse penser que cette dition avait vu le jour dans le
monde iranien. Par surcrot nous possdons une preuve supplmentaire du fait
que la version ancienne de ce mariage attribuait l'initiative Ansirwn.
99
Ces pripties ont t invontes p a r l'auteur do la Nihyah dans le but d'expli-
quer la politique que i'arwz suivra ensuite vis--vis de la dynastie d ' a l - f j r a h .

N o u s trouvons c e t t e histoire dans une riwyali d ' I b n Balhi 1 0 0 qui est ind-
p e n d a n t e de celles que nous venons d ' e x a m i n e r et qui r e m o n t e son t o u r
l ' p o q u e sassanide, ainsi que le prouve l'accord sur ce p o i n t entre Scbos 1 0 1 e t
al-Mas'd. 1 0 2 Al-Mas'd comme le faisait noter M a r q u a r t , a donn la prin-
cesse t u r q u e le nom de F q . m (variantes Faq.r, Fl.g), mais il a p p a r a t de
Sbos et d ' I b n B a l h que F q . m (Sbos : Kayn, I b n Balh : Qq.m) t a i t
en ralit le nom d u pre de cette princesse. Cependant c e t t e riwyah t a i t
incertains propos d e la nationalit de ce Fq.m qui, selon les uns t a i t le
H q n des Turcs, selon les autres l'un des Rois hazar d e la rgion a t t e n a n t e
Al-Bb wa-l-Abwb. 1 0 3 P r o b a b l e m e n t d a n s sa forme originaire elle p a r l a i t
d ' u n Roi des H a z a r car, dans le rcit de Sbos, K a y n t a i t le roi des T ' e t a -
l a ' i k , un terme, soulignait Marquart, que l'historien armnien a employ p o u r
les H e p h t h a l i t e s et les Hazar, nous a j o u t e r o n s aussi pour les barbares en gnral,
m a i s jamais pour les Turcs.
V) Episode: P a r v e n u la f i n d e sa vie, Ansirwn prfra H u r m u z
ses autres fils cause de ses vertus et d e la noblesse de sa mre. De nos trois
t r a d i t i o n s seulement celle du ps. I b n - a l - M u q a f f a ' contient ce renseignement
qui revient chez t o u s les historiens qui en dpendent, m m e dans l'histoire
d'ad-Dainawar, q u o i q u e ce dernier a v a i t prcdemment racont que H u r m u z
la t t e d ' u n e a r m e perse avait forc Singib vacuer la Transoxiane e t la
F e r g h a n a . At-Tabar, il est, vrai, d o n n e lui aussi cette notice au dbut du cha-
p i t r e sur le rgne de H u r m u z , mais il l ' a interpole dans la tradition I en l'em-
p r u n t a n t au ps. Ibn-al-Muqaffa', qu'il utilisait largement, car la phrase Et sa
m r e tait la fille d u Grand Hqn 1 0 4 n'existe pas dans l'Anonyme Sprenger.
On n o t e r a cependant que Sbos et al-Mas'd citent eux-aussi la riwyah d u
mariage d'Ansirwn avec la fille de K a y n / F q . m propos de H u r m u z p o u r
n o u s assurer qu'il t a i t n de ce mariage.
E t enfin che les auteurs qui d p e n d e n t du ps. Ibn-al-Muqaffa c nous t r o u -
v o n s deux riwyah isoles. 105 Celle d'ad-Dainawar qui a t t r i b u a i t H u r m u z
100 Frgnmah, E d . LE STRANGE & NICHOLSON P. 94 1. 8 95 1. 6. La lacune d a n s

c e t t e riwyah signale par M. WIDENGREN (1. c. p . 81) peut probablement tre comble
-
par le rcit d'al-Baldur (194/13) propos des dissensions qui s'levrent entre Anuir-
w n et le H q n des rgions nord du Bb-al-'abwb.
101
Voir MARKWART : Historische Glossen zu den alttrkischen Inschriften. W Z K M
X I I . MARKWART a donn aussi la traduction de ce passage do Sbos, p . 36 que n o u s
reproduisons.
Murg-ad-dahab ch. X X I I p. 211 tumma inalaka b a ' d a h u - b n u h u H u r m u z
m i n - 'Ani rwn b. Qubd wa ' u m m u h u F q . m ibnatu H q n maliki-t-turk wa qla b a l
malikui) min mulki-1- B a z a x mimm yal al-Bba wa-l-'Abwb.
Voir note 102."
101
wa k n a t ' u m m a t u h u - b n a t u Hqni-l-'akbar.
o pour notre b u t , nous pouvons fairo ici abstraction des riwyah se r a p p o r t a n t
a u x B a z a r , qui r e m o n t e n t elles-aussi au Bodynmah, ainsi que le prouve leur prsence
d a n s toutes nos trois traditions, aux expditions contie Sarandb que BaTam et a t -
T a ' l i b localisent d a n s l'Inde et contre la Chine (Ibn Balh) et la suzerainet iranienne
impose la Chine (at-Ta'lib) et au Tibet (al-Mas r d). Il s'agit de lgendes dpourvues
de t o u t e historicit et qu'on ne rencontre que chez les auteurs les plus rcents.

LA CHUTE D E L ' E M P I R E HEPHTHALITE 239
d'avoir oblig Singib se retirer de la Transoxiane et de la Ferghana et d o n t

nous ignorons la source. Probablement elle appartenait la premire rdaction
de la Nihyah qui, parfois, devait donner deux versions diffrentes du m m e
pisode, comme il en est le cas pour le t h t r e de la bataille entre Bahrm Gr
et le Hqn des Turcs. E t surtout la riwyah de Firdaus que M. Widengren
attribue la ligne IV et dont nous venons de prouver qu'elle contient u n pi-
sode (le mariage d'Ansirwn avec la fille du Hqn) emprunt de n o u v e a u
au ps. Ibn-al-Muqaffa'.
Or cette riwyah pose un problme qui nous oblige revenir sur notre
article Deux documents nouveaux propos de la Lgende de Buzurgmihr
paru dans le tome X X V I des Acta Antiqua. Nous maintenons notre p o i n t
de vue, savoir que le plus souvent le t e x t e de Firdaus remonte une version
en prose persane trs proche de la Nihyah et qui doit donc driver de l ' a r a b e
et non pas directement d'un Uodynmah en pahlv. D'ailleurs nous avons
constat cette affinit mme dans les pisodes du rgne d'Ansirwn. Mais nous
avions tort quand, dans la seconde partie de cet article, pour laquelle nous nous
tions imprudemment appuy l'tude de Christensen, 10 nous avions cru
pouvoir avancer que mme les Conseils de Buzurgmihr dans le Shnmah
de Firdaus drivent du texte arabe conserv par Ibn Miskawayh. Ne connais-
sant pas le pahlv, nous prsentions nos conclusions comme une simple hypo-
thse qu'il a p p a r t e n a i t aux iranologues d'examiner. Ensuite le prof. M a h y a r
Nawab nous a dmontr notre erreur. Maintenant, d a n s l'pisode d'Anir-
wn et de la guerre contre les Hephthalites nous nous trouvons en prsence
d'un cas assez analogue. La description de la mission charge de choisir p o u r
Ansirwn sa femme entre les filles du H q n a un p e n d a n t exact d a n s le
chapitre de B a h r m Cbn dans la Nihyah. 107 P a r contre, le rcit de la guerre
entre les Turcs et les Hephthalites ne correspond aucune de nos traditions.
Elle donne au Roi des Hephthalites le nom de Gatfr qui rappelle invitable-
ment celui de KaxovXcpo dans le f r a g m e n t de Mnandre. Elle attribue la d f a i t e
des Hephthalites aux seuls Turcs, elle prtend que, par dsespoir, la noblesse
hephthalite dcida de se soumettre au Shenh de l ' I r a n et que le H q n
vacua la fin la Transoxiane et la F e r g h a n a pour s'assurer l'amiti d'Ansir-
wn. Nous noterons que ce rcit de Firdaus, qui l'attribue un dehqn illustre,
est celui qui rabaisse le plus le rle de l ' I r a n dans l'limination de l ' E m p i r e
hephthalite. Il y a des changes qu'il nous ait conserv une ancienne riwyah
106
Dans son tude La lgende du Sage Buzurgmihr (A. 0 . 1 9 2 9 p. 81 123) CHRISTEN-
SEN se rapporte a u x texte du P a n d n m a k dit par Dastr Peshotanji en dpit d u fait
que WEST dans son article publi dans le Muson, Notes sur quelques petits t e x t e s poh-
levs avait d j prvenu que le texte original s'tend seulement treize lignes au del d u
paragraphe 119 do l'dition du Dastr P e s h o t a n j i et que la continuation donne p a r ce
d a s t r ( 121 169) appartient un autre ouvrage.
107
E d . de Calcutta p. 16971700 (L'offre d ' u n maiiage) et p. 1 7 0 0 - 1 7 0 3 (La
mission de Mihrnsitd).

240 M. GRIGNASCHI
sassanide i n d p e n d a n t e de nos trois t r a d i t i o n s et qui refltait fidlement la

m a r c h e des vnements. Toutefois n'en sommes-nous pas certain. E n effet,
d a n s les pisodes qui nous ont t conservs, la tradition du ps. I b n - a l - M u q a f f a '
ne p a r l e pas explicitement de la guerre perso-hephthalite. Elle donne seulement
u n e liste des provinces recouvres p a r A n i r w n au d b u t de son rgne e t
d o n t elle savait qu'elles avaient t sous la domination des Hephtlialites. 1 0 8
Mais ensuite elle f a i t aussi allusion au fait q u e la plus grande p a r t i e de l'arme
h e p h t h a l i t e avait t crase par le TJqn. 1 0 9 E t , par consquent, rien ne s'op-
pose ce que les pisodes raconts par le seul F i r d a u s (la guerre turco-hephtha-
lite, l'lection de F a g n i s , l'arrive d ' A n s i r w n dans le Horassn, l'vacuation
d u Sogd, de S a m a r k a n d e t de g par le H q n et l'hommage de Fagnis
Ansirwn) aient f i g u r d j dans le rcit d u ps. Ibn-al-Muqaffa r . N o u s
s o m m e s d ' a u t a n t plus enclin le croire que F i r d a u s nous a conserv certains
pisodes qui doivent a p p a r t e n i r au R o m a n d e Bahrm Cbn, que nous ne
lisons plus dans notre Nihyah mais qui d e v a i e n t figurer d j dans la pre-
mire dition de ce livre.
Nous devons dsormais nous d e m a n d e r quelle est la valeur historique de
ces riwyah, d o n t nous avons nous sommes efforc jusqu'ici d ' t a b l i r l'origine
e t les rapports, et nous commencerons p a r l'pisode V de la tradition du ps.
I b n - a l - M u q a f f a ' (l'origine t u r q u e de H u r m u z ) car, mme s'il s'agit d ' u n pro-
b l m e marginal ne c o n c e r n a n t pas la c h u t e d e l'Empire hephthalite, il p e r m e t
de constater c o m m e n t les historiens perses connaissaient m a l les vnements
d e la Cour sassanide e t c o m m e n t ils faisaient circuler des relations tendancieuses
d a n s des b u t s propagandistes. On sait que t o u t e s les versions conserves d a n s
les sources arabo-persanes font de la mre d ' A n s i r w n la fille d ' u n dehqn, la
classe de la petite noblesse favorise p a r Q u a b d I et p a r JJosr I alors, que
d ' a p r s Procope, elle t a i t la sur de l ' A s p a h b e d , c'est--dire d ' u n m e m b r e de
la p l u s haute noblesse. 1 1 0 Or, nous nous t r o u v o n s en prsence d ' u n cas assez
analogue. Nous ne connaissons pas la source de la riwyah d ' a d - D a i n a w a r
qui, t o u t en a c h e v a n t son chapitre sur A n s i r w n par le rcit d u choix de H u r -
m u z comme f u t u r Shensh cause du r a n g de sa mre, la fille du H q n ,
a v a i t a u p a r a v a n t a t t r i b u au mme H u r m u z le mrite d'avoir guid une arme
perse contre le f j q n des Turcs et de l'avoir forc vaquer la Sogdiane et la
F e r g h a n a . Cependant il ne serait pas impossible que cette riwyah ait a p p a r t e -
n u une tradition a u j o u r d ' h u i perdue se p r o p o s a n t d'exalter la figure de H u r -
m u z , d ' a p r s les a u t r e s traditions perses u n t y r a n cruel, digne de son aeul
t u r c . Sbos le d i t en t o u t e s lettres : Il t a i t encore plus grand que ses ascen-
d a n t s paternels et plus g r a n d et plus sauvage q u e ses ascendants maternels. 1 1 1
108
Voir l'pisode 3.
109
Voir l'pisode 3.
110
De hello persico Livre I eh. 11.
111
Traduction MARKWART : 1. c. Rappelons que Firdaus revient dans plusieurs
pisodes sur l'origine t u r q u e de Hurmuz pour en expliquer la cx'uaut.
Acta Antiqua Acudemiae Scientiarum Hungrii 28, 1980

E t , dans nos extraits du Hodynmah, on relve encore deux tendances

opposes que les historiens musulmans se sont efforcs de concilier : le dsir
de mettre en relief la protection que ce Roi accordait au menu peuple, un dsir
conforme l'idologie du TJodynmah qui s'efforait toujours de voiler les
dfauts des Rois lgitimes et de leur attribuer des mrites, et le jugement
svre qu'en donnait le R o m a n de Bahrm Cbn qui, t o u t en reprsentant ses
hros comme les victimes de la fatalit, ne cachait pas leurs dfauts.
Or, la possibilit que Hurmuz ait jou un rle quelconque dans la guerre
perso-hephthalite pour qu'on lui confit en 559/60 le commandement, ne
ft-ce que nominel, de l'arme contre les ex-allis turcs il aurait d tre n vers
545 au plus tard ne saurait tre retenue la lumire d u rcit de J o h a n n e s
d'Ephse. E t sur ce point cet historien mrite notre confiance puisqu'il rap-
porte des vnements, dont le dernier acte s'tait droul Constantinople
son poque. E n p a r l a n t du choix de Hurmuz (Ormazd IV) comme successeur
de Hosr, il emploie les termes de puer arrogans et immanis et rationis exi-
guus. 112 De son ct, Michel Syre, reprenant les renseignements de Johannes,
note qu'il tait un jeune homme bouillant, cruel et de peu de cervelle. 113
On prfrera les termes de la traduction de Chabot, savoir qu'au moment de
son arrive au trne en 579 Hurmuz tait encore un jeune homme puisque,
quand en 590 son fils Hosr I I se rfugia dans la Syrie byzantine, il tait ac-
compagn par les femmes de son harem avec leurs ourissons. 114 Donc en 579
H u r m u z ne pouvait pas tre proprement parler, un puer.
E t d ' a u t r e part, les renseignements qu'on t r o u v e dans l'histoire de
Thophylacte et certain dtails du R o m a n de Bahrm Cbn rendent difficile
d'accepter l'historicit de la riwyah qui fait de H u r m u z le fils d'Anirwn
et de la fille d'Istemi, mme si cette possibilit ne saurait tre tout fait carte.
E n 589 Hurmuz proposait la noblesse perse de mettre sa place son fils cadet,
un yeigxiov dit Thophylacte, 1 1 5 donc un garon de 15 ans au moins, l'ge
auquel commenait la majorit d'aprs le Pandnmak-i-Zaradut. l i e D a n s
cette anne-l Hosr I I , dj pre de plusieurs nourissons, devait avoir au
moins 17 ou 18 ans. E t , dans le Roman de Bahrm Cbn on raconte l'tonne-
ment du hros la vue de l'allure royale et de la barbe qui ornait dsormais le
visage du nouveau roi Parwz. Ce dernier devait donc tre n au plus t a r d en
571 et il est difficile de croire que son pre ait pu tre le fils de la princesse
turque qu'Ansirwn avait pouse en 556 si l'on prte foi at-Tabar, en 560
si on accepte l'indication d'Eutychius. E t d'ailleurs Johannes d'Ephse et
Michel Syre ne savent rien de la mre de Hurmuz et nous racontent une his-
112
Hist. Ecel. P a r s I I I Livre VI ch. X X I I (trad. BROOKS p. 243).
113
Livre X ch. 20. Trad. CHABOT tome I I p. 353.
114
Histria L. IV ch. 10.
115
Historia Livre IV ch. 4.
116
E d . et t r a d . FREIMANN 27. Le texte n'est c e p e n d a n t pas explicite.

242 M. G R I G N ASCHI
toire fort diffrente de celle de la tradition du ps. Ibn-al-Muqaffa'. D'aprs

ces auteurs c'tait l'assemble de la noblesse perse (avyxXtjro) qui avait impos
U s r I la dsignation d'Ormazd, le fils cadet et Hosr I avait donn de
l ' a r g e n t son fils an qu-il prfrait pour qu'il se s a u v t et qu'il ne f t pas
mis mort 117 . P l u s tard un impostor quidam astutus ex ipsis Persis, puer
adulencens 1 1 8 arriva Constantinople et se fit reconnatre comme l'hritier
lgitime de Hosr. A la fin il f u t dmasqu par le Spatharius du Roi de Perse
q u i avait trouv son tour refuge la Cour byzantine. 1 1 9
Ainsi les riwyah recueillies par le ps. Ibn-al-Muqaffa' sur la participa-
tion de Hurmuz la guerre turco-liephthalite et sur son origine turque n ' o n t
a u c u n e chance d'avoir une base historique. E t on ne saurait p a s non plus prter
foi la tradition de Sbosal-Mas r d car les sources byzantines et le Krn-
mag-i-AnSirwn s'accordent prsenter les populations nord du Caucase
c o m m e des peuples barbares sans organisation et ayant leur t t e des roitelets
d n u s de prestige, avec lesquels le Shensh ne se serait j a m a i s emparent.
Q u a n t Istemi, il ne fit son apparition dans le Caucase qu'aprs 572 et, de toute
vidence, Hurmuz ne pouvait tre issu d'une union scelle cette poque.
T o u t e s ces riwyah doivent appartenir des romans populaires tels que celui
d e B a h r m Cbn et on voit combien la prudence s'impose d a n s l'utilisation
de nos sources, m m e quand elles remontent, comme il en est dans notre cas,
l'poque sassanide. P a r contre on peut faire confiance la riwyah des
traditions d'at-TabarAnonyme Sprenger, EutychiusIhn Q u t a y b a h lorsqu'
elles nous racontent qu'Ansirwn avait pous la fille d ' I s t e m i en vue d'obtenir
son appui contre les Hephthalites. Ce rcit cadre bien avec t o u t ce que nous
savons de la marche des vnements et, surtout, il est confirm p a r un document
conserv par Moses Dasxuranci, un historien qui possdait des sources arm-
niennes contemporaines pour l'histoire de Hraclius et de H s r ( 1 II- Dans
Fhistoire des Agouans nous lisons le t e x t e d'une lettre envoye par Hosr
I I a u R q n des Turcs, o le Roi perse rappelle ce dernier les mariages qu'unis-
s a i e n t la Maison des H q n celle de l ' I r a n mais o il ne p r t e n d pas toutefois
d ' a v o i r lui-mme d u sang turc. 120
Cependant ce ne sont pas seulement les riwyah d r i v a n t des romans
populaires de l'poque sassanide qui sont suspectes. Nous constatons le mme
m a n q u e d'historicit dans la liste des provinces hephthalites annexes par
Ansirwn et cela quoique sur ce point toutes nos traditions soient au fond
concordes et qu'elles a t t r i b u e n t toutes Ansirwn d'avoir annex la Trans-
oxiane et la F e r g h a n a elles-mmes. L a liste la plus dtaille des conqutes
d'Ansirwn nous a t conserve par le ps. Ibn-al-Muqaffa' et elle n'inspirait
117
Michel Syre Livre X eh. 20 Trad. CHABOT t. I I p. 353.
118
Johannes d'Ephse Hist. Eccl. pars I I I , Livre VI ch. 29.
119
Michel Syre L. X ch. 20.
120
Trad. PATKANOV Livre I I ch. 12.

p a s de la confiance Nldeke. 1 2 1 E t , en fait, cette tradition renferme une contra-

diction de fond. Elle prtend t o u t d ' a b o r d qu'Anirwn a v a i t conquis le Sind,
le Bost, l'Arachosie, le Zablistn, le Tohristn, le K b u l i s t n et le Dardis-
tn. 1 2 2 Ensuite elle raconte incidemment que le TJqn des Turcs avait conquis
t o u s les pays des Hephthalites l'exception de ceux qu'Anirwn avait d j
occups. 1 2 3 Il est vident que si Anirwn avait annex toutes les rgions
numres dans cette riwyah, I s t e m i n ' a u r a i t fait q u ' a t t a q u e r un Roi qui
a v a i t d j perdu le centre de son E m p i r e e t qui ne se serait m a i n t e n u que d a n s
la Transoxiane. E t nanmoins cette tradition n'arrive pas cacher que l ' a t t a q u e
principal contre les Hephthalites avait t men par les Turcs. Aussi serait-on
t e n t d ' a t t r i b u e r cette liste quelque p a r t i s a n de la S u ' b i y y a h , un m o u v e m e n t
auquel participait le ps. Ibn-al-Muqaffa r . E t cela d ' a u t a n t plus que, ainsi
que le faisait remarquer Marquart, 1 2 4 le t e r m e de K b u l i s t n a p p a r t i e n t au
no-persan et q u ' l'poque sassanide on parlait du Royaume de Kbul.
Cependant cet indice n'est pas dcisif, car nous ne possdons que des t r a d u c -
tions du I I sicle et du I I I sicle de l'hg., dont les a u t e u r s pouvaient t o u t
naturellement remplacer l'expression ancienne par sa f o r m e moderne. Ainsi
nous rencontrons le terme de Kbulistn jusque dans la traduction d u
Krnmag-i-Ansirwn, un ouvrage a p p a r t e n a n t sans contredit possible
l'poque sassanide. Rien ne s'oppose donc ce que cette liste remonte relle-
m e n t l'poque perse et ce qu'Anusirwn, qui faisait imprimer sur ses mon-
naies la lgende ghn apbm kartar, 1 2 5 ait fait aussi circuler quelque
fathnmah n u m r a n t ses prtendues conqutes. 1 2 6
Q u a n t au problme de savoir si la liste originale contenait rellement le
D a r d i s t n et le K b u l i s t n dans les mss. d ' a t - T a b a r I 894 on lit D.rwstn e t
K r . s t n nous ferons noter que la leon Kbulistn est confirme p a r les
ditions les plus rcentes du ps. I b n - a l - M u q a f f a ' , mais nous avons d j con-
s t a t comment ces ditions remanient souvent le t e x t e original q u ' a t - T a b a r
a v a i t encore sa disposition. D a n s ces conditions, il sera p e u t - t r e plus p r u d e n t
d ' a c c e p t e r la leon de BaTam Gargistn, la province sur le h a u t M u r g b
qu'Anirwn a certainement annexe et la lecture propose par M a r q u a r t
Zablistn au lieu de Drwstn. 1 2 7 Il est p a r contre probable que la liste originale
contenait aussi le Oagniyyn, ainsi qu'il en ressort d'ad-Dainawar. M. Widen-
121
NLDEKE : Geschichte der Perser u n d der Araber zur Zeit der asaniden
p . 156 n. 1.
122
Episode I.
123
Episode I I I .
121
EranSahr p. 32 n. 3.
125
Celui qui dlivre le monde do la pour. Voir WIDENGREN 1. c. p. 94 n. 2.
126
Rappelons que la Nihftyah nous a conserv, ne ft-ce que sous uno forme
modifie, le fathnmah que Bahrm Gr adressa ses peuples aprs la victoire sur le
H q f m turc. Mmo cette pice semble tre emprunte un roman populaire.
127
Voir MARKWART : Eranahr p. 32 n. 3 et HARMATTA : L a t e Bactrian Inscrip-
tions. A c t a Antiqua 17 (1969) p. 401 et 402 et note 71.

244 M. G R I G N ASCHI
g r e n considre c o m m e extrmement probable que les provinces indiques

d a n s la tradition d u ps. I b n - a l - M u q a f f a ' o n t t occupes rellement p a r An-
irwn mais qu'elles f u r e n t perdues ensuite, peut-tre l'poque de H u r m u z d
IV. 1 2 8 E t le s a v a n t sudois justifiait sa conviction p a r la description des fron-
tires de l'Iran d a n s la Lettre d e T a n s a r qu'il considre avec Christensen com-
m e a p p a r t e n a n t e l ' p o q u e de Hosr I 1 2 9 mais que nous prfrons a t t r i b u e r
a u x toutes dernires annes du R o y a u m e Sassanide. 1 3 0 Cependant a u j o u r d ' h u i ,
g r c e au Krnmag-i-Ansirwn et a u x sources byzantines, il est possible
d e prouver que la plus grande partie de ces provinces n ' a p p a r t e n a i e n t pas
l ' I r a n en 568 et on ne saurait penser qu'elles aient t vacues dans la priode
560 568 car, ainsi q u e nous l ' a p p r e n d Mnandre, p e n d a n t cette priode les
r a p p o r t s perso-turcs s'taient dtriors cause des obstacles crs par la Perse
a u commerce turco-sogdien de la soie, mais on n ' t a i t p a s arriv une guerre
ouverte. 1 3 1 L'pisode I X du Krnmag-i-Ansirwn nous apprend qu'
c e t t e d a t e (468) Anirwn reut les hommages du seigneur d u Rwrizm et d u
R o i de l'Inde, de D w a r et du Kbulistn. 1 3 2 Donc ces p a y s n'avaient t jamais
a n n e x s l'Iran, ce qui est par surcrot confirm p a r le fait que les missions
des Napki-Malka de K b u l ne prsentent aucune interruption cette poque 1 3 9
e t que nous possdons t o u t e une srie de monnaies des Rois du IJwrizm avec
la lgende M.RA M . L K HAR.ZM et les noms, malheureusement d ' u n e lecture
d o u t e u s e , des souverains qui les mettaient. 1 3 4 E t le m m e pisode I X du
Krnmag raconte que, dans cette anne-l, Ansirwn autorisa des popula-
t i o n s turques, sans d o u t e principalement des Sabir mais peut-tre aussi des
t r i b u s pseudo-Avar, entrer dans l ' I r a n et qu'il c a n t o n n a les auxiliaires enrls
p a r m i ces nomades d a n s les territoires des Alains, d a n s l ' A z e r b a d j n et d a n s
la rgion de Marw. Difficile de se soustraire l'impression qu'il s'agissait
d e rgions frontalires e t que, mme en 568, les confins de l ' I r a n ne dpassaient
la rgion de Marw, Marw-i-rd et T a l a q n . De plus, en 570 Mnandre enrgistre
u n e invasion t u r q u e d e la Mdie. 135 Comme en 569, au m o m e n t o Zemarchos
r e n t r a i t Constantinople, les troupes perses contrlaient encore le territoire
128
L. c. p. 9 0 - 9 1 .
129
L. c. p. 84 et n o t e 2.
130
Voir r e m a r q u e s au Testament d'Ardasr dans le J A I960.
131
De legationibus gentium frag. 18.
132
Nous profitons de cette occasion pour corriger notre note 74 do la page 41
d u J A 1966. Mrne si at-Ta'lib (d. ZOTENBERG p. 615) et B a l ' a m (Ed. ZOTENBERG
I I p . 221) situent S a r a n d i b dans la valle infrieure de l'Indus, il est plus probable que
l ' a u t e u r sassanide songeait aux souverains des grands centres commerciaux sur la route
de la Chine : Ceylan et la ville de Kelt.
133
Voir GHL 1. . Vol I p. 68 et suiv. Nous devons cependant rappeler que M.
MITCHINER dans son article Who were Napki Malik (East and West X X V 1976) est enclin
a d m e t t r e l'historicit do la conqute perse et considre que les missions de Napki
Malka s'achvent vers 560.
134
M. MITCHINER : The early coinage of Central Asia p. 48.
135
Mnandre De Leg. Gentium fr. 23 (d. DE BOOR p. 459).

des Alains 138 et que Derbent tait puissamment fortifi, il est peu prs certain
que l'arme t u r q u e f i t irruption dans la Mdie par la route allant de Qmis
R a y . E t ce renseignement de l'historien byzantin est confirm par le Krn-
mag. Dans l'pisode X, qu'une allusion aux embassadeurs byzantins envoys
auprs d'Istemi permet de dater de 571 ou 572, Ansirwn rappelle un revers
subi par les Perses au cours d'une a t t a q u e rcente des Turcs et il parle des me-
sures prises pour m e t t r e en t a t de dfense les villes et les forteresses du f f o -
rassn. Mme si la Gographie Armnienne inclue Balh et le Tohristn d a n s
le fforassn, il est difficile de croire qu'Ansirwn parlait ici de ces rgions
loignes.
E t enfin la mme date (568) Maniach, l'ambassadeur sogdien d ' I s t e m i
Constantinople, assurait Justin I I que les villes des Hephthalites p a y a i e n t
le tribut aux Turcs. 137 Cette dclaration corrobore p a r le fait qu'Istemi se
servait d'un ambassadeur d'origine sogdienne, prouve dfinitivement, comme
l'enseignait Nldeke, que la Transoxiane n ' a p p a r t e n a i t point l'Iran.
Ainsi de toutes les provinces de notre liste Ansirwn n'avait recouvr en
ralit que l'Arachosie et le Zablistn, 1 3 8 qui d'ailleurs taient probablement
sous l'autorit des dynasties hephthalites locales, ainsi que la Nihyah le
laisse entendre. De plus, il avait sans doute rtabli son autorit sur les rgions
de H r a t et de Bagds. 1 3 9 A notre avis, il y aurait une seule interprtation qui
permettrait de considrer ces listes des provinces recouvres comme quelque
chose d ' a u t r e q u ' u n faux mis en circulation de propos dlibr. Une p a r t i e
importante de la noblesse hephthalite, telle que le roi Eagnis de la riwyah
de Firdaus et le prince KaxovXtpo de l'histoire de Mnandre, avait sans d o u t e
prfr se soumettre l'Iran, dont elle avait absorb la culture, 1 4 0 plutt q u ' a u x
barbares turcs. E t ds lors il serait concevable qu'Ansirwn se soit considr
comme le suzerain lgitime des rgions que les Rois hephthalites avaient info-
des ces princes.
136
137
Mnandre De leg. gentium frag. 21 d. DE BOOR p. 453 4.
Mnandre De leg. gentium frag. 18 E d . DE BOOR. p. 451 452. On notera q u e
Maniaeh affirme seulement que les Turcs avaient oblig les Hephthalites leur p a y e r
des tributs. Il ne parle p a s d'une occupation directe de leurs villes qui taient videm-
m e n t restes sous l'autorit de la noblesse locale hephthalite. De son ct, la t r a d i t i o n
nationale tou-kiue (inscription de Kltegin cte est ligne 9) considrait les Portes d e
Fer comme le point extrme de l'expansion turque l'poque de la Grande D y n a s t i e .
138
Voir GBL : 1. c. Vol I p. 68 et suiv. ; en particulier propos de l'occupation
de Kapia la fin du V I sicle par les Napki qui possdaient d j Gazn. Il est intressant
de noter que mme le D w a r faisant partie du Zablistn n ' a p p a r t e n a i t pas Anirwn
quoiqu'on 508 Qubd I et pris d'assaut le sanctuaire de cette rgion, le Tovveeg
(MARQUART : D a s R e i c h Z a b l p . 2 7 9 2 8 7 ) .
no Voir MARQUART: Eranahr p . 61 62 propos d u problme de l'absence des
vques nestoriens de H e r a t aux conciles nestoriens de la fin du V sicle et de la premire
moiti du VI sicle. Cette rgion, ajoutons-le, tait occupe par le QadiS (Eranahr
p. 77) qui, ainsi que CZEGLKDY l'a prouv (Acta Orient. H u n g a r i c a t. IV fasc. 1 3 p . 63),
appartenaient aux tribus chionites.
140
La langue des fragments hephthalites dchiffrs p a r O. HANSEN est u n
dialecte iranien.

L e bilan de ce que les sources perses p e r m e t t e n t d'tablir, u n e fois car-

tes les riwyah tendancieuses ou dfigures p a r les remanieurs et les faux
d e la propagande officielle, semble premire vue assez maigre. Mais elles
n o u s o n t conserv q u a n d mme quelques renseignements prcieux qui per-
m e t t e n t d'entrevoir le droulement des vnements. Elles nous livrent le
t e r m i n u s post quem p o u r la chute de l ' E m p i r e Hephthalite e t elles compltent
ainsi les sources byzantines. Elles nous o n t appris que ce f u t Ansirwn qui
p o u s s a Istemi contre les H e p h t h a l i t e s en d e m a n d a n t la m a i n de sa fille, une
d e m a n d e qui devait f l a t t e r le H q n t u r c qui avait t j u s q u ' 552 un obscur
vassal des Jeou-jan et qui obtenait ainsi u n prestige comparable celui que le
m a r i a g e avec une princesse des Si-We a v a i t assur son frre an Tou-men.
Elles nous ont aussi rvl que ce f u r e n t les armes turques qui p o r t r e n t t o u t
le p o i d s le la guerre et qui occuprent la p l u p a r t des territoires hephthalites.
Mais le succs r e m p o r t p a r le S h e n s h n ' t a i t pas moins g r a n d . Il avait
a n a n t i l'ennemi traditionnel de l ' I r a n on remarquera que les paroles du
plnipotentiaire perse Yezdegusnar Zieh a u dlgu byzantin P e t r u s magister
o f f i c i o r u m , prises la lettre n'impliquent p a s de conqutes territoriales 1 4 1 ,
il a v a i t recouvr des territoires aussi i m p o r t a n t s que Hrat, Bagds, l'Arachosie
e t le Zablistn et il a v a i t obtenu l'acte d e soumission d ' u n e p a r t i e considrable
d e la noblesse h e p h t h a l i t e . E t , parmi nos sources, la plus i m p o r t a n t e est sans
c o n t r e d i t possible le Krnmag-i-Ansirwn, dont l'auteur a t probablement
le seul crivain perse a y a n t accs a u x archives royales, t a n d i s que, entre les
t r a d i t i o n s drivant d u H d y n m a h , celle d u ps. I b n - a l - M u q a f f a ' est la plus
riche, m m e si elle risque de fourvoyer les historiens p a r son caractre htro-
gne, ses e m p r u n t s a u x romans populaires et les remaniements, auxquels elle
a t soumise. 142
Trieste.
141
Mnandre : De lgat. R o m . f r a g m . 11 (d. DE BOOR p. 177).
142
A cette tradition appartient aussi a t - T a ' l i b qui, dans son chapitre sur An-
irwn, donne deux notices provenant de la tradition du ps. Ibn-al-Muqaffa' (Ed. ZOTEN-
BERG p . 615 Ta'lib I et I I I de WIDENGREN) et une notice a p p a r t e n a n t la tradition
d ' a t - T a b a r A n o n y m e Sprenger (Ed. ZOTENBERG p. 615 Ta'lib I I de WIDENGREN).
On n e s'tonnera pas trop, car il est bien connu qu'at-Ta'lib, tout en suivant uno ver-
sion parallle celle de Firdaus, a employ aussi la Chronique d ' a t - T a b a r . Nous ne
s a u r i o n s partager entirement l'opinion de WIDENGREN (1. c. p. 89), savoir que l'ana-
lyse d e s riwyah se r a p p o r t a n t la chute des Hephthalites ne confirme p a s la thorie
de ZOTENBERG qui a d m e t t a i t que Firdaus et at-Ta'lib ont suivi des narrations dis-
t i n c t e s remontant une source commune. E n fait les deux auteurs ont donn seulaments
u n relief diffrent aux pisodes IV du ps. Ibn-al-Muqaffa'. Ce q u ' o n ne trouve pas
d a n s le rcit d'at-Ta'lib c'est la description de la guerre turco-hephthalite qui caractrise
la version do Firdaus. On pourrait voir d a n s c e t t e lacune dans la narration d'at-Ta'lib
uno r a i s o n pour attribuer cette dernire riwyah non pas au ps. Ibn-al-Muqaffa' mais
u n e tradition indpendante.
Acta Antiqua Aeademiae Scientiarum Hungarieae 28, 1980

LA CHUTE D E L ' E M P I R E HEPHTHALITE 247
APPENDICE
LA MISSION D E M I H U N S I T D
(DANS L E MS L^*.) D ' A P R S LA NIHYAH

248 M. G R I G N A S C H I : LA C H U T E D E L ' E M P I R E HEPHTHALITE
Cambridge ms. Qq 255 d a t 1024 h (1615) = i!

B r i t i s h L i b r a r y Arabes Add. 23 298 = J
Acta Antiqua Academiae Scientiarum Hunaaricae 28.198 )

I. ECSEDY
WESTERN TURKS IN NORTHERN CHINA

IN THE MIDDLE OF THE 7TH CENTURY
1. EAST AND W E S T I N ANCIENT CHINA
E a s t and W e s t are s e c o n d a r y in a n c i e n t Chinese o r i e n t a t i o n . Their p r o p e r

i n d i c a t i o n d e p e n d e d o n t h e m a i n d i r e c t i o n s of South a n d N o r t h , i.e. t h e s t a r t -
i n g a n d r i t u a l direction of S o u t h r e f e r r i n g t o early times if o n l y for cultic r e a s o n s
on t h e one h a n d , a n d t h e politically i m p o r t a n t N o r t h w i t h a symbolic imperial
rule of t h e P o l a r S t a r over t h e s k y a t n i g h t o n t h e o t h e r h a n d . N a t u r a l l y , t h e
l a t t e r , i.e. t h e e x a c t d e t e r m i n a t i o n of t h e N o r t h u p o n t h e P o l e - S t a r c a n n o t b e
earlier t h a n t h e a p p e a r a n c e of t h e P o l a r i s in t h e position of o u r N o r t h P o l e -
S t a r in t h e late 2nd millennium B . C . a t least. The b e g i n n i n g s of t h e Chinese
s c r i p t a b o u t t h e 1 7 t h 1 6 t h c e n t u r y B . C. show a S o u t h - f a c e d o r i e n t a t i o n ,
b e c a u s e t h e p i c t o g r a m m of t h e w o r d f r o m t h e Y i n p e r i o d (prior t o t h e f i r s t
m i l l e n n i u m B.C.), Vzi ' N o r t h ' r e p r e s e n t s t w o h u m a n f i g u r e s back t o b a c k ,
a n d it m e a n t ' N o r t h ' as well as ' b a c k side', ' t u r n t h e b a c k ' e t c . 1 I n t h e t i m e
p r e c e d i n g t h e celestial rule of t h e P o l e - S t a r , i.e. in t h e South-facing p e r i o d ,
S o u t h e r n constellations, n a m e l y A n t a r e s a n d its n e i g h b o u r h o o d in t h e Scorpio
could be t h e base of o r i e n t a t i o n , b u t t h i s is a n o t h e r p r o b l e m , being b e y o n d t h e
scope of this p a p e r . 2
T h e direction of N o r t h S o u t h o u g h t t o be k n o w n in C h i n a for o f f i c i a l
p u r p o s e s a n d in a u t h e n t i c f o r m , also f o r building houses, f o u n d i n g a s e t t l e -
1
Cf. . KARLGREN : Grammata Serica Recensa. BMFEA 19 (1967) No. 909ad ;
e) : when being added the classifier key of 'body', the character jUf pei meant 'the back,
posterior part' or 'turn the back on, cheat', from the Book of Songs (Shih-king) on;
and, f) added another key, i.e. that of 'man', pei meant 'turn the back on, abandon',
'not to face straightly', fiom the Han-period onwards (as it occurs in the book of cere-
mony Li-chi).
2
On an old cult of Antares ('Star of Fire') see H . MASPERO : La Chine antique.
Paris 1966. 197(198), 510( 511) ; B. KARLGREN : Legends and cults in ancient China.
BMFEA 18 (1946) 243 244. On a Sino-Tibetan or Tibeto-Burmese folklorie tradition
concerned and the related Chinese records see m y papers : On a few traces of ancient
Sino-Tibetan contacts in the early Chinese mythic tradition : Proceedings of tho Csoma
de Krs Memorial Symposium, held a t Mtrafred, Hungary, 2430 September 1976,
E d . by L. LIGETI. Budapest 1978. 89 99; Far Eastern sources on the history of the
steppe region: Memorial Volume devoted to Paul Demiville. B E F E O L X I X (1981)
268 269 (and Note 11).
Acta Antiqua Acadcmiae Scientiarum Hungaricae 28, 1980

250 I. ECSEDY
m e n t , fixing r i t u a l points or lines etc., 3 b u t t h e exactitude of E a s t e r n or West-

e r n direction could seem less i m p o r t a n t , and, as seen above, its astronomic
d e t e r m i n a t i o n m u s t have been secondary, and still l a t e r t h a n t h a t of t h e
N o r t h South P i v o t of t h e Universe, b o t h d a y and n i g h t . T h a t is why t h e
a n c i e n t practical m e t h o d used in China, too, when d e t e r m i n i n g the cardinal
directions according t o t h e sun, to its light a n d shadow on t h e e a r t h as concerns
S o u t h (and North), 4 a n d the celestial r o u t e of the sun as concerns E a s t a n d
W e s t , could r e m a i n widespread in case of t h e latter for a longer period, as it is
r e f l e c t e d as late as e.g. in the historical records of the 1st millennium A. D .
T h e starting and e n d i n g points of the a p p a r e n t way of t h e s u n could serve for
a practical fixing of E a s t and West respectively, the place of sunrise being in
t h e E a s t , while the place of sunset showing West.
We can see it e.g. in the letter of t h e Persian ruler sent t o the Chinese
e m p e r o r around 520 A. D.; having s u r v i v e d in Chinese records, its Chinese
t e x t (translation) calls China's ruler t h e Son of Heaven in t h e Orient, expres-
sing E a s t as sunrise place, i.e. the place where the sun rises I tried t o show
its background in a p a p e r submitted t o t h e former meeting of this same Con-
ference, in 1976. 5 A similar wording is characteristic of a n o t h e r famous letter
of diplomacy in t h i s period, namely t h e l e t t e r of the ruler of J a p a n sent t o t h e
Chinese E m p e r o r in 608; it became of ill-fame in Chinese historiography for
t h e impertinence calling the emperor Y a n g - t i of the Sui D y n a s t y the Son of
H e a v e n of the place (i.e. land) of sunset, while calling himself to be exact :
herself the Son (Child) of H e a v e n of t h e place (land) of t h e sunrise. The
l e t t e r sent by E m p r e s s Suiko can well reflect a foreign, say, J a p a n e s e wording,
if n o t translating a message from J a p a n e s e , as in case of t h e Persian ruler's
l e t t e r when a similar supposition could be raised. 6
3
On the ancient Chinese cities, capitals of principalites etc., divided by North-
S o u t h directed lines a n d representing or symbolizing a universal order, see. P. WHEATIEY :
T h e Pivot of the F o u r Quarters [of the Universe], A Preliminary I n q u i r y into the Origins
a n d Character of t h e Ancient Chinese City. Chicago 1971. 423 436, 460 468.
4
Even in the middle of the Chou period (11th 3rd centuries B. C.), in the histoti-
cal time of astronomical observation and orientation, as shown by ritual books (compiled
a s late as the Han-time, 206 B. C. 220 A. D.), the credit of their traditionally respected
t e x t s being supported by related astronomical records found by recent archaeological
excavations in China concerning the Chou period, too.
5
Cf. Early Persian envoys in the Chinese courts (5th 6th centuries A. JD.).
Studies in the Sources on the History of Pre-Islamic Central Asia. E d . by. J . HARMATTA
(Collection of the Sources on the History of Pre-Islamic Central Asia. Series I. Vol. I I .
E d . b y J . HARMATTA.) Budapest 1979. 153 162 (East as the place of sunrise : 160161,
a n d Note 33).
' Here I express m y thanks to Professor ROY ANDREW MILLER (University of
Washington, Seattle) for his private letter of August 15, 1980, drawing my attention to
t h i s interesting parallel of the Persian ruler's message, shown in m y paper mentioned

WESTERN TURKS IN N O R T H E R N CHINA 251
We may well suppose a Japanese world-view behind the addresses of

the two rulers residing in the Far East from our angle , because it is hid-
den even in the name of country of the envoys in question 0 Nihon
(Nippon) root or the sun, i.e. the place where the sun originates (the Euro-
pean name Japan reflecting a Chinese pronunciation Jih-pen of the same
Chinese characters). The inhabitants of J a p a n might consider their country the
easternmost land of the world known for them, every other country being
but West of it. J a p a n , however, borrowed the Chinese script for its language,
involving many cultural or just linguistic borrowings as well, the expression
the place of sunrise for the eastern direction must have been at least familiar
to the Chinese using it in other records, too and it could reflect a method
of orientation of at least one millennium old in China, as late as at the beginning
above. The letter of the Japanese ruler, written in Chinese, was preserved in the c h a p t e r
on ancient J a p a n [j . Wo-kuo chuan of the official Chinese Dynastic work Sui-shu
(and in the less creditable texts of the Wei-shu and Pei-shih as well) : |][ J g ^ q ^ g s jf^jf
0 K iML N. -f- Jih-ch'u-ch'u t'ien-tzu, chih shu jih-mo-ch'u t'ien-tzu, um yang, translated
by Goodrich and Tsunoda (as it is understood today in J a p a n , too), like this : The
Son of Heaven in t h e land where t h e sun rises addresses a letter to the Son of H e a v e n
in tho land where the sun sets. We hope You are in good health, cf. Rysaku Tsunoda,
translator L. CARRINGTON GOODRICH, editor : J a p a n in the Chinese Dynastic Histories,
Later H a n Through Ming Dynasties (South Tusadena 1951), p. 32. R . A. MILLER adds: The
Sui-shu says this was a [kuo-shu letter of country (importance)], Goodrich et
Tsunoda have 'official message' for this, which is probably fine ; but I have sometimes
also wondered if it means t h a t the t e x t was in Chinese script but in a Japanese-style
Chinese, i.e., what struck the Chinese as being Japanese national script [gSjj : kuo-shup,
and he finds the phrase |Jj g jih-ch'u-ch'u (The land where the sun rises) interesting.
This impertinent (2(/| wu-li ceremony-less) letter from the country of the Root of
the Sun ( H JC Nihon/Nippon) arrived to t h e court of ffiffrrfr Sui Yang-ti in the 4th year
of his - Ta-yeh period, on the zE/J jen-hs day of the 3rd month (April 9th, 608,
cf. P . HOANG : Concordance des chronologies nomniques chinoise et europenne.
Varits Sinologiques X X I X , Shanghai 1910), according to t h e chronicle Tzu-chih-t'ung-
chien by Ssu-ma K u a n g (1019-1086). Vol. VI. Peking 1956. 5 6 3 7 - 5 6 3 8 . The head of
the delegation, a certain ^ J l J fi, i t Jffi T o-li-ssu-pi-ku is entitled -F wang in the record
like tho sender of the Persian letter , thus ho must have been considered a royal
person, some kind of prince at least. I n J a p a n , however, t h e records say that in t h e
period f f j v , during the reign of the Empress Af Suiko (554 628; she ruled
from 592), the governing regent Shtoku-taishi (Prince Shtoku), initiator a n d
protector of Buddhism a n d Chinese culture in J a p a n , sent delegations to China, in 608
sending another high ranking person of the court /j\ P^^jcY" Ono-no-Imoko as a leader of
t h e envoys. Sui Yang-ti, hurt by the written message, told t h e )||1)( hung-lu-ch'ing,
i.e. t h e head of tho court office dealing with foreigners t h a t he should not receive a t all
these Barbarians (of South and E a s t man-i) on t he following occasion ; b u t t h e
efforts of Japanese diplomacy continued, and not without success. See the visits of envoys
from J a p a n in the Sui court in 600, 607, 608, 610, 614 : f - H s Hsien-yao, V \ i \ 4 Sc
no ^ ni tsuite (Zui-Wa kokko no taiisei ni tsuite) : Bunka 29 (1965) 2, 247 - 2 7 8 .
Acta Antiqua Academiae Scientiarum Hungarirae 28, 1980

252 I. ECSEDY
of t h e 7th century B.C. 7 All this happened n o t long after t h e period when the
neighbouring Turk Empire was divided into two parts (after 581 A.D.), and
t h e Chinese records called their rulers E a s t e r n and Western Turks respectively.
(As f a r as I know, none of their own records shows this kind of distinction,
a d o p t e d probably for facility reasons by European historians.)
T h e points of sunrise and sunset, however, are changing place, as well
k n o w n , b o t h according t o t h e day of t h e year and the geographic position of
t h e observer, and therefore they cannot provide a firm basis of determination
of t h e directions, at least not without t h e due astronomical knowledge and
i n s t r u m e n t s etc. B y the way, it is a commonplace in China t h a t her two main
s t r e a m s divide the land by their flows f r o m West to East ; b u t t h e lines of the
H u a n g h o and Yangtzu-kiang go from North-West to South-East instead, and
in m y opinion this could also contribute to a slight t u r n of the pivot
of Chinese 'universe', i.e. geographic view of historical records, too. E.g. the
above envoys of J a p a n were called b y t h e Chinese emperor Barbarian with
t h e ill-name man-i, i.e. (Southern) Barbarian and (Eastern) Barbarian
or (South-East) Barbarian also for simplicity sake, remembering t h e southern
islands, too, where a p a r t of J a p a n ' s population originated f r o m , connected
with China since a longer period of time. B u t it should be mentioned t h a t this
geographic view a n d traditional terminology did not change even when the
Chinese officials regularly met the J a p a n e s e themselves, recording their organ-
ized, centralized, administered etc. c o u n t r y (i.e. a civilized world in Chinese
view, too) a part of t o d a y ' s J a p a n as a land lying east of China.
N a t u r a l l y , this is not the only sign of different types of troubles in the
history of Chinese orientation. 8 To mention only one more example of the above
inclination of axis : t h e period when t h e T u r k s appeared in the history, in the
7
Cf. the chapter 3=|_fll. Yao-tien, on t h e organization of seasonal works a n d observa-
tion of t h e culmination of t h e four seasons (each of them being connected with a star,
direction a n d sun's position) in t h e classical book Shu-king : B . KARLGREN, The
Book of Documents : B M F E A 22 (1950), p. 2(3). As concerns this t y p e of title of China's
e m p e r o r , expected to be used by his foreign subjects, it was recorded as early as e.g.
in t h e middle of the 1st century A. D., when in Ytinnan the Chinese officials offered to
t h e c o u r t three poems of t h e recently subdued n a t i v e population, in their n a t i v e language
(with Chinese characters) a n d in Chinese translation, respectively. I n a passage concerned
t h e e m p e r o r is addressed 0 jjj vf jih-ch'u chu, i.e. t h e Lord of Sunrise while t h e sur-
r e n d e r i n g people seem to have called themselves 0 A i'| jih-ju chih pu (A Tribe of
Sunset). See Hon H a n - s h u (So-yin po-na-pen erh-shih-ssu shih I I I , Peking 1958), chap.
L X X X V I , lieh-chuan L X X V I , p . 1283a; c f . W . SOUTH COBLIN: A n e w s t u d y of t h e P a i -
lang songs. Ch'ing-hua Hseh-pao (The Tsing-Hua Journal of Chinese Studies, T'aipci)
X I I N . S. 1979: 12. 179216.
8
I n case of E a s t arid West, e.g., t h e old Chinese anthropomorphous designations
left (side) and right (side) respectively are responsible for certain difficulties of inter-
p r e t a t i o n , their usage depending on the position of the observer. I n ancient China of
various k i n d s of ceremonies a n d prescriptions, these sides and t h e orientation concerned

W E S T E R N T U R K S IN N O R T H E R N C H I N A 253
t i m e of t h e S o u t h e r n a n d N o r t h e r n D y n a s t i e s (^ Nan-pei-ch'ao,
3 r d 6 t h c e n t u r i e s A . D . ) . S o u t h a n d N o r t h i m p l i e d i n t h e n a m e of t h e p e r i o d
in t h i s , h i s t o r i c a l o r d e r of s e q u e n c e a r e t h e m a i n d i r e c t i o n s , w h i l e E a s t a n d
W e s t a s r e g a r d s p o s i t i o n s o r m o v e m e n t s ( c h a n g e s o f c a p i t a l s ) of t h e d y n a s t i e s
concerned are inclined to t h e m : the West t o w a r d N o r t h and the E a s t
t o w a r d S o u t h . T h e l a t t e r connection, also s h o w n b y t h e a b o v e q u o t e d j o i n t
e x p r e s s i o n of t h e B a r b a r i a n s of t h e s e d i r e c t i o n s (man-i), could be illustrated
b y t h e n a m e s of t h e s t a t e f f Chin ( 2 6 5 4 1 9 ) , o n e o f t h e S o u t h e r n a n d N o r t h e r n
D y n a s t i e s . T h e s e c o n d p a r t of i t s r u l e , w h e n it h a d t o l e a v e i t s N o r t h e r n h o m e -
land, a n d to move its court from L o y a n g ( H o n a n province) to Chienk'ang
( t o d a y ' s N a n k i n g Southern Capital, in K i a n g s u ) , w a s called by t h e p o s t e r i t y
j j t f f Tung Chin E a s t e r n Chin ( f r o m 317 o n ) . M e a n w h i l e t h e f i r s t p a r t o f
i t s r u l e , in t h e N o r t h , i.e. Chin, n o t y e t n e c e s s a r y t o b e d i s t i n g u i s h e d , is c a l l e d
s o m e t i m e s t o s e p a r a t e i t f r o m i t s E a s t e r n c o n t i n u a t i o n W e s t e r n Chin
(gf f f H si Chin).
T h e r e l a t e d s l i g h t i n c l i n a t i o n in t h e r e c o r d s o f t h e o t h e r e n d of t h e
pivot of C h i n a ' s f o u r d i r e c t i o n s , c a n b e e x p l a i n e d a l s o b y g e o g r a p h i c a l
r e a s o n s , a n d n o t a t all m e r e l y b y t h e a b o v e - m e n t i o n e d false E a s t W e s t l i n e
of t h e t w o b i g r i v e r s . T h e g e o g r a p h i c a l f a c t o r of h i s t o r i c a l i m p o r t a n c e c o u l d b e
C h i n a ' s w e s t e r n g a t e , i.e. t h a t o p e n e d f r o m a n d t o w a r d s C e n t r a l A s i a a n d t h e
were connected with a kind of social hierarchy and their order of sequence was a t t r i b u t e d
ritual importance, but several times in an ambiguous way, the significance of right
being stressed now, a n d left being esteemed then, or vice versa; a n d the connection of
sides with E a s t a n d West was not constant and consequently applied. Instead of a kind
of conscious practice of this terminology, tradition and social convention seem to be t h e
main reasons for this trouble of right or left, i.e. sides a n d directions in China, right-
handed from the pictograms and m y t h i c tradition to t h e earliest artistic representa-
tions , as pointed out by M. GRANET : La droite et la gauche [1933] : t u d e s sociologi-
ques sur la Chine. Bibliothque de Sociologio Contemporaine. Paris 1953. 261 278,
a n d La pense chinoise. Paris 1934. 360 373. The task r e m a i n s when interpreting
a historical record to scrutinize the right or left if they mean East a n d West re-
spectively or in the way round, according to the period, source etc. of a passage, cf.
P . DEMIVIJLLE : Gauche et droit en Chine [1968]. Choix d ' t u d e s sinologiques (1921
1970). Leiden 1973. 518 532. In m y opinion, the old change from South-faced to N o r t h -
faced orientation, mentioned above, could well bring forth a confusion of sides, h a n d s
etc., a n d East a n d West connected with them in traditional order , East being on
the left side, when the observer looks southward, while it is on the right one, when h e
faces North, cf. my book A knai llam kezdetei [On t h e Beginnings of the Ancient Chinese
S t a t e ] Budapest 1984. Luckily, in the case of the Chinese records on the Turks or o t h e r
foreigners, the above problem rarely m e a n s serious difficulties of interpretation ; a n d if
certain passages still reflect it (cf. left a n d right wings of armies, see later, part 3.), t h e y
should be analyzed within t h e textual a n d historical context concerned, by no m e a n s
mechanically relying on a common knowledge, i.e. only one kind of tradition as regards
pairing of East-West and right-left or left-right (sides).
Acta Antiqua Acudemiae Scientiarum Hungaricue 28, 1080

254 I. E C S E D Y
s t e p p e region. N a m e l y , t h e large roads leading westward from China, went to

t h e N o r t h , avoiding t h e western region of mountains, the old mythic lands of
immortals, visited b y Buddhist pilgrims and foreign merchants only from a
comparatively late period. Anyhow, as an indirect consequence, it was also
noticed by Chinese readers of China's historical records t h a t although sepa-
r a t e t e r m s had designated the four B a r b a r i a n s of the four directions ( -fj
ssu-fang) of the world : man of t h e South, fX ti of the N o r t h , i of the
E a s t and ^ jung of t h e West , as regards China's B a r b a r i a n neighbours,
W e s t m a y also mean N o r t h , or Western something like Northern,9 and this is to
be considered in case of Chinese records e.g. concerning the Western Turks, too.
2. E A S T E R N A N D W E S T E R N T U R K S I N C H I N E S E H I S T O R I C A L
RECORDS
The Chin-shan, i.e. Altai region, where the mythic ancestors of the
T u r k s entered into t h e horizon of Chinese historians, is lying west or north-
w e s t of the related courts and their imperial sphere of power. The " T'u-
cheh, i.e. Turks, however, are n o t registered in China among Western
B a r b a r i a n s (jung or hsi-jung), b u t among the northern ones (J\tik pei-ti),
connecting them w i t h a famous predecessor of Northern Barbarians, the
JX Hsiung-nu-s, a n d thus indirectly with all the alleged ancestors of
t h e i r s , told to h a v e been Northern.
Naturally b o t h t h e mythic nucleus of the Turks and their first historical
representatives, unified under the f o u n d e r of the first empire J-.PE1 T'u-men
(Bumn) in 551 A.D., were recorded simply as Turks, i.e. as an undivided people
or political unit, if only because no f u r t h e r details of their organization had
r e a c h e d China so f a r . B u t in the Chinese sources concerning t h e following
decades no more distinction can be f o u n d either, although their envoys of
diplomacy and their raiding troops arrived regularly to the Western imperial
house Chou called Northern Chou , and then to the eastern one
Ch'i Northern Ch'i in the northern p a r t of the Chinese E m p i r e unified in
581/589 A.D. by t h e Sui House. A p p a r e n t l y the allies and enemies of the
parallel and successive Chinese dynasties came from one and the same, still
undivided world of t h e Turks. 1 0
9
Cf. e.g. W ( (Jjl TS'EN CHUNG-MIEN'S s t u d y on those components of the Chinese
( f f f i f j f c han-tsu 'Han-ethnic'), which can be considered of western origin : |
A ^ t J ik-liE Han-tsu i pu-fen hsi-lai chih ch'u-pu k'ao-cheng (in his volume f f f )
Chou wen-shih lun-ts'ung, Shanghai 1968), to t h e effect (p. 27) t h a t in the
r e l a t e d records IV est (em) and North(ern) could be alternately used, a n d both of t h e m
indicated but a vaguely known tribe or peoplo, originating from a distant land.
10
See the details of t h e early Turk history in my paper : Trade-and-war relations
b e t w e e n the Turks a n d China in the second half of the 6th century. Acta Orient. H u n g .
21 (1968) 131 - 1 8 0 .

WESTERN TURKS I N NORTHERN CHINA 255
The germs of social division and political separation, however, must h a v e

developed meanwhile, since the turbulences that led to t h e split of the empire
in 581 A.D. had been started by revolts of discontent leaders of strong tribes.
This is the period when one can shed light on the social network, clan hierarchy
and birth of tribe, i.e. various social elements being in statu nascendi.n As none
of t h e heads of clans (tribes) could get the upper h a n d above these new or
strengthened social forces, a part of the most ambitious tribes (clans) sought
for a better chance westward as far as in Western Asia or Eastern Europe ,
giving by this an opportunity to t h e Chinese to designate them as Western
Turks. No Chinese record concerning the former period would suggest more
t h a n one kind of Turks, and even the ancestor of Western Turks "^Tii'l
Shih-tien-mi (IStmi) occurs only in passages containing references to ante-
cedents of later Western Turk actions on historical events connected with t h e m .
Consequently, in case of the Turks, recorded by Chinese sources, the adjective
Western seems to be first of all a chronological t e r m of ante quem non t y p e ;
it shows the Turks a f t e r 581 A.D., strong enough to wage conquering wars even
westward of their cradle, i.e. the Altai region, b u t n o t necessarily separated
f r o m it completely in later periods either.
One of the most active tribes in the events of 581 A.D., t h a t of p i j ^ bjFT
A-po k'o-han (Apa qayan) may well illustrate b o t h t h e lack of an original
separative difference between a Western Turk group a n d an Eastern one a t
t h a t time, and the direction of t h e possibilities of a social or ethnic change.
Namely, ^jSf. Ta-lo-pien, or, by his given title : Apa qayan, a disregarded
and discontent member of the ruling A-shih-na clan organized and re-
organized his tribe several times, now losing his troops consisting of E a s t e r n
Turk subjects and then gaining armed forces from t h e Western Turk ruler
against the common enemy, i.e. the ruling Khagan 3fr Sha-po-lio (ISpara),
considered Eastern from the time of division of his empire. In every case
Apa's tribe, Apa's people is mentioned, apparently irrespective of t h e i r
origin. On the other hand, Apa's mother (her clan) seems to be lower in clan
hierarchy t h a n those of his princely relatives, so probably his maternal rela-
tives are of a relatively distant origin or distant land, relation etc. When in
his home his mother (i.e. the women, children) are killed in a decisive b a t t l e
(in 583 A.D.), it t u r n s out from t h e record t h a t his homeland is in the N o r t h ,
and it is a kind of old land to be re-conquered by ISpara. Thus Apa's northern
relations can also be made responsible for a tension, contributing to the politi-
cal separation of his people, i.e. his Western and, in some way Northern T u r k s
from ISpara's E a s t e r n Empire.
11
See a related trial ill my papers Tribe and tribal society in the 6th century T u r k
Empiro. Acta Orient. H n n g . 25 (1972) 245 262, and Tribe a n d empire, tribe and society
in the Turk age. Acta Orient. Hung. 31 (1977) 1 13.
Acta Antiqua Acudemiae Scientiarum Hungaricue 28, 1080

256 I. E C S E D Y
This distant N o r t h e r n land of old or new relations or joining people m a y

be m e t also in the Chinese description of m a n y Northern Barbarians, registered
side b y side with the T u r k s or after t h e m , within their empire etc. 1 2 This is then
a n o t h e r objective distinctive feature of Western Turks : those T u r k s who con-
q u e r o r gain etc. new subjects or allies, are the Western Turks, while the
E a s t e r n ones, in China's protective and dangerous neighbourhood, get isolated,
do n o t increase or j u s t lose importance. 1 3 Obviously it is t h e E a s t e r n Turk t h a t
m u s t be distinguished, first and then mostly in relation to t h e Western brother ;
a n d t o be sincere, in several cases this distinction, if indicated a t all, seems b u t
m e r e l y mechanic. I t is characteristic, a f t e r all, t h a t when only Turks are
m e n t i o n e d in a separate event : raid, visit etc., we can hardly clear up, whether
W e s t e r n or Eastern T u r k s are concerned. N o wonder t h a t t h e reader is inclined
t o simplify the question to a geographical distinction, seeking e.g. Eastern
T u r k s in the East east of Altai a n d t o find or suppose only Eastern Turks
w i t h i n the field of Chinese authority.
L e t me quote now a counter-example, t h a t of the K a r l u k s , representing
t h e related problems in t h e records concerning the 7th century at least.
3. T H E CASE O F A W E S T E R N T U R K T R I B E , ALLY OR S U B J E C T :
THE KARLUKS
I n a recent s t u d y of mine, on t h e basis of re-examined old a n d a few new

a r g u m e n t s and Chinese records I could conclude t h a t the t r i b a l structure, i.e.
t h e so-called three-tribe K a r l u k could be known for the Chinese not earlier
t h a n t h e beginning of t h e 8th century, t h e time of the wars of Mo-ch'o
(Bg cor) (Q an gay an) against t h e m , and contrary to t h e common view
based on Chinese records, even the names of provinces established for them by
t h e Chinese occurred f i r s t at the same time. Thus their history, prior to t h a t
period, remained open t o new research, t h e date 657 658 A.D. of their alleged
whole defeat and settling by the Chinese included. 14
12
E.g. in the m y t h of origin of the ruling family of the T'o-pa (tabyac) Northern
Wei dynasty, controlling Northern China in the centuries prior t o T u r k history, cf.
Wei Shou, Wei-shu (So-yin po-na-pen erh-shih-ssu shih VIII), chap. I, p . 18a.
13
A f t e r this p a p e r h a d been delivered (in 1980, Budapest), t h e problem of East-
W e s t division of the T u r k s was treated upon Chinese sources, with a conclusion t h a t only
a n E a s t e r n Turk E m p i r e is t o be taken into consideration, while those various leaders
a n d t r i b e s which fought for power on its western peripheries a n d even westward of it,
i.e. t h e so-called W e s t e r n Turks a p a r t f r o m Apa khaghan's short period of trials
failed t o found an empire proper. See WANG IIUAS: Apa Qaghan, founder of the Western
T u r k i s h Khanate, the splitting up of the Turkish K h a n a t e a n d t h e formation of t h e
W e s t e r n Turkish K h a n a t e . Social Sciences in China (Peking) 1982: 4. 124154.
14
Cf. my paper A contribution to t h e history of K a r l u k s in t h e T'ang peiiod.
A c t a Orient. Hung. 34 (1980): 1 - 3 . 2337.

WESTERN TURKS IN N O R T H E R N CHINA 257
As regards the first mentioning of the Karluks, we read t h a t t h e y belong-

ed to the tribes of the Western Turks and lived according to the same customs,
slightly differing in their language, on the slopes of the Altai, later on moving
slightly southward. This first record about them occurs in the chapter on t h e
Uigurs (of the HsinT'ang-shu), i.e. the inheritors of the Western Turks, so their
appearance on t h e stage of history at the beginning of the 7th century is
over the chronological limit, when a part of the Turks became Western, b o t h
in t h e sense t h a t t h e y sought for new territories westward, and t h a t t h e y in-
creased by new (Western) tribes, allies or subjects. The cultural unity, i.e. t h e
way of life seems t h e most important cohesive force, while the common or
related language is b u t second, and no other tie with the ruling clan(s) being
mentioned, the cohesion seems to be first of all political (judging from the Chi-
nese terms in general : Country of Western Turks etc.), realized in joint mili-
t a r y actions. N a t u r a l l y the time of division of t h e T u r k Empire does not m e a n
the beginning of K a r l u k history on the slopes of t h e Altai, and the m o v e m e n t
to South can concern their whole population and a p a r t of them as well ; as it
t u r n s out, the latter case can be supposed. Anyhow, t h e y participated as a
unit, a tribe among the other subjects of the Western Turks around 630 A.D.
in t h e smaller or bigger battles, causing troubles t o the recently established
T ' a n g dynasty. B u t these same . Ko-lo-lu (qarluq ; in transcriptions of a
probably later period : Ko-lo-lu, Ko-lu etc.), i.e. Karluks sent their
leaders to take p a r t in sacrifices at Chinese sacred mountains as far as in
Shantung, E a s t China.
As a rule there is no indication, whether E a s t e r n or Western T u r k s are
concerned in the sporadic records of daily events in the northern sphere of t h e
T ' a n g Empire, f r o m t h e Altai region practically t o Easternmost China. In most
cases, however, t h e participant leaders or the names of tribe Karluk, / g p j
ch'u-yiie (igil) f U j g ch'u-mi (comul) etc. reveal t h a t mostly or exclusively
Western Turks are present in the warlike activities in this period, u n d e r t h e
rule and leadership of the K h a g h a n jf^gi Ho-lu or at least acknowledging his
authority. I t is only this period, when the K a r l u k s can balance between t h e
Eastern and Western Turks as told in their s u m m a r y description quoted
above , since later on almost no role is played by the offsprings of the T u r k s
called Eastern, who lived under the protection of their great eastern neigh-
bour. The turning-point in the events is indicated by the defeat, capture a n d
execution of Ho-lu khaghan by the Chinese in 649 650 A.D., followed in 657
658 by the establishment of new provinces for t h e Karluks as well as for t h e
other tribes who took part in Ho-lu's campaigns. As seen from a so f a r so t o
say hidden record of the geographical chapters of the H sin T'ang-shu, sever-
al tribes, groups etc. t h a t appear later on in the West even westward of t h e
Altai are settled under Chinese control in inner N o r t h China, in Shanhsi,
Shantung provinces.
17 Acta Antiqua Academiae Scientiarum Hunjfaricae 28,1980

258 I . E C S E D Y : W E S T E R N T U R K S I N N O R T H E R N CHINA
T h e record in question contains an interesting detail an indirect infor-

m a t i o n as regards t h e localization of those original Karluks, which gave
troops t o Ho-lu'a campaign. Namely, their two (left and right) wings, are de-
scribed i.e. a military organization, like t h a t of the tribal groups of the
W e s t e r n Turks, instead of a mere tribal division by t h e n a m e s of their
h o m e l a n d s : one of t h e m originated f r o m t h e tkn region, while the other
f r o m t h e neighbourhood of Pei-t'ing. T h a t is to say, a part of t h e m must have
m o v e d eastward (or : slightly southward), and thus Western Turks, repre-
sented a n d followed b y t h e Karluks, could be present, in a significant number
and vigour, in the entire T u r k territory east of the Turk homeland in the Altai,
at least till the middle of the 7th century. A f t e r their defeat and f a t a l pro-
tection by China in t h e 650's, they i.e. t h e Karluks of the two above lands
of t h e i r homelands appeared in t h e Chinese borderlands only sporadically,
and apparently with small forces. H a v i n g no chance against t h e T ' a n g military
power, t h e tribal armies armies of t h e already three-tribe Karluks, as seen
f r o m l a t e r records must have sought for another land t o conquer and to
tax. 1 5
4. CONCLUSIVE S U P P O S I T I O N S F O R A E U R A S I A N A S P E C T
O F T H E W E S T E R N T U R K H I S T O R Y I N T H E 7TH C E N T U R Y C H I N A
I t m a y be of interest to emphasize, especially in Hungary, t h a t Chinese

records a n d East Asian history can serve as a background of the steppe history
as a whole, also in t h e time when the second wave of the rulers of t h e Avarian
E m p i r e appeared in t h e Carpathian Basin. If a t any time in the 7th century,
it could be possible a n d reasonable to suppose a Central Asian military move-
m e n t a n d action just in t h e 660's before 670, the second Avarian invasion,
t h e d i s p u t e d date being t h u s supported b y t h e eastern events in question ,
no m a t t e r how far this action could reach a t all, pushing other peoples west-
w a r d , b u t a possibility to be considered perhaps going as far as the Car-
p a t h i a n Basin. If so, it could be very similar t o the events of the middle of the
6th c e n t u r y , when t h e ft/ftr Juan-juan-s, the so-called Asian Avars were
d e f e a t e d by the Turks in 551, and a p a r t of t h e m tried its f o r t u n e in North
China a n d got protection in 554,18 while t h e first wave of Avars appeared in
E a s t e r n Europe in 558 A.D.
Budapest.
15
On the role of K a r l u k s in the f u r t h e r history of the region of 4t,!i Pei-t'ing see
m y p a p e r Uigurs and Tibetans in Pei-t'ing. A c t a Orient. Hung. 17 (1964) 83 104 (esp.
p . 86 s q q ) .
16
Cf. Trade-and-war relations . . . A p p e n d i x I . 174179.
Acta Antiqua Academiae Scientiarum Hunqaricae 28 ,1980

UTAS
THE PAHLAVI TREATISE A V D H U S A H K H 7

S A K I S T AN OR WONDERS AND MAGNIFICENCE
OF SISTAN
I n a previous paper, entitled Non-religious Book P a h l a v i literature as a

source to the history of Central Asia, 1 1 have sketched something of the histor-
ical and pseudo-historical contents of four of the few e x t a n t non-religious Book
Pahlavi works, namely Sahristnih Ern (Cities of Iran), VicriSn catrang
(The explanation of chess), Krnmak ArtaxSr Ppakn (The chronicle
of Artaxsr, son of Ppak) and Ayytkr Zarrn (The memoir of the Zarr
family). From a strictly historical point of view the contents of those works
m a y be considered insignificant, b u t t h e y can serve a purpose by turning our
attention to the intricate blend of reflections of historical events, primordial
m y t h s and religious reconstructions which is peculiar to m a n y of our sources on
t h e history of Pre-Islamic Central Asia.
Here follow a few comments on another work of a similar t y p e : the short
P a h l a v i treatise known under the title Avdh sahkh Sakistn or Wonders
and magnificence of Sistan. I t is a sister t e x t to those just mentioned, not only
because it is similar in contents (especially so to the Sahristnih Ern), b u t
also because it belongs to the same t e x t tradition : t h a t of J a m a s p - A s a n a ' s
famous Codex MK, dated 691 A.Y., corresponding to 1322 A.D. This compre-
hensive manuscript, which was published by J a m a s p - A s a n a in his Pahlavi
Texts (vol. I I I , B o m b a y 1897 1913), is our main and often only source for
some t h i r t y short b u t important Pahlavi works. 2
Avdh sahkh i Sakistn is t h e third work in t h a t collection, appearing
already in the first volume of 1897. I t occupies only two pages there (pp. 25
26), or 37 lines. I t offers substantial problems, however, as t h e t e x t is obviously
corrupt in m a n y places. In his edition, J a m a s p - A s a n a gives a few variants
f r o m t h e manuscript designated J J , dated 1136 A.Y., corresponding to 1767
A.D. (i.e. 445 years a f t e r MK), b u t t h a t is of little avail, since t h a t manuscript
is a copy of Codex MK.
1
Acta Ant. Hung. 24 (1976) 116124 ; also publ. in J . HARMATTA (od.) : Studies
in the sources on the history of Pre-Islamic Central Asia. Budapest 1979. 119128.
2
Cf. my contribution to Monumentum H . S. Nyberg. Tehran Lige, Acta Iranica,
1976, vol. I I , pp. 399 f., and Acta Orientalia, Copenhagen, 37 (1976) 76.

260 UTAS
Jivanji J a m s h e d j i Modi published a transliteration of this text in G u j a -

r a t i script with English and Gujarati translations and notes in Gujarati in his
Aiydgr-i-Zarirn, Shatrih-i-Airn, and Afdiya va sahigiya-i-Sistn,
p r i n t e d in B o m b a y in 1899 (pp. 123127). N e x t in time, in t h e t r e a t m e n t of t h e
t e x t , comes a hand-written transliteration and translation b y E d w a r d W. West,
published posthumously by A.V. Williams Jackson in the Journal of the Amer-
ican Oriental Society of 1917. 3
Heinrich J u n k e r paraphrased t h e contents of the treatise in his compre-
hensive article Mittelpers. fraSmurv 'Pfau' in Wrter und Sachen of 1929, 4
a n d a t about the s a m e time Ernst Herzfeld gave a translation of it in the second
volume of his Archaeologische Mitteilungen aus Iran3 Finally, Harold Bailey
adduced parts of it in an argument on t h e memorizing of t e x t s , giving a trans-
lation of about half t h e treatise, in his Zoroastrian problems in the ninth-century
books.6
These scholars generally agree on t h e comparatively recent origin of the
t e x t , putting its compilation to around t h e t e n t h century A.D., but their inter-
pretations differ r a t h e r much from each other. Without d o u b t this is mainly
due t o the poor shape of the available t e x t material. A n u m b e r of problems still
r e m a i n to be solved, b u t I shall venture to present a transcription and a t e n t a -
tive translation of t h e complete t e x t , before taking u p a few points of historical
relevance for discussion.
I t seems t h a t t h e title of the treatise, Avdh sahkh Sakistn, does
not occur in the manuscripts. I t has obviously been a b s t r a c t e d by J a m a s p -
A s a n a from the f i r s t line of the t e x t . I t is not unlikely, however, t h a t the com-
position of this little piece is a reflection of the 'Aj'ib, Wonders or Marvels,
genre, so common in Arabic and New Persian literature in later centuries, b u t
known already f r o m t h e fourth century A.H., i.e. the t e n t h certury A.D., in
t h e work 'Aj'ib ul-buldn, Wonders of the countries, written by the poet
Abu'l-Mu'ayyad al-Balxi for the Samanid prince N h b. Mansur. 7 Incidentally,
J u n k e r ' s point of paraphrasing the contents of the treatise was to prove t h a t
there are no Wundern in it. He argues t h a t avd should be translated merk-
wrdig, sonderbar. H i s problem m a y have been connected more with Ger-
manEnglish semantics t h a n with I r a n i a n .
3
Vol. 36. 115121.
4
Vol. 12. 1 3 2 - 1 5 8 (this subject : 134).
3
Berlin 1930. 94.
Oxford 1943. 161.
7
I t is also k n o w n a s 'Aj'ib-i barr bahr, Wonders of land a n d sea ; v. F. TAUER
in J . RYPKA (ed.), H i s t o r y of Iranian literature, Dordrecht 1968. 461, and G. LAZARD
in Ydnme-ye J a n R y p k a . P r a h a 1967. 95 96. This work b y al-Balxi seems to be lost
b u t it obviously contained ample material on Sistan, as is shown b y references in t h e
11th century work Trlx-i Sistn, History of Sistan (ed. M. BAHR, Tehran 1314. 13 ;
transi. L. P. SMIRNOVA, Moscow 1974. 30 31, 58).

T H E PAH LAVI T R E A T I S E AVDB V SAHIREN I SAKISTN 261
Transcription (paragraph n u m b e r s are inserted according to the n u m b e r -

ing in Palliavi Texts) :8
pat nm i yazdn
1. avdh u sahkh i zamik i sakistn hoc aprik Sahrih t rd aprtar u vh.

2. vak n k rt i htmand u var i frazdn u zrh i kayns u gar [t] uSd-
(da?)Str9 andar zamik i sakist[n\.
3. zyiSn u parvariSn hu(r)Star u huStarmh [m] s(ay)Sns i zartuStn
i spitmn u-S rist-xz kartan.
4. vak n k patvand u thmak kayn dahyupatn (t) pat n kiSvar vizand
aviS mat
5. hac frazandn i frtn : salm k kiSvar i Tirm u tz k turkistdn pat xuatyh
dSt rc rn dahyupat bt u-S b zat.
6. u hac frazandn [] rc b kanik- an kas b n mnd.
1. u pas frtn var [] frazdn nayt10 u pat nihn dSt tdy [da~[humn patvand
ka hac hn kanik pus zyt.
8. pas frtn var [] frazdn Sut u-S hac, artvsr andht yaft xust u pat
apdc drdstan [] rdn-Sahr u xuarrah [i] kayn aprik yazdn hac andar sa-
kistn gh mahmdnh apartar yaft vindt apk mnuScihr u ySn rn
frin.
9. vak n k viStsp-Sh dn pat var i frazdn kart ravkeh fratom pat sakis-
tn u pas pat aprik Sahrih.
10. u viStsp-Sh pat hampursakh zartuSt u sn [] ahmst[t]dn i bustk
cign-iS hviStn i zartuSt fratom pat sar-hviSth y but hnd.
11. dn [i] sakistn cSt ravk dStan rd nask nask12 pat dtak [i] vhn
frc raft.

As far as possible the original text of codes MK has been restored from t h e edi-
tion of JAMASP-ASANA ; m y additions to the text are given between square brackets [],
while letters which I suggest to be excluded are p u t in parentheses () ; emendations
are marked with a plus + and are explained in foot-notes. The transcription system is,
with some slight modifications, that employed by H . S. NYBERG in his Manual of Pahlavi,
vols. I I I (Wiesbaden 196474); it must be stressed t h a t it is purely conventional,
representing t h e archaistic Pahlavi orthography as faithfully as possible while supple-
menting it with vowels and substituting Iranian equivalents for Aramaic ideograms.
9
Both MK and J J infix an unnecessary extra sign.
19
The ending is written with t h e ligature -x, in MK and -x2 in J J (JAMASP-
ASANA substitutes -t) ; cf. m y notes on these ligatures in Acta Orientalia, Copenhagen,
37 (1976) 9699.
" T h e original text has only hwm; both JAMASP-ASANA and WEST supplement
d{a)-; this seems justified considering Bundahin X X X I : 11 (transi. W E S T : S B E ,
vol. V, 1 3 3 - 134 ; Ir. BundahiSn, translit. & transi. ANKLESARIA, X X X V : 13, 294 295 ;
face, of MS. TD,, Tehran 1348/1970. 197).
12
P o s s i b l y t o b e r e a d i n s t e a d ( u ) 50 u 50, a s b y JAMASP-ASANA, MODI, HERZFELD
and BAILEY.
Acta Antigua Academiae Scientiarum Hungaricae 28, ISSO

262 BO UTAS
12. nask- +bagdn-ic13 Trnend cign sen [u\ burzmihr i zartuStdn pat virds-
takh i hn bt () +dSndk.M
13. ka gizistak alaksandar i hrm[y] rdn-Sahr mat dySdn k pat brh15
mg-marth raft grift zat.
14. mart u rtak -cand sakistn mat hnd.
15. nask- bt zandn bt () apurndyk- nask- +bagdn-ic13 xuastw u varm
kart estt pat ce17 hdn rdh18 dn andar sakistn apc gaSt u rst u vrdst ndk
ndk {ndk) b pat sakistn adak-iS19 apdrik giydk n varm.
16. +avar20 pat hdn giydk hamdk-dn- yaznd Sdth gdhrkdnh rad htdxt-
framdynd yaStan21
Colophon :
frazaft pat drdt [u~\ Sdth [M] rmiSn St u farraXu u dr-ziviSn u prdzkar u
ahlav [M] kdmak-hanjdm bavt k nipiSt k xuS u k xudnt. a$dm.
Translation :
In the name of t h e gods
(1) The wonders a n d magnificence of t h e land of Sakistn are greater a n d

better t h a n those of other lands for these reasons :
(2) One [reason] is this t h a t the river H t m a n d and the lake F r a z d n and
t h e sea K a y n s a n d the mountain Udtr [are situated] in the land of
Sakistn ;
(3) [and] the birth a n d upbringing of H u s t a r and Hustarmh [and] Ssns
of the family of Z a r t u s t Spitmn a n d t h e resurrection [are] t o occur from
it.
(4) One [reason] is this t h a t the offspring a n d seed of the K a y n rulers are
in this clime ; disaster came upon t h e m (5) through the sons of Frtn :
13
W i t h t h e e m e n d a t i o n proposed b y B a i l e y (op. cit., 160, . 4) : + ' R l J Y ' - n c '
w i t h reference to Denkart (DkM 692, 15 ; W e s t : S B E , vol. X X X V I I , 34 ; t h e reference
t o Frahang i Pahlavik 1 : 3, however, seems m i s t a k e n ) . M o d i reads dvsirjo, W e s t
guvah-sn-ico (the witness is even Sen), H e r z f e l d dvasana (die beiden Sen), a n d
J u n k e r professes himself u n a b l e to read it.
14
With t h e e m e n d a t i o n of B a i l e y ; t h e original t e x t seems to h a v e akn, Arsacid
( t h u s Modi), b u t is r e a d ykn, the coming ones, b y W e s t ; or Skr
15
Or brh, a s B a i l e y reads it, suggesting t h a t it stands for brahm, garment,
manner.
18
Or si, firm, reliable ; B a i l e y e m e n d s xunnd.
17
Possibly for -id, a s r e a d by B a i l e y .
18
Possibly for brh in t h e sense of brahm, a s emended by B a i l e y (cf. . above)
becauso t h e t e x t h a s I'h, while rah normally is w r i t t e n Vs.
19
T e x t : 'DYN-S; m a y also be read (with B a i l e y ) ' d y n y ' for 'yny' = ny.
20
Tentative e m e n d a t i o n : *LPN(M)H, for t h e a p p a r e n t L MNW = man k, me
who, of t h e t e x t which is difficult to reconcilo w i t h t h e c o n t e x t .
21
May also be r e a d SGYTWN = rav, go ! (in t h a t case p r o b a b l y with t h e read-
i n g man k in the beginning of the paragraph (cf. t h e preceding n o t e ) .
Acta Antiqua Academiae Scientiarum Hunqarirae 28, 1980

T H E PAHLAVI T R E A T I S E AVDF.H U SAHIREN I SAKISTN 263
Salm who [ruled] the clime of R o m e and Tz who ruled Turkistn, [while]
r was the ruler of Ern ; a n d he was slain by t h e m .
(6) And of the children of Er except one girl nobody else remained ; (7) a n d
t h e n she was brought by F r t n to Lake Frazdn and was held in conceal-
ment till the [ten]th generation, when a boy was to be born from t h a t girl.
(8) Then Frtn went to Lake F r a z d n , and he asked a boon from Artvlsr
Anhit, and for the restoration of Ern-Sahr and the K a y n xuarmah t h e
other gods, through [this] dwelling-place in Sakistn, gained a superior
boon with MnuSihr, and their blessing was with E r n ( ?).22
(9) One [reason] is this t h a t K i n g Vitsp propagated the religion a t L a k e
F r a z d n , first in Sakistn t h e n in other lands; (10) and King Vitsp
was in consultation with Zartust and Sn, son of A h m s [ t ] t , of Bust, as
of t h e disciples of Zartust he was the first in head-discipleship ( ?). 23
(11) In order to keep the religion of Sakistn current for the teaching, nask
a f t e r nask passed on in the families of the Zoroastrians, 2 4 (12) one nask
[being] the one they call *Bagn 2 5 for the arrangement of which Sn a n d
Burzmihr, son of Zartust, were well-known.
(13) When the accursed Alexander, the Roman, came t o rnahr, those who
went in the splendour 2 6 of t h e Magians were seized [and] slain.
(14) A few men and boys came to Sakistn.
(15) There was one nask which was among the women, [ a n d ] a child had been
trained 2 7 and had memorized [this] one nask, t h a t same Bagn. 25 I n such
a way the religion returned in Sakistn, and it was ordered and arranged
afresh. Except in Sakistn, t h e n there was no memory of it in any other
place.
(16) Come( ?), in t h a t place they perform a complete religious ceremony [ a n d ]
t h e y order a H d x t to be recited for the happiness of t h e nobility( ?).
The text ends with a s t a n d a r d colophon which unfortunately adds noth-

ing to our knowledge of the background and origin of this treatise :
Completed with benediction, happiness [and] satisfaction ; may he be
h a p p y and f o r t u n a t e and long-lived a n d victorious and righteous and success-
ful who wrote [it], who owns [it] a n d who reads [it]. . 28
22
T h e somewhat uncertain s y n t a x of this p a r a g r a p h allows m a n y variations in
t h e order of words a n d phrases, b u t t h e general p u r p o r t seems clear.
23
T h e verb is formally in t h e plur., they were, p e r h a p s referring t o b o t h Sen
a n d his f a t h e r A h m s t t , hardly to K i n g ViStsp.
21
Or possibly : a hundred in t h e families of t h e nobles w e n t forth (BAELEY,
161; cf., n . 12 above).
25
I.e. The gods ; cf. n. 13 a b o v e .
28
Or fate or garment, manner ; cf. n. 15 above.
27
F o r xuast in this sense, v. N y b e r g , Manual of P a h l a v i , vol. I I . 221, a n d xMast
as trampled path in Burhn-i qfi', ed. M u ' n , vol. I I I , 791. I f st (cf., n . 16 a b o v e ) ,
t r a n s l a t e : had memorized firmly!
28
I.e. t h e Avestan AJa-prayer.
Acta Antiqua Arademiae Scientiarum Hungaricae 28, 1980

264 UTAS
As marked by t h e introductory phrase vak , t h e t e x t falls into three

p a r t s : the three reasons for considering Sistan the most remarkable of lands.
These reasons are derived from purely Zoroastrian traditions. I t is noticeable
t h a t t h e not very m u c h younger New Persian work Trix-i Sistdn, History of
Sistan, begins with an enumeration of t h e excellencies (fadd'il) and won-
ders ('ajd'ib) of t h e same province. 29 N o n e of these outstanding features, how-
ever, coincides with those mentioned here, in spite of t h e fact t h a t the anony-
m o u s author of Trix-i Sistdn mentions t h e Pahlavi BundahiSn (in the form
Ibn Dahitl !) as one of his sources for this passage. 3 0 According to the first
pages of this Tdrix-i Sistdn, there should also have existed a work directly
entitled Fadd'il-i Sistdn, composed by an otherwise unknown Hill [ibn]
Y s u f Uq (Auql?), a title mirrored in t h e later work Fadd'il-i Balx, an anon-
y m o u s Persian t r a n s l a t i o n of a work f i r s t written in Arabic in the beginning of
t h e 7th/13th century. 3 1
The first reason for pre-eminence is geographical : Sistan is presented as
t h e land of certain legendary topographic features, namely the holy river
H t m a n d (Av. hatumant-), the holy lake K a y n s (Av. kqsaoya-) and the holy
m o u n t a i n Usdtr (Av. uSi.darana- or uSi.dam-).32 W i t h o u t doubt, this im-
plies the arena for Zoroaster's teaching of the religion. I t m a y be considered
an archetypal geographic trinity ; for a parallel compare, e.g., Christ, the
R i v e r Jordan, L a k e Gennesaret and t h e Mount of Olives. 33 From a historio-
graphical point of view, however, this information is rather trivial. The young-
er Zoroastrian tradition, in general, identifies Kayns with t h e lake H m n ,
U d t r with the m o u n t a i n Kh-i X v j a h and, naturally enough, H t m a n d
w i t h the river H i l m a n d , all parts of the modern geography of Sistan.
I t is not entirely correct, however, t o describe this first reason as a
geographic trinity, because also a f o u r t h name is mentioned, var frazdn,
t h e identification of which is more uncertain. The list of lakes in Chapter
X X I I (XII) of the BundahiSn places it in Sistan, 3 4 and in 36 of the Sahristd-
nih Erdn it is mentioned in connection with Bust, which is situated in the
e a s t e r n p a r t of Sistan in its more narrow sense. 35 The tradition about Frazdn
goes back to Yat 5 : 108, according to which K a v i Vstspa sacrificed a hundred
stallions, a thousand oxen and ten t h o u s a n d sheep in f r o n t of( ?) dp- frazddnu-.
29
E d . BAHAB, 918 ; transi. SMIRNOVA, 5460.
3 0
E d . BAHAR, 1 6 1 7 ; t r a n s i . SMIRNOVA, 6 9 6 0 ; f r o m t h e context in Trix-i
Sistn it might be inferred t h a t BundahiSn was known to t h e a u t h o r only through the
i n t e r m e d i a r y of A b u ' l - M u ' a y y a d al-Balx (cf. . 7 above).
31
Cf. . A. STOREY, Persian literature. Vol. I. London 1953. 1 2 9 6 - 1 2 9 7 (transi.
& r e v . by Y u . . BREGEL', Moscow 1972. Vol. I I . 1 0 5 3 - 1 0 5 4 ) .
32
Cf. Yat 19: 66.
33
C f . H E R Z E E L D , op. cit., 95.
31
Transi. WEST, S B E , vol. V, 86 ( X X I I : 5 ) ; Ir. BundahiSn, translit. & transi.
ANKEESARIA, 114115 ( X I I : 6).
35
Cf. my Non-roligious Book Pahlavi literature, Acta A n t . H u n g . 24 (1976) 118.

T H E PAHLAVI T R E A T I S E AVDH V SAHIKN I SAK1STN 265
One of the difficulties in placing it in the geography of Sistan is to find a suitable

lake (if var must be t a k e n to mean exactly lake). Some modern scholars, like
E r n s t Herzfeld, 36 have cast their vote for Gaud-i Zirah, a huge marshy depres-
sion to the south-east of Lake H m n . So does Gherardo Gnoli, who has elab-
orated the connections between legendary and Sistanian geography in great
detail, lately in his richly documented book Zoroaster's time and homeland.31
I n spite of the great ingenuity applied by Herzfeld, Gnoli a n d others to definite
geographic identifications of topographic features figuring in Zoroastrian t r a -
ditions, it seems impossible to avoid the conclusion t h a t F r a z d n , Kayns,
Udtr, etc. basically remain mythical places. 38
To this geographic section of t h e text is added a reference to the three
eschatological saviours who are t o be born of the water of K a y n s from t h e
seed of Zoroaster, i.e. uxSyat.srdta-, uxSyat.nsmah- and saoSyant- (for astvat.
-) of the A vesta. 3 9
The second reason for t h e pre-eminence of Sistan belongs to legendary
K a y a n i a n history, as incorporated into Zoroastrian tradition. This section men-
tions the role of Sistan, and especially Lake Frazdn, in securing the legitimate
succession to the throne of Ern-ahr a f t e r the fratricide among the sons of
F r t n (Av. hrataona-). When the younger son Er, who s a t on the throne of
I r a n , was slain b y his brothers, Salm of the western realm (Hrm) and T z
of t h e north-eastern (Turkistn), t h e succession of MnuSihr t o the throne of
I r a n and the passing of the xuaronah to him was achieved b y hiding a daughter
of Er at Lake F r a z d n .
Here we meet the eponymous heroes of the main peoples of the Iranian
political universe, roughly described as the Shem, H a m , a n d J a p h e t of t h e
I r a n i a n legend by E . G. Browne. 4 0 These peoples are, in fact, only three o u t of
five in the old list given in Yat 13 : 143 144 (Av. airya-, tirya-, sairima-,
sinu-, dhi-), b u t these three completely dominate in later traditions. 4 1
There is, however, a certain discrepancy in the tradition concerning the succes-
sion to the Iranian throne a f t e r Er. Originally (cf. Yast 13 : 131), Mnusihr
(Av. manuS.cihra-) appears as t h e direct son of r (Av. airyva-), b u t in
later tradition he is often, as in t h e Shndmah, presented as t h e son of the daugh
36
Op- cit., 91.
37
= Istituto Universitario Orientale, Seminario di studi asiatici, Series minor, 7 ;
Naples 1980; on Frazdn/Gaud-i Zirah, v. esp. 133.
38
See, e.g., t h e conclusions drawn by J . KELLENS in his p a p e r L'Avesta comme
source historique: la liste des kayanides, Acta A n t . H u n g . 24 (1976) 37 49 (on geog-
r a p h y , v. 41) ; repr. in J . HARMATTA (ed.) : Studies in the sources on the history of
Pre-Islamic Central Asia. Budapest 1979. 41 53.
39
Cf. Yat 13: 128129, Yat 1 9 : 9 2 and Vendidad 1 9 : 5 (where kqsaoya- is
placed in the east ; east of what?).
40
A literary history of I r a n , vol. I, 1902, etc., 116.
41
On the formation of t h e cponyms, v. A. CHRISTENSEN: E t u d e s sur le zoroastrisme
de la Perse antique. Copenhagen 1928. 23.

266 UTAS
t e r of Erc. 42 The present text seems to suggest an e x t r a retardation of ten gen-

erations, in relative concordance with t h e younger Zoroastrian tradition found
in BundahiSn X X X I : 9 11 (cf. p. 261, n. 11 above). The most interesting
piece of information in this passage of the Avdh sahkh Sakistn is, how-
ever, the role given t o F r a z d n and Sakistn. They are n o t generally associated
w i t h these events in o t h e r sources, b u t it is interesting t o notice t h a t Trix-i
Sistn (ed. Bahr, p . 6) writes t h a t AfrdGn, up to t h e time of Mancihr( ?),
e n t r u s t e d Manihr t o Nariman, until he went away a n d took revenge for his
f a t h e r Iraj. N a r i m a n was the great-grandfather of R u s t a m and thus a hero
of Sistan. I t is u n f o r t u n a t e t h a t the t e x t of 8 seems too confused to yield
more exact information on the relation between Artvisr Anht, the other
gods and Sakistn.
The third and last reason for the pre-eminence of Sistan is directly related
t o central Zoroastrian traditions on t h e propagation a n d survival of the faith.
T h e starting-point is once more the mysterious lake(?) of F r a z d n and the
activities of King ViStsp. The latter is here presented as a prime propagator of
t h e religion on a p a r w i t h Zoroaster himself and Sn (Av. sana-, Y t . 13 : 97),
t h e son of A h m s t t of Bust. The opaque syntax of 10 makes it difficult to
determine exactly w h a t relationship is intended between these three heroes of
t h e faith. 4 3
The following description of the restoration of the religion after the devas-
t a t i o n brought a b o u t b y Alexander has a little more verisimilitude t h a n t h e
legendary material presented so far. This restoration is said to have t a k e n
place in Sistan t h r o u g h t h e memorizing of one special nask of the Avesta, which
possibly, with the e m e n d a t i o n proposed by Bailey, could be identical with the
Bagn yaSt described in t h e 15th chapter of Book V I I I of Denkart (cf. above).
This must have been a p a r t of the original Avesta giving particulars about the
divinities, and obviously the circumstances of its transmission are far from the
full story of the survival of the religion. Similarly the localization of this crucial
memorizing must be t a k e n both as symbolic and as relative to the ambitions
of t h e author and his local tradition.
The final p a r a g r a p h of the t e x t , as it stands now ( 16), is somewhat
confusing. Even w i t h t h e rather uncertain emendation suggested above, it is
difficult to see w h a t it is aiming at. Given t h e abruptness of this paragraph and
t h e shortness of t h e preceding list of wonders, in comparison, for instance, with
t h e great number of 'aj'ib presented in t h e beginning of Trix-i Sistn, one
42
E d . (paper-back ed. Tehran), vol. I, 8385; ed. BERTEL'S, vol. I,
107110.
43
On the various aspects of the_complex figure of Vitsp, v. J . MARKWART:
A catalogue of the provincial capitals of E r n s h a h r , ed. G. MESSINA, Rome 1931. 3637,
51 (a water-god out of I n d o - I r a n i a n myth), a n d iecent evaluations by J . KELLENS : op.
cit., 3839, and GH. GNOLI : op. cit., 3, 9495 (with n. 15) et pass.

T H E PAHLAVI T R E A T I S E AVDEH U .S A HI EX 1 SAKISTN 267
could suspect the present text of being only a f r a g m e n t or just an improvisation.

However this m a y be, there is something incidental about its form and message.
I t clearly had its origin in a narrowly religious environment and betrays no
ambition of really presenting a portrait of Sistan. I t is perhaps understand-
able t h a t there is no reference to t h a t great hero of Sistan, R u s t a m , who in
37 of Sahristnih Ern is presented as rtastahm l sakistn-Sh,u b u t it is
striking t h a t there is no mention of the K a r k y fire, for instance. T h a t fire-
temple is not only mentioned in Sahristnih Ern ( 38),45 but even the Mus-
lim Trix-i Sistn devotes a substantial section to it. 46
Now, w h a t we can glean from this t e x t is not what we would expect a
Sistani of the centuries after the Arab conquest t o say on the marvels of his
province, neither geographically nor historically. I t is rather a few f r a g m e n t s
of late Zoroastrian traditions with bearing on decisive moments in the history
of this faith.
University of Uppsala, Sweden.
44
Cf. m y Non-religious Book P a h l a v i literature. A c t a A n t . H u n g . 24 (1976) 118.
48
Ibid., 118.
44
E d . B A H R , 3 5 3 7 ; t r a n s i . SMIRNOVA, 6 9 7 0 .

M. M A R O T H
DIE XALA IN DEN ARABISCHEN QUELLEN
F r die Akten dieser Tagung habe ich den zentral-asiatischen Teil der
Weltkarte von al-Xwrizml vorbereitet. Es bedeutete gewisse Schwierigkeiten,
die K a r t e nach den Angaben des Textes zu zeichnen, aber hinter dieser Arbeit
stecken im Stoff noch weitere, im engeren Sinne genommen nicht mehr geogra-
phische Probleme. In diesem Zusammenhang mchte ich kurz nur ein Beispiel,
das Problem der Xalag, anfhren, und zwar hauptschlich auf Grund der
arabischen geographischen Schriften.
Die bekannteste Textstelle finden wir bei al-Istaxri.
Das Land al-Dwar ist ein fruchtbares Gebiet, die Grenze von al-Gr,
Bagnin und Xalag und Pislang und X . . . Bagnin und Xalag und Kabul und
al-Gr und jene Distrikte nach diesen, haben den Islam angenommen . . .
und die Xalag sind eine Art der Trken, und in den uralten Zeiten sind sie
nach diesem Land gekommen, das zwischen Indien und den Gegenden von
Sigistn liegt, hinter den al-Gr. Sie besitzen Weidevieh, nach der Art der
Trken, und so sind auch ihre Kleider und ihre Sprache. 1
Muqaddasi, als er ber die Flsse von Sigistn spricht, lt den Hilmand
hinter den al-Gur (j>JI ji J ) entspringen. Auf diese Weise knnen wir die
Xalag um die Quellen des Hilmand lokalisieren. 2
Ihn Hawqal 3 wiederholt fast buchstblich, was al-Istaxri geschrieben hat.
Die Quelle von al-Istaxri scheint als Grundlage auch fr al-Idrisi gedient zu
haben, weil er auf die folgende Weise schrieb : Wir sagen, da der stliche
Teil von Sigistn an al-Gr grenzt, und das nchste Distrikt nach den al-Gr
al-Dwar heit. Es ist ein weites, reiches, fruchtbares Gebiet, es ist die Grenze
von al-Gr und Bagnin und Xalag und Pislang . . . Und weiter : Dieses L a n d
wird von einem Stamm namens al-Xalag bewohnt. Sie sind eine Art der Trken,
sie sind nach diesem Ort in uralten Zeiten gekommen. Ihre Wohnsitze grenzen
1
Al-ltaxr : Kit.b al-maslik wa '1-mamlik. E d . M. J . DE GOEJE. Lugd. B a t .
1870. 244.
2
Al-Muqaddasi : Ahsan al-taqsim fi ma'rifat al-aqlini. E d . M. J . DE GOEJE.
Lugd. Bat. 1877. 329.
3
l b n R a w q a l : K i t b al-maHlik wa '1-mamlik. E d . M. J . DE GOEJE. Lugd. B a t .
1873. 302.
Acta Antiqua Aeademiae Scientiurum Hungaricae 1980

270 M. MAROTH
im N o r d e n an Indien, (dann) an den R c k e n von al-Gr u n d a n einige Gebiete

des stlichen Sigistn. 4
Dieselbe geographische Lage d e r al-Xalag lag auch d e m T e x t v o n
M a s ' d i zu Grunde, der den Streifzug von ' A b d a r r a h m n M o h a m m a d al-
A ' a t geschildert h a t . 5
D e r auffallende wrtliche E i n k l a n g zwischen den W o r t e n der oben
e r w h n t e n geographischen W e r k e n lassen u n s d a r a n denken, d a die Verfas-
ser, o b w o h l sie die wirkliche Lage beschrieben haben, doch aus einer u n d der-
selben Quelle s c h p f t e n .
E s m u aber a u c h eine andere Quelle vorgelegen haben, die in der Regel
in d e n verschiedenen I t i n e r a r i e n belegt ist. Bei I h n X u r r a d d b e h f i n d e n wir
z. B . die folgende Beschreibung des Weges : Von Tarz n a c h d e m u n t e r e n
B a r x n 3 Farsax, v o n d o r t nach K e s r B a s 2 F a r s a x . . . d o r t sind die Winter-
sitze d e r Xarluxen, u n d in ihrer N h e die Wintersitze der Xalag. 8
Al-Idris! beschreibt denselben W e g , die E n t f e r n u n g e n werden aber in
Meilen angegeben. b e r die Xalag sprechend f g t er hinzu, d a sie eine A r t
(ein Zweig) der T r k e n sind. 7
D i e Xalag, die zwischen a l - D w a r u n d Oaznin lebten, sind von allen
Schriftstellern einschlgig f r T r k e n gehalten, u n d die X a l a g , die im Norden
im T a l a s - T a l lebten, w u r d e n von Al-Idrisi ebenso als T r k e n bezeichnet.
D i e Xalag sind also in zwei Quellengruppen in zwei verschiedenen, von
e i n a n d e r weit e n t f e r n t e n Gegenden e r w h n t .
E s gibt aber eine d r i t t e A r t v o n N a c h r i c h t e n . M o h a m m a d ihn Nagib
B a k r n schreibt in seinem persisch a b g e f a t e n Werk, d a die X a l a g ein trki-
sches Volk ist, d a sie v o n den X a r l u x - G r e n z e n an die Grenzen von Zbu-
l i s t n k a m e n , u n d in d e r U m g e b u n g v o n G a z n a in der Steppe wohnten. 8 Diese
N a c h r i c h t kann n a t r l i c h wahr sein, sie scheint jedoch eine K o n t a m i n a t i o n
d e r zwei oben e r w h n t e n Quellen zu sein.
V. Minorsky v e r t r a t die Meinung, d a die Texte der zwei Quellen einan-
d e r wiedersprechen u n d deshalb sie wahrscheinlich verbessert werden sollten. 9
Die Xalag-Gebiete u m G a z n a sind v o n d e m Talas-Tal so e n t f e r n t , d a die Nach-
r i c h t sicherlich u n a n n e h m b a r ist.
4
Idris : Opus geographicum. N e a p o l i R o m a e Brill 1974. Fase. I V . 466.
5
M a s ' d i : Les prairies d'or. E d . . BARBIER DE MEYNARD. P a r i s 1869. V. 302.
j! 1La !J IT.... ^Jx- y -u y -- J-"-"^-! (" l
gWI J J > l l r+l J t L <JyJI ^J tiyi'I
8
I b n X u r r a d d b e h : K i t b al-maslik wa '1-mamlik. E d . DE GOEJE. Lugd. B a t .
1889. 28.
' Idris : op. cit. N e a p o l i R o m a e . Brill 1976. Fase. VI. 715.
8
Mohammad ibn N a g i b B a k r n , G i h n n a m e . Moskva 1960. E d . J . E . B o R E V S K i j .
17 b .
9
V. MINORSKY : R u d u d al-'lam. O x f o r d 1937. 347 348.

D I E XALAG I N D E N ARABISCHEN Q U E L L E N 271
Um diese Schwierigkeit zu beseitigen, mssen wir ihm gegenber anneh-

men, da sich diese zwei Quellen auf zwei verschiedene Gruppen der Xalag
beziehen.
Das Bild wird aber mehr verwickelt dadurch, da Al-Ksgari die Xalag
f r Oguz-Trken hlt, Al-Xwrizmi dagegen behauptet, da sie berreste
der Hephthaliten sind. 10
Diese zwei Behauptungen schlieen einander nicht aus, aber die Erfor-
schung der trkischen Sprache einer dritten Xalag-Gruppe, die im heutigen
Iran zwischen Hamadan und Qum lebt, fhrte zu dem Ergebnis, da diese
trkische Sprache einen selbstndigen Zweig der Trk-Sprachen darstellt, sie
ist also der Mundart der Oguzen nicht unter-, sondern nebengeordnet. 11
Vorausgesetzt, da der Name Xalag in diesen drei Fllen drei verschie-
dene Gruppen eines Volkes bezeichnet, schlieen die Ergebnisse der modernen
Forschungen die Richtigkeit der Nachricht von Al-Kgari aus. In diesem Fall
knnte man die Worte Al-Kgaris durch die Annahme erklren, da die zweite
Quelle die Xalag auf dem von den Oguzen bewohnten Gebiet erwhnte, und so
konnte er sie fr Oguzen halten. Das ist aber nur eine Vermutung, die sich
nicht beweisen lt.
Einige Forscher haben angenommen, da die arabischen Quellen nur
in einem weiteren Sinne behaupten, da die Xalag Trken ( J l ^ l y D ^ )
sind. Die Kabgina, die bei al-Xwrizmi zusammen mit den Xalag als ber-
reste der Hephthaliten erwhnt sind, werden von den meisten Forschern nicht
f r Trken gehalten. 12 Das Wort wrde in diesem Fall so viel bedeuten, wie
Nomaden. Die Quellen, die die Lebensfhrung der Xalag schildern, beschrei-
ben sie wirklich als Nomaden, die ihres Vieh in der Steppe zwischen al-Dwar
und Gazna weiden.
Unter der oben genannten Voraussetzung mssen wir das Wort jC
in bezug auf die Xalag jedoeh in engerem Sinne verstehen. Al-Istaxri sagt
nmlich ausdrcklich, da die sdlichen Xalag sich einer trkischen Spra-
che bedienten, und nicht einmal Al-Ksgari kann voreilig einer so unverant-
wortlichen Behauptung verdchtigt werden, da er die nicht existierende
trkische Sprache der Xalag in eine bestimmte Gruppe der trkischen Spra-
chen einordnete, ohne sie gehrt zu haben, ohne etwas ber sie zu wissen.
Gleichzeitig stellen die heutige Existenz einer Xalag-Gruppe und ihre trkische
Sprache schwerwiegende Beweise dar.
Diese drei Umstnde lassen uns also die Worte Al-Xwrizmis im engeren
Sinne verstehen, als er ber die Al-Xalag spricht.
10
C. E . BOSWORTH G. CLOWSON : Al-Xwrizmi on tho Peoples of Central Asia.
J R A S 1965. 2 12. Kgar : Divn lugat al-turk. I I I . 4 1 2 - 4 1 6 .
11
V. MINORSKY : The Turkish Dialect of K h a l a d j . BSOS 10 (1940) 417437.
G. DOERFER : K h a l a d j . Language. Eno. of Islam IV. 918 m i t Literatur.
12
A. D. H . BIVAR : H a y p l a . E n c . of Islam I I I . 303 ff. m i t L i t e r a t u r .

272 M. M A R O T H : D I E XALAG I N D E N ARABISCHEN Q U E L L E N
D a die t r k i s c h e H e r k u n f t der K a b g l n a jedoch n u r d u r c h die W o r t e

al-Xwrizms u n t e r s t t z t wird, so knnen wir nicht entscheiden, was die ge-
n a u e Bedeutung des W o r t e s in bezug auf sie sei.
Die angenommene trkische H e r k u n f t der X a l a g s t e h t nicht im Wider-
s p r u c h mit der N a c h r i c h t von al-Xwrizmi, die besagt, d a die Xalag ber-
reste der H e p h t h a l i t e n sind. 1 3 Diese B e h a u p t u n g bedarf aber einer Erklrung,
die wir n u r h y p o t h e t i s c h geben knnen. E s gibt nmlich eine Meinung u n t e r
d e n heutigen F o r s c h e r n , die die H e p h t h a l i t e n f r T r k e n hlt. 1 4 In diesem
F a l l ist es natrlich, d a auch ihre R e s t e T r k e n sind.
T r i f f t jedoch diese Meinung nicht zu, selbst d a n n h a b e n wir unsere Fol-
g e r u n g nicht zu v e r w e r f e n . Die H e p h t h a l i t e n , wie die N o m a d e n immer, haben
aller Wahrscheinlichkeit nach auch andere Vlker oder Teile von anderen
Vlkern in den V e r b a n d ihrer Stmme aufgenommen u n d auf ihre Wanderun
gen mitgenommen. Diese Annahme k n n t e erklren, wie einige Xalag-Grup-
p e n im Norden geblieben sind, w h r e n d andere im Sden als Abkmmlinge
der Hephthaliten g a l t e n . Diese A n n a h m e gibt uns die Mglichkeit, alle ber-
reste der H e p h t h a l i t e n f r Nomaden zu halten, die teilweise trkischer Ab-
s t a m m u n g sein k o n n t e n , teilweise einer anderen Sprachgruppe angehren
konnten.
Jedenfalls scheint es so, da die angebliche hephthalitische A b s t a m m u n g
d e r X a l a g durch die ewig wiederkehrende B e m e r k u n g unserer geographischen
Schriftsteller, wonach die Xalag in u r a l t e n Zeiten nach ihren sdlichen Wohn-
sitzen kamen, indirekt u n t e r s t t z t wird.
Die arabischen Quellen knnen also gewisse F r a g e n ber die X a l a g
b e a n t w o r t e n , f r a n d e r e Fragen geben jedoch n u r eine hypothetische Lsung,
die im Lichte anderer Quellen weiter u n t e r s u c h t werden soll.
Budapest.
13
AL-KHWRIZM : Mafatlh al-'ulm. E d . V. VLOTEN. Lugduni-Batavorum. 1895-
119120.
11
A . D. H . BIVAR : H a y t i l a .
Acta Antiqua Academiae Srientiarum Hunqaricae 28, 1980

ILYA GERSHEVITCH
THE BACTRIAN FRAGMENT

IN MANICHEAN SCRIPT*
I n J u n e 1958, while cataloguing in Berlin the I r a n i a n manuscripts writ-

ten in Manichean script, Professor Mary Boyce made t h e remarkable discovery
t h a t one small fragment was inscribed on both sides in an unknown I r a n i a n
language. She was anxious t o i m p a r t the news at once to Henning in London,
in a manner t h a t might enable him to identify language and contents. I t
would take time for the two sides of the fragment to be photographed. There-
fore, as the Verso side was h a r d to read, the Recto less so, she transliterated
t h e Recto only, posting the outcome without delay. I t was an outcome never
m e a n t to be definitive. Some indistinct letters later proved to be different
from what she at first thought. F o r some others she left gaps. B u t as a provision-
al first shot this initial revelation, private and to her teacher only, deserves
nothing b u t praise.
The transliterated 19 lines reached Henning on 20 J u n e 1958. They did
m a k e him sit up. He replied on t h e same day, having t a k e n at least a full
hour (his underlining) to make u p m y mind on the language. I t was a language,
he explained, t h a t stands between P a r t h i a n , Sogdian, and Pashto. Hencc, he
continued, it will be the language of the intervening area, Balx, presumably
the same language, at a later stage, as the one in which t h e French had found an
inscription at Surkh Kotal, perhaps also the same language as t h a t of the so-
called Hephthalite fragments.
I n his letter, which consists of a single sheet written on both sides, Hen-
ning then proceeded to details. H e had found translatable four sequences, of
which I quote only the words he correctly translated from the spelling here
reproduced as at the time it stood before him.
First sequence : mwwl myg(d)yg 'ivwd 'sp(r)gmyyg 'yryynd 'wd cgwwng
'ww w(S) 'wd . . . 'cyy ' [ ] *wd g"w Ihyyh t'(')d pn'yyd p'yyg 'wd
rwgn 'y(r) [ roots (mla) fruits and flowers produce ; and as the (<.hau)
grass and . . . which to (<.abi) [sheep] and cows you( ?) give, then . . . cheese
(-r ?) milk (payah-) and b u t t e r t h e y produce.
* (Paper delivered at tho Conference The Sources of the History of Pre-Islamic

Central Asia in Budapest on 25 March, 1976.)

274 I. GERSHEVITCH
Second sequence : S(.)g qyrdg'n wy(p) 's l'(hw)'n (. .)jyd good (Brak ?)
action all from (las < ats) gifts arise.
Third sequence: ywg p(w.) hz'r pwnyng rg (,)yryyd 'wd wy(p) nrh
wwhyyd 'wd pwwn pydrw[.]d y'wyd'njyg A single (aiwaka-, Pashto _y_)
p u n y a a thousand p u n y a - f r u i t s (Sogd. r'kw) produces and wipes out (? cf.
krmSwhn) all hells and . . . ( ? cf. P a r t h . pdrwb-) puny as eternally.
F o u r t h sequence : ('cy)yd Smn'n '(d)hn xwynynd whom Buddhist monks
call arhant, postulating a d o t over the d of 'dhn. The word occurs once more on
the R e c t o , again w i t h o u t visible dot over t h e d.
Of individual words Henning recognized in lyynl(.)yg"n the plural of
t h e word meaning an Electus, a dnddr, which in fact occurs in the singular,
as lynlyryg, on the Verso ; and he of course compared t h e noun I'hw'n with
P a r t h . d'hw'n.
I n addition his letter contains remarks on phonology. H e noted the pala-
tization of s in wysp, expressed, as Professor Boyce had told him, by a special
diacritic, two dots placed over the usual Manichean sign for S. The same dia-
critic placed over the j sign, in y'wyd'njyg for example, is used, he surmised,
to indicate t h a t thus m a r k e d the sign stands for dz. He pointed out the voicing
a f t e r vowel of Olran. a n d t ; the change of d t o the outcome S of str in wS ;
the prothetic vowel of ' c y which, with which he compared Sogd. 'cw ; the s of
's from, from the of O l r a n . hac ; a n d t h e interior of -aya- present stems,
exemplified by xwyn- to call and 'yr- to bring, produce.
I t was a rich communication, a t o u r de force achieved in one hour, with
no Verso as yet to compare and from a transliteration not quite perfect. B u t
this first pronouncement of Henning's on the fragment was also his last, be-
cause thereafter he was t o communicate on it only with himself, to whom he
stood in no need of i m p a r t i n g his thoughts in writing. I t was his habit to think,
a n d when he had finished thinking, t o compose the final article at his type-
writer, ready for the p r i n t e r . This explains why Professor Boyce never received
f r o m him any further details about the t e x t , despite its having been his inten-
tion to publish it under b o t h their names in a joint article.
H e r own share, however, of the planned article, Professor Boyce promptly
b r o u g h t to completion : on 12 J u l y she posted to him a painstaking trans-
literation of both R e c t o a n d Verso, supplying an apparatus in which she dis-
cussed possible alternative readings and t h e spacing of gaps, and enclosing
photographs. This was a great service rendered, by which she crowned her
discovery of the f r a g m e n t with the provision of every help possible t h a t could
be given from the distance. The next and final step it was for him to take. He
never took it, in t h e eight and a half years still left to him.
I n 1968, one year a f t e r Henning's death, Professor Boyce wrote to ask
if I would care to t a k e over what was t o have been their joint publication.
I replied t h a t I was willing to try. Henning's dear wife, who unhappily one
Acta Antiqua Academiae Seientiarum Hungaricae 28, 1980

T H E BACTRIAN F R A G M E N T I N MANICHEAN S C R I P T 275
year later was also to die, having agreed and the German Academy, through
t h e good offices of D r Sundermann, having kindly approved of the transfer,
I received from Professor Boyce the letter from which I have quoted, from Mrs
Henning her husband's file on the f r a g m e n t , and from D r Sundermann two
enlarged photographs, one each of R e c t o and Verso.
Henning's file contained only one sheet written entirely in his hand,
namely his own transliteration, and even t h a t only of the 19 lines of the Recto,
with a succinct apparatus. There was no translation, except indirectly of t w o
words. One, restored by him as p(rdyj)g, he had glossed with N P plz. I t
occurs in the line preceding the first sequence quoted above. Against the other
word, xwyc, he h a d pencilled, with a question mark, Av. -. Other-
wise the file contained only Professor Boyce's two contributions : her first,
provisional transliteration of the Recto, which had prodded him into the t o u r
de force she by t h e n had sent me and I h a d eagerly read ; and her second, final
transliteration of both sides of the sheet. This second contribution is of inval-
uable help for reading the difficult Verso written in a different hand from t h e
Recto, and much more faded and smudged, requiring extensive restoration.
Against a few of Professor Boyce's readings of the Verso Henning had j o t t e d
down in pencil alternative readings, thrice with a translation each of one word :
recalls with a question mark, against ' y ' s ' d ; that against kwd\ and German
hiesig against mT(m)'(n)g. Moreover, against two of her readings left untouch-
ed b y him, he h a d entered a translation, namely vov against lyrng fr, a n d
he calls him against ' w d y h xwynyd.
These are small crumbs from the table of one of the best-endowed brains
t h a t ever turned t o Iranian studies. W h a t e v e r they were crumbs of, perished
on 8 J a n u a r y 1967, together with t h a t brain.
When I had t a k e n stock of the position, my heart sank. I t was of course
clear t h a t the Recto contained two R u v n e g n similes new to us, a vegetable
one and an animal one. The laymen's gifts to the Elect produce pupyas a n d
salvation, as irrigation induces roots t o produce fruits and flowers, as grass
a n d water (y', t o replace the three dots a f t e r wS 'wd in the first sequence above)
given to cattle produce meat (p'yyd, to replace the five dots after i'(')d),
cheese, milk and butter.
I t was also clear, and not only from the vov jotted down by Henning
against lyn'ng fr, t h a t the Verso was of esehatological purport. Quite a n u m b e r
of words even on the Verso make immediate sense, such as 'z 'ydym I have
come, or rStyg l'dr wyspz'nyndyy the j u s t all-knowing judge.
B u t the greater part of the Verso seemed to consist of words either un-
known or illegible, or of whose reading there could be no certainty ; a n d
nowhere on the Verso was it clear how sentences, let alone periods, were t o be
construed. W h a t , for example, did Henning think, w h a t was I to think, of t h e
mysterious drwfr in line 13 of the Verso, whose occurrence also as first word of

276 I. G E f t S H E V I T C H
line 10 of the Recto he had recognized with uncanny perspicacity ? W h a t did

he m a k e of yw'g lyynl(yr)yg' 'n, even though in the longer of these two words
he h a d at once recognized, as we saw, t h e Electi ? W h a t did he think of the
sequence 'wdyh 'Iwg nyst qyrd w(), which is clearly a self-contained clause,
seeing t h a t it is followed by a sequence beginning with 'wdyh and ending with
a p a s t participle? A n d what were t h e Buddhist monks doing in the three
b o t t o m lines of t h e Recto, with whom Henning had ingeniously grappled in
his t o u r de force letter, but on t h e premise of a transliteration not wholly
correct, and in any case with the beginning of all three lines lost in a hole?
This sramana passage is crucial in deciding whether or not the eschatological
Verso, written in a different hand a n d deprived at the t o p of one or perhaps
t w o lines, was at all t h e continuation of t h e Ruvnegn statement of the Recto,
or constituted a new text under a new, lost headline.
To each of these vital questions Henning was bound to have given anxious
t h o u g h t . If clues to t h e answers occurred in Manichean texts in other languages,
published or unpublished, nobody was better equipped t h a n he to recognize
t h e m . Nor could there be any doubt t h a t , having in 1960 revealed in a printed
footnote t h a t he was about to publish the fragment, and in the same year
authorized Mary Boyce to reveal in her Catalogue his identification of its
language as Bactrian, he would b y all means have proceeded to publication
w i t h the urgent speed characteristic of all his major discoveries, if only he felt
satisfied t h a t he h a d t h e right answers t o the above questions and to the m a n y
more which the t e x t poses. The conclusion was inescapable : what had stopped
him f r o m publishing it was awareness t h a t what he had written to Professor
Boyce within the first hour, and a n y t h i n g t h a t had occurred to him since,
t h o u g h it did go some way towards clarification, was as yet insufficient t o
w a r r a n t publication under his name.
The conviction t h a t this had been t h e cause of his silence was to me, who
knew very well his dread of posthumous publication, an insuperable bar to
bringing into the open what had been entrusted to me of his. I had a right to
bring it into the open only if I saw it lead to a breakthrough, which evidently
it h a d not done even under his own, exceptionally high-powered care. As to
myself, throughout t h e five years f r o m 1968 to 1973, despite m y time and again
t u r n i n g t o the f r a g m e n t , I could see no prospect of breaking through. Words
which I recognized unaided by a n n o t a t i o n s of Henning's, such as p'yyd for
meat or y' for water, or hys of which as yet I saw no meaning but took for
g r a n t e d the identity with what in Hephthalite is written vaitfo, did not really
f u r t h e r an overall understanding. I t was unthinkable t h a t such trifles of recog-
nition had by-passed Henning, b u t clearly they did not drive him into print.
T h e y must therefore n o t drive into p r i n t me either, especially as the immediate
sequel to hys, namely (. . )ng pwwn drwfr, with (. )ng as a possible alternative
t o (. . )ng, remained, a p a r t from the word for punya, incomprehensible.

B y 1973 I h a d no illusion as t o w h a t Henning would have wanted me t o

do : destroy everything I had of his. B u t this was ruled o u t b y the trust M a r y
Boyce and the G e r m a n Academy h a d shown me. H a d t h e y t h o u g h t me capable
of t h u s dealing with his notes, t h e y would have m a d e sure of keeping t h e m
safely out of my reach. I was caught in a conflict of loyalty, r u n aground in a n
impasse.
The fact is t h a t , as the observations of Henning's which you have h e a r d
struck me as obvious f r o m the m o m e n t I saw them, y e t t h e f r a g m e n t remained
largely incomprehensible, I had come t o underestimate his achievement. T h a t
achievement had been b r e a t h - t a k i n g a t the time when he reported on it a f t e r
one hour. B u t given a few days t o t h i n k , let alone weeks months or years,
almost a n y Iranianist, or so I began t o think, would h a v e achieved as m u c h .
The conclusion seemed obvious : if even Henning in y e a r s a f t e r the miracu-
lous feat achieved no progress, or no significant progress as his failure t o
publish left no d o u b t h a d been t h e case , t h e n the difficulties must simply
be insoluble beyond t h e most elementary stage. H a v i n g drawn this con-
clusion by the time of the B u d a p e s t conference of 1973, if I had been able
t o a t t e n d it I would have been r e a d y t o declare in t h i s very Kisterem, t h a t
I was giving up.
B u t some m o n t h s later in 1973 I was able to a t t e n d t h e International
Congress of Orientalists in Paris. And in Paris something happened which
completely changed m y outlook.
A t one of the Iranological sessions a paper was delivered which included
a wholly convincing demonstration t h a t the Bactrian word for wine was mal.
There followed a discussion. And in t h e course of the discussion, a declaration
was made before t h e packed international audience t h a t no, it was not mal,
it was mul as in Sogdian, seeing t h a t t h e Bactrian word for wine occurs in t h e
Berlin fragment, w r i t t e n mwwh. T h e speaker evidently h a d in mind the f i r s t
word of the first sequence quoted above.
I was too d u m b f o u n d e d even t o open my m o u t h . While others were
challenging the speaker as t o how on e a r t h he had got hold of the Berlin frag-
m e n t , I fell into a kind of trance, lost in mourning and nostalgic admiration for
t h e unfailing greatness of t h e scholar it h a d been m y s u p r e m e privilege to h a v e
as teacher.
N o t once in t h e five years h a d it occurred to me, w h o a m not unfamiliar
with Sogdian, t h a t t h e first word of t h e quoted sequence might mean wine.
The context precluded t h e very t h o u g h t of such a thing. B u t then, I had never
been in a position t o t r y and unravel the context b y myself. Straightway I
was presented with H e n n i n g ' s t o u r de force letter, so t h a t it was quite impos-
sible for me to go wrong on mwwl a n d its context when later the photographs
arrived, b u t alas possible, in t h e light of the d i s a p p o i n t m e n t a t Henning's
notes offering help with so little else, t o give him v e r y m u c h less t h a n his

278 I. GEftSHEVITCH
d u e for his immediate grasp of the first sequence's purport. I t took the Paris
u p r o a r to shake me o u t of this unappreciativeness and m a k e me realize w h a t
a horrible t r a p mwwl posed, and w h a t a test this word constituted of the t r u e
q u a l i t y of would-be decipherers. I h a d literally to hear t h e t r a p snap, to con-
v i n c e myself t h a t it was not true t h a t given days weeks m o n t h s or years, others
w o u l d also find o u t w h a t Henning saw in the first hour.
And so, there a n d then in the P a r i s Kisterem, the realization took hold
of me t h a t if I were n o t to publish w h a t I h a d from Henning, even though he
n e v e r thought it w o r t h bothering t o p u t it into print by itself, years if not
decades might pass sucking casualties into not only this t r a p , b u t also other
t r a p s equally insidious, which Henning with sure step h a d avoided.
I returned t o Cambridge a reformed character, and at once set to work in
t h e mood of one j u s t emerged from an illness. I was no longer troubled by not
understanding crucial passages and key-words of the Recto, a n d by my inabil-
i t y t o construe even one period of t h e Verso. Henceforward I was going to be
n o more t h a n an amanuensis. All I would do was to publish the text in the
reading of Mary Boyce with the improvements Henning h a d entered here and
t h e r e , and the addition of such alternative readings as here a n d there had come
t o seem to me w o r t h considering. The t e x t would be followed by the bits of
translation and t h e few hints I had f r o m Henning, and b y a few self-evident
t r i f l e s which had occurred to me in five years. I no longer felt it incumbent on
m e to achieve progress.
B u t no sooner h a d I begun to proceed in this strictly expository manner
insouciant of progress, when progress no longer prayed for took to beckoning
f r o m all sides. I t was as if the f r a g m e n t , for so long a d e a d weight round m y
n e c k , had suddenly sprung to life. One d a y I saw t h a t t h e key-word drwfr
occurring, you will remember, once on the Recto and once on the Verso,
w o u l d make good sense if it were a metathesized outcome of the Zoroastrian
t e r m ratufrl-. We h a d not had this word before in Manicheism. I t qualifies
p u n y a in the f r a g m e n t , and thereby sheds new light on the Manichean doctrine
of redemption, indeed on the very essence of the Ruvnegn institution. Surely
n o religion could survive which preached as sole duty of laymen the feeding of
its Elect. Priority m u s t be given at least in theory, to religious behaviour,
p u n y a s . B u t religious behaviour on t h e part of laymen, w i t h o u t their assid-
uously attending t o gifts of food, would have exposed the poor Elect to star-
v a t i o n . Hence t h e clever invention of a rider : the p u n y a s must be ratu-
pleasing, t h a t is, inducing the ratus t o grant forgiveness of sins ; and what
renders them ratu-pleasing is the giving of gifts of food t o the Elect. The
vegetable simile a n d the animal simile of t h e Recto thus emerge fraught with
doctrinal subtlety : without roots there can be no fruit, w i t h o u t cows no milk
or beef, without p u n y a s no redemption. B u t roots will n o t give rise to fruits
unless they are watered, cows will n o t yield milk or beef unless they are fed.

And punyas will not lead to redemption unless they secure ratu-pleasure, are
RATified as it were, by the care t a k e n of the Elect.
Once drwfr made sense, it became clear t h a t the sequence hys (.. )ng
or hys (. )ng quoted above, is like drwfr a qualification of pwwn, or rather,
qualifies pwwn drwfr taken together. A likely meaning of it will therefore be
of very m a n y kinds, something on t h e lines of Sogd. zngzng'n, yrf znq'nt,
wyspw znk'n, O P vispazana-, Aram, wspzn. As soon as one thus realizes t h a t
the context could do with the presence of a z, one actually sees t h a t the traces
preceding ng tally with z, and with nothing else. To account for hyS (the
Hephthalite vaioo) expressing large quantity one m a y relate the word t o
K h o t . harpbisa, P a r t h . and Pers. ambls heap. The base will be the bai- of Av.
bavar- ten thousand, the suffix t h a t of Av. navas by nines, nine times
and Sogd. IIsw (wasu) at the rate of two, the prefix t h a t of Greek na;
quite all, all possible (as distinct from OIran. hama-, present in hambisa a n d
ambis). Literally the adjective will t h u s mean thoroughly-hugenumber-fold.
On another day the daringly primitive thought struck me t h a t t h e puzzl-
ing yw'g which precedes lyynl(yr)yg"n, seeing t h a t in P a r t h i a n the Elect are
often called pw'g pure, might be t h e very same word preceded by the article
after which its initial p would in this language have suffered voicing, as hap-
pens in Celtic with feminine nouns. I n Welsh, for instance, a pipe is pip, b u t
the pipe y bib, with soft m u t a t i o n (and assimilation). Thereby also t h e
existence of the article i, familiar f r o m the Nokonzok inscription, would be
assured for Manichean Bactrian.
From yw'g lyynl(yr)yg"n meaning the pure Electi it would follow t h a t
t h e verb pydrw- corresponds only formally to P a r t h . pdrwb- to render dis-
traught (cf. Skt rpyati, L a t . rumpo), since in the f r a g m e n t its meaning would
have to be to garner, reap, in agreement with the simple rwp- in Sogdian.
Soon afterwards it was the B u d d h i s t monks a t t h e bottom of the Recto,
and the Arhants, who fell into place, a t least to the e x t e n t of no longer compel-
ling the assumption t h a t the eschatological Verso constitutes a separate t e x t
pursuing a different trend of t h o u g h t from the Verso.
Of course each individual step forward was a p t to require a reappraisal
of the whole. Thus gradually the Verso came to unfold a growing n u m b e r of
meaningful stretches, until one day it at last dawned on me t h a t my inability
t o construe periods had been duo to nothing more alarming t h a n the abandon-
ment b y the language of the oblique ending of nouns. I t was then t h a t at long
last the meaning of the complete clause ' w d y h 4wg nyst qyrd w() emerged,
a n d with it a remarkable Manichean doctrine hitherto missed, about t h e over-
riding role played by Beauty, with capital B, in the redemption of Man a n d
indeed the World. When the evidence is in print,* it will be seen t h a t t h e
* And already preliminarily, in a lecture which I gave in Budapest fifteen d a y s
later, on 9 April 1976, to t h e Society for Ancient Studies.

280 I . G E R S H E V I T C H : T H E BACTRIAN F R A G M E N T I N MANICHEAN S C R I P T
supremacy of B e a u t y also explains at last why the Ruvnegn, the alms-gifts,

were called by t h a t n a m e . I t is an explanation which could have been found
ever since 1960, f r o m t h e moment Mary Boyce published Henning's view t h a t
t h e Middle Persian n a m e xwSn of t h e Living Soul meant Beauty. B u t t h e
o p p o r t u n i t y was missed because this view failed to get credit, seeming t o
require independent confirmation. T h e Bactrian clause now provides it :
and he O F grain consigned-to-perdition t h e Beauty (namely by eating t h e
grain himself instead of giving it to t h e Electi). The word w, which had teased
me, a n d no doubt H e n n i n g , for years, is nothing else b u t the Middle Bactrian
outcome of O l r a n . vafuS, whose identity of form and meaning with Vedic
vapus was recognized b y Andreas and Wackernagel half a century ago, and is
now confirmed beyond all reasonable d o u b t b y inter-Iranian Manichean evi-
dence.
Naturally as a result of the f r a g m e n t t h u s springing to life and revealing
far-reaching novelties of Iranian Manichean thinking, it took me longer t h a n
I h a d foreseen on m y r e t u r n from Paris, t o prepare it for publication. B u t b y
now it is possible t o offer a connected translation of the Recto, and to grasp
t h e h a n g and partially even the s y n t a x of t h e Verso, although on t h a t side t h e
indifferent state of preservation will keep us reduced to guess-work on m a n y
details for so long as no parallel version in another language is found.
Cambridge.
Acta Antiqua Aeademiae Scientiarum Hungaricae 28, 1980

ILYA GERSHEVITCH
BEAUTY AS THE LIVING SOUL

IN IRANIAN MANICHEISM*
T h e one b u r n i n g concern which p e r v a d e d all M a n i c h e a n thinking, w a s

t h e c a t a s t r o p h e which befell t h e L i g h t t h a t had gone f o r t h i n t e n d i n g t o d e f e a t
t h e P o w e r s of D a r k n e s s , b u t i n s t e a d w a s itself d e f e a t e d a n d swallowed u p .
I t is t r u e t h a t p l e n t y more L i g h t w a s left in t h e H i g h e s t H e a v e n , t h e
s e a t of t h e F a t h e r of G r e a t n e s s . B u t t h e loss of L i g h t h a d b e e n of cosmic p r o -
p o r t i o n s , a n d t h e H i g h e s t Aeons can h a v e no peace u n t i l t h e missing L i g h t is
r e s t o r e d t o t h e m , t h e L i v i n g Soul, t h e A n i m a V i v a d e s p e r a t e l y longing f o r
liberation f r o m c a p t i v i t y in D a r k n e s s a n d for r e t u r n h o m e . The whole of
M a n i ' s highly e l a b o r a t e s y s t e m a m o u n t s t o one g i g a n t i c rescue operation, of
t h e A n i m a Viva s u f f e r i n g t o r t u r e in t h i s dismal d o m a i n of H y l e in which w e
live.
I m u s t spare y o u t o - n i g h t t h e E m a n a t i o n s , t h e c o s m o g o n y , t h e cosmol-
ogy, t h e m a n u f a c t u r e of M a n , all t h e devices aimed a t t h e rescue of t h e L i v i n g
Soul. B u t we m a y join t h e L i v i n g Soul, or r a t h e r t h e p a r t i c l e s of L i g h t w h i c h
c o n s t i t u t e it, a t t h e m o m e n t when, f r o m edible p l a n t s , c u c u m b e r s for e x a m p l e ,
or t h e w h e a t of which b r e a d is m a d e , t h e particles of L i g h t p a s s into t h e bodies
of p u r e Manichean E l e c t i engaged in e a t i n g c u c u m b e r o r b r e a d .
T h e Electi a c t as h u m a n filters for t h e particles of L i g h t . As t h e y e a t
a n d digest t h e foods c o n t a i n i n g L i v i n g Soul, t h e l a t t e r escapes t h r o u g h t h e i r
bodies t o t h e C o l u m n of Glory. Along t h a t Column t h e particles of L i g h t
m o u n t f r o m E a r t h t o H e a v e n , invisible t o t h e h u m a n e y e despite their a s c e n t
in c o l u m n - f o r m a t i o n , because t h e y consist of n o t h i n g b u t light a t its p u r e s t .
Visibility intervenes only w h e n a t t h e u p p e r end of t h e C o l u m n t h e Moon-ship
arrives and, visibly t o all of us, fills u p , over a period of f i f t e e n days, w i t h
particles of L i g h t as t h e y d i s e m b a r k f r o m t h e Column a n d r e - e m b a r k in t h e
Moon-ship. W h e n t h e Moon-ship is loaded, it t a k e s i t s passenger-particles t o
t h e Sun-ship, for f u r t h e r , f i n a l c o n v e y a n c e t o t h e R e a l m of everlasting L i g h t .
N a t u r a l l y t h e r e a r i n g of p u r e Electi w a s a crucial p a r t of Mani's plan f o r
t h e liberation of t h e L i v i n g Soul. F o r only Electi w e r e c a p a b l e of acting as
* Lecture delivered in Budapest on 9 April 1976 at a meeting of the Society for

A n c i e n t Studies, Professor J . HARMATTA in t h e chair.

282 I. GERSHEVITCH
f i l t e r s for the Light. B u t of course no Electus was pure a n d could act as filter
if he himself should in any way have h u r t or harmed t h e Living Soul, for
i n s t a n c e by cutting bread with a knife or, Heaven forbid, trampling on some
c u c u m b e r in a field. I n fact it was so risky to let him do anything to the
L i v i n g Soul other t h a n eat it, t h a t t h e whole preparation of food, from cultiva-
t i o n of crops to t h e actual cooking of meals had to be left t o the Auditores,
t h e Manichean laymen.
These were n o t sufficiently pure t o act as filters for t h e Living Soul. B u t
t h e y could nevertheless hope for salvation in the other world. Their impurity
a n d their sins, including the dreadful one of hurting the Living Soul, would be
forgiven them if t h e y made sure t h a t vegetarian food daily reached the mouths
a n d bodies of the Elect, so as to m a i n t a i n a constant u p w a r d stream of liber-
a t e d Living Soul along the Column of Glory.
This duty which the Manichean Auditors had, of feeding the Elect, and
incidentally also clothing them and providing them with shelter, was known
u n d e r the term of alms-giving or gift-giving. One third of every Auditor's
working-day had t o be devoted to alms-giving pursuits, in fields, workshops
or kitchens.
Of course also t h e Auditors ate. Their eating brought t h e Living Soul no
d i r e c t benefit. B u t for obvious reasons it was not considered sinful for them
t o e a t , provided t h a t t h e y did not eat portions which t h e y should have made
over t o the Elect. If t h e y did t h a t , or indeed, if they fed Electi who wrongly
claimed to be pure, t h e y became guilty of grievous sin against the Living Soul,
consigning it to perdition at the very moment when it h a d reason to expect
t o be a t last about t o be filtered.
I t was necessary t o dwell on these details, because in t h e only surviving
Manichean fragment written in Bactrian language, which I a m preparing for
publication and on which I had t h e honour to report here in Budapest a fort-
n i g h t ago at an International Conference convened by Professor H a r m a t t a ,
t h e remarkable sentence occurs, and of grain he consigned t o perdition the
beauty.
The Manichean t e x t in question is a so-called R u v n e g n text, Ruvne-
gn being the Iranian Manichean term for the institution of alms-giving to the
E l e c t i on the part of Auditores. The s t a t e m e n t I have quoted, and of grain he
consigned to perdition the beauty, appears to have as subject an Auditor
who h a d been remiss in his Ruvnegn duties. What he consigned to perdition
was therefore the Living Soul imprisoned in grain. B u t it is of quite unusual
i n t e r e s t to find the Living Soul here being referred to under t h e simple name
of Beauty.
You may of course say t h a t I am wrong, and t h a t t h e Bactrian word
wa which I translate as beauty, means something quite different. I t is a
B a c t r i a n hapax. Most Bactrian words of any interest are hapaxes.
Acta Antiqua AcademiueScientiarum Hungaricae 28, 1980

B E A U T Y AS T H E LIVING SOUL I N I R A N I A N MANICHEISM 283
B u t wa is comfortably etymologized as corresponding to Rigvedic

vapus shape, beautiful shape, beauty, which in Avestan occurs as vafuS.
The Bactrian sentence therefore immediately reminds one of a suggestion
offered by the late Professor Henning to Mary Boyce and published b y her in
1960, to the effect t h a t the Middle Persian name xwSn of the Living Soul
meant beauty. When after Henning's death in 1967 I enquired of Professor
Boyce what reason other t h a n etymological ho had given her, she t h o u g h t
none, and even t h a t subsequently Henning himself gave up the idea. B u t
now t h a t Bactrian wa seems t o confirm its correctness, we have a more
pressing incentive t h a n Henning appears to have felt, t o cast round our minds
in search of even further confirmation, t h a t with any luck might clinch
t h e matter.
I believe such luck to have come m y way, in the form of a reappraisal,
forced on me by wa, of the meaning of t h e names which the Column of Glory
bears in P a r t h i a n and Sogdian. I n P a r t h i a n texts the Column is called h'mcyhrg,
in Sogdian texts "wkrSny. B o t h names are compounds, each having as first
term t h e P a r t h i a n , respectively Sogdian, outcome of Old Iranian hma-, to
which in Greek etymologically corresponds 6/-. The second compound-term,
cyhr in Parthian, krSn in Sogdian, means shape, form, yoocprp). Accordingly it
was thought until now t h a t in b o t h languages the literal meaning of the
compound was homomorphic, without any convincing explanation being
available as to why homomorphic should have become a technical t e r m for
t h e Column of Glory. B u t you have now seen three distinct Iranian words for
shape, form : the Old Iranian vafuS whose Vedic equivalent in addition to
shape meant beautiful shape, a n d P a r t h i a n cyhr as well as Sogdian krSn,
adjectival derivatives of both of which mean shapeLY, beautiful by t h e same
semantic evolution as from L a t i n forma produced formosus. The Column of
Glory, as we saw, consists uniformly and exclusively of particles of Light which
are Viva Anima. B u t if t h a t Light, if t h a t Living Soul, was also called Beauty,
t h e n surely the yogcprj of the term homo-morphic stands for shapeLIness,
and t h e name of the Column m e a n t not incomprehensibly homomorphic,
b u t T R A N S P A R E N T L Y that which evenly consists of Light, the t e r m for
Light = Living Soul here used being Beauty.
We may now take it for granted, on the strength of Manichean writers
in four distinct Iranian languages each applying to the Living Soul a t e r m for
beautiful shape peculiar to his own language, t h a t really in Iranian Mani-
cheism, and so far as the evidence goes only in Manicheism Iranian, t h e Living
Soul was regarded as sheer and exclusive Beauty. B u t to recognize this as a
fact means to reach the core of the problem. Why is t h e Anima Viva Beauty,
and why is she Beauty only in Manicheism Iranian ?
There can be no doubt t h a t the term Living Soul had been Mani's
own, coined by him in the Aramaic language of Babylonia in which he first
Acta Antiqua Academiae Scientiaruin Hungaricae 28, 1980

284 I. GEftSHEVITCH
f o r m u l a t e d his doctrine. F r o m there St. Augustine got his Anima Viva, as in-
deed also in Iranian Manicheism the Living Soul is often called, by Iranian
t e r m s meaning respectively living and soul. B u t it is well known t h a t when
Mani, a Persian grown u p in Babylonia, preached to Iranians, he often trans-
l a t e d his own religious terminology into a quite different, typically Iranian
w a y of thinking.
Bearing this in m i n d , if we wish to u n d e r s t a n d why Mani presented t h e
L i v i n g Soul to Iranians as B e a u t y , we m u s t not fail to ask if by any chance
earlier, pre-Manichean, I r a n i a n religious thinking could have served him as
model for renaming t h e Living Soul. Many of you will b y now have guessed
w h a t I am making for : t h e famous, age-old belief related in the HaSxt
N a s k of t h e Avesta, t h a t every man has a D a n , a personification or live illus-
t r a t i o n of his soul in t h e other world, whom he will meet and see for the first
t i m e a f t e r death. A m a n who has lived a good life will encounter after death
his D a n in the shape of a beautiful young woman. A m a n who has led an
evil life will find t h a t his D a n has the looks of an ugly old hag.
Men will meet their respective Dans only a f t e r death, when of each
t h e s u m of all stages of behaviour throughout life has been drawn and the total
is, as it were, frozen, no longer alterable, reflected in a Dan immutably beauti-
ful, or i m m u t a b l y ugly, or of appropriate intermediate looks likewise immutable.
B u t it stands to reason t h a t if it were possible for us to slip out from time t o
t i m e i n t o the other world while we are still alive, just to have a peep at how
our D a n is faring, we might find her on certain days uglier or more beautiful
t h a n on other days, according as our balance of actions stood on the day in
question. This is just how Dorian Gray was able stage b y stage to observe t h e
deterioration of his soul, b y furtively mounting to the room he kept locked
a n d watching the uglification of his portrait. I cannot help i t thinking t h a t
Oscar Wilde must have d r a w n the inspiration for his Picture of Dorian Gray
f r o m t h e Zoroastrian story of the Dan.
B u t now, to convince you t h a t we are not dealing with mere coincidences,
b u t are on the track of something solid, grounded on age-old Iranian tradition,
let m e bring you face t o face with the problem as yet unsolved, as to why t h e
Manichean institution of alms-giving by Auditors to Electi is called in Iranian,
a n d only in Iranian, R u v n e g n .
T h e word ruvn, which in Avestan has t h e form urvan-, is the pan-Iranian
word for soul, so t h a t literally Ruvnegn means that which has reference
t o t h e soul. Of course in religion almost anything has reference to the soul,
b u t it has never been clear why a term so ostentatiously sporting the soul,
should be the very one chosen to denote the p r e t t y materialistic concern with
t h e feeding, clothing and housing of Manichean monks. For it is not the Living
Soul which was alluded t o by the term R u v n e g n , seeing t h a t in Iranian Mani-
chean terminology t h e word used for the anima of Anima Viva is not ruvn,

B E A U T Y AS T H E LIVING SOUL IN I R A N I A N M A N I C H E I S M 285
b u t another word for soul, grw. Therefore all t h a t Iranologists have sofar
been able to suggest is t h a t the term Ruvnegn alludes to the spiritual wel-
fare and salvation of the Auditor who conscientiously performs his alms-
giving duties.
This definition of the implications of the term R u v n e g n is certainly not
incorrect, b u t it does by-pass w h a t to my mind is a pointer to a more exact
definition. This pointer I would recognize in the epithet hu-ruvn which in
Middle Persian Manichean texts is frequently bestowed on Auditors who excel
in performing their religious duties. I n his 1933 publications Henning trans-
lated hu-ruvn as gut-seelig, referring to the Zoroastrian, already Avestan
term hurunya- of which Bartholomae's explanation reads : Gutseeligkeit,
eigentlich Zustand eines *huruvan-, des Seele gut ist. L a t e r , however, Henning
noticed t h a t the Sogdian Manichees translated t h e Middle Persian term
huruvdn as if to them it meant soul-loving, which is w h y he himself in sub-
sequent publications translated into English even Middle Persian huruvdn as
soul-loving. Yet there can be no d o u b t t h a t a compound of hu- and ruvcin
can strictly speaking mean only he whose ruvn is hu-, never he who is hu-
towards his ruvn.
The catch, as I see it, lies in w h a t has been done with hu- in this com-
pound. Of course hu-ruvn can mean des Seele gut ist. B u t if this was a term
used already in Zoroastrian antiquity, in the same Avestan scripture which
tells us t h a t a man who is good, hu-, in this world, will f i n d in the other world
t h a t his Dan is not good, ethically, but beautiful, aesthetically, visually
attractive, then surely Avestan hurunya- is not the Zustand of one whose
soul is good, which in any case would be a pretty meaningless Zustand, b u t the
Zustand of one whose Dan in t h e other world, whose picture of Dorian
Gray in the locked room, is beautiful. I n Avestan usage hu- can have either
meaning : the name Hutaos, for instance, means not she whose thighs are
good, b u t she who has beautiful thighs.
So we see t h a t when Iranian Manichees, and quite likely Mani himself,
describe dutiful Auditors as huruvn, they consciously or subconsciously fol-
lowed an age-old Iranian tradition we already find in primitive Zoroastrianism,
which translated ethical goodness into physical beauty, which loved goodness,
so to speak, with eyes.
Those of you familiar with Zoroaster's Gathas, will remember t h a t in
certain verses the prophet uses the term dan- interchangeably with the term
urvan-. I t is in fact t h a n k s to this partial interchangeability t h a t we know t h a t
to him already the D a n was a kind of soul. I t has of course long been noticed
t h a t etymologically dan- could very well belong t o the root of Persian
didan to see, so t h a t one might regard a man's D a n as a kind of Platonic
idea of him in the other world, his visible soul-prototype, the word idea also
belonging to a base meaning to see, vid-.

286 I. GERSHEVITCH
I would go further, and suggest t h a t the confusion in modern discussions

of t h e term dand-, which from oldest times down to the present also served as
general word for religion, arises from there having existed in antiquity two
homonyms dand- : one belonging to the base did- to see and denoting the
visible outline of a man's soul ; the other belonging to the base of Sanskrit
dyti to bind, Greek dem, with suffix -ind-, and therefore quite literally mean-
ing re-ligio.
And to cap it all, I would add t h a t it is not for nothing t h a t up to now
no agreement has been reached as to what was the real original meaning of the
pan-Iranian word for soul, urvan-\ruvdn. We have been looking in the wrong
direction, seeking explanations in the light of our own experience with anima,
spiritus, yvxp. But we are dealing with Iranians, who distinguished between the
breathing soul as principle of life, and the soul whose S H A P E they would see
in the other world. The most natural analysis of ruvn is t h a t a suffix -an was
added to a base -. And if you want a base - with a meaning t h a t fits what
we have seen to be characteristic of the Iranian V I E W of the soul, well, San-
skrit has rilpa shape, form and beautiful form, base - with suffix -pa.
Whether or not this etymology commends itself, there remains the fact
t h a t already to Zoroaster urvan- was a variant of dand-, so t h a t it is safe to
define the Manichean huruvn Auditors as laymen entitled to expect their
souls, their ruvns, to look attractive. Our gain from this definition is that at
last we understand why to Iranians Mani presented the alms-giving institu-
tion under the name of Ruvnegn. The institution must have been a harsh
one for new converts to swallow. Laymen asked to toil and sweat for the Elect
might well say but what do we get out of it, are not the beneficiaries only the
theoretical Living Griw, and the lazy Elect, fed clothed and sheltered at our
expense? Mani's answer would set them completely at r e s t : Oh no, it is
not only the far from theoretical Living Griw, pure divine quintessence tortured
in the grip of dread Darkness, nor the far from lazy Elect distilling themselves
ascetically to Filters, who benefit from your forced labour within the Ruvne-
gn institution. Not for nothing its name is Ruvnegn : each of you safeguards
his own ruvn by submitting to the institution's prescriptions and discipline.
J u s t think, what CAN be more wonderful than union forever with a soul shaped
b y your own toil to breath-taking beauty, what CAN there be drearier than
eternal union with her disfigured, allowed by your own neglect to grow ugly ?
B u t our now knowing, or at least believing that we know, why Mani called
the alms-giving institution Ruvnegn, exposes us to the danger of incurring
a fallacy in respect of the reason why he called the Living Soul Beauty.
Would it not be tempting to infer from this name t h a t he had pressed to its
last consequence the Dan-myth : a good man finds his soul in the other
world beautiful ; but no human being is so good t h a t his soul's beauty will
not have to put up with some imperfection ; only God has goodness perfect ;

B E A U T Y AS T H E LIVING SOUL I N I R A N I A N MANICHEISM 287
ergo the soul that is Beauty, par excellence, can only be God's own. If we were
to read such a reasoning into the wonderful name of the Anima Viva we have
discovered, we should soon find ourselves drowning in vertiginous Living-
Soul/God-Soul/World-Soul speculations on behalf of Mani.
The reason for not drawing such inferences lies partly in its being grw,
and not dn (from dan-) or ruvan, which designates the Living Soul as soul ;
but chiefly the reason lies in the fact t h a t the soul which is called Beauty, the
Living Soul, consists of Light. U p to now in this lecture we have been looking
exclusively in the Living Soul's being a soul, for an explanation of her name
Beauty. Although we did find an explanation, it would be rash to close the
inquiry without asking also if perchance within Iranian tradition the Anima
Viva might not deserve the name Beauty on account of her being LIGHT.
We shall do well to remind ourselves why she is Light. The difference
between Zoroaster's and Mani's dualism, to p u t it in a nutshell, is that Zoro-
aster opposes Right to Wrong, while Mani opposes Light to Darkness. Zoro-
aster's dualism is ethical, Mani's is optical. I t is that by dint of his make-up,
his genes, which even a prophet cannot escape. Mani's system is a cerebral
elucubration, but its elucubrator was an artist, a visual artist. He did write
books, but he also drew a picture book. In Persia, long after his writings were
forgotten, wiped out by Islam, Mani remained a household word for an artist,
the prophet survived as painter. Combining these two vocations of his we might
call him the prophet of chiaroscuro, inexorably fated to translate Zoroaster's
Right into Light.
Having so translated it, what prompted him to call the lost particles of
Light, the ones swallowed up by Darkness, the Living Soul ? There is no prob-
lem about Living, Viva. The particles were alive, Darkness had swallowed
them but not killed them. But why did Mani not call them the Living Light,
which surely would have been, indeed is, their truest definition?
If it is only to Iranians that Mani gave Beauty as name of the Living
Soul, then he must have known that this name would spark off in their minds
the recognition of a notion which the tradition of non-Iranians lacked. B u t
Mani was himself an Iranian. The notion in question was therefore one with
which he himself had grown up. Might then not this be the notion which impel-
led him, long before he turned to Iranians, while he was still addressing only
Babylonians, to call the lost particles of Light not the Living Light but the
Living Soul?
The fact is t h a t the requisite notion makes its appearance centuries be-
fore Mani, in the A vesta. I dealt with it at some length years ago, in my Mithra
book, totally unaware at the time of its relevance to Manicheism as disclosed
by the Bactrian fragment. The Avestan notion is that light, more specifically
daylight and the sun inasmuch as he is daylight, is the most beautiful crea-
ture and has of (all) shapes the most beautiful shape.

288 I. GERSHEVITCH: BEAUTY AS T H E LIVING SOUL IN I R A N I A N MANICHEISM
Light is B e a u t y without f u r t h e r qualification, absolute Beauty. W h a t

notion can there be, more compellingly appealing to a prophet born painter.
B u t t h e prophet in question, witness his use of the t e r m s huruvn a n d Ruv-
, was heir also to A N O T H E R notion : the more dutiful the man, t h e more
b e a u t i f u l his soul. I n the HaSxt N a s k in fact the Dan meeting her deceased
paragon of virtue is said to be in shape of the same b e a u t y as the most beauti-
f u l creature. The most beautiful creature in the Avesta, we have just learned,
was Light. So an Avestan Dan, even though paired with a being no more
t h a n human, was not below vying with t h e unsurpassably ravishing b e a u t y of
L i g h t itself !
Would not Mani have been familiar with the Avestan statements in
point ? And if we are reminded of t h e m on seeing him call Beauty t h e lost
particles of Light we were wondering w h y he calls Soul, is this not because it
was he who had in mind these very same statements when he decided t o call
t h e Light soul?
We have come round full circle. W e started in astonishment a t learning
f r o m the Bactrian fragment t h a t Mani called the Living Soul, the lost particles
of Light, Beauty. We looked for a reason, and found one. B u t the reason
boomeranged on us, and we end b y discovering t h a t it was B E A U T Y which
induced Mani to call t h e lost particles of Light Living Soul ! W h a t we took for
effect has t u r n e d o u t to be cause : in B e a u t y Mani the painter saw the common
denominator of Light and Soul, through B e a u t y he merged them.
Such aesthetics of course do n o t m a r , how could they, the ethics which
as a prophet Mani felt bound to uphold. They in fact greatly enhance their
appeal. I t is Mani at his most beguiling whom Lord H e n r y unwittingly mimics
in his repartee : I can't bear the idea, Dorian Gray tells him, of m y soul
being hideous ; to which the perceptive Lord replies A very charming
artistic basis for ethics, Dorian.
Cambridge.

289 W. SUNDERMANN
PROBLEME DER INTERPRETATION

MANICHISCH-SOGHDISCHER BRIEFE*
D a sich u n t e r d e n s o g h d i s c h e n T u r f a n t e x t e n in m a n i c h i s c h e r S c h r i f t
e i n e R e i h e h i s t o r i s c h b e m e r k e n s w e r t e r B r i e f e b e f i n d e n , h a t als e r s t e r W . B .
H e n n i n g vor m e h r als vierzig J a h r e n mitgeteilt.1 H e n n i n g m a c h t e auch eine
k n a p p e I n h a l t s a n g a b e d e r S c h r e i b e n , so wie e r sie d a m a l s v e r s t a n d , u n d h o b
i h r e w i c h t i g s t e n P u n k t e h e r v o r . E i n e V e r f f e n t l i c h u n g d e r B r i e f e sollte d a s
5. K a p i t e l seines B u c h e s Sogdica, L o n d o n 1940, b i l d e n . H e n n i n g s B e a r b e i -
t u n g w a r o f f e n b a r so w e i t g e d i e h e n , d a die B r i e f e i m I n h a l t s v e r z e i c h n i s a n g e -
k n d i g t w e r d e n , a b e r n i c h t w e i t g e n u g , u m in d e n T e x t e i n e s u n t e r s c h w i e r i g e n
Zeitumstnden entstandenen Werkes noch rechtzeitig E i n g a n g zu finden.2
S t a t t d e s s e n h a t t e I . G e r s h e v i t c h die M g l i c h k e i t , v i e l e w i c h t i g e W r t e r u n d
F o r m e n d e r B r i e f e in s e i n e m A r t i k e l Sogdian C o m p o u n d s , T P S 1945, S. 137
* In seiner gedruckten Form ist dieser Vortrag des Ergebnis einer berichtigenden
und verbessernden berarbeitung, in deren Verlauf ich Gelegenheit fand, die Original-
texte in der Orientabteilung der Staatsbibliothek Preuischer Kulturbesitz zu studie-
ren. Wertvolle Richtigstellungen, Hinweise und Anregungen verdanke ich Herrn Prof.
D. N. MACKENZIE (im folgenden MK) und Herrn Dr. N. SIMS-WILLIAMS (im folgenden
SW), sowie allen Fachkollegen, die meinen Vortrag am 2. 10. 1980 in Budapest hrten
und zu ihm Stellung nahmen. Eine wertvolle Hilfe waren m i r der R a t und Beistand
meiner Kollegen Dr. . THILO und Dr. P. ZIEME, sowie meiner F r a u . Ihnen allen, und
mit ihnen Herrn Dr. D. GEORGE, dem Direktor der Orientabteilung der Staatsbibliothek,
sei an dieser Stelle gedankt. [Vgl. auch Anhang I I . ]
1
W. B. HENNING : Neue Materialion zur Geschichte dos Manichismus. ZDMG
90 (1936) S. 16 ff. Ein Verzeichnis der mutmalich zu dieser Textgruppe gehrenden
Stcke bei M. BOYCE : A Catalogue of the Iranian Manuscripts in Manichean Script in
the German Turfan Collection. Berlin 1960. S. 148. Die Zugehrigkeit von M 891 a er-
scheint mir fraglich. hnlichen Schriftcharakter habon u. a. die Fragmente M 5885 9
und T I I I D 271 (nicht im Katalog). In ihrem Fall ist mit Zugehrigkeit zu Briefen zu
rechnen. Auch das zutreffend als school-exercise charakterisierte Fragment M 7391
(Catalogue, S. 131) enthlt Briefformoln. Das vermutliche Brieffragment M 858 a wird
an den Zeilenendon ergnzt durch das von A. N. RAGOZA publizierte Leningrader Stck
K r TV
S I i n v . No. 3599 (Sogdijskie fragmonty cential'no-a/.iatskogo sobranija instituta
852
vostokovedonija. Moskva 1980. S. 52 u. 152).
2
Vgl. I. GERSHEVITCH i n : W. B. Hnning Memorial Volume. London 1970.
S. X I I .
19 Acta Antiqua Aeademiae Scientiarum Hungarirae 28, 1980

290 Vi. SUNDEBMANN
ff., u n d in seiner Grammar of Manichean Sogdian, O x f o r d 1954, zu klren.

W a s aber war m i t H e n n i n g s Manuskript geschehen? Sein Schicksal ist mir
u n b e k a n n t geblieben. I c h m u gestehen, d a ich auch nicht systematisch ver-
s u c h t habe, es aufzuklren, denn es erschien mir sowieso wenig sinnvoll, mich
a n eine Arbeit zu m a c h e n , die H e n n i n g ohne Zweifel bereits f r h e r viel besser
geleistet hatte.
Wenn ich mich n u n doch entschlossen habe, an einem berufenen O r t
m e i n e Vorstellungen von den beiden grten u n d wichtigsten F r a g m e n t e n ,
d e r e n H a u p t s t c k e M 112 u n d M 119 sind, 3 mitzuteilen, so 1. weil ich Gelegen-
h e i t finde, das Urteil der F a c h l e u t e zu hren, 2. weil das Schicksal des wissen-
schaftlichen Nachlasses P r o f . Hennings nach wie vor u n k l a r ist u n d die
E x i s t e n z eines von ihm ausgearbeiteten Manuskripts zweifelhaft zu sein scheint,
u n d 3. weil jngst in einer Reihe erschienener und noch unpublizierter Arbeiten
die Geschichte des Manichismus in Zentralasien und China m i t wertvollen
n e u e n R e s u l t a t e n u n t e r s u c h t wurde, 4 f r die auch die manichisch-soghdischen
Briefe, genauer gesagt : H e n n i n g s Angaben von 1936 ber diese Schriften, eine
gewisse Rolle spielen. Ich meine, d a m a n heute einiges m e h r d a z u sagen k a n n ,
u n d d a ich Zugang zu d e n iranischen T u r f a n t e x t e n habe, f h l e ich mich ver-
p f l i c h t e t , es auch zu t u n . Ich tue es nicht, ohne d a n k b a r anzuerkennen, d a
es v o r allem die R e s u l t a t e des wissenschaftlichen Lebenswerkes W . B. Hennings
s i n d , die diesen F o r t s c h r i t t ermglichten. Henning u n d Gershevitch folgend
bezeichne ich M 112 usw. als den ersten, M 119 usw. als d e n zweiten Brief.
Mit Henning b e t r a c h t e ich die Schriftstcke als Briefe, wenngleich ihre
F o r m dies nicht von vornherein erwarten lt. Sie sind jedenfall keine leicht
e r k e n n b a r e n Faltbriefe wie e t w a die soghdischen Alten Briefe 5 oder einige der
M u g h - D o k u m e n t e , sondern einseitig beschriebene Rollen. Sie sind auch nicht in
3
Vgl. dazu M. BOYCE : Catalogue [s. A n m . 1]. S. 10.
4
P . ZIEME : Ein uigurischer Text ber die Wirtschaft manichischer Klster im
Uigurischen Reich. I n : Researches in Altaic Languages, Budapest 1975. S. 331 ff., Kong
Shi-min ( : > 0 / If- SC f f b l (Notes on a n Uigur Government
C h a r t e r issued to a Manichaean Monastery), in : Kaogu Xuebao 4 [1978], S. 497 ff. (Eine
neue Gesamtbearbeitung dieses wichtigen Textes wird gegenwrtig von mehreren Gelehr-
t e n vorbereitet.), S. N. C. LIEU : A Lapsed Chinese Manichaean's Correspondence with
a Confucian Official in t h e L a t e Sung D y n a s t y (1265) : a Study of t h e Ch'ung-Shou-Kung
Chi b y H u a n g Chen. I n : The J o h n R y l a n d s University Library of Manchester 1977.
S. 397 ff., ders. : The Religion of Light. An Introduction to the H i s t o r y of Manichaeism
in China. Hong Kong 1979. Ders. : Polemics Against Manichaoism a s a Subversive Cult
in Sung China (A. D. c. 960 c. 1200). I n : The John R y l a n d s University Library of
Manchester 1979. S. 132 ff., ders. : Nestorians a n d Manichaeans on t h e South China Coast.
I n : Vigiliae Christianae 34 (1980), S. 71 ff., bes. S. 75 mit A n m . 23 a, d e r s . : Precept
a n d Practice in Manichaean Monasticism. I n : The Journal of Theological Studies NS
32 (1981) S. 153 ff.
5
H . REICHELT : : Die soghdischen Handschriftenreste des Britischen Museums. I I .
Heidelberg 1931. S. 5.

PROBLEME DER INTERPRETATION MAN.-SOGHD. B R I E F E 291
der gelufigen soghdischen Schrift zu Papier gebracht, sondern in einer verein-

fachten Variante der sonst der sakralen L i t e r a t u r v o r b e h a l t e n e n manichischen
Schrift. W a s f r Hennings Auffassung spricht, ist ihr I n h a l t , und die formalen
Eigentmlichkeiten d r f t e n sich so erklren, d a gewichtige und lange Schrei-
ben an hochgestellte Persnlichkeiten vorliegen. D a k o n n t e m a n keinen all-
tglichen Faltbrief schicken, sondern whlte die altehrwrdige Schriftrolle,
und zwar nicht die Rckseite von chinesischen T e x t e n sondern jungfruliches
Papier. D a h e r auch die Verwendung der manichischen s t a t t der soghdischen
Kursivschrift." Die verfeinerte F o r m der Schreiben m a c h t es auch unwahrschein-
lich, d a sie lediglich E n t w r f e oder zum Verbleib beim Absender b e s t i m m t e
Abschriften darstellen. Man k a n n daher von der A n n a h m e ausgehen, d a ihr
wahrscheinlicher F u n d o r t Q060 7 zugleich der W o h n o r t oder Amtssitz des
E m p f n g e r s w a r und nicht umgekehrt der Ort des Absenders. I n h a l t u n d
Schrift sprechen auch f r Hennings Charakterisierung der Briefe als D o k u m e n t e
manichischer H e r k u n f t . 8
I m folgenden gebe ich eine knappe Beschreibung dessen, was ich als
I n h a l t der beiden Briefe v e r m u t e und als Probleme erkenne. Meine Lesung u n d
bersetzung des Textes ist diesem Aufsatz als A n h a n g beigefgt.
8
Vergleichbar damit ist die an uigurischen Brioffragmenten gemachte Beobach-
tung, da amtliche Schreiben an hochgestellte Persnlichkeiten sorgfltiger ausgefertigt
werden als Privatbriefe (S. TEZCAN U. P . ZIEME : Uigurischo Brieffragmente. In : Studia
Turcica. Budapest 1971. S. 451 ff.). W a s die Art des Briefverschlusses betrifft, so nohmen
die Bearbeiter f r lngere Schreiben die Rollenform an (S. 452).
7
Da sowohl M 112 wie M 119 alte, von F . W . K . MLLER gegebene M-Signaturen
haben, mssen sie von der ersten Turfanexpedition (Nov. 1902 bis Mrz 1903) in Qo6o
gefunden oder g e k a u f t worden sein (BOYCE : Catalogue [s. A n m . 1], S. X X I ) u n d m i t
grter Wahrscheinlichkeit der Ruine A entstammen (BOYCE : ebd. Anm. 3). Diese
Annahme wird f r M 112 aber durch den Umstand in F r a g e gestellt, da das zugehrige
F r a g m e n t M 162 a (BOYCE: ebd., S. 13) die F u n d s i g n a t u r T I D 51 trgt, was die
Mglichkeit anderer H e r k u n f t innerhalb Qoos offen lt (BOYCE: ebd., S. X I I I ) . W a s
M 119 betrifft, so konnte ich auf dem in der Staatsbibliothek in Westberlin befindlichen
Original sehen, d a seine Rckseite die alte Signatur T I D I I a (so!), d. h. T I D, T I a,
t r g t . Das Stck m u also whrend der ersten Turfanexpedition in Qoo in der R u i n e a
geborgen worden sein. Damit s t i m m t berein, da das zugehrige Fragment M 1226
die Signatur T i n (BOYCE: ebd., S. 69), das ebenfalls zugehrige Stck M 1867 die
Signatur T I D (BOYCE : ebd., S. 82) trgt. Die d a n n naheliegende Annahme, d a
das Gebude a auch der Amts- u n d Wohnsitz jenes hohen manichischen Geistlichen
war, an don die Briefe gerichtet wurden, ist verlockend aber nicht sicher beweisbar.
Sie ist unwahrscheinlich angesichts der Erklrung der R u i n e als Rest eines buddhisti-
schen Heiligtums (A. v. LECOQ : Die buddhistische S p t a n t i k e in Mittelasien. Die rnani-
chaeischen Miniaturen, Berlin 1923, S. 28 ff.).
8
Neue Materialien [s. Anm. 1], S. 16 ff. Die einmal erwogene Alternativmglich-
keit, da nestorianische Briefe vorliegen (M. BOYCE : H a n d b u c h der Orientalistik, 1.
Abt., 4. Bd. Iranistik, 2. Abschn. Literatur, Lieferung 1, Leiden Kln 1968, S. 73),
trifft mit Sicherheit nicht zu.

292 W. SUNDERMANN
Erster Brief
(Ein aus den n u r teilweise zusammensetzbaren Stcken M 112, 146 a,

162 a, 336 a, 336 b u n d 336 bestehender Briefrest in Rollenform, die Rck-
seite t r g t einen uigurischen Text in s p t e m Duktus. 9 ber die Lokalisierung
der Teile lt sich n u r sagen, da M 146 a u n d M 336 im mir vorliegenden
T e x t m i t M 112 zusammengefgt sind u n d d a M 336 a in grerem Abstand
M 112 usw. voranging. W o M 336 b u n d M 162 a einzuordnen sind, ist unbe-
k a n n t . I n dieser Arbeit gehren die Zeilen /1 13/ zu M 336 a, /14 32/ zu
M 112 + 146 a + 336 , / 3 3 - 3 7 / zu M 336 b u n d / 3 8 - 4 0 / zu M 162 a.
Die geschichtlich bedeutsamste Aussage dieses Schreibens, von Henning
b e r e i t s bekanntgemacht, 1 0 ist die E r w h n u n g der beiden P a r t e i e n der syrisch-
manichischen Kirche, der m y h r y ' n d (Zeile 5) und mkl'sykt (Zeile 10), im ara-
bischen Text des Fihrist mihriya und miqlsiya.11 Es scheint, d a der soghdi-
sclie Verfasser eines ihrer W o r t e in mp. Sprache zitiert, dessen unvollstn-
diger Erhaltungszustand sehr zu bedauern ist. Es lautet in Zeile 3 : 'yg h m ' g
d y n ' m ' h hwm der ganzen Religion/Kirche sind wir. D a n a c h d r f t e n die
Syrer, vielleicht insbesondere die in Zeile 5 genannten Mihrynd, als Sachwalter
oder Sprecher der manichischen Gesamtkirche aufgetreten sein, was insofern
verstndlich ist, als die Mihriya jene P a r t e i der manichischen Kirche Meso-
p o t a m i e n s darstellte, von der die Miqlsiya sich abspaltete u n d deren Ober-
h a u p t in Babylon seinen Sitz hatte. D a in Zeile 10 dann noch die Miklsikt
in u n k l a r e m Zusammenhang genannt sind, so scheint es, da Anhnger beider
P a r t e i e n sich damals in Zentralasien befanden. Der Briefschreiber d r f t e aber
wenig Unterschiede zwischen ihnen bemerkt u n d beide Gruppen mit derselben
Geringschtzung b e t r a c h t e t haben. Seine Bemerkung in den Zeilen 11 12
denn v o n einem Gesetz u n d einem U r s p r u n g . . . sind sie l t sich jedenfalls
so d e u t e n . Hennings allgemein akzeptierte Annahme, der Briefschreiber habe
allein gegen die Mihriya polemisiert, 12 ist nicht widerlegbar. Der I n h a l t des
zweiten Briefes spricht aber nicht fr diese Auffassung (vgl. S. 297).
Von den zuvor g e n a n n t e n Syrern ist d a n n gewi auch im folgenden er-
h a l t e n e n Hauptteil des Briefes die Rede. W a s zunchst gegen diese gedank-
liche E i n h e i t des Gesamtschreibens zu sprechen scheint, ist die Tatsache, da
d a s O b e r h a u p t der kritisierten Gemeinschaft den persischen N a m e n Mihr-
9
M. BOYCE : Catalogue [s. Anni. 1], S. 10.
10
Neue Materialien [s. A n m . 1], S. 16 ff.
11
G. FLGEL: Mani, seine Lehre und seine Schriften. Leipzig 1862. S. 68. Z. 1.,
der P l . al-maqlia S. 68. Zeile 13 u. 15. vgl. K i t b al-Fihrist I. ed. G. FLGEL, Leipzig
1871. S. 335, Z. 19, 27, 29. The Fihrist of al-Nadim, ed. B. DODGE, B d . I I . New York,
L o n d o n 1970. S. 793. ber die beiden Religionsparteien vgl. FLGEL: Mani, S. 66 ff.,
97 ff., K i t a b al-Fihrist I. S. 334, Z. 3 ff., DODGE : S. 792 ff.
12
Neue Materialien (s. A n m . 1), S. 17.
Acta Antiqua Academiae Scirntiarum Hungaricae 28, 1980

P R O B L E M E D E R I N T E R P R E T A T I O N MAN.-SOGHD. B R I E F E 293
pdr f h r t und da die Electae dieser Gemeinschaft ebenso nvarnc (Sg.)

genannt werden, wie die Erwhlten-Schwestern der zentralasiatischen Den-
warya-Gemeinde. Aber beide Einwnde sind nicht zwingend. E i n syrischer
Manicher konnte ebensowohl Mihr-pdr heien wie der Begrnder der
Mihriya-Partei den persischen N a m e n M ihr f h r t e und der eigentliche Be-
grnder der Miqlsiya Zd-Hormezd hie. U n d Begriffe wie dnwar und
dnwarft haben die zentralasiatischen Manicher offenbar nicht zur Abgren-
zung von ihren westlichen Glaubensbrdern bentzt und daher ihrer eigenen
Gemeinschaft vorbehalten. Wenn sie diese bezeichnen wollten, so sprachen sie
von der Kirche von Churasan. 1 3
U n t e r der Voraussetzung also, d a die Zeilen 14 ff. von den zuvor ge-
nannten Syrern handeln, nehme ich an, da der Briefschreiber nunmehr die
hohen Ansprche der Fremden kritisch an ihrem Glaubenswandel und diesen
an den Forderungen der fnf f r die Elekten geltenden Gebote mit. Ihre
Taten werden gewogen u n d zu leicht befunden.
Als erstes k o m m t das Zeile 13 genannte Gebot der Wahrhaftigkeit
(rty'k cxS'pSS), also des Bekenntnisses der manichischen Glaubenslehren,
zur Sprache, doch das dazu Gesagte ist leider ganz verlorengegangen, was u m
so bedauerlicher ist, als es f r eine genauere Kenntnis der Auffassungen von
Mihriya u n d Miqlsiya ohne Zweifel sehr bedeutsam gewesen wre. Mit Zeile
14 ff. setzt der Text des Briefes wieder mitten im Zweiten Gebot, pw'zrmy'h
das Nichtverletzen, ein, in dem T a t e n einer oder mehrerer Erwhlten-
Schwestern der angeklagten Gemeinde beschrieben werden, die der in den irdi-
schen Elementen und Organismen verstreuten viva anima Leid u n d Schaden
bereiten und die eigentlich nur von Hrern htten verrichtet werden drfen.
F r alle Beschuldigungen werden Augenzeugen genannt, so d a die ungelieb-
ten fremden Glaubensbrder lngere Zeit kritisch beobachtet worden sein
d r f t e n . D a hat jemand gesehen, da j e m a n d pflanzliche P r o d u k t e abhieb u n d
daraus etwas nhte (Zeilen 1516). E i n anderes Mal h a t ihre Erwhlten-
Schwester mit einem Grabegert den Boden bearbeitet (Zeile 18). Auch wurde
von den Beklagten Mehl gemahlen und Holz geschnitten (Zeilen 1920).
Groen R a u m (Zeilen 20 22) n i m m t die skandalse Beobachtung ein, d a
ihre Erwhlten-Schwester einmal Blut nahm (xwrnyy "syyskwn, Zeile 21),
was wohl dasselbe bedeutet wie . xn gereftan, wtl. Blut nehmen, nmlich
" D y n 'y xwr's'n u. . in M 4 (f) / R / 1 0 - 1 1 / , M 82 /V/1819/, M 83 /I/V/17/,

M 190/4/, M 235/J/R/9/, M 494/3/(7), M 1368/4/, M 1863/14/. Dagegen finde ich p a r t h .
d y n ' w r y f t n u r ein einziges Mal belegt in M 6020 /3/, publ. bei HENNING : A Grain of
Mustard. AION-L 1966, S. 30 f. HENNING erklrte seine dort gegebene bersetzung
the m a n who is in the Dnvariyya berzeugend als Hrer der (Eastern) Manichaean
church. Diese Selbstbezeichnung ist aber gewi nicht als Parteienname zu verstehen,
der die eigene Gemeinschaft vom R e s t der manichischen Kirche abgrenzen sollte. W e n n
borhaupt, so mochte sie der Abgrenzung gegen Christen, Buddhisten u n d andere dienen.

294 W. SUNDERMANN
z u H e i l z w e c k e n B l u t s c h r p f e n o d e r z u r A d e r l a s s e n . 1 4 D i e s h a b e sie g e t a n u n d
d a n n verwerflicherweise en " p y ' p r y w k m ' s n ' y y s k w n , mit W a s s e r d a s k m '
gewaschen. L e i d e r i s t k m ' ein H a p a x l e g o m e n o n u n b e k a n n t e r B e d e u t u n g .
M a n erwartet, d a es entweder ein v o n B l u t beschmiertes chirurgisches Messer
o d e r das Schrpf i n s t r u m e n t oder die bei der B l u t e n t n a h m e e n t s t a n d e n e W u n d e ,
vielleicht auch ein Reinigungstuch15 ist. D a s A b w a s c h e n des Blutes m u t e d a n n
d a s Wasser verunreinigen, und eben dies wurde der Heilpraktikerin vorgewor-
fen. Die Anklage s t i m m t berein m i t d e m , was wir aus a n d e r e n Quellen b e r
die H o c h a c h t u n g u n d R e i n h a l t u n g d e s W a s s e r s bei d e n Manichern wissen.16
14
J . A. RUBINIK : Persidsko-russkij slovar'. I. Moskva 1970. S. 689 xn gere/tan
m e d . puskat' krov'. D a s kann sowohl zur Ader lassen wie schrpfen bedeuten.
So erklrt M. MO'N : A Persian Dictionary I . Teheran 1963. S. 1464 - als rag-zan,
fa??d Aderlasser, haffljm Aderlasser, Schrpfer. I m vorliegenden Text d r f t e d a s
Zeile 25 ff. beschriebene dreifache Blut-Nehmen aber eher ein Schrpfen als ein Aderla
gewesen sein. Das Schrpfen (hetfmat) wird in Mo'in Wrterbuch, S. 1341, so beschrie-
b e n : Dem Krper wird so Blut entzogen, d a eine besondere Klinge ein Stck H a u t
zwischen den Schulterknochen aufschneidet u n d mit einem besonderen Instrument (dem
' S c h r p f h o r n ' ) die aufgeschnittenen H a u t t e i l e abgesaugt imd die gengende Menge Blut
d e m Krper entzogen wird. Zu Schrpfhorn (Sh-e heijmat) sagt Mo'in : Das E n d -
s t c k eines Rindshornes, dessen Inneres ausgehhlt worden ist. Beide Formen des Blut-
entzuges haben in d e n Heilkunden des Orients eine alte Tradition. F r Indien vgl.
J . JOLLY in : Grundri der Indo-Arischen Philologie und A l t e r t u m s k u n d e . 3. Bd., H e f t
10, Medizin. Straburg 1901. S. 35, f r die islamische Welt : M. ULLMANN : Die Medizin
im Islam. I n : H a n d b u c h der Orientalistik. 1. Abt., E r g b d . 6, 1. Abschn. Leiden, Kln
1970. S. 16. Auch in Zentralasien m u der Blutentzug praktiziert worden sein, wie aus
trkischen Texten hervorgeht (vgl. R . MLLER : Ein Beitrag zur rztlichen Graphik
a u s Zentralasien (Turfan). I n : Archiv f r Geschichte der Medizin 15 (1923) S. 21 ff.,
G. R . RACHMATI : Z u r Heilkunde der Uiguren. I I . S P A W Phil.-hist. Kl. 1932. X X I I .
Berlin 1932. S. 419).
15
Darf in diesem Fall np. kam a sleeve ; old cloth sewed together, whereon bread
is laid to be baked in t h e oven (F. STEINGASS: A Comprehensive P e r s i a n - E n g l i s h
Dictionary, London 1963. S. 1046) verglichen werden 1
18
Die groe griechische Abschwrungsformel sagt den Manichern das Verbot
n a c h , ihien Schmutz m i t Wasser abzuwaschen, damit nicht, wie sie sagen, das Wasser
befleckt wird (Die Gnosis I I I , Der Manichismus, unter Mitwirkung von J . P. ASMUSSEN
ed. v. A. BHLIG. Zrich, Mnchen 1980. S. 299, Text bei A. ADAM : Texte zum Mani-
chismus. Berlin 1954. S. 100, 114 f.). D a der Besuch der ffentlichen Bder den
Manichern grundstzlich als verwerflich galt, folgt auoh a u s Augustine Beschuldigung :
Multi in vino et carnibus, multi lavantes in balneis inventi sunt (De . Man., K a p .
19, 68, in : Patrologiae cursus eompletus, sor. latina prior. B d . 32, Paris 1861. Sp. 1374).
U m g e k e h r t verleiht die errterte Briefstelle einer Mitteilung Li Chao's ( ^ - ) in seinem
Tang guo shi bu (ffj- ^ -j(|[]) ( l . H t . des 9. J h . ) Gewicht, wonach die Manicher das Wasser
verehrten, vgl. ED. CHAVANNES und P . PELLIOT : Un T r a i t Manichen retrouv en
Chine (deuxime partie), in : JA 1913, S. 268. Die Richtigkeit von ils respectent
h a t t e n die Autoren w e n n nicht zurckgewiesen, so doch in Zweifel gezogen und gemeint,
d a der Text ursprnglich sie trinken gehabt habe (ebd., S. 274 A n m . 3). Vgl. auch
S. 349 Anm. 2.

PROBLEME D E R I N T E R P R E T A T I O N MAN.-SOGHD. B R I E F E 295
W e n n die Fremden ihre Electa mit der Entschuldigung verteidigen, sie habe
Brunnenwasser genommen u n d dies sei tot, d. h. wohl, es enthalte nichts
von der Lebendigen Seele, so wurde dies in Mittelasien zweifellos als Beweis
f r die Laxheit der Moral der Syrer aufgefat. Uns erinnert es aber auch an
die wichtige Scheidung zwischen stehendem und flieendem Wasser beim T a u f -
ritual der Mander, 17 so da sich in dieser H a l t u n g der beschuldigten Erwhl-
ten-Schwester vielleicht eine altererbte, unter den mesopotamischen Tufern
verbreitete Auffassung k u n d t u t . Eine Verletzung des Zweiten Gebotes ist es
auch, wenn Heilpflanzen zerstoen u n d Feuer gelscht( ?) werden (Zeilen 22
23).
Die Zeilen 24 ff. bis zum E n d e des Fragments behandeln Verste der
Syrer gegen das Dritte Gebot, d y n c y h r y f t cxs'pSS, Religionsmiges Verhal-
ten, d. h. keuscher, ja streng enthaltsamer Lebenswandel in sexuellen Dingen.
W a s der Briefschreiber hier vorzubringen hat, sind Klatsch- und Tratschge-
schichten reinsten Wassers, deren Wichtigkeit f r den Manicher strikter
Observanz aber durch chinesische Nachrichten ber die radikale Trennung der
Geschlechter bei den manichischen Klerikern verdeutlicht wird. 18 Unser
Briefschreiber nun wei von Mihr-pdr, dem Oberhaupt (s'r'r) der inkrimi-
nierten Gemeinde (Zeile 25), zu berichten, da er eine Lohnmagd (z'k'nc
kr'y'nc) zu sich kommen lie u n d mit ihr hinter verschlossener Tr alleine
blieb, damit sie sein Fuleiden durch Blutentzug, d. h. wohl Schrpfen, heile.
Die eigenen Electi Yazd-Arymn, Drist-RSn, Mihr-Wahman und Wahman-
Sh stellten darauf das Mdchen zur Rede, das sich aber nicht bufertig zeigte,
sondern sogar erwiderte, sie msse noch ein weiteres Mal zu Mihr-pdr gehen
(Zeilen 2 5 - 2 9 ) .
Eines anderen Verstoes gegen das Dritte Gebot machte Mihr-pdr
sich schuldig, als er einen heftigen Streit zwischen einer Dienerin und einer
Erwhlten-Schwester seiner Gemeinde dadurch gewaltsam schlichtete, d a
er die Dienerin am Arm packte u n d somit krperlich berhrte (Zeilen 29 31).
Bedenken wir, da noch das Vierte und das F n f t e Gebot behandelt
worden sein drften, so mu der Brief ein sehr umfangreiches Sndenregister
enthalten haben, aus dem hervorging, da eine tiefe Diskrepanz zwischen den
hohen Ansprchen der Syrer u n d ihrem Glaubenswandel bestand.
17
K . RUDOLPH : Iconography of Religions X X I . Mandaeism. Leiden 1978. S. 8.
18
Vgl. besonders, was Lu (11251209) in seinem Lao xue ang bi ji
als Meinung der Anhnger der Religion des Lichts (d. h. der Manichor) in F u - K i e n
zu dieser Frage mitteilt : Qu'hommes et femmes ne soient p a s spars, voil ce qui
est dmoniaque ; q u ' h o m m o s et femmes ne se donnent rien en se t o u c h a n t la main, voil
la [rgle de la] religion de la lumire (CHAVANNESPELLIOT : Trait I I . [s. A n m . 16],
S. 342 mit Anm. 2, Text S. 343). Vgl. auch ebd., S. 353 f., 362, LIEU : Polemics against
Manichaeism [s. A n m . 4], S. 153.
Acta Antiqua Academiae Scieniiarttm Hungaricae 28, 1980

296 W. SUNDERMANN
Zweiter Brief
( E i n a u s d e n S t c k e n M 119, 1 1 9 a , 119 b , 1 2 2 5 , 1867 a u n d 1867 b

b e s t e h e n d e s , e i n s e i t i g b e s c h r i e b e n e s R o l l e n f r a g m e n t , 1 9 d e s s e n Z e i l e n sich w i e
f o l g t z u s a m m e n s e t z e n : / 1 4 / : M 1 2 2 5 , / 4 6 / : M 119 a , / 5 - 1 9 / : M 119,
/ 1 8 1 9 / : M 1867 b , / 1 9 21/ : M 1 8 6 7 a , / 1 9 - 2 0 / : M 119 b . D e n s e l b e n D u k t u s
u n d Z e i l e n a b s t a n d h a t ein kleines, drei Zeilenreste u m f a s s e n d e s F r a g m e n t
T I I I D 271 K , d a s n i c h t i m K a t a l o g v o n M . B o y c e e r s c h e i n t , h i e r M 119
g e n a n n t . Seine z w e i t e Zeile l a u t e t yy&ty[y] k r j w ( r z . . .). I m G e g e n s a t z z u
a n d e r e n S t c k e n d e s z w e i t e n B r i e f e s t r g t M 119 a u f s e i n e r R c k s e i t e u i g u r i -
s c h e n T e x t , doch s c h l i e t dies Z u g e h r i g k e i t n i c h t aus. N u r ein Teil des Briefes
k n n t e j a zur W i e d e r b e s c h r i f t u n g v e r w e n d e t worden sein.
Als Sprecher e i n e r v o n Zwistigkeiten zerrissenen G e m e i n d e w e n d e n sich
d e r o d e r die Schreiber d e s zweiten B r i e f e s a n eine hochgestellte, n u r x y p S ' w n d
H e r r s c h e r g e n a n n t e ( Z e i l e n 5, 14), h f l i c h y H e r r a n g e r e d e t e (Zeile 5)
P e r s n l i c h k e i t . D i e S c h w i e r i g k e i t e n s i n d o f f e n b a r so g r o g e w o r d e n , d a sie
d e r B e a c h t u n g d e r a n g e r u f e n e n A u t o r i t t a n e m p f o h l e n w e r d e n , d a m i t sie
S c h l i m m e r e s v e r h i n d e r e (Zeile 7, 17). 2 0
F r w e n d e r B r i e f s c h r e i b e r s i c h v e r w e n d e t , w i r d i n d e n Z e i l e n 10, 11
u n d 12 g e s a g t . E s s i n d m ' x x w y s t r t u n s e r e O b e r e n u n d ' s k w n c y k t x w y s t r t
19
Vgl. M. BOYCE : C a t a l o g u e [s. A n m . 1]. S. 10.
20
F r m ' y " m s ' ' k t y y bzw. " m s ' f r m ' y 'krtyy. HENNING h a t t e zuvor den Sinn
des Satzes dahingehend zusammengefat, d a der Erzbischof die Missetter verwarnen
u n d ihnen anbefohlen solle, den frher in diesem Lande so angesehenen religisen Ober-
h u p t e r n die gebhrende Achtimg entgegenzubringen (Neue Materialien [s. Anm. 1],
S. 16), h a t t e f r m ' y also im blichen Sinne als befehlen verstanden. Eine Bedeutung
Gehorsam, Gefolgschaft f r " m s ' vorausgesetzt, erschien m i r diese Auffassung sinn-
voller als jene von I. GERSHEVITCH, der please, p a y obedience bersetzte (A Grammar
of Manichean Sogdian. Oxford 1961. 1682). Ich war daher H e n n i n g in der ursprnglichen
F a s s u n g dieser Arbeit gefolgt. MK und SW belehren mich jedoch m i t Recht, da " m s '
tatschlich Beobachtung, Aufmerksamkeit bedeutet, MK u. a. mit Hinweis auf A. N.
RAGOZAS Sogdijskie f r a g m e n t y [vgl. Anm. 1], N r . 37, Z. 18, S W u. a. mit Hinweis auf
HENNINGS richtigstellende bersetzung bei J . P . ASMUSSEN : X s t v n i f t . Studies in
Manichaeism. Copenhagen 1966. S. 238, A n m . 6. Die zutreffende Bedeutung von chr.-
soghd. ' m s ' wn- faire a t t e n t i o n , observer d r f t e E . BENVENISTE als erster festgestellt
h a b e n (JA 239 [1961], S. 118), der das W o r t mit p a r t h . pdms- verstehen u n d p d m ' s
Verstand, Urteil verglich. Die von MK erwogene Herleitung von *-\-mn-sko- ist damit
g u t vereinbar. Zur E t y m o l o g i e von " m s ' vgl. weiterhin D. WEBER : Die Stellung der
sog. Inchoativa im Mitteliranisehen. Gttingen 1970. S. 102 f. W e n n aber in den frag-
lichen Briefstellen zur A u f m e r k s a m k e i t aufgefordert wird und nicht zum Gehorsam, so
gewinnt die von GERSHEVITCH vertretene Auffassung von f r m ' y als einer hflichen Bitte
sehr an Wahrscheinlichkeit u n d ist mit MK u n d S W gewi zu bevorzugen. D a sie im
Soghdischen nicht unbelegt ist, betont SW m i t Hinweis auf V J , Z. 179, M 617 /V/8/
in GMS 733 und 927 u n d auf 2/13. V 15 in seiner noch ungedruckten Arbeit ber
diese Handschrift, wo f r ] ( m ) ' ( y ) qty please do (it) syr. s'wr do (it,) entspricht.

die gegenwrtigen Oberen. Offenbar wird ihr Schicksal mit dem einiger fr-
herer Lehrer und anderer Electi verglichen, von denen es heit, da ihnen
Lehrer u n d Bischfe der angeklagten P a r t e i auf den Kopf gestiegen seien
(pr . . . sry . . . stty, Zeile 18, [prw] . . . sryy . . . st'nd, Zeile 19 20). Hier
liegt wohl eine idiomatische Wendung vor. Wenn ein Vergleich mit np. bar sar
madan superare ; incementum afferre, augere ; elevari, 21 wtl. auf den K o p f
kommen, erlaubt ist, so knnte dies bedeuten, da die Vertreter der Gegen-
partei jenen der eigenen an die Seite getreten sind, sie vielleicht auch berwun-
den, bedrngt oder verdrngt haben, ihnen jedenfalls mancherlei Schwierig-
keiten bereitet haben. D a mehrere Lehrer der eigenen Partei erwhnt werden,
darunter ein Newnzdag (Zeile 19), verdient Hervorhebung.
Wer ist die Gegenpartei, die zum Gehorsam gebracht werden soll? Die
Zeilen 15 ff. nennen und bewerten sie. E s heit d a : Dieser schmutzigen u n d
gemeinen Syrer Brauch und K u n s t ist so : in Spaltung u n d Streit sind sie er-
fahren [und] gebt, denn 'geistige Spaltung' herrscht hier. Die wesentlichen
Teile dieser wichtigen Stelle h a t t e Henning schon 1936 bekanntgemacht. 2 2
Sie lassen erkennen, da es Glaubensgenossen fremder Nationalitt, Syrer,
d. h. Manicher aus Mesopotamien, waren, die U n r u h e stiftend bei ihren
zentralasiatischen Glaubensbrdern a u f t r a t e n . D a tatschlich die als Gegener
der eigenen Geistlichen genannten Persnlichkeiten Syrer gewesen sein knnen,
wird durch den syr. Namen Gavryav (Zeile 19, auch bereits Zeile 3) besttigt.
Auch der Bischof ktwn' knnte einen syrischen N a m e n getragen h a b e n
(Zeile 20). Dagegen f h r t e n die Lehrer Saxt (Zeile 20)23 und Mhddd (Zeile
18) persische Namen. Die den Syrern pauschal vorgeworfene Streitsucht be-
zieht sich wohl auf die Auseinandersetzung zwischen Mihriya und Miqlsia,
deren Parteien der erste Brief nennt. E s scheint also, da die wegen ihres laxen
Glaubenslebens angeklagten Gegener des ersten Briefes derselben H e r k u n f t
sind wie die Unruhestifter des zweiten Schreibens. Ich habe auch den E i n d r u c k ,
da beide Briefautoren den Parteiungen der syrischen Glaubensbrder mit
gleicher Distanz gegenberstehen (vgl. S. 292). Sie sind nicht Mihrynd und
Miklsikt, ihre Partei ist eben die Kirche von Churasan, die die westlichen
Manicher als Dnwarya bezeichneten. Den Gegnern schreibt der zweite
Brief die nigmatische Behauptung zu : Die Oberhupter steigen ein wenig
empor, und dann steigen sie wieder h i n a b . . . (Zeilen 1617). Problematisch
ist an diesem Satz fast alles, 'yw p ' z k y ' wurde bisher m. W. nicht belegt. 24
Was aber bedeuten hinauf- und hinabsteigen? Mein Erklrungsversuch
21
J . A. VULLERS : Lexicon Porsico-Latinum. I. Bonn 1855. S. 51.
22
Neue Materialien [s. A n m . 1]. S. 16.
23
Zu diesem Namen vgl. F . JUSTI : Iranisches Namenbuch. Marburg 1895. S. 280
s. v. Saxtyeh.
24
Vgl. A n m . 47 zum A n h a n g I .

298 W. SUNDERMANN
s e t z t voraus, d a beide Verben eine A r t Wechselbewegung bezeichnen, u n d

ferner, d a sie die den manichischen E l e k t e n vorgeschriebenen bestndigen
Reisen im Dienste ihrer Religion, 25 ihr K o m m e n und Gehen von Ort zu Ort,
z u m Gegenstand haben. Chinesische Quellen bezeugen sie f r China ausdrck-
lich u n d sprechen von einem K o m m e n aus der F r e m d e u n d einer R c k k e h r
in d e n Heimatort. 2 6 I n Zentralasien d r f t e es nicht anders gewesen sein. Aber
A u f - u n d Absteigen ist nicht K o m m e n u n d Gehen. Beide Bewegungen k n n t e n
n u r u n t e r der Bedingung identisch sein, d a das Reiseziel hoch ber dem Aus-
g a n g s p u n k t der Reise lag oder eine idiomatische Verwendung der W o r t e nach-
weisbar ist. E s lassen sich verschiedene Situationen denken, in denen dies zu-
t r i f f t , z. B. : wenn die Reise vor der Seidenstrae im Einzugsgebiet des T a r i m
aus ber den Tien-Schan in das Uigurische Steppenreich f h r t e . Die T u r f a n -
oase e t w a liegt 154 m u n t e r dem Meeresspiegel, whrend die hchsten Gebirgs-
gipfel bei U r u m t s c h i die 5000 m-Grenze bersteigen. Eine Reise in dieser Rich-
t u n g ist in der T a t ein Hinaufsteigen. 27 A b e r es knnte auch anders gewesen
sein, wenn hinaufsteigen z. B. das Gehen in eine Metropole bezeichnete. Sollte
eine derartige idiomatische W e n d u n g in d e m manichischen Brief vorauszu-
setzen sein, so wrde das Aufsteigen der Syrer ihre A n k u n f t in der H a u p t s t a d t
des Uigurischen Steppenreiches oder in Qoco oder in der H a u p t s t a d t eines
anderen Kleinstaates a n der Seidenstrae, j a vielleicht sogar Chinas bedeuten.
W i e Th. Thilo mir mitteilt, ist es im Chinesischen blich, das Gehen in die
H a u p t s t a d t als Hinaufgehen zu bezeichnen, das Gehen aufs L a n d als Hinab-
gehen. I n der meinem Vortrag folgenden Diskussion besttigten mir auch J .
H a r m a t t a und I. Gershevitch, da Bewegungen nach und von Budapest, b z w .
in englische U n i v e r s i t t s s t d t e oder aus ihnen fort in analoger Weise bezeich-
n e t werden, u n d es lieen sich gewi noch weitere Beispiele in anderen Spra-
chen finden.
25
Vgl. ED. CHAVANNES u. P . PEILIOT : U n Trait Manichen retrouv en Chine.
J A 1911. S. 572 f.
2e
Vgl. den K o m m e n t a r des Hu Sanxing Vf) zum Zi zhi tong jian ( 'ici ilfi j g )
(1285) : Les grands [religieux] manichens . . . n e font la voyage q u ' u n e foie en plusieurs
annes pour venir de leur p a y s d'origine ou y retourner ; les petits [religieux] alternent
chaque anne (CHAVANNES PEHLIOT : Trait I I [s. A n m . 16], S. 266 mit A n m . 2, T e x t
S. 267). Seine Quelle, Li Chao's Tang guo shi bu (vgl. Anm. 16) wird so zitiert : Les
g r a n d s [religieux] manichens . . . ne se remplacent q u ' a u bout de plusieurs annes,
allant d a n s le royaume de Milieu et en revenant ; les p e t i t s [religieux] alternent chaque
anne. Vgl. auch LIEU : Polemics against Manichaeism [s. Anm. 4], S. 145 f.
2
' Vgl. A. v. LECOQ (in entgegengesetzter R i c h t u n g reisend) : Nachdem wir d a s
Gebirge [bei U r u m t s c h i ] durchquert hatten, stiegen wir in die merkwrdige Niederung
von T u r f a n herab, . . . (Auf Hellas Spuren in Ostturkistan, Leipzig 1926. S. 35). Zu soghd.
sn- besteigen vgl. z. . 87 = T I I I 61 4 // [ q w ] (y)rys'r snnt cn zyrny p y d ' [ r ] auf
den Berg steigen sie des Goldes wegen.

Welche Sachvorhalte wir auch immer hinter den W o r t e n des Briefes ver-
m u t e n mgen, d a herauf- u n d hinabsteigen in diesem Fall dasselbe wie
kommen und gehen bedeutet, ist sehr wahrscheinlich, u n d der fragliche
Satz knnte besagen, da die syrischen Gste zur B e r u h i g u n g der G e m t e r
b e h a u p t e t e n , sie wrden, gewissermaen ihrem a m b u l a n t e n Lebensstil ver-
pflichtet, nicht f r immer bleiben, u n d sie kmen in der besten Absicht, d a s
Wohl der Kirche zu mehren (Zeile 16 18). Es berrascht freilich nicht, d a
der Verfasser des zweiten Briefes ihnen m i t r a u t . Dies W o r t ist ganz ver-
derblich, sagt er (Zeile 17). E r b e f r c h t e t also, da die unvertrglichen Syrer
sich f r die D a u e r in der Gemeinde einnisten und sogar die Gewalt a n sich
reien werden. Dies sagt er offenbar in den Zeilen 9 ff.: Sie steigen empor u n d
vermischen sich mit uns, u n d es wird einige J a h r e d a u e r n , . . . . nach(?) dem
Willen der gegenwrtigen Oberen wird es sein, . . . ungesetzlicher und unzeiti-
ger Befehl wird zu den gegenwrtigen Oberen e m p o r k o m m e n . . . [Jene], die
ihre Spione waren, 2 8 werden fest an die Stelle unserer Oberen t r e t e n u n d unser
. . . U n d sie werden die H a r t - B e d r n g t e n in ihrem Selbst zerbrechen u n d den
Oberen groe N o t bereiten und im B l [ u t . . . ihre H n d e ( ? ) ] waschen. . ..
Als ein schlechtes Beispiel, das Schlimmeres befrchten lt, scheint s o d a n n
in mir unklarem Z u s a m m e n h a n g das Wirken eines gewissen Yazd-Mh ge-
n a n n t worden zu sein (Zeile 14).
Diese Syrer also waren dabei, d u r c h aufsssige Streitsucht die A u t o r i t t
eines zuvor etablierten Gemeindeklerus zu untergraben. W a s sie b e h a u p t e t e n ,
m u im verlorenen Anfang des Briefes gestanden haben. I n den erhaltenen
Teilen ist, wenn meine Textergnzungen zutreffen, lediglich von w'xst c k s t y t
hlichen Worten (Zeile 7) die R e d e . Ein weniges wissen wir von deren I n h a l t ,
wenn die B e h a u p t u n g auf einem dem F r a g m e n t M 119 a zugefgten E t i k e t t
zutrifft, 2 9 d a dies die rechte E c k e von M 119 bilde, u n d zwar offenbar so, d a
M 119 a /2/ den Anfang der ersten Zeile von M 119 darstellt. I n diesem F a l l
h t t e j e m a n d b e h a u p t e t : n w m p 8 k ' mn(') I mywn x c y y das Gesetz ist mir
ebenso. 30
28
Meine bersetzung folgt GERSHEVITCH : Grammar [s. A n m . 20] 1679.
29
Vgl. M. BOYCE: Catalogue [s. A n m . 1], S. 10.
30
I m y wj i zu man.-soghd. 'yw mywn, das von GERSHEVITCH : Grammar [s. A n m .
20] 1317 altogether bersetzt wird, dazu buddh.-soghd. 'yw m y ' w n (E. BENVENISTE :
J A 1936. S. 218) und christl.-soghd. 'yw mywn, als dessen B e d e u t u n g N. SIMS-WILLIAMS
in seiner noch unpublizierten Edition der Sammelhandschrift 2 like nachweisen
k o n n t e (Anm. zu p. 65 /V/21/). Die Bedeutung des Ausdruckes ist gleichwohl fraglich.
S W erwgt The law is the same for mo (as for you). Sollte aber eine der Sonderbedeutung
des deutschen gleich im Sinne von egal, gleichgltig entsprechende Entwicklung auch
f r das soghdische Wort angenommen werden drfen, so ergbe sich die provokatorische
Feststellung Das Gesetz ist mir egal [Vgl. aber S. 309, Anm. 31.].

300 W . SUNDERMANN
Nach dieser Beschreibung der Gegenpartei lt sich das ber die eigenen
Oberen Gesagte noch etwas konkretisieren. Sie waren keine Syrer sondern
o f f e n b a r Iraner, Soghder oder Perser, wie mindestens aus dem Namen Nwn-
zdag (Zeile 19) hervorzugehen scheint.
Man wird dem Verfasser des zweiten Briefes Betroffenheit von einem
schwerwiegenden gesellschaftlichen Problem nicht absprechen, wie immer
m a n seinen S t a n d p u n k t beurteilen mag. Aus dem ersten Brief lt dieser Ein-
d r u c k sich nicht u n m i t t e l b a r gewinnen. I m brigen aber haben beide Schrei-
ben so viele Gemeinsamkeiten, da die Annahme, sie seien gleichen Alters u n d
handelten auf ihre Weise von denselben Dingen, gestattet ist. Die noch offenen
F r a g e n , wann, von wem u n d an wen die Briefe geschrieben wurden, sollten da-
her f r beide D o k u m e n t e im Zusammenhang zu beantworten versucht werden.
W a n n wurden die Briefe geschrieben? Zweifellos, wie bereits Henning
b e t o n t e , nach der E n t s t e h u n g der Mihriya- und Miqlsiya-Y&rteien der meso-
potamischen Manicher, d. h. nach dem Zeugnis des Fihrist allerfrhestens
in der 1. Hlfte des 8. Jahrhunderts. 3 1 Ein terminus ad quem lt sich weniger
leicht gewinnen. F r H e n n i n g war er durch die R c k k e h r der Miqlsiya zur
Mihriya gegeben, die e t w a 880 erfolgte. 32 Henning selbst betonte jedoch die
Unsicherheit einer solchen Zeitbestimmung. Die Mglichkeit fortdauernder
Parteienkmpfe, zumal im fernen Zentralasien, lie sich j a nicht ausschlieen. 33
Epigraphische Grnde sprechen nach Henning f r eine E n t s t e h u n g der Schrei-
ben frhestens um die Mitte des 9. Jh. 3 4 Mir selbst erscheint in diesem Zusam-
m e n h a n g der rein iranische Charakter der Briefe hervorhebenswert. Ihre Spra-
che ist soghdisch, nicht trkisch. Die genannten Personen tragen iranische
oder syrische Namen. Nirgendwo stt m a n auf einen trkischen N a m e n oder
ein trkisches Wort. Die gelufige Sprache mindestens des Adressaten in
Qoco, vielleicht auch des Absenders, wahrscheinlich aber jene des Briefschrei-
bers d r f t e also die soghdische gewesen sein.
Bedenkt man, d a die Vernichtung des Uigurischen Steppenreiches durch
die Kirgisen im J a h r e 840 zur Auswanderung uigurischer Stmme nach
Oaochang, Oanzhou u n d in andere Gegenden Chinesisch Turkistans f h r t e u n d
d a m i t zur schlielichen Turkisierung dieser Gebiete, 35 so erscheint eine Datie-
r u n g der Briefe vor M i t t e des 9. J h . oder doch nicht viel spter sehr wohl mg-
lich.
Der Schriftduktus widerspricht nicht dieser Mglichkeit. Als Briefe
mochten sie in einem vereinfachten Stil geschrieben worden sein, neben dem
31
Neue Materialien [s. A n m . 1]. S. 17.
32
Neue Materialien [s. A n m . 1]. 8. 17.
33
Neue Materialien [s. A n m . 1]. S. 17 A n m . 4.
34
Neue Materialien [s. A n m . 1]. S. 17 A n m . 4.
35
Vgl. A. V. GABAIN : D a s Leben im uigurischen Knigreich von Qoo, Wiesbaden
1 9 7 3 . S . 1 9 , VANNES P E L L I O T : T r a i t I I [ s . A n m . 1 6 ] . S . 3 0 7 .

P R O B L E M E DER I N T E R P R E T A T I O N MAN.-SOGHD. B R I E F E 301
die ornamentale Schrcibform blich war u n d blieb. D a beide Stile lngere

Zeit nebeneinander beherrscht wurden, ist jedenfalls wahrscheinlicher als d a
sie einander einfach ablsten.
An wen sind die Briefe gerichtet? An eine hochgestellte Persnlichkeit
von Kompetenz u n d Autoritt in geistlichen Dingen, also eher an einen Kle-
riker als an einen weltlichen Herren. D a er im zweiten Brief xypS'wnd
Herrscher genannt wird, spricht nicht dagegen. Das F r a g m e n t M 7440 stellt
die Anschrift eines Briefes dar u n d nennt den Adressaten in der F o r m ' t '
y'n 'nywn p'scn xypS'wnd mzyx Synmyc srSng An den gttergleichen
(oder: An Vayn-anyn?), den ehrwrdigen Herrscher, das groe O b e r h a u p t
der Kirche.
Henning vermutete im Adressaten der Briefe den Lehrer, der vielleicht
als Oberhaupt der manichischen Kirchenprovinz Churasan in Qoo residierte
u n d dessen Wrde er vergleichend als die eines Erzbischofs beschrieb. 38 I c h
halte dies fr richtig. Allerdings sind wir ber die hierarchische Gliederung der
manichischen Kirche Zentralasiens k a u m unterrichtet. Aber es ist ja denk-
bar, da der mutmalich in Qoo residierende Lehrer als Lehrer der Provinz
Churasan und Oberhaupt von ahr-Tuyristm, bzw. als Groer Lehrer von
*Trt-Tuyri die Stellung eines Primas aller Kleriker der Dnwariya-Gemein-
schaft, mindestens aber des ahr-Tuyri-la^ndes, einnahm. 3 7 Die Frage ist nur,
ob ein solcher Wrdentrger bereits f r die Zeit vor Begrndung des uigurischen
Reiches von Qoo angenommen werden darf. Mglich ist das durchaus doch
bisher nicht beweisbar. Der Zeugenwert eines uigurischen Textes zugunsten
dieser Annahme hngt von der Frage ab, ob der Buyuy-qayan des Textes m i t
dem Bg-qayan, der den Manichismus im uigurischen Reich einfhrte, gleich-
gesetzt werden darf. 3 8
Gewi aber war Qoco Metropole des zentralasiatischen Manichismus
in der Zeit nach Begrndung des uigurischen Reiches von Qoo, also seit Mitte
des 9. Jh. 3 9
Was wissen wir von den Absendern der Briefe ? Ihre Namen und Wohn-
orte sind uns u n b e k a n n t . Unter den oben gemachten Voraussetzungen sollte
ihre Heimat aber die Dizese des Erzbischofs von Qoo gewesen sein, also
vorzugsweise das Cahr-Tuyriih&nd, vielleicht aber auch das Uigurische
Steppenreich oder China. Vom Verfasser dos zweiten Briefes lt sich vermuten,
da er nicht der hheren Geistlichkeit angehrte, da er sich frbittend f r die
Gemeindeoberen verwendet. E r knnte sogar ein Vertreter des Laienstandes
36
Neun Materialien [s. Anin. 1]. S. 16 A n m . 2.
3
' W . B. HENNING: Argi and the Tokharians. BSOS 9 (1938) 651 f.
38
Neue Materialien [s. Anm. 1]. S. 15 A n m . 2. Der Text wurde verffentlicht von
A. v. LECOQ in der Festscqrift Vilhelm Thomson, Leipzig 1912. S. 147, 149 ff.
39
VANNESPELLIOT : T r a i t I I [s. A n m . 16]. S. 196 f.

302 W. SUNDERMANN
der Hrer gewesen sein. Die manichische Schrift htte er d a n n zwar schwer-
lich beherrscht, doch knnte er seinen Brief ja diktiert haben. Der erste Brief
d r f t e durchaus demselben Milieu e n t s t a m m e n . Mehrfach ist in ihm von Electi
u n d Electae die Rede, ohne da der Verfasser sich ausdrcklich einer dieser
Gruppen zurechnet.
Die hier behandelten Dokumente sind Privatbriefe genannt worden. 4 0
D a s t r i f f t insofern zu, als sie keine Staatsschreiben darstellen. I m brigen
handeln sie aber von Gemeindeangelegenheiten ffentlichen Charakters.
Worin liegt die Bedeutung der Briefe f r die Geschichte des zentralasia-
tischen Manichismus ? Zweifellos in der Bezeugung syrischer Manicher in
Zentralasien, in den Informationen ber sie und ber ihr Zusammenleben mit
Angehrigen der zentralasiatischen Dndwarya-Gemeinde. Syrer werden sie
im zweiten Brief genannt, von den Mihrycind und Miklsikt, die in Mesopo-
t a m i e n zu Hause waren, spricht der erste. Aber dies bedeutet wohl mehr, d a
m a n es mit Fremdlingen aus Sristn, d. h. Mesopotamien, zu t u n hatte. Die
N a m e n der gegenerischen Partei sind weitgehend persisch (vgl. S. 297). Persisch
sind aber auch die zitierten Worte der Gegner, und schlielich haben sich lite-
rarische Nachweise einer Prsenz manichischer Syrer in der Turfanoase bis-
her nicht gefunden. Der schon Flgel aufgefallene betrchtliche Anteil irani-
scher Manicher am Parteienkampf der mesopotamischen Kirche d r f t e hier
eine Besttigung finden. 4 1
Bisweilen wird das Erscheinen syrischer Manicher in Zentralasien auf
Manicherverfolgungen im islamischen Gebiet zurckgefhrt. 4 2 Dem Eihrist
zufolge geschah in nachsasanidischer Zeit die erste groe F l u c h t der Manicher
aus Mesopotamien nach Churasan u n t e r dem Druck der Verfolgungen al-
Muqtadirs zu Anfang des 10. Jh. 4 3 Es liegt nahe, das Vordringen der manichi-
schen Syrer nach Zentralasien dann mit diesem Ereignis zu verbinden u n d in
das 10. J h . zu datieren. Sprache und I n h a l t der Briefe sind damit aber nicht
vereinbar, die wie oben gezeigt, ein hheres Alter erfordern.
Tatschlich h a t G. V a j d a eine lange Zandiqa-Verfolgung, die auch
Manicher erfate, in der Zeit von 163 d. H . (779/80) bis 179 (795/96) und auch
noch in den folgenden J a h r e n nachgewiesen. 44 Mit den oben gemachten Datie-
rungsversuchen ist dieses Ereignis als Auslser der Flucht von Manichern
besser vereinbar als die Verfolgungen u n t e r Muqtadir.
40
M. BOYCE: Catalogue [s. Anm. X]. S. 148. H a n d b u c h der Orientalistik (s.
A n m . 8). S. 73.
41
Mani [s. Anm. 11]. S. 322.
42
LIEU : Nestorians a n d Manichaeans (s. A n m . 4). S. 75.
43
FLGEL : Mani [ s . A n m . 11]. S. 77, 105 f., K i t b al-Fihrist I [ s . A n m . 11]. S. 337.
44
Die Zindlqs im Gebiet des Islam zu Beginn der 'Abbasidenzeit, in : Der Mani-
chismus. E d . G. WIDENGREN. D a r m s t a d t 1977. S. 426.

Da die syrischen Manicher im Gebiet der ihnen entfremdeten oder

fernstehenden Dnwarya-Gemeinde Zuflucht suchten, knnte aber auch
durch die folgenschwere Bekehrung des Chans des Uigurischen Steppenreiches
Bg-qayan in Lo-yang im Jahre 762/63 zum manichischen Glauben motiviert
gewesen sein. 45 Sie f h r t e nachweislich zu einer Reihe von Missionsreisen
manichischer Geistlicher durch die uigurischen Lande, von denen die Zeilen
8 bis 10 der chinesischen Version der Inschrift von Qara-Balgasun (um 810) 4e
in der bersetzung von E d . Chavannes und P . Pelliot folgendermaen berich-
ten : [Le qaghan] emmena quatre religieux, dont -Q. Jouei-si, et les intro-
duisit dans son royaume ; ils dveloppaient et exaltaient les deux sacrifices 47
et pntraient profondement les trois moments. D'ailleurs le matre de la Loi
tait merveilleusement instruit d a n s la doctrine de la Lumire et comprenait
parfaitement les sept ouvrages. Ces capacits taient plus hautes que le pic
maritime ; son loquence tait comme une cascade. C'est pourquoi il p u t initier
les Ouigours la vraie religion. . . Le roi de la religion, ayant appris que
[les Ouigours] avaient accept la vraie religion, loua fort leur respectueuse. . .
[vertu], conduisant des religieux et des religieuses, e n t r a dans le r o y a u m e
pour y rpandre et y exalter [la religion]. Ensuite la foule des disciples du
mou-ch parcourut les pays en tous sens de l'Est l'Ouest et alla et vint en
prchant la religion. 48
Danach m t e n mehrfach, mindestens drei Mal, verschiedene G r u p p e n
manichischer Glaubensboten in das uigurische Reich gekommen sein : 1. vier
Missionare, die Bg bei seiner R c k k e h r aus Lo-yang begleiteten. I n unkla-
rem Zusammenhang mit ihrer Reise wird das Wirken eines matre de la Loi
erwhnt, 2. ein roi de la religion m i t Electi und Electae, 3. viele Schler eines
Lehrers, die von Ost nach West d a s uigurische Reich durchdrangen. I n dem
roi de la religion (jÊE fa-wang) sahen Chavannes und Pelliot le chef suprme
de la religion manichenne, der lange in Babylon residierte. 49 Sollte dies zu-
treffen, so h t t e n wir den Beweis in der H a n d , da die Bekehrung des Uiguren-
chans und seines Volkes zum Erscheinen manichischer Missionare aus Meso-
potamien in Zentralasien f h r t e .
" V g l . z. B. A. v. GABAIN : D a s uigurische Knigreich von Chotscho 8501250.

SDAW, Kl. f. Sprachen, Lit. u. K u n s t 1961 Nr. 6. Berlin 1961. S. 24.
16
L. BAZIN in Philologiae Turcicae F u n d a m e n t a I I . E d . L. BAZIN U. a. Wiesbaden
1964. S. 205.
" D a z u s. CHAVANNESPELLIOT : T r a i t I I [s. A n m . 16]. S. 190 A n m . 2. L I E U :
Polemics against Manichaeism [s. A n m . 4]. S. 156 Anm. 6.
48
CHAVANNES PELLIOT : Trait I I [s. Anm. 16]. S. 190 ff. Der ganze T e x t bei
G. SCHLEGEL : Die chinosischo Inschrift auf dem uigurischon Denkmal in K a r a Balgas-
sun. Holsingfors 1896. S. 43 ff.
48
Vgl. CHAVANNESPELLIOT : T r a i t I I [s. A n m . 16]. S . 195 A n m . 1.
Acta Antiqua Academiae Scieniiarttm Hungaricae 28, 1980

304 W. SUNDERMANN
Aber dies ist nicht mehr als eine Hypothese. Da der roi de la religion
der Archegos der manichischen Kirche in Babylon sei, geht aus dem chinesi-
schen Text selbst nicht hervor. Man knnte in ihm auch den Lehrer der Kirche
v o n Churasan in Qoco oder einen anderen Lehrer vermuten. Nichteinmal die
Richtigkeit der Lesung fa-wang roi de la religion darf als vllig sicher
gelten. Eine Lesung fa-zhu ( j ^ ) chef de la Loi wrde fast gleich geschrie-
ben. 5 0 Sollte fa-wang im Text stehen, so wre nach seiner genauen Bedeutung
als manichischer Terminus zu fragen. D a der Begriff nicht ohne Kenntnis
seiner Verwendung im buddhistischen Sprachgebrauch, nmlich als quiva-
lent von aind. dharmarja und Bezeichnung Buddhas 5 1 , gebraucht wurde, ist
sicher. Als ein manichischer Kleriker d r f t e der fa-wang e t w a das sein, was die
soghdischen Manicher m z y x Synmyc srSng das groe Oberhaupt der Kirche
n a n n t e n . Wir wissen aber nicht, ob dies der Titel eines bestimmten Wrden-
t r g e r s war oder ob er von vielen geteilt wurde.
Auf Grund der hier vorgetragenen Erwgungen halte ich es fr mglich,
d a die soghdischen Briefe die folgenden historischen Situationen vorausset-
zen : Entweder : Die Bekehrung der Uiguren zum Manichismus fhrte dazu,
d a a m Ende des 8. u n d zu Beginn des 9. J h . soghdische Manicher in das Uigu-
rische Steppenreich zogen, um die heidnischen Trken zu bekehren und eine
manichische Kirche zu begrnden. Manicher aus Mesopotamien, die in ihrer
H e i m a t Verfolgungen ausgesetzt waren, folgten ihnen, um am Werk und seinen
F r c h t e n teilzuhaben. Sie gerieten bald mit ihren soghdischen Glaubensbr-
d e r n in Konflikt. Diese riefen den Beistand des Lehrers in Qoco an. Oder :
Die Existenz eines manichisch-uigurischen Staatswesens in Qoo zog in der
2. H l f t e des 9. J h . manichische Geistliche aus Mesopotamien an, die sich in
d e n H a u p t o r t e n der kleinen Staaten an der Seidenstrae niederlieen und mit
der ansssigen, noch soghdischen Geistlichkeit in Konflikt gerieten. Der Lehrer
von Qoo als P r i m a s des manichische Klerus in Zentralasien wurde von den
Soghdiern um Beistand angerufen.
Berlin.
50
VANNES PELLIOT : T r a i t I I [s. A n m . 16]. S. 192 A n m . 1 u n d T r a i t I
(s. A n m . 25). S. 581 A n m . 4.
51
V g l . SCHLEGEL [S. A n m . 4 8 ] . S . 64.

P R O B L E M E D E R I N T E R P R E T A T I O N MAN.-SOGDH. B R I E F E 305
ANHANG I
E D I T I O N D E R B R I E F E 1 und 2
Erster Brief
III [ ]nd(s)kwn w(n)[ ] /2/ [ ]' kww Sstww [ ]

/3/ [ ] 'yg h m ' g dyn ' m ' h hwm /4/ [ ](c) m ' x yrw'n mwj'(k)
/5/ [ ](w)m't'nd mynd (my)hry'nd k r t r t y ( ' ) 1 /6/ [ ]m'x
s y t m ' n Syn'rt ' t y 'wr8p(')[r] /7/ [ Sy]nS'ry' cx'pSS t ' p y y /8/ [ ]
m'yzym n y x y ' ' t y /9/ [ ](. )dyy ' t y xwp m ' x t y s y y /10/ [ ](. )
kyy en mkl'syktyy /11/ [ ](. ) p ' r t y I pSkyy I ' w s t n y y /12/ [ ]
(x)nd p ' r t y prw n m w pr cxs'pSS /13/ [ 'f](tmww) 2 prw rsty'k '
[Lcke von unbestimmter Weite)
(14/ [ ]wyy 7 '[(?) ] /15/ [ 3 - 5 ](c)[ 1 0 - 1 5 yn']-

()[r']nc[ 'p](r)[y](w)( ?)3 wyyn k t wys'ndy [ 7 ' '] /16/ [ 2 - 4 ]p'xw'y
('t)[y 8 10 ](.)swnr'rt 4 t y m m ' x 8yn'r'[nc w y y n k t wyndy] 5 /17/
i^yn'](r')nc xwsyc[ e 5 - 7 p]rsm'skwn( ?) ms " z ' d d w x t (xyp) [ 8
12 ] /18/ [ 2 - 4 pw] (ry)st( ?)7 w'nw wyn'nd kt wy[S'ndy] 8 yn'r'nc tsycq
n y ' t y y ' t y z'y kn('sk)[wn] /19/ [ 1 2 - 1 6 ](rwrt)t 9 nxw(')[yn](ds)kwn
' t y ms 'rwk ' t y p ' s t ' t'snds(kwn) /20/ pww pcqwyr t y m ms m ' x yn'r't
w'nw wyn'nd kt c'nw wys'ndy yn'r'nc /21/ x w r n y y "syyskwn cn " p y y
'pryw k m ' sn'yyskwn ' t y m ' x yn'r'st nxrysndskw(n) /22/ wys'nd w'nw
p'tcynyy kwn'nd k t c'tyk " p m w r t y y xcyy m'y p c ' y t t ' t y ms /23/ rwr'
x(w)tyy(?) nxw'y'ndskwn ' t y "t(r) xwtyy 'wst'ndskwn 'tysn prywy8
w'xs yw'(n)[ ' t y ] /24/ k m b w n y y n y y sm'r'ndskwn 'rty t y m
d y n c y h r y f t cxs'pyy w'x xcyy /25/ xwycq 'ty pww s f ' r p'rtysn [s](')r'r
k y y myhr p'd'r 1 0 xcyy xwycn'k w m ' t ' t y sy /26/ c's'r p y k r'f wm't ' t y I (z)-
1
G e r s h e v i t c h las k r t r t y y (GMS 249 A n m . 1). Vom l e t z t e n Buchstaben ist n u r
ein oberer R a n d erhalten, der zu einem y oder ' gehren k n n t e . E i n ' wrde den a m
Zeilenende zur Verfgung stehenden R a u m besser fllen als ein y .
2
Spitze eines t, oberer R a n d des m mid obere Teile der letzten Buchstabon erhalten.
3
P u n k t ber dem r erhalten, w durch Ri im Papier zerstrt.
4
Der dem vorangehende B u c h s t a b e knnte ein w gewesen sein, also '](w)Swm8'rt
5
Meine E r g n z u n g modifizierte inen Vorschlag von SW, Syn'r'[St w y n ' n d k t ]
zu losen, m i t dem Ziel, d a s erwartbaro Possessivpronomen wy(')ndy in der vorhandenen
Lcke unterzubringen.
Die GMS 1012 erwogene Mglichkeit, xwycq zu lesen, wird durch den erhal-
tenen Text ausgesclilossen, der die F o r m des finalen c deutlich b e w a h r t .
7
Kopf und P u n k t des r erhalten, y versehrt, auch w mglich.
8
So wahrscheinlicher als wyS'nd (SW).
9
N u r r t t sicher erkennbar.
10
Li /25/ wie m w h r p ' d ' r geschrieben, in /30/ deutlich m y h r , so da das, w a s
in /25/ wie der Aufstrich eines w aussieht, in Wirklichkeit d a s linke untere E n d e
eines m ist.
20 Acta Antiqua Academiae Scientiarum Hungaricae 28, 1UH0

306 W. S U N D E R M A N N
[ ' ] ( k ) ' n c k r ' y ' n c w n y y s ' r t y s ' t y y y r t r n y j t y m [ ' x ] /27/ s y t m ' n y y ' p t y y ' p
c ' n w x w n y y k r ' y ' n c n y j t y y ' t y S y n S ' r t n y x y ' k w n ( ' ) n d ' t y /28/ k r ' y ' n c k w w
y z d ' r y ' m ' n d r y s t r w s n m y h r w h m n ' t y w h m n ' h s'r w ' n w w y t w S ' ( r ) [ t ] /29/ k t
I I p r w r t y y psyy Sry' xwrnyy " s t w w 1 1 'ty t y m I prwrtyy "smk'm ms
12
my(n)[dy] /30/ x y p S I ' s p ' s k r ' n c I ayn'r'nc 'pryw y'r "8w p8w
[w]()'13 ' t y m y h r ''()14 /31/ s'r'r w n y y 'sp'skr'nc 'z' n y y ' s ' t y cn y ' r y y
w y x n m s x z ' n y ' t y h /32/ ( )
{Zeilenfragmente unbekannter Zugehrigkeit) /33/ [ ](.)[
/34/ [ ] ( c w ) x t [1 2]()[ ] /35/ [ ] pyrnm

prw xyp(e)[ ] /36/ [ ](.)rt kwnyskwn (w')[ ]
/37/ [ ](s)ty ty(.)[ 2-4 ](.)'d[ ]
Zeilenfragmente unbekannter Zugehrigkeit) /38/ 8yn'(r)[ ]
/39/ nwmpk'[ ] /40/ [ 1 - 3 ](ps' ')[ ]
Erster Brief (bersetzung)
/!/ [ ] zur H a n d [ ] der ganzen Kirche sind wir. [

u n s e r * g l a u b e n s s t a r k e r 1 5 L e h r e r /5/ [ ] w a r e n d i e s e A n h n g e r M ihr s
11
Nicht " s t w (GMS 690).
12
E r g n z t g e m GMS 1396 (Hinweis von SW).
13
I n GMS 463 p 8 w ( s ) \ D a s W o r t e n d o ist durch ein L o c h im Papier zerstrt,
d a s breiter ist als ein s. Von d e m so gelesenen Buchstaben ist n u r die linke obere Spitze
e r h a l t e n . D a sie spitz a u s l u f t u n d n i c h t r u n d ist, spricht aber fr s t a t t s.
" V g l . A n m . 10.
15
yrw'n in / 4 / u n d im zweiten Brief /18/ ist unpubliziert wiederholt belegt :
M 635 /I/V/67/ ' r t y 'yS yrw'n p r [ n ] ( x ) w n d y y w m ' t k y y Und dieser glckselige .
w a r (es), der . . ., M 4 4 3 6 / 1 6 / yrw'n ' f t ' ( g ) [ ' n ] der . Bischof oder Bischof ., M 6673
/ 5 / ] y n y y yrw'(n) dieser .. Ist vayruwn appellatives A t t r i b u t oder Eigenname ?
I c h h a l t e die erstere Mglichkeit f r die wahrscheinlichere, d a m a n u n g e r n einen namens-
gleichen Bischof u n d L e h r e r a n n e h m e n wird, whrend andererseits dieser N a m e nicht
s i c h e r bezeugt zu sein scheint. D a die Wortstellung m ' x . m w j ' k s t a t t m ' x m w j ' k .
o d e r . m ' x m w j ' k gegen die A n n a h m e eines Eigennamens spricht, b e t o n t mir gegen-
b e r S W . Vayruwn l t sich entweder als eine Gottesseele (habend) (d. h. gttlich )
v e r s t e h e n oder eine Seele (wie der Gott) V a y (habend) oder als Zusammensetzung
m i t d e r unbezeugten soghd. E n t s p r e c h u n g des m p . belegten W o r t e s b g *sicher (Mir.
M a n . I I , S. 338), womit auch p a r t h . (und m p . ) 'bg *unoasy, nervous ( H e n n i n g , BSOS 9
[1937], S. 80) v e r b u n d e n wird u n d wozu H . W . B a i l e y chot.-sak. bagalagv gallants( ?),
oss. bgu surely usw. (Dictionary of K h o t a n Saka, Cambridge 1979, S. 264) stellt.
Z u g u n s t e n der U b e r s e t z u n g eine Gottesseele habend liee sich auf die Bedeutung des
s o g h d . Personemiamens yyfrn, yyprnw God-luck (I. G e r s h e v i t c h , T S P 1945, S. 139,
D . W e b e r , I F 77 [1972], S. 196) hinweisen. Die letztgenannte Mglichkeit erscheint m i r
a b e r doch etwas berzeugender, wenn m a n . als Appellativum u n d nicht als Eigen-
n a m e n a u f f a t . J e d e n f a l l s verbinden sicher d e u t b a r e K o m p o s i t a m i t ruwn als Zweit-
b e s t a n d t e i l dieses W o r t m . W . m i t A d j e k t i v e n oder gleichwertigen P r f i x e n : m p . anSru-
wn mit unsterblicher Seele, d. h. seelig verschieden, huruwn gutseelig, d. h.
fromm, p a r t h T . duiruwrnjt belseeligkeit, d. h. Schlechtigkeit, Gottlosigkeit.
( E i n e Verkrzung von AnSruwn ist nach P h . G i o n u x , P a r o l e jde l'Orient 6 7
[197576], S. 522 syr. Sirwn.) U n t e r dieser Voraussetzung sollte auch im Fall von
y r w ' n Zusammensetzung m i t einem A d j e k t i v erwartet werden.
Acta Aruiqua Academiae Scientiamm Hunaricae 28, 1980

PROBLEME DER INTERPRETATION MAN.-SOGDH. B R I E F E 307
*Ttig[keit( ?) ie ] all unsere E r w h l t e n und darberhinaus [ ]

Erwhltenschaft, Gesetz, Siegel [ ] wir begannen *Nachforschung 1 7
18
und [ ] u n d gut ? /10/ [ ] die von den A n h n -
gern des Miqls [ ] denn von einem Gesetz u n d einem U r s p r u n g
[ ] sind sie. Denn in Gesetz und Gebot [ ] Als erstes : Gegen
das Gebot der Wahrhaftigkeit [sndigen sie
(Lcke)
[ ] /15/ zusammen mit der Electa sah, da ihre [Electa ]

abhieb und [ ]nhte. Wiederum [sah] unsere Elec[ta, da ihre]
Electa ? [p]rete. 19 Ferner sahen die jungfrulichen( ?) [Tchter]
der zd-duxt, d a ihre Electa eine *Hacke 2 0 nahm und E r d e ausgrub. [ ]
" K r t r t y ' in /5/. Zur Lesung vgl. A n m . 1. GERSHEVITOH bersetzte insidious,
cunning (GMS 249 A n m . 1). F r die B e d e u t u n g des Wortes als ein negatives A d j e k t i v
eprechen die Parallelen M 133 /R/i/1819/ yrf q r j r jyqt n y z ' n y y j many cunning,
sinful passions (GMS 1668) und M 6912/3/ q r t r t Sywt listige Dmonen. Damit u n v e r -
einbar ist aber die Verwendung desselben W o r t e s in dem christl.-soghd. Text 3 =
T I I 35 + T I I 14 / R / 1 3 1 5 / s ' t wySnt byy p t r y n b n t c y q q r t r t 'nt'St 4 f r y S t y t
k r w b y t w s r ' p y t 4 s ' t (s)p'dy'nt qy sqwnt s m ' n y sqys'r alle diese bei Gott d e m V a t e r
befindlichen k. Scharen und Engel, die Cherubim und Seraphim u n d alle Heerscharen,
die ber dem H i m m e l weilen. Hier liegt eine bersetzung mchtig, stark, a k t i v , lob-
haft nahe, die f r g u t e und bse Wesen zutreffen k a n n . Dies vorausgesetzt, b e t r a c h t e t
SW q r t r als soghd. E n t s p r e c h u n g von p a r t h . k y r d y r , m p . k r t y r usw. If these d e r i v e
from *k(a)rtrya- (with SZEMERNYI) there is p r o b a b l y no phonetical problem, cf. w y t r
(3 sg. impf.) < *wi-a-tryat. Zu soghd. -r < -fya vgl. genorell GMS 185. Wie P r o f .
H o t b a c h a m 2. 10. 1980 in Budapest mitteilte, k o n n t e er den N a m e n k r t ' y r in einer
der soghd. Shatial-Biidge-Inschriften lesen. Beide Schreibungen, k r t r und k r t ' y r , sind
u n t e r der Voraussetzung vereinbar, d a sie f r kartlr stehen.
17
N y x y ' i n / 8 / u n d / 2 7 / . HENNINGS bersetzung Sorgsamkeit, Aufmerksamkeit
(BB, S. 86 zu 727, S. 104 zu f. 56) erfat gewi jene Bedeutung des Wortes, die der i m
vorliegenden Text vermuteten Wiedergabe durch Untersuchung, Nachforschung zu-
grunde liegt.
18
M ' x t y s y y in / 9 / ist, wie die Verlngerung der unterzoiligen Horizontalen des
X u n t e r die folgenden Buchstaben erkennen lt, ein Wort. Seine Bedeutung Mondein-
gang o. . und seine F o r m (Ohl. ?, -aka Erweiterung ?) sind unklar. Sollte der astro-
logische Eigenname einer Person vorliegen ? ( E t w a : geboren zur Zeit des E i n t r i t t e s d e s
Mondes in sein Haus, oder : zur Zeit des E i n t r i t t e s des Mondes in seino Planetenstunde.)
Besteht vielleicht sogar eine Verbindung m i t dem etymologisch nicht ganz klaren per-
sischen F r a u e n n a m e n Mahsat o. . (vgl. F . MEIER : Die schne Mahsat I . Wiesbaden
1963. S. 43 ff.) ?
19
Die E r g n z u n g sali und die f r das Verstndnis des ganzen Satzes entschei-
dende Beziehung von [p]rsm'skwn auf ihre Electa s t a t t auf unsere verdanke ich S W .
20
T6ycq in /18/ wurde von HENNING ursprnglich spade bersetzt (GMS 1012),
spter von ihm aber, worauf MK mich hinwies, mit chwar. tyckyk axe verglichen
(Z. V. Togan'a A r m a g a n , Istanbul 1956, S. 435). Wie M K mir weiterhin aus seiner K e n n t -
nis des Chwar. m i t t e i l t , entspricht n p . qadm, ta adze. E r folgert : If t h e K h w a r .
is a diminutive in -'k, the Sogd. word a tool for digging, r a t h e r t h a n cutting will
bo a ,mattock, Hacke'. Besttigend kann auf d a s etymologisch verwandte n p . ta hin-
gewiesen werden, d a s Beil und Hacke b e d e u t e t (STEINGASS : S. 342). Dazu g e h r t
aber auch p a s c h t u tiyo plough-share, d a s G. MORGENSTIERNE : An Etymological
Vocabulary of P a s h t o . Oslo 1927, von *tay- ableitet u n d mit n p . ta vorgleicht, sowie
wakhi ti ieleznyj nakonenik pachotnogo o r u d i j a , k o v a n y j sonik, lemech (A. L .
GRJUNBERG I . M. STEBLIN-KAMENSKIJ : V a e h a n s k i j jazyk. Moskva 1976. S. 473). Ange-
sichts der hnlichkeit von primitivem H o l z p f l u g (Hackenpflug) und Hacke (vgl.

308 W. SUNHERMANN
H e i l p f l a n z e n z e r s t o e n sie u n d s c h n e i d e n a u c h o h n e S c h e u H o l z u n d * G e r t . 2 1
/20/ F e r n e r sahen u n s e r e Electae a u c h , wie ihre E l e c t a B l u t schrpfte22 (und)
23
mit Wasser das ? w u s c h . U n d u n s e r e E l e c t a e t a d e l t e n sie, sie g a b e n
z u r A n t w o r t : B r u n n e n w a s s e r ist t o t , so ist es gestattet. U n d a u c h H e i l p f l a n -
z e n z e r s t o e n sie s e l b s t , u n d d a s F e u e r l s c h e n sie s e l b s t ( ?). 24 U n d sie s e h e n i n
d i e s e r S a c h e n i c h t S n d e n o c h M a n g e l . U n d ferner : Z u m G e b o t d e s reli-
g i o n s m i g e n V e r h a l t e n s i s t d a e i n e S a c h e , /25/ o f f e n b a r u n d s c h a m l o s . 2 5
D e n n i h r O b e r h a u p t , w e l c h e s Mihr-pdr26 ist, w a r k r a n k , er h a t t e ein L e i d e n
u n t e r d e m F u , u n d ein gemietetes M d c h e n 2 7 t r a t bei i h m ein, u n d s p t e r ging
sie h i n a u s . U n s a l l e e r g r i f f V e r d a c h t , 2 8 a l s j e n e L o h n d i e n e r i n 2 7 h e r a u s g i n g ,
A. NEUBURGER: Die Technik des Altertums. Leipzig 1919. S. 86) berrascht es nicht,
d a H a c k e und Pflugschar mit etymologisch verwandten Begriffen bezeichnet werden,
D a die Electa sich also einer Hacke bediente, d r f t e MK mit R e c h t annehmen. Wenn
die m i t dem Gert a u s g e f h r t e Arbeit ein regelrechtes Graben war (kan-), so lt sich
a n eine Grabhacke (ketman) der Art denken, wie sie bis in unsere Zeit in Chinesisch-
T u r k i s t a n beim Ausschachten der Erde in Gebrauch sind (vgl. L. GOLOMB : Die Boden-
k u l t u r in Ost-Turkistan. Posieux, Ereiburg/Sehweiz 1959. S. 61, 63.). Die Electa knnte
die H a c k e dann etwa f r Arbeiten an einem Bewsserungskanal verwendet haben. Sollte
kan- hier aber in freierer Verwendung n u r eine Lockerung des Bodens bedeuten, so wre
eher a n eine Spitzhacke zu denken (GOLOMB : aaO., S. 63).
21
P ' S t ' in /19/, vgl. GMS 973, 1268, ein hlzernes Gert nicht genau bestimm-
b a r e r Bedeutung, zu buddh.-soghd. p'Styt (PL.), vgl. HENNING : BSOAS 11 [1946] S. 728.
22
Oder : zur Ader lie. Zur Bedeutung vgl. Textanmerkung 14.
23
K m ' in /21/, vgl. dazu T e x t a n m e r k u n g 15.
24
" t r . . . ' w s t ' n d s k w n in /23/, wtl. sie setzen das Feuer. Zur Bedeutung vgl.
d a s n p . te neSndan to extinguish a fire, w t l . Feuer setzen, vgl. STEINGASS : S. 13.
S W gibt aber zu bedenken, da " t r . . . 'wst- im Soghd. vielleicht auch light a fire
b e d e u t e n knne, denn this m a y have been regarded as an improper activity for electi,
since it would cause pain to t h e fuel. Ich meine, d a auch diese Mglichkeit in B e t r a c h t
gezogen werdon mu, zumal f r die Manicher d a s Feuer eine aus oinom guten u n d einem
bsen Element bestehende Erscheinung war.
25
SW schlgt vor : In the m a t t e r of t h e commandment dyncyhryjt it ( = t h e Syr.
c o m m u n i t y ) is open a n d shameless. Wenn, wie SW mit Recht betont, pww f'r n u r
schamlos bedeuten k a n n , so erinnert die Begriffsverbindung xwycq 'ty pww s f ' r an
p w 'r r ' z y h in Or. 8212/83, Zeile 6 (bei N. SIMS-WILLIAMS : I I J 18 (1976), S. 49), das
SW zweifelnd so gelesen u n d without shame (or) secrecy bersetzt hat (aaO. S. 50
A n m . 33). Xwycq ' t y p w w Sf'r durfte also in positiver Wendung dasselbe besagen und
d a m i t f r die Richtigkeit der Lesung von S W sprechen.
23
J e nachdem, ob m a n pdr als Ableitung eines nomon actoris von pdan
schtzen oder pdan verweilen, warten a u f f a t (MACKENZIE : S. 62), liee dor N a m e
Mihr-pdr (in /25/ u n d /30/) sich sinnvoll als Beschtzer der F r e u n d s c h a f t , Liebe/ des
Vertrages oder als (wie/in/durch) Mithras bestndig, vielleicht, falls pdr auch Die-
ner sein konnte, als Diener des Mithras auffassen. K a u m denkbar ist Beschtzer der
Mithras. Wenn Mihr selbst Name ist, so wre zu fragen, ob hier eine Gottheit gemeint
ist (mpT. spiritus vivens, p a r t h T . der legtus tertius als Sonnengott) oder der Begrnder
der Mihriya-Partei der mesopotamischen Manicher, der auch Mihr hie (vgl. S. 292).
27
Z ' k ' n c k r ' y ' n c in /26/, k r ' y ' n c in /27/ u n d /28/. K r ' y ' n c ist gewi nicht von np.
kerye hire, price paid for labour, fare, rent (STEINGASS : S. 1020) zu trennen, das aller-
dings m . W . nie die gemietete oder entlohnte P e r s o n bezeichnet. D a s Verhltnis zwischen
n p . kerye und soghd. k r ' y ' n c d r f t e daher dasselbe wie etwa zwischen einem Ortsnamen
(*KS Kaschghar) u n d einem fem. Zugehrigkeitsadjektiv (q'ti'nc Kachgarienne) sein,
vgl. GMS 1048. Danach liee sich auch eine m a s k . Adjektivbildung * k r ' y ' n y gemietet,
entlohnt vorstellen.
28
Sn' hier ehor Verdacht als lediglich Mitrauen, wie S W mir gegenber mit
H i n w e i s auf HENNING in BSOAS 11 [1946], S. 719 u n d soino eigene, noch unpubliziorte
Arbeit ber die christl.-soghd. Handschrift 2 b e t o n t .
Acta A ntiqua Academiae Scientiarum Hungaricae 28, 1980

u n d die Electi stellten eine *Nachforschung17 an. U n d die L o h n d i e n e r i n s p r a c h

so z u Yazd-Arymn, Drist-RSn, Mihr-Wahman u n d Wahman-Sh: Zweimal
h a b e ich h i n t e r der Tr29 B l u t *geschrpft,30 u n d n o c h einmal werde ich
schrpfen. A u c h z e t t e l t e /30/ eine ihrer D i e n e r i n n e n m i t einer E l e c t a e i n e n
S t r e i t a n . U n d Mihr-pdr,26 (ihr) O b e r h a u p t , e r g r i f f d e n A r m j e n e r D i e n e r i n
u n d schuf Befreiung vom Streit. Auch im *Herbst u n d [
(Lcke)
[
/35/ [ ] vor im eigen [ ] tue [
(Lcke)
/38/ E l e c t [ ] Gesetz [ ]
Zweiter Brief
l [ ]y(x)'p8 8(')r(y)[ ] /2 'r]t[y]T (mys'ndy .)

[ ] /3/ [ ]( 6 - 9 m'x) gryjh ] /4/ ( 20-25
k'm(?) .s 15-20 )[ ] / 5 / (w'n)w wyt-
wS'rt n w m p8k' mn(')31 I mywn xcyy 'rtyy xyp8(')wn(d)[
/6/ [ ](w' r w c p r w s t y y w y y m y ' n d y x y p ) 8 3 2 r w ' n g ( r y w ')
[ ] /7/ [ ](y't) w ' n w f r m ' y " ( m s ) ' ' k t y y (cndn(?)33 w ' x s t
2s
P y y Sry' in /29/. GERSHEVITCH bersetzte in GMS 373 *backdoor, gewi
weil pyy keine regelrechte Prposition sein kann. Als solche h t t e sie unbedingt die
Postposition s'r nach sich ziehen mssen. Andererseits erscheint p y y Sry' auch n i c h t
unter den GMS 1143 aufgezhlten regelrechten Komposita, wahrscheinlich, weil m a n
in diesem Fall eher *p'Sr(y) erwarten wrde. E s verbleiben d a n n die Mglichkeiten,
in pyy Sry' eine altertmliche, erstarrte idiomatische W e n d u n g mit dem Nomen im
L o k a t i v zu sehen oder vielleicht ein open compound (GERSHEVITCH : Compounds.
S. 148 f.), im Deutschen durch hintertrs nachbildbar. Da das Skandalse der Beschul-
digung des Briefes darin besteht, d a ein manichischer Kleriker eine Zeit lang, hinter
verschlossener Tr mit einer F r a u allein blieb, so wird der Sinn ties Ausdruckes eher
hinter der Tr als an der Hintertr sein. Damit lt sich das von GERSHEVITCH zitierte
osset. faesduar behind the door (GMS Add. 1143), in weiteren Sinne etwa np. pytaxt
Hauptstadt, wtl. (Ort) zu Fen des Thrones vergleichen.
30
Oder : zweimal habe ich hinter der Tr zur Ader gelassen, und nochoinmal
werde ich zur Ader lassen, vgl. T e x t a n m e r k u n g 14.
31
SW teilt mir mn(d) I mywn als Lesung von 1. GERSHEVITCH mit. Da am R a n d e
eines Risses im Papier sowohl eine rechte obere wie eine linke u n t e r e Ecke erhalten ist,
so betrachte ich mn(') als die wahrscheinlichere Lesung. [Vgl. aber S. 299,'Anm. 30.]
32
E r g n z t gem GMS 1396 und 1435 (Hinweis von SW). I n senem heutigen
Zustand, d. h. nach Abtrennung des F r a g m e n t s M 119 a von M 119, ist dieser Text nicht
in vollem Umfange lesbar. Mit Hilfe der Originalschriften h a t t e ich lediglich p r w s t y y
sicher erkennen knnen. Angesichts der unmotivierten Verwendung von pers. rwc f r
soglid. myS Tag m u aber auch gefragt werden, mit welcher Sicherheit die ergnzten
Textteilo vor dem Krieg tatschlich lesbar waren.
33
Alle Buchstaben versehrt, nur -nd u n d ein vorangehendes sicher. Cndn las auch
GERSHEVITCH : G M S 1 6 0 9 .
Acta Antiqua Academiae Scientiarum Uungaricae 28, 1980

310 W. SUNDERMANN
ckstyt)[ ] /81 c k s t y t 'skwnd s ' t r y j w8yy cynd(r)p'r 'styy w'nw

w ' t xw'n (.)[ ] /9/ (sn'nd ' r t y m ' x 'pryw) wryS'nt 'rty I c'f srSyt
s k w ' m n d y y w('t .) [ ] /10/ (k'm prm(?) r y j 'skwncykt 3 4 xwyst)rtyy
(w)'t ' r t y mys'nd (ms)[ ] / l l / (n' pSkcy)q ' t y (n') kry'cy(q) f r m ' n
35
(sn't kww ')s(kwncyqt x)wystrty (s'r)[(?) ] /12/ (my)'ndyy
y w s y t t kyy w'nd m ' x xwystrt(yy) sxww p'ckrt 'ws(t')ndk'm ' t y (m'x m)
[ ] /13/ (')ty (nyz)'r p(y)tytyy prw yryw 'wsyndndk(')m ' t y (xwy)-
t r t y yr'n (t)fsnwk k w n ' n d k ' m ' t y prw xw(r)[nyy 3 6 ] /14/ f s n ' y n d ) -
k ' m 3 7 m'y88 c'nw r y m n y t yzd m ' h n ' m y jw'(n)wtr ' k r t w S ' r t xwtyy xyS
n y ' m y y xyp8(')[wnd ] /15/ w m ' t y s p(')rty (m)ys'nd r y m n y t k m b y t
s w r y k t y y 'ndyk ps'k (m)'y8 xcyy pr xtwnyy ' t y p(c'w)' (y)w(x)t(yt)[
' r t y ] /16/ (fs)ytyt xnd p ' r t y y (w')xs(yg) b x t g y y mSyy ' x s y t skwn ' r t y cw
w ' n w w''nd skwn k t sr8(ng)t p r ' y w p ' z k y ' s n ' ( n d ) / 1 7 / [ ' r ] t y Styk 'w(x)z(')nd
8 y m y 8 w'x I p ' r y k n's xcyy ' r t y w'nw n(y)x " m s ' f r m ' y ' k r t y y kyy 3 8 pyrn-
m s t r s t ' n d /18/ [cndn( ?)] (wr' 39 ' t y f r t r y ) ' 'krtwS'rnd p ' r t y (pr(?) )yrw'n
m y h ( . , 4 0 mwj)'kyy sryy m ' h d ' d m w j ' k s t t y y cn(w(?) pr)[w 24 8yn] /19/
(8)'rty ' t y p(r)w n y w ' n z ' d g m w j ' k y y s(ryy) gbryhb ( 3 5 s ) n ( ' . . ) [ ]
() [ ]( +7 ) /20/ s(r)yy sxt(w)yy (mwj)'k ' t y k t w n ' (')-
f t ' 8 ' [ n ] (st')n(d . . . ) [ ] /21/ sryy rym(ny)y f(rw)x'(y) 'ywt(yy)
[ ]
Zweiter Brief ( bersetzung)
/11 [ ] und, Herr, diese [ ]

41
Oavryav [ ] /5/ sprach so : Das Gesetz ist mir gleich( ?). Und,
H e r r , Herrscher, [ ] der Tag mag sich gewendet haben(??), 4 2 ihre
eigene Seele (und) das Selbst [ ] so geruhe zu beachten, 4 3 wieviel
hliche Worte [ ] hlich sind, alles Verlangen ist darinnen. So wird
sein der Tisch(?) [ ] sie steigen empor und vermischen sich mit
34
Alle Buchstaben durch R i im Papier sehr stark versehrt. S t a t t p r m auch p n c
mglich.
35
GERSHEVITCH las 'skwncyk (GMS 1248), Hinweis von SW. 'skwncykt ist
a u s R a u m g r n d e n wahrscheinlicher als die Singularform.
36
SW weist mich darauf hin, da GERSHEVITCH offenbar noch xwrnyy (unver-
sehrt ?) lesen konnte (GMS 1168 Anm. 1).
37
GERSHEVITCH las berzeugend sn'ynd k ' m (GMS 756, Hinweis von SW).
Vor dem s ist aber noch ein nach links oben gefhrter Aufstrich erkennbar, so da ich
' s n ' y n d k ' m fr wahrscheinlicher halte.
38
Nicht ky (GMS S 1682).
39
So GERSHEVITCH : GMS 973, Hinweis von SW. Ich k a n n aber nur die oberen
R n d e r der Buchstaben u n d einen diakritischen P u n k t orkennen.
40
Auf h drfte ein b oder folgen, danach oder x. Myhr- erscheint nicht mglich.
[Vgl. S. 315.]
41
Zur Bedeutung vgl. Toxtanmerkung 30.
42
Ganz unsicher, vgl. A n m . 32.
43
Vgl. Textanmerkung 20.

P R O B L E M E DER I N T E R P R E T A T I O N MAN.-SOGDH. B R I E F E 311
u n s , u n d es w i r d einige44 J a h r e d a u e r n [ ] / 1 0 / n a c h ( ?) d e m W i l l e n
d e r g e g e n w r t i g e n O b e r e n wird es sein, u n d diese [
ungesetzlicher und unzeitiger Befehl wird zu den gegenwrtigen Oberen
emporkommen [ ] die ihre *Spione waren,45 w e r d e n fest a n die Stelle
unserer Oberen treten und unser[ ] u n d sie w e r d e n die H a r t - B e d r n g -
t e n in i h r e m Selbst z e r b r e c h e n u n d d e n O b e r e n groe N o t bereiten u n d im
Blfut ] w e r d e n sie [ i h r e H n d e ( ?)] w a s c h e n , s o w i e ? ? ?4e
Selbst zu dieser Stunde, Herr, [ ] /15/ w a r s t d u , d e n n dieser s c h m u t -
zigen, g e m e i n e n S y r e r B r a u c h u n d K u n s t ist so : in S p a l t u n g u n d S t r e i t s i n d
sie e r f a h r e n [ u n d ] g e b t , d e n n geistige S p a l t u n g h e r r s c h t hier. U n d w e n n sie
s a g e n : D i e O b e r h u p t e r s t e i g e n e i n w e n i g ( ? ) 47 e m p o r , u n d d a n n s t e i g e n s i e
wieder hinab,48 so ist dies W o r t g a n z v e r d e r b l i c h . U n d so g e r u h e a u f j e n e
sorgfltig a c h t z u g e b e n , 4 3 die f r h e r e m p o r g e s t i e g e n sind,49 wieviel G e w i n n u n d
F r d e r u n g sie b e w i r k t e n . 5 0 D e n n b e r d e n * g l a u b e n s s t a r k e n 1 5 L e h r e r Mih..
51
i s t d e r L e h r e r Mhdd g e k o m m e n , wie(?) ber [ ] die Electi und
b e r d e n L e h r e r Nwnzdag d e r Oavryav [ ] /20/ b e r [
44
1 c'f in / 9 / b e r t r g t wrtlich in ST I , S. 8, Z. 15 syr. h d k m ' wievielmehr
u n d m u daher andere Bodeutungsmglichkeiten dieser W e n d u n g nicht ausschlioon.
Ich verstehe I c'f im Sinne des gleichbedeutenden n p . yek-Sand somewhat ; a little
(STEINGASS : S. 1533), rap. w <5and some, a few (MacKenzie, S. 31).
45
Meine bersetzung folgt QMS 1679.
48
Der ganze Satzteil c ' n w r y m n y t yzd m ' h n ' m y j w ' n w t r 'krtwS'rt in /14/ ist m i r
unklar u n d scheint oline A n n a h m e einos Fehlers ( r y m n y t f r r y m n y y oder ' k r t w S ' r t
f r 'krtwS'rnt) nicht bersetzbar. W a s bedeutet j w ' n w t r Sicher eine negative Eigen-
s c h a f t , Person oder Gruppe. Darf a n den K o m p a r a t i v (-tr) eines von 'nwt Beistand
abgeleiteten A d j e k t i v s jw-anut < *duz-anut der blen Beistand hat (vgl. GMS 287)
gedacht werden?
47
P ' z k y ' , bisher unbelegt, b e t r a c h t e ich in V e r b i n d u n g m i t 'yw als Nominal-
ableitung von man.-soghd. 'yw p ' z y y , I p ' z y y . P ' z y y bedeutet HENNING zufolge in
diesem Z u s a m m e n h a n g Teilchen ; ein wenig, bichen (BB, S. 81 zu 674). Der Sinn d e r
B e h a u p t u n g wre d a n n , d a die Syrer n u r in geringer Zahl, f r kurze Zeit oder nach u n d
nach zu k o m m e n beabsichtigten. E i n e sehr einleuchtende Altornativlsung erwgt jedoch
SW : If 'yw p ' z k y ' is a compound its basis need not be p'zyy ,bit' b u t can equally well
he p'z ,face, f r o n t ' , whence a bahuvrihi *'yw-p'zyy with *-aka- as compound-suffix (cf.
N P - beside -()), a b s t r a c t 'yw-p'zky'. T h e m e a n i n g of N P yak-r'i .unani-
m i t y , sicerety, conciliation, friendship' would f i t very well here.
48
Zu diesem Satz vgl. S. 298 f.
49
GERSHEVITCH : GMS 1682, stellte hier die bersetzungen who wore before
und those who are foremost zur Diskussion, r u m t e abor auch die Mglichkeit ein,
d a im P r d i k a t eine Ableitung von sn- rise vorliegen knno (GMS 780). Die T a t s a c h e ,
d a von E m p o r - u n d Herabsteigen bereits in den Zeilen 1617 die Rede ist, spricht
aber d a f r , eine entsprechende B e d e u t u n g auch im vorliegenden Fall a n z u n e h m e n .
Besttigend weist SW darauf hin, d a neben s t ' n d die F o r m e n 'styy ist in /8/ u n d s t t y
stieg auf in /18/ auch f r die im T e x t stehende bersetzung sprechen.
50
Meine bersetzung folgt weitgehend einem Vorschlag von SW : Please n o t e
carefully [whether] (those) who came u p previously brought a b o u t profit a n d improve-
ment. Lediglieh folgt aus K e n n t n i s des Originaltextes, d a die von S W vorgeschlagene
E r g n z u n g c w t y nicht mglich ist, d a in diesem Falle a m T e x t r a n d der Aufstrich eines
t erhalten sein m t e .
51
Cnw Fehler f r c ' n w (vgl. BST I, ed. O. HANSEN, Zeile 116)?

der Lehrer Saxt u n d der Bischof K t w n ' , ber [ ] der schmut-

zige, *zerstrungsschtige 5 2 [ ]
ANHANG II
E s ist mir eine ehrenvolle P f l i c h t , in den Kreis der Gelehrten, die durch
ihre kritische Hilfe zur Verbesserung meiner Arbeit beigetragen haben u n d
denen ich tiefen D a n k schulde, auch Dr. Gershevitch (IG) einzuschlieen.
E r stellte mir auf das grozgigste seine eigenen Bemerkungen zu den Briefen
wie auch Aufzeichnungen der ihm mndlich mitgeteilten Interpretationsvor-
schlge Hennings (H) zur Verfgung (Schreiben vom 15.1.1981 u n d 9.6.1982)
und gestattete mir ihre Verwendung trotz der Bedenken, die Dr. Gershevitch
prinzipiell gegen jede von Henning deplored and dreaded unauthorized publi-
cation of great men's titibits hegt, d e n n Henning selbst war ja a staunch
defender of a scholar's right to be judged by posterity exclusively on t h e
s t r e n g t h of what he h a d personally approved for printing (vgl. I. Gershevitch :
W a l t e r Bruno Henning 1908 1967. I n : W. B. Henning Memorial Volume.
L o n d o n 1970. S. xxiii, und : Proceedings of the British Academy. London,
Vol. 65 (1979) [erschienen 1981], S. 717). Wenn ich diese Bemerkungen auch
nicht mehr in den T e x t meines Artikels einfgen konnte, so gab mir Prof.
H a r m a t t a doch die Mglichkeit, als einen zweiten Anhang zu meiner Arbeit
in k n a p p e r Form jene Bemerkungen mitzuteilen, die ich akzeptiere oder zur
Diskussion stellen mchte. Ich hoffe zugleich, auf diese Weise deutlicher wer-
den zu lassen, zu welchen Resultaten die Edition der soghdischen Briefe ge-
f h r t h t t e , wenn es Henning vergnnt gewesen wre, sie zu besorgen.
1,4, 11,18 : Auch H . betrachtete yrw'n als Titel, IG betont die Proble-
m a t i k aller bersetzungsversuche. Neben anderen Mglichkeiten erwgt er
'der gottselige' in t h e sense of 'the late, le feu', cf. M P 'nwslwb'n ? ?.
1.5 : Die Lesung k r t r t y y vorausgesetzt, leitet IG das Wort provisorisch
von *kartara- ab, das er mit chot. lcdara sword verbindet und cutting
bersetzt. Im bertragenen Sinn knnte das Wort vielleicht als Bezeichnung
gttlicher Mchte incorruptible, safe, erlst bedeuten, als Bezeichnung von
D m o n e n irredeemable. In M 133 sei q r t r vielleicht Adverb u n d knnte
irredeemably (sinful passions) bczcichnen.
1.6 : 'wrSp'r dortig, being there.
1,9 : M'xtysyy ein Wort, f r das H. eine eindrucksvolle E r k l r u n g
gefunden hat : arab. Maqdis a man f r o m Jerusalem, d. h. a man who had
52
F r w x ' y 'ywstyy in /21/ : in f r w x ' y sehe ich Infinitiv eines sonst frxw'y-, frxwst
geschriebenen Verbs zerstckeln (BB, S. 59), zur Metathese von -xw- zu -- s. GMS
417. F r w x ' y 'ywtyy wre dann wrtlich vom/zum Zerstckeln erregt, was m a n
vielleicht als zerstrungswtig oder spaltungsschtig deuten knnte. Sollte jedoch
Inversion eines eigentlich *'ywst f r w x ' y lautenden Kompositums vorliegen (vgl. G e r s h e -
v i t c h : Compounds, S. 147), so liee sieli einfach Zerstrung erregend bersetzen.

PROBLEME DER INTERPRETATION MAN.-SOOHD. B R I E F E 313
lived in Aramaic surroundings. X wre dann als Versuch zu betrachten, arab.

q wiederzugeben, whrend spter in manichisch geschriebenen np. Texten
hierfr das aus dem Trkischen vertraute Zeichen q verwendet wurde. Whrend
ich m'xtysyy als Eigennamen ansehe, betonte IG mir gegenber die Mglich-
keit, da auch ein Appellativum vorliegen knne.
1.11 : 'wstnyy kann auch als P a r t . Prs. aufgefat werden (GMS 892).
In diesem Fall wre zu bersetzen of the law there is one establisher.
1.12 : p ' r t y . . . cx'p88 : H bersetzte for, as regards (the fulfilment)
of) the law (in general) [they are weak] in (observing the) commandments.
1.14 : ]wyy f r n]wyy neu ?
1.15 16 : Syn'r'nc . . . swmS'rt : H bersetzte our nun saw t h a t their
nun cut [a garment] and sewed (it).
1,1617: t y m m ' x 8yn'r'[nc]. . . [p]rsm'skwn : gegen meine ber-
setzung wendet IG ein : Introduced by tym, one expects to see a report of yet
another misdemeanour of nuns. E r deutet daher prsm- massieren als tappen
und in freierer und idiomatischer Verwendung ertappen, erwischen. I n
xwsyc vermutet IG ein groes, tnernes (xuS aus *xurmS) Wassergef. I n
ihm habe die erwischte Nonne gebadet, und dies sei ihre Snde gewesen.
1,17 : Im letzten erkennbaren, von mir (xyp8) gelesenen Wort, dessen 8
wirklich sehr zweifelhaft ist, sieht IG vielmehr einen weiteren, zdduxt fol-
genden Namen.
1,19 : ]rwrtt bereits in GMS 492. Die Lesung geht auf H zurck.
1,19 : 8'rwk ' t y p ' s t ' nach IG eher blocks of wood und chips of wood
splitt off a block, mit Hinweis auf die Anhang I, Anm. 21 zitierte E r k l r u n g
Hennings.
1.23 : " t r . . . 'wst'ndskwn eher lay (aufschichten) a fire wie englisch
to lay a fire und in Anbetracht der berlegung, da gewi eher Anla bestand,
ein Feuer zu entfachen als zu lschen (IG).
1.24 25 : ' r t y t y m . . . pww sf'r : Henning bersetzte And as regards
the third commandment, it is loose and shameless (with them). IG erklrt,
da H offenbar w'xs als Sache verstand und mit vorangehendem dyneyh-
ryft cxs'pSyy verband, (in) the m a t t e r of the third commandment.
I, 26 : IG wendet gegen meine bersetzung von c's'r p8yk r'f als ein
Leiden unter dem Fu ein, da soghd. Fu pd u n d nicht sei. H habe
daher aw. pacta- place, spot verglichen und bersetzt a lower-part illness,
wahrscheinlich im Sinne von eine Unterleibskrankheit. Es sei dann ferner
naheliegend, psyy 8ry' als backdoor zu bersetzen, wie GMS geschehen
(vgl. Anh. I, Anm. 29), und dies als einen Euphemismus f r anus zu erklren.
Schlielich sei s t a t t psyy 8ry' auch psyyS ry' als Lesemglichkeit erwogen
worden, was mir nicht ganz unmglich erscheint. Allerdings ist das 8 mit dem
zusammengeschrieben. F r c's'r downwards (GMS 459) mte mit nicht
korrekter Verwendung im Sinne von unten, unterhalb gerechnet werden, was

aber nicht unmglich ist, vgl. christl.-soghd. cyns' innerhalb s t a t t hinein

in ST I I , 3 /9/.
1.26 : z'k'nc k r ' y ' n c : H verglich kr'wr'n und kwr'ynk, suggesting t h a t
t h e z'k'nc may have been a girl from Kroraina.
1.27 29: ' t y SynS'rt . . . " s m k ' m : H bersetzte they [die Electi]
paid attention ( = heard) t h a t ('ty) she tould four persons (viz. Yazd Arya-
m a n etc., all of them, as Henning stessed, bearing Persian names) t h a t twice
I took blood. . .. Erklrend f h r t I G dazu ferner aus, da die vier Personen
nicht notwendigerweise verhrende Electi waren sondern vielleicht lediglich
B e k a n n t e der Heilpraktikerin, denen sie von ihren Taten berichtete, ohne zu
ahnen, da sie belauscht wurde.
I,34 : Statt meines ](cw)xt las IG ']wzxt (ebensowohl mglich), s t a t t
](.)[ : ](w)[.
1,37: Statt t y ( . ) [ : t y [ m (naheliegend, da ein Interpunktionszeichen
vorausgeht), statt ] ( . ) ' d : ]'r (ich sehe keinen r-Punkt).
I I , 3 : vor m ' x : ] ( . s t ' 'ty) (IG konnte st' ' t y lesen).
11.4 : IG konnte seinerzeit ]mt 'skw'm ' t y rwsn m w j ' k lesen, d. h. ich
bin u n d der Lehrer Rsn (kaum der lichte Lehrer, weil in diesem Fall wohl
ein soghd. Wort verwendet worden wre).
11.5 : Wenn m n d I mywn zu lesen ist, dann ergeben sich mit Henning
die mglichen bersetzungen 1. it is altogether in accord with the religious
law (mnd = mimetic affixation of M P -mand to the Sogdian open compound
nwm-pSk'), 2. the (observance of the) law is ceasing (mnd = privative pre-
fix). E s knnte aber auch, wenn I mywn mit SW like bedeutet, mand-wma-
yn unlike, different oder incompatible sein.
11.6 : IG konnte lesen : . . . t w' rwc prwstyy wyy my'ndyy xypS
r w ' n yryw '[. E r erklrt rwc prwstyy als ein adjektivisches offenes Komposi-
t u m in der Bedeutung one for whom destiny has turned (Lehnbersetzung
eines pers. *rz-bargaSte, vgl. np. roz bargard-jbargaSt, Wolff, S. 447 unter
N r . 43) oder einfach unfortunate, desperate, abandoned (Hinweis auf Np.
bargaMe-rzgr, bei Johnson, Steingass, Haim). Das vorangehende w' ist IG
zufolge Elativprfix. F r die Zeilen 5 7 schlgt IG als bersetzung vor :
And Lord [in order not only to save ourselves, but also to prevent t h a t ]
their own soul's self become so-very-much ill-fated [as a result of their Laxity,]
it is necessary to order t h a t etc..
11.7 u. 8 : Falls m a n s t a t t ckstyt cxstyt lesen darf (in beiden Fllen
finden sich P u n k t e ber dem s, die man vielleicht auf das vorangehende
beziehen kann), so liegt vielleicht ein P a r t . P r t . caxSt- received vor, das
neben cayt- existierte (IG, vgl. GMS 551 (c)).
11.8 : F r xw'n schlgt IG Ableitung von aw. *hu-na- gutes Atmen
gewhrend > wohlbehaglich vor u n d verweist auf Bartholomaes berset-
zung von aw. x-dra- als (gutes Atmen sva) Wohlbehagen.

P R O B L E M E D E R I N T E R P R E T A T I O N MAN.-SOOHD. B R I E F E 315
11.9 : S t a t t es wird einige J a h r e dauern bersetzte H there is staying

( = t h e y stay) for several years.
11.10 : S t a t t prra las I G pnc u n d bersetzte five desires shall be t o t h e
present teachers, auch das vorangehende k ' m v e r s t a n d H als desire.
11,12 : S t a t t kww las IG ky (wenn richtig, so f r kyy), meine d u r c h kww
angeregte E r g n z u n g der letzten Schriftspuren dieser Zeile als s'r h a t I G folg-
lich nicht.
11.12 : Neben der von IG mitgeteilten, auf H zurckgehenden berset-
zung im Text erwog H die Mglichkeit, da in diesem Fall auch Hrer, also
Angehrige des Laienstandes der Kirche, gemeint sein knnten. Ich halte das
u n t e r der Voraussetzung f r denkbar, d a der Briefschreiber hier die Hrer
der feindlichen P a r t e i absichtlich herabsetzt und v e r s p o t t e t , sie also sozusagen
Horcher s t a t t Hrer nennt.
11.13 : n y z ' r . . . ' w s y n d n d k ' m : H bersetzte they shall break d o w n on
t h e body those who are (already) broken by Elend. On the body e r k l r t I G
als physically.
11.13 14: H e r g n z t e : ' t y prw xwrn[yy xypS yryw] s n ' y n d k ' m
and t h e y shall wash themselves in blood. Weder H s noch meine T e x t e r g n -
zung (xypS Sst') wrden das Zeilenende normalerweise ganz fllen, so d a mit
mindestens einem weiteren W o r t zu rechnen ist.
11.14 : Wie I G mitteilt, betrachtete H r y m n y t als Fehler f r r y m n y ,
so wie ich dies in A n h a n g I, Anm. 46 auch v e r m u t e t h a b e . E r sah in j w ' n w t r
einen Titel u n b e k a n n t e r Bedeutung. Yzd m ' h n ' m y e r k l r t IG berzeugend als
Bahuvrihi-Kompositum . . . whose-name-is Yazdmh.
11,14 f. : H bersetzte und ergnzte At t h a t t i m e (viz. the time of Yazd-
mh) you yourself, Lord, were (bishop?).
11,16 : S t a t t w'x(yg) b x t g y y las IG w'x ' y b x t g y y , was H the spirit
of schism bersetzte. I n seinem heutigen Zustand g e s t a t t e t der Text eine solche
Lesung aber nicht mehr.
11,17: S t a t t [ ' r ] t y a m Zeilenanfang ergnzte I G berzeugender f ' ] t y ,
zumal andernfalls ein r - P u n k t erhalten sein sollte. Ebenso d r f t e 11,15 am
E n d e mit IG ' t y zu lesen sein.
11,18 A n f a n g : S t a t t [cndn] ergnzte IG c(w) what. In seinem heutigen
Z u s t a n d g e s t a t t e t das Manuskript beide Mglichkeiten, wenn m a n breite
Schreibung des a n n i m m t .
11,18 21 : I n diesen letzten erhaltenen Zeilen des Briefes, die heute aus
vier voneinander getrennten Stcken bestehen, k o n n t e n H und IG einst weitaus
m e h r lesen als dies heute mglich ist.
11,18: S t a t t m y h ( . . m w j ) ' k y y : m y h r f ' n m w ] ( j ) ' k y y . Damit wre der N a m e
des dort a n g e f h r t e n Lehrers der eigenen Partei als Mihrn wiederherstellbar.
I n seinem heutigen Z u s t a n d gestatten allerdings weder das Photo, mit dem ich
arbeite, noch der Originaltext eine solche E r g n z u n g (vgl. Anhang I, Anm. 40).

316 W. SUNDERMANN: P R O B L E M E D E R I N T E R P R E T A T I O N MAN.-SOGHD. B R I E F E
1 1 , 1 8 - 1 9 : S t a t t cn(w?) (pr)[w 24 8 y n ] / (8)'rty . . . : cndn f [ r t r y '

' k r t w ] / 8'rt wieviel Besserung h a t er bewirkt? Diese einleuchtende Lesung
d r f t e grundstzlich den Vorzug verdienen. Insbesondere c n d n ist mglich,
w e n n m a n a n n i m m t , d a das finale n mit d e m linken E n d e des c-Bogens zu-
s a m m e n f i e l . F r f [ r t r y ' ' k r t w ] reicht das E n d e von 18 k a u m hin. IGs Ergn-
z u n g scheint mir aber doch mglich, wenn m a n sie etwas v e r k r z t ; etwa :
( p r ) [ w ' k ' k r t w ] / (8)'rt er h a t verleumdet.
11.19 : Nach g b r y h b konnte I G zweifelsfrei ' f t ' 8 ' n lesen u n d danach
[ s ] t t y (Prt.) s t a t t meines sn' (Imperf.) ergnzen. Gavryav w a r also ein Bischof !
Die folgenden Worte, von denen ich gar nichts zu lesen vermochte, waren offen-
b a r ebenfalls weniger zerstrt als heute, doch auch seinerzeit n u r zweifelhaft
d e u t b a r . IG hat mir g e s t a t t e t , seinen folgenden, sehr zweifelhaften R e k o n s t r u k -
tionsversuch dessen, w a s er glaubt, einst m i t H gelesen zu h a b e n , mitzuteilen :
[ s ] t t y c n d n n' ? . . . (p/y)t[w]8[']rt [prw N N m w j ' k y y ] . E r g n z e n d teilt er mir
m i t , d a n' immer ganz u n k l a r war und d a das Verb von H als he has given
a w a y as a present bersetzt, also als ytwS'rt verstanden w u r d e .
11.20 : Nach s t ' n d : cnd[n wie viel, heute nicht mehr erkennbar.
11.21 : S t a t t f r w x ' y : f r w x ' n , dazu die bersetzung the d i r t y E a r r u x n
t h e a g i t a t o r (or : ' t h e disturber of the h a p p y ones'). Eine einleuchtende u n d
d a h e r wohl richtige L e s u n g . Ich k a n n aber n u r wiederholen, d a das, was ich
zu s e h e n vermag, als f(rw)x'(y) erscheint. D e r letzte Buchstabe h n e l t am mei-
sten einem in der Mitte Versehrten y.
A n h a n g I, Anm. 44 : I G verweist mich noch auf paschtu yau several
m i t B e z u g n a h m e auf GMS S. 251, zu 1586. Leider h a t t e ich v e r s u m t , in meiner
A n m e r k u n g auch 1586 selbst zu zitieren.
A n f a n g 1982 erhielt ich d a n k freundlicher Vermittlung v o n D r . N. Sims-
W i l l i a m s durch Herrn D r . G. Gropp aus der P h o t o s a m m l u n g des Seminars f r
Geschichte und K u l t u r des Vorderen Orient der Universitt H a m b u r g ein aus-
gezeichnetes Photo, d a s den H a u p t t e i l des zweiten Briefes (Zeilen 1 19) in
seinem weniger Versehrten Vorkriegszustand zeigt. Beiden H e r r e n danke ich
an dieser Stelle f r ihre Hilfe, die um so grer ist, als sie eine R e i h e problema-
tischer Lesungen definitiv entscheidet, u n d zwar :
11.5 : mn(d) I m y w n (so IG),
11.6 : ](ty w' rw)c pr(w)styy wyy m(ys)'ndyy xyp8 r w ' n gryw (')[
(so I G ) ,
11.7 : ]ywt w ' n w (so IG),
11.9 (erste Stelle) : ('ty) s t a t t ('rty),
11.10 : s t a t t p r m eher pne (so IG),
11,16 : w'xs y b x t g y y (so IG),
11,19 nach g b r y h b : [ ](-)^ (c)ndn 'n(w.)[, nicht s n ' !
Acta Antiqua Academiae Scicntiarum //angaria*r 28, 1980

323 M. M A R T H
PTOLEMAIC ELEMENTS AND GEOGRAPHICAL

ACTUALITY IN AL-HUWRIZM'S DESCRIPTION
OF CENTRAL ASIA
INTRODUCTION
The geographical knowledge a n d geographical science of the Arabs, as

shown by J . H . K r a m e r s , 1 have developed in several stages until they a t t a i n e d
t h e high level known to us. This development h a d several factors. One of its
earliest components is the practical knowledge of n o m a d i c people wandering
in t h e desert a b o u t their own c o u n t r y , the Arab peninsula. 2 The p o e t r y of t h e
ghiliyya has preserved a few practical directions a n d descriptions showing
how t o get f r o m one place to another. L a t e r on, these simple items of knowledge
were summarized b y the lexicographers on a higher level. At this t i m e geog-
r a p h y was not a s e p a r a t e science, b u t it represented already part of scientific
knowledge. These items of practical knowledge were included in their alpha-
betically compiled works. One of the earliest p r o d u c t s of this t r e n d was
al-Asma'i's work d a t e d from t h e 8th century t h a t , beside the Arabic vocab-
ulary, also contained the geographical names of t h e A r a b peninsula. Of t h e
works known to us most useful are al-Bakri's work entitled Geographical
dictionary a n d Y q t ' s K i t b m u ' g a m al-buldn.Beside the old geographical
names, b o t h contain also t h e t o p o n y m s of the newly conquered territories.
I t was an i m p o r t a n t achievement, when t h e A r a b s got acquainted w i t h
theoretical geography. The f i r s t signs of this can also be gathered a l r e a d y
f r o m the K o r a n . T h a t knowledge, however, can be ascribed to foreign influence. 3
Much more significant t h a n these modest beginnings was the f a c t t h a t t h e
Arabs adopted t h e achievements of the theoretical geography of the Greeks,
reaching them in three routes. I t came via I n d i a in t h e form of t h e so called
Sind-Hind literature, and through I r a n in the form of t h e ZIg literature. T h e
known world m a p s of al-Huwrizin and S u h r b c a m e into existence as a
result of direct Graeco-Arab relations (although this word direct does not
m e a n t h a t we would forget a b o u t the intermediary a c t i v i t y of the Syrians).
T h e m a p of a l - R u w r i z m l is very i m p o r t a n t because it came into existence
early, in the f i r s t half of the 8th century. The cartographical a c t i v i t y of t h e
Arabs cannot be understood w i t h o u t the knowledge of the work of P t o l e m y
1
J . H . K r a m e r s : D j u g h r f i y a . E I S u p p l e m e n t u m b a n d . 62.
2
F. B u h l : Das Leben Muhaineds. Leipzig 1930. 61.
3
E I . Leiden 1965. I I . 5 6 7 5 - 5 5 7 8 .

318 M. M A R T H
e n t i t l e d ] vcprjyr/ai. In f a c t , t h e activity of the Arabs was closely

connected with this work. I n a short time even three Arabic versions of it were
p r e p a r e d . The work contains the detailed description of t h e world, thus i t
could serve as the f r a m e w o r k of t h e activities of the Arab cartographers, in
w h i c h t h e changes t a k e n place since P t o l e m y had only to be inserted as well
as t h o s e items of knowledge t h a t were acquired by the Arabs on t h e basis of
i n f o r m a t i o n s more accurate t h a n those of P t o l e m y . Unfortunately, neither of
t h e t h r e e translations mentioned above has been preserved, t h u s we do not
k n o w whether they were simple translations or rewritings.
T h e third group of sources is represented by the geographical writers.
T h e geographical writers of the 9th and 10th centuries borrowed from several
sources. Their activity, on the one hand, was the continuation of the earlier
P e r s i a n geographical science, and on the other hand they also borrowed from
t h e F a d ' i l and H a d i t literatures, from t h e practical knowledge of the travel-
lers a n d post-officials, as well as from t h e d a t a of different itineraries. The
geographical writers can be divided into two groups. The f i r s t group is the
so called Iraqi school. Members of this are I h n Hurraddbeh, Qudma, al-
Y a ' q b , al-Mas'di, etc. The other is t h e so called Balhi school, its famous
representatives are al-Istahri, Ibn H a w q a l and Muqaddasi. 4
I n the work following here I shall examine the p a r t of al-Huwrizm's
work, t h e K i t b S r a t al-Ard, relating to Central Asia. I t m e a n t a help t h a t
several scholars already m a d e similar investigations earlier, a n d a t the same
t i m e , even the limits of m y works are determined by their research work.
I n t h e West I s t a r t e d m y investigations a t t h e territory elaborated by J .
E c k m a n n . 5 I finished it in the E a s t a t t h e same line with t h e elaboration of
H . Mzik, a t the border of the Ptolemaic Scythia extra Imaurn.6 The southern
b o r d e r was meant all along by India. The mountain range stretching between
I n d i a a n d Central Asia is slightly ascending towards the E a s t . Therefore, if
I w a n t e d to represent t h e cities of Central Asia in the West, it could not he
a v o i d e d to put the northernmost parts of I n d i a on the map. T h e examination
of t h e s e territories, however, was not m y aim, therefore the formations of the
s u r f a c e to be found there were t a k e n into consideration only if t h e y helped
m e t o determine the formations of the surface in the territory examined.
T h e primary aim of m y study is t h e reconstruction of al-Huwrizm's
lost m a p . Besides, a f t e r drawing the m a p , I tried to examine t h e points repre-
sented. I n the course of this work I compared al-Huwrizm's m a p with the
4
Cf. notes 1 a n d 3.
5
J . E c k m a n n : K e l e t e u r p a s N y u g a t z s i a a legrgibb arab t r k p e k e n (Eastern
E u r o p e a n d Western Asia on t h e Oldest Arabic Maps). Fldrajzi K z l e m n y e k 57 (1929)
N o . 67. P p . 9 1 - 1 0 5 .
6
. Mzik : Parageographische E l e m e n t e in den Berichten der arabischen Geo-
g r a p h e n ber Sdostasien. Beitrge zur historischen Geographie usw. F e s t s c h r i f t f r
E . O b e r h u m m e r . E d . H . 2 1929. 172 202.

AL-HUWRIZbll'S D E S C R I P T I O N OF CENTRAL ASIA 319
c o r r e s p o n d i n g p a r t s of P t o l e m y ' s m a p , a s w e l l a s w i t h o t h e r A r a b i c maps,
n o t a b l y w i t h t h a t of S u h r b t h a t s i m i l a r l y h a d t o b e r e c o n s t r u c t e d first.
F i n a l l y , if i t w a s p o s s i b l e , I t r i e d t o r e a s o n o u t t o w h a t t h e m a p p r e p a r e d
w i t h m o r e p r i m i t i v e m e a n s t h a n t h o s e of t o d a y a n d t h e r e f o r e f r e q u e n t l y
d e f o r m e d corresponds in reality according to our present knowledge.
I n t h e course of t h e w o r k f i r s t I t r a n s l a t e d a n d c o r r e c t e d a l - H u w a r i z m i ' s
t e x t frequently preserved in a distorted f o r m . Thereafter follows t h e identifica-
t i o n a n d o b j e c t i v e c o m m e n t i n g of t h e c i t i e s , m o u n t a i n s , n a m e s of c o u n t r i e s
a n d r i v e r s . T h e r e s u l t s of t h e r e s e a r c h c o u l d b e d r a w n o n l y a f t e r t h e s e . 7
AL-5UWRIZM ON CENTRAL ASIA
1. C i t i e s
Climate V
(398) U S r u s a n a t 9110' 3640'

1
(399) U o g a n d a 9230' 3710'
(400) B a n k a t h 2 9430' 3830'
(401) A h s l k a t h 3 9630' 3640'
(402) T r a b a n d 9630' 3640'
4
(403) I s b g b 9810' 3950'
5
(404) a l - T a r z , c i t y of t h e m e r c h a n t s 10030' 4024'
' My research started o u t f r o m t h e following works of f u n d a m e n t a l importance :

Das K i t b urat al-Ard des A b G a ' f a r M u h a m m a d ibn M s al- war izmi.
E d . H . v. M2ik. Leipzig 1926.
D a s K i t b 'Ag'ib al-Aklim as-Sab'a des Suhrb. E d . h . v . MZik. Leipzig 1930.
Claudii Ptolemaei Goographia. E d . C. F . A. N o b b e . Leipzig 1898.
Tabulae in Claudii Ptolemaei Geographiam. E d . C. M l l e r . Parisiis 1901.
1
I n al-ffuwrizm, in t h e figure denoting longitude, t h e d o t is missing from t h e
. I n Suhrb t h e figure occurs in its correct f o r m : . In t h e latitude, on the o t h e r
h a n d , Suhrb omitted a dot f r o m t h e j , while t h i s was correctly w r i t t e n by a l - f j u w r i z m
as j l .
2
I n a l - f j u w r i z m i t h e latitudinal d a t u m appears ncorrectly in t h e form J ^ J .
This ought t o be read as 3330'. A t t h e same place in Suhrb we f i n d J^J , this however,
h a s to bo interpreted as 3830'. On t h e basis of t h e situation a s compared with each
o t h e r and t h e system of t h e cities, Suhrb's d a t u m has to be accepted a s correct, accord-
ingly al-Uuwrizm h a s to be corrected.
3
Differing from al-ffuwarizmi, in S u h r b t h e longitudinal co-ordinate is 9430'.
1
The second member of t h e figure representing the longitudinal co-ordinate of
t h e city in a l - f j u w r i z m i is a n d in Suhrb r p Starting out f r o m t h e system formed by
t h e cities, surely we have to accept Suhrb's d a t u m as correct, accordingly a l - f j u w r i z m
lias to be corrected.
5
I n al-TJuwarizrni t h e longitude is indicated by J , and in S u h r b by ._). S u h r b ' s
d a t u m h a s to be corrected according to t h e other, apparently correct d a t u m .

320 M. M A R T H
(405) a l - T u b b a t 130 38
(406) A t r q r 7 14910' 3750'
(407) C i t y b e s i d e a m o u n t a i n a n d a s p r i n g 8 I4915' 39
Climate VI
(467) H w r i z m 9 9150' 4210'

(468) a l - H a z a r 1 0 93 45
(469) N a w k a t h 1 1 104 44
(470) C i t y 1 2 110 4445'
(471) C i t y 1 3 11245' 42
Climate VII
(497) S i n i s t n 14810' 4644'
Climate V I I I
(536) j U I c 131 5810'

(537) - L.15 14130' 5I20'
6
I n the case of t h e longitudinal d a t a in connection with t h e J both in al-Uuwr-
izm a n d in Suhrb t w o d o t s were left out f r o m the J . Al-uwrizmi's defective lati-
t u d i n a l d a t u m J m u s t be interpreted by all means as 38, because in Suhrb
we f i n d 40.
' I n the case of t h e latitudinal d a t a Suhrb writes o j . This has to be corrected
according to al-Ruwarizmi's correct jJ .
8
I n the case of this a n d also of t h e previous point in t h e longitudinal co-ordinates
f r o m t h e 3 the two dots were left out both in al-uwrizmi a n d in Suhrb.
9
Both in al-Uuwrizrn and in Suhrb the to be found in t h e latitudinal d a t a
h a s t be completed with one dot. The correct form of the figure is ^ .
10
Both in al-IJuwrizrr and Suhrb one dot is missing f r o m t h e figure ^ denoting
t h e longitude. The longitude of the city if definitely 93 a n d n o t 98, because t h e con-
ditions given at the description of the direction of t h e course of t h e Syr-Darya are per-
formed only by this interpretation of t h e figure. The correct form of the figure is
11
According to S u h r b the d a t u m of al-Ruwrizmi has to be corrected into A.
12
The correct form of the longitudinal d a t u m is J . Above t h e 3 the two dots are
missing with both al-!fuwrizm and Suhrb.
13
The ambiguous figure ( A. <_~J ) representing the longitudinal co-ordinate can be
established correctly so t h a t we take into consideration Suhrb's m a p . I n his work, here
several cities appear a f t e r t h e city given in point 471, a n d t h e longitudinal d a t u m of
each of t h e m is unambiguously smaller t h a n 150. Thus, the a m b i g u o u s figure has to bo
corrected into .
14
A name of uncertain reading.
15
A name of uncertain reading.

A L - H U W A R I Z M I ' S D E S C R I P T I O N OF CENTRAL ASIA 321
2. M o u n t a i n s
Climate V
( 8 1 7 818) Mountain, stretching between this mountain a n d Samarkand

88 36 106 3920'
d i r e c t i o n of p e a k s : N o r t h c o l o u r of m o u n t a i n : r e d
(819820) Mountain connecting with the previous one
106 3920' 12435' 3945'
d i r e c t i o n of p e a k s : N o r t h c o l o u r of m o u n t a i n : i n -
different blue
( 8 2 1 822) Mountain16
12650' 3950' 13930' 38
North i n d i f f e r e n t iron colour
( 8 2 3 824) M o u n t a i n s i m i l a r t o t h e f o r m e r one 1 7
137 72 13920' 38
d i r e c t i o n of p e a k s : W e s t colour : indifferent iron
colour
(825826) Mount Emodon
14030' 37 14920' 38
d i r e c t i o n of p e a k s : N o r t h colour : indifferent iron
colour
Climate VI
18
(865-866) Burg IJigarat
10155' 415' 10155' 4215'
d i r e c t i o n of p e a k s : E a s t colour : ( )
18
On the basis of comparison with Suhrb's data, the j h in al-IJuwrizm h a s
to be corrected into jYi , because t h e picture t h u s received is closer to that of P t o l e m y .
Similarly on t h e basis of comparison with Ptolemy, t h e latitude given in point 822 c a n n o t
be read as 33, but only 38. I m a o n : P t o l e m y j) vtpr/yrjat. VI. 131.
" I n Suhrb t h e co-ordinates are 140, 36144', 686', respectively. Of t h e
differing data of Suhrb and al-Ruwarizmi those of the latter h a v e proved to be correct
on the basis of comparison with Ptolemy. This comparison furnished the basis also for
t h e definition of t h e ambiguous figures in absence of the p u n c t u a t i o n .
18
Contrary to al-Ruwarizmi, in Ptolemy (VI. 13 2), t h e Al&ivo lvgyo is re-
presented only as one point. Taking this into consideration, we succeeded in selecting
t h e adequate figures from the values t h a t are ambiguous because of the absence of punc-
tuation. The longitude of point 866, as collated with that of p o i n t 866, had to bo inter-
preted as <Jli, and the latitude of point 866, as collated with t h a t of point 866, had to be
interpreted as <o ^ , because we could get the smallest extension this way. Unfortunately,
on account of the differences between the d a t a of Ptolemy and al-uwrizm (East-West
shifting of degrees, in a more detailed form in t h e concluding p a r t ) t h e comparison of t h e
two is a rather uncertain method, and still we had to rely upon this, because S u h r b
does not give any hint.

322 M. M A R T H
(867 868) Mountain

11750' 4530' 12220' 4150'
direction of p e a k s : South colour : concentrated
red
(869-870) Mount Asfatf18
12455' 4950' 13440' 43
d i r e c t i o n of p e a k s : North colour : yellow
( 8 7 1 872) Mount Qsiy20
141 4110' 147o45' 4;7o
d i r e c t i o n of p e a k s : North colour : red
Climate V I I
( 8 8 5 886) Mountain21
96 4555' 10050' 4950'
d i r e c t i o n of p e a k s : N o r t h colour : pale yellow
( 8 8 7 888) Mountain22
102 47 10720' 47
d i r e c t i o n of p e a k s : N o r t h colour : pink
After Climate VII

23
( 9 2 7 928) Mount Rmiq
88 57 9325' 55
d i r e c t i o n of p e a k s : N o r t h colour : pink
19
The minutes belonging to the western longitude can be interpreted accurately
according to Suhrb's t e x t . The ^ of the eastern latitude stands without punctuation
b o t h in al-fluwrizm a n d Suhrb, thus it represents number 8. Still on the basis of
comparison with Ptolemy (Askataghas. VI. 1314 ; 1413), correctly it has to be inter-
p r e t e d as eastern latitude 43 instead of 48, b o t h in al-Huwi izmi a n d Suhrb a dot
m u s t be p u t into the ^ . T h e correct form is >
20
I n the longitude E a s t in al-Huwrizm we find j , but this has to he corrected
according to Suhrb's J , because this figure is closer to Ptolemy (Kasia. VI. 15-2). Very
likely al-fjuwrizm's j is only clerical error. W i t h Suhrb t h e western longitude is
1415' a n d the eastern latitude is 475'.
21
Contrary to al-Ruwiirizmi, the co-ordinates of the western extreme point in
S u h r b are 965' a n d 4545'.
22
Suhrb's western co-ordinates are 1025' a n d 475', t h e eastern latitude with
h i m is 475'. The differences do not render the correction of al-fjuwrizm's text necessary.
I n his work, however, in t h e ease of t h e western latitude and t h e eastern longitude a n d
l a t i t u d e the dot m u s t be p u t on t h e letters j , because their correct form is j . (Thus y
23
I n Suhrb the two latitudes are 575' a n d 555', respectively. I t is important
t h a t in his work in the case of t h e d a t a of t h e eastern longitude there is a dot in t h e
t h u s it renders sure t h e value 9325'. This dot m u s t be put t h e r e also in the case of
al-Buwar2zmi-

A L - H U W R I Z b l l ' S D E S C R I P T I O N OF CENTRAL ASIA 323
(929-930) Mountain 2 4
8935' 61 9450' 61
direction of peaks : N o r t h colour : indifferent
sky-blue
(931 932) M o u n t Arsn
9540' 5510' 10050' 5920'
yellow
(933-934) Mountain 2 5
102 6020' 10820' 5930'
direction of peaks : N o r t h colour : i n d i f f e r e n t iron
colour
(935 936) Mountain, begins f r o m t h e fore-part of previous mountain 2 6
9720' 6730' 102 6030'
direction of peaks : N o r t h colour : pink
(937 938) Mount Asqsiy 2 7
10425' 56 o 40' 10840' 5220'
direction of peaks : N o r t h
(939 940) Aln-mountain colour : sky-blue
10745' 6225' 6220'
direction of peaks : N o r t h 11435'
(941 942) Tafr-mountain colour : yellow
10920' 5545' 11540' 5120'
direction of peaks : N o r t h colour : olive
(943 944) Sny-mountain 2 8
11415' 5820' 124 63
direction of peaks : N o r t h colour : concentrated
red
24
I n S u h r b t h e latitude is 616'.
25
I n S u h r b t h e western longitude is 1025'. T h e value of t h e eastern l o n g i t u d e i s
dubious. I t is w r i t t e n as j ^J, in t h e ^ t h e r e is no dot either in S u h r b or in a l - f j u w -
rizmi. T h e r e is n o P t o l e m a i c m a t e r i a l for comparison, t h u s I i n t e r p r e t e d the ^ as 8.
26
S u h r b ' s d a t a differ f r o m a l - f f u w a r i z m l ' s values a t t w o points. W i t h h i m t h e
western l a t i t u d e is 6030', a n d tho oastern longitude is 102 5'. I n t h e western l o n g i t u d e
t h e j is incorrect. T h e d o t h a s been o m i t t e d f r o m it both in S u h r b a n d a l - j u w r i z m .
Correctly w r i t t e n it is Aye .
27
I n a l - f f u w r i z m l the eastern longitude, erroneously, is p u n c t u a t e d so t h a t i t
o u g h t t o bo i n t e r p r e t e d as 103. On t h e basis of comparison with P t o l e m y (Aspisia ore
V I . 14-6), we h a v e t o accept S u h r b ' s 103 as correct. Thus, a l - f j u w r i z m ' s t e x t cor-
rectly is yd .
28
Tho eastern longitude in S u h r b , in c o n t r a s t t o a l - f j u w r i z m i , is 1245'. I n
a l - f f u w r i z m i t h e western longitude is p u n c t u a t e d . W i t h t h e h e l p of this t h e <u to be
f o u n d w i t h S u h r b can be corrected i n t o <u . I n t h e case of t h e eastern latitude, h o w e v e r ,
t h e d a t u m of a l - f f u w r i z m i can be corrected i n t o on t h e basis of S u h r b ' s t e x t . I n
f a c t , t h i s value can also be p r o v e d on t h e basis of P t o l e m y ' s t e x t (Sueba ore V I . 14-8).

324 M. M A R T H
(945 946) M o u n t a i n connecting w i t h this 2 9

118 6745' 125 63
direction of peaks : N o r t h colour : sky-blue
(947 948) A n z y m o u n t a i n range
12210' 5910' 12840' 5425'
direction of peaks : N o r t h colour : sky-blue
(949 950) M o u n t a i n 3 0
12645' 6130' 13520' 57
yellow
(951 952) Mount Awzaqiyn 3 1
14140' 5120' 14850' 58
direction of p e a k s : N o r t h colour : indifferent iron
colour
(953 954) Mount L a n t i t 3 2
14740' 65 15555' 6010'
3. Names of Countries
1585 C o u n t r y of Ss a n d T r a b a n d , its m i d d l e longitude : 98 l a t i t u d e : 42

1600 C o u n t r y of Osqniy, land of t h e T u r k s , its middle l o n g i t u d e : 114
l a t i t u d e : 5930'
1601 C o u n t r y of Osqthiy, country of t h e Toguzguz, its middle longitude :
143 l a t i t u d e : 5930'
4. R i v e r s
River Btis
I t s beginning is a spring a t l o n g i t u d e : 146 a n d l a t i t u d e : (2046)

3 9 4 0 ' . F r o m here it flows t o longitude : 153 a n d l a t i t u d e : 4010'... (2048)
A spring meets this river, its beginning is a t longitude: 14720' and
l a t i t u d e : 4520'. I t meets t h e river a t l o n g i t u d e : 149 a n d l a t i t u d e
4130'. (2051)
A n o t h e r spring also m e e t s river Btis, its beginning is between two
cities a n d a m o u n t a i n a t longitude : 14930' and l a t i t u d e : 3840'. (2052)
I t m e e t s the river a t l o n g i t u d e : 153 a n d l a t i t u d e : 40. (2053)
29
I n the case of t h e eastern latitude the p u n c t u a t i o n is omitted both in al-ffuw-
rizm a n d Suhrb. On t h e basis of the text it h a d t o be adjusted to t h e previous point,
viz. p o i n t 944. I n accordance with this the latitudinal figure has to be corrected into -
in b o t h A r a b authors.
30
I n Suhrb both latitudes are different. T h e western latitude is 7145' and the
eastern 575'.
31
I n Suhrb t h e eastern longitude is 15350'. The collation with P t o l e m y (Auxakia
ore. V I . 15-2) justifies t h e latitudinal d a t u m established here.
32
In Suhrb t h e western latitude is 6 5 0 5 ' .

A L - H U W R I Z b l l ' S D E S C R I P T I O N O F CENTRAL ASIA 325
River Balh
I t s beginning is a lake a t longitude : 100 and latitude : 48. (2196)

The dimension of t h e river is one p a r t . I t flows to the corner of t h e
m o u n t a i n a t l o n g i t u d e : 102 and l a t i t u d e : 47. (2197)
F r o m there i t t u r n s back to the point a t longitude : 9640' and lati-
t u d e : 4430', and t h e n from there it flows t o longitude : 9630' a n d (2198)
l a t i t u d e : 42. (2199)
T h e r e a f t e r it comes and crosses the city of IJwrizm, flowing out from
t h e r e it reaches t h e point longitude : 9130' and l a t i t u d e : 3940'. (2200)
I t flows along in t h e vicinity of the city of Balh and meets the B a t i h a a t
l o n g i t u d e : 88 and l a t i t u d e : 3910'. (2201)
F r o m River Balh flows a river, its beginning is a t longitude : 9130'
and l a t i t u d e : 3940'. (2206)
I t flows, a n d t h e n it crosses a long m o u n t a i n , it passes along t h e cities
of U s r s a n a and IJoganda and arrives a t a place a t longitude : 9230'
and 3440'. (2207)
A spring flows o u t from t h e m o u n t a i n a t longitude : 9940' and lati-
t u d e : 39. (2211)
It flows along in t h e vicinity of the city of merchants to the place a t
l o n g i t u d e : 100 and latitude : 4130'. (2212)
I t flows f a r t h e r and meets River Balh a t longitude : 9640' and lati-
t u d e : 4430'. (2213)
I n t o this river flows a river rising f r o m a spring, its beginning is at
l o n g i t u d e : 10140' and latitude : 3920'. (2214)
I t flows touching t h e city of merchants and meets t h e river rising from
a spring a t longitude : 100 and l a t i t u d e : 4130'. (2215)
A river rising f r o m mountains falls in (the water) of the first spring. I t s
beginning is a t longitude : 10430' a n d l a t i t u d e : 475'. (2216)
I t flows into t h e river originating f r o m a spring a t longitude : 9720'
and l a t i t u d e : 445'. (2217)
Long river
I t s beginning is a spring a t longitudine : 12930' and latitude :

46. (Suhrb mentions 415'). (2218)
I t flows to longitude : 118 and l a t i t u d e : 465'. (2219)
F r o m there it flows to longitude : 1075' and latitude 5030'. (2220)
Then f r o m t h e r e it makes for longitude : 10030' and latitude : 51. (2221)
I t passes by the city of t h e K h a z a r s a n d between the sea and t h e city
of f j w r i z m , its water flows into the l a k e of River Balh a t longitude :
9 0 5 ' a n d l a t i t u d e : 4130'. (2222)

326 M. M A R T H
I n t o this river flows a river from the m o u n t a i n of Arsn, its beginning

is a t longitude : 99 a n d latitude : 5730'. (With Suhrb longitude :
995'.) (2223)
I t flows into t h e big river a t longitude : 925' a n d l a t i t u d e : 455'.
( W i t h S u h r b l a t i t u d e : 445'.) (2224)
A river coming f r o m Mount Arsn also flows into t h e big river, its
beginning is a t longitude : 100 and l a t i t u d e : 595'. (Suhrb : longi-
t u d e : 10040'.) (2225)
I t flows into t h e river a t longitude : 9 4 5 ' a n d l a t i t u d e : 465'. (2226)
A river flows into t h e big river from M o u n t Asqsiy, its beginning is
a t l o n g i t u d e : 10440' a n d latitude : 565'. (2227)
I t flows into t h e river a t longitude: 10030' a n d latitude: 515'. (2228)
A river f r o m M o u n t Asqsiy similarly flows into the big river, its be-
ginning is a t longitude : 10630' and l a t i t u d e : 545'. (With Suhrb
l a t i t u d e : 5440'.) (2229)
I t flows into t h e river a t longitude : 10430' a n d l a t i t u d e : 51. (With
S u h r b l a t i t u d e : 515'.) (2230)
A n o t h e r river f r o m M o u n t Asqsiy also flows into t h e big river, its be-
ginning is a t longitude : 1 0 8 5 ' a n d l a t i t u d e : 5230'. (2231)
I t s m o u t h a t t h e big river a t longitude : 1 0 7 2 0 ' a n d l a t i t u d e : 5030'. (2232)
T h r e e rivers f r o m a m o u n t a i n also flow i n t o t h e big river.
T h e source of t h e f i r s t river is a t longitude : 103 a n d l a t i t u d e : 4830'. (2233)
I t flows into t h e big river a t longitude : 1 0 3 1 0 ' a n d latitude : 5130'. (2234)
T h e source of t h e second one is a t longitude : 10330' and l a t i t u d e :
4730'. (2235)
I t s m o u t h a t t h e big river is a t longitude : 10430' and latitude : 5040'. (236)
T h e source of t h e t h i r d one is a t longitude : 10445' a n d l a t i t u d e :
485'. (2237)
I t flows into t h e river a t longitude : 1065' a n d l a t i t u d e : 5020'. (2238)
T h r e e rivers f r o m M o u n t Taghr similarly flow into t h e big river, the
source of t h e f i r s t one is a t longitude : 112 a n d l a t i t u d e : 5810'. (2239)
(With Suhrb longitude : 1 1 2 5 ' a n d l a t i t u d e : 535').
I t s m o u t h into t h e river is a t longitude : 10920' a n d l a t i t u d e : 505'. (2240)
T h e source of t h e second one is a t longitude : 1145' a n d l a t i t u d e :
525'. (2241)
I t flows into t h e river a t longitude : 1145' a n d l a t i t u d e : 4930'. (2242)
T h e source of t h e t h i r d one is a t l o n g i t u d e : 11530' a n d l a t i t u d e :
5130'. (With S u h r b l a t i t u d e : 5120\) (2243)
I t flows into t h e river a t longitude : 1 1 4 5 ' a n d l a t i t u d e : 4830'. (2244)
I n t o this big river flows a spring, its source is a t longitude : 11940'
a n d l a t i t u d e : 4430' (With Suhrb longitude : 1195'.) (2245)
I t s m o u t h a t t h e river is a t longitude: 1185' and l a t i t u d e : 465'.

AL-HUWRIZMI'S D E S C R I P T I O N OF C E N T R A L ASIA 327
(With S u h r b longitude : 11840'.) (2246)

A spring flows similarly i n t o t h e big river, its source is a t longitude :
12430' a n d l a t i t u d e : 495'. (With S u h r b l o n g i t u d e : 1245'.) (2247)
I t flows into t h e river a t longitude : 123 a n d l a t i t u d e : 4 6 2 0 ' . (With
S u h r b l a t i t u d e : 1235'.) (2248)
R i v e r Yhardis
I t s source is a spring, a t longitude : 14530' a n d l a t i t u d e : 47. (2249)

F r o m t h e r e it flows t o l o n g i t u d e : 15230' a n d l a t i t u d e : 5130'. (2250)
. . . I n t o this river flows a spring, its source is a t l o n g i t u d e : 14730'
a n d l a t i t u d e : 5230'. (2257)
I t flows into t h e river a t longitude : 1 5 2 4 0 ' a n d l a t i t u d e : 5130'. (2258)
. . . I n t o this river flows a river f r o m t h e m o u n t a i n , its source is a t
longitude : 9230' a n d l a t i t u d e : 61. (2389)
I t s m o u t h into t h e river is a t longitude : 8 1 3 0 ' a n d l a t i t u d e : 5830'. (2390)
. . . I n t o t h e former river flows a river f r o m M o u n t R m l q , its source is
a t l o n g i t u d e : 91 a n d l a t i t u d e : 565'. (2399)
I t flows into t h e river a t longitude : 88 a n d l a t i t u d e : 54. (2400)
A river f r o m M o u n t A r s n also flows i n t o t h e f o r m e r river, its source
is a t longitude : 97 a n d l a t i t u d e : 56. (2401)
I t flows into t h e f i r s t river a t longitude : 8720' a n d l a t i t u d e : 53. (3402)
II. EXPLANATION AND IDENTIFICATION OF THE

GEOGRAPHICAL POINTS
(398) lyj-l l o n g i t u d e : 9110' a n d l a t i t u d e : 3 6 4 0 ' . T h e co-ordinates

given b y a l - f f u w r i z m i accurately tally w i t h t h o s e of Suhrb. T h e n a m e
l - j d e n o t e s a p r o v i n c e . 1 I t s n a m e occurs also in t h e forms S u t r s a n a a n d
Sursana. 2 T h e n a m e of t h e c a p i t a l of t h e province was B u n g i k a t h . 3 l-JJM
c a n be identified w i t h t h e surroundings of p r e s e n t - d a y U r a - T b e . 4 A l - R u w -
rizml in his w o r k , however, denotes b y t h e n a m e l - j n o t t h e province,
b u t its capital. T h e r e f o r e b y t h e p o i n t given B u n g i k a t h m u s t be u n d e r s t o o d .
T h e one t i m e city can b e identified w i t h t h e r u i n s of t h e medieval S a h r i s t n . 5
(399) l o n g i t u d e : 9230' a n d l a t i t u d e : 3710'. S u h r b also
gives e x a c t l y these co-ordinates. T h e city was s i t u a t e d a t t h e border of t h e
1
I b n R a w q a l , 379 : lip. da. lp ^ - J j ^li'VI -U-JI l S ^VI li L-jj-il LI
VI, Itahr: . l i p L l L r J j jd5VI (-I -uJI l S jv-JLVI ^ ( i L - j ^ l Lli
2
HA. 354.
3
At each of t h e places quoted so far.
4
T u r k e s t a n , 16416G.
6
Op. cit. 1 6 6 1 6 7 . A . N . N e g m a t o v S . G . H m e l n i t s k i : 191 197.
Acta A ntiqua Academiae Scientiarum Hungaricae 28, 1980

28 M. M A R T H
province Farghna, b u t i t belonged still to Ss. 6 I t can be found beside t h e

P t o l e m a i c Alexandreia Eschate, b u t their names have nothing to do with
each other. 7 Ibn R u s t a , in contrast to al-JIuwrizm, places t h e city not in
Climate V, but in Climate IV together w i t h t h e cities of Osrsana and Earghna. 8
On its place we can f i n d today the city named ffogend. 9
(400) i V b l o n g i t u d e : 9410' a n d l a t i t u d e : 3830'. The city can
be f o u n d also in Suhrb, b u t at longitude : 9455'. I n the A r a b geographic
l i t e r a t u r e its name occurs also in t h e form d-TL> , b u t B a r t h o l d remarked
t h a t t h e form i S is correct. 1 0 I t occurs in this form also in al-Huwrizm
a n d Suhrb. Besides t h e variant mentioned above the form is also
k n o w n . 1 1 The city was situated in t h e one-time province of Ilq a t the bank
of R i v e r Angren, near its mouth into t h e Syr-Darya. I t s one-time existence
is witnessed by the present-day Sarkiya ruins. 1 2
(401) longitude : 9630' a n d latitude : 3640'. W i t h Suhrb t h e
c i t y can be found a t longitude : 9430' and latitude : 3630'. I t was the
c a p i t a l of province F a r g h n a , therefore in the Arabic itineraries it is fre-
q u e n t l y mentioned like t h i s : i n x . . 1 3 The city is described by the Arab
a u t h o r s differently in regard to its size. I t was situated in t h e territory of
t h e present-day villages Ahsi and Shand. 1 4 In contrast to al-Huwrizm and
S u h r b , Ibn R u s t a places it into Climate IV. 1 5
(402) xjjlk l o n g i t u d e : 9630' a n d l a t i t u d e : 3935'. W e find the same
d a t a also with Suhrb. The city is also mentioned by the names xaj\Jk> and
,ysliJI > ,16 The l a t t e r name originates from the circumstance t h a t for a
t i m e it was the capital of province Ss. This name can also be found with
t h e historians, al-Tabarl a n d al-Baldri. W i t h the exception of al-Huwrizm
a n d Suhrb, the other geographical writers mention CSCJ as capital of Ss. 17
6
H A . 355. The opposite is said b y I b n R u r r a d d b e h , p . 30.
' T h o m s o n : 126. J . M a r k w a r t : Die Sogdiana des Ptolemaios. Orientalia 15
(1946) 125.
8
I b n R u s t a , 97 : -uJI ^ LjJ jSLi u L I ^ ^ . s t j L . . .
J j 4J Iiji ,
9
Turkestan. 164.
10
E I . . 671.
" V. M i n o r s k y : A d d e n d a to t h e R u d i i d al-'lam. BSOAS 17 (1955) 262.
12
N o t this city w a s t h e capital of I l q , b u t T u n k a t . Cf. Ljtahr, 331.
13
I b n B u r r a d d b e h , 30. Qudma, 208. I s t a h r , 333. Y a ' q b , 294. The latter uses
t h e expression b liji JU , b u t he refers it t o a n o t h e r city : J i b dJUl LjJjj b l c j bj-Uj
uir.
14
Turkestan, 161 162. E I . I. 247.
15
I b n R u s t a , 97.
18
The XJJU occurs with the historians. Baldr, F u t h al-Buldn, Leiden, 1866.
421. T a b a r : Ta'rh a l - R u s u l wa al Mulk. I I . 1571. ^dldJI can be found in t h e
itineraries.
17
Muqaddasi, 276, a n d also with others.

A L - H U W R I Z b l l ' S D E S C R I P T I O N OF C E N T R A L ASIA 329
The reason of t h e difference is very likely t h a t t h e cartographers, in c o n t r a s t

to t h e o t h e r geographical writers, recorded t h e older conditions in their works.
I n B a r t h o l d we find the n a m e B i r k a t h instead of B i n k a t . 1 8 Y q t in his w o r k
accepts t h e co-ordinates known f r o m a l - f f u w r i z m i a n d Suhrb, a n d writes
t h a t t h e c i t y is also known b y t h e n a m e s T u r r a b a n d , T u r r and Otrr. On t h e
basis of this t e x t T r a b a n d can be identified with present-day Otrar. 1 9
(403) v W " 1 longitude: 9810' and l a t i t u d e : 3950'. I n S u h r b t h e
l a t i t u d e is 397'. I n Y a ' q b i t h e n a m e of t h e city can be found in t h e f o r m
-20 T h e n a m e equally denoted the city a n d t h e province s i t u a t e d
around i t , whose capital it was. 2 1 I t was situated in t h e valley of R i v e r Aris,
in t h e neighbourhood of imkent. I t can be identified with p r e s e n t - d a y
Sayram. 2 2 According to Istahri i t was a small town. 2 3
(404) jl>II l o n g i t u d e : 10030' a n d l a t i t u d e : 4024'. In S u h r b t h e
l a t i t u d e is 4025'. I t is mentioned b y the Chinese sources as a f a m o u s com-
mercial city. 2 4 I t is also mentioned b y al-Huwarizmi a n d Suhrb like t h i s :
jUdlSqju. I t was situated in t h e valley of River Talas, near p r e s e n t - d a y
Auliya-Ata. 2 5
(405) longitude: 130' a n d l a t i t u d e : 38. I n Suhrb t h e l a t i t u d e
is 4025'. I t is also known from t h e Arabic itineraries. C. A. Nallino identifies
it with Tibet. 2 6 Very likely, also here as already in several cases, a l - f j u w r i z m i
indicates b y t h e n a m e of the province its capital. A t this time Tibet h a d no
p e r m a n e n t capital, its place was always changing. T h e changing places of t h e
capital, however, can always be f o u n d in t h e region of present-day Lhassa. 2 7
The n a m e of L h a s s a can also be found in t h e HA. 2 8 I b n R u s t a places also
this city i n t o Climate IV. 29
(406) Ijll>1 longitude: 14910' and l a t i t u d e : 3750'. I n S u h r b t h e
longitude is 14950'. The city can be identified w i t h t h e city 'OTZOQOXQQO.
t o be f o u n d in P t o l e m y , exactly on t h e basis of t h e similarity of n a m e a n d
t h e site. According to Mzik, the transliteration of t h e n a m e of the city p o i n t s
18
E I . IV. 745.
19
Y q t : K i t b Mu gam al-Buldn. Leipzig ] 8661873. I I I . 524.
20
Y a ' q b , 295.
21
Muqaddas, 262.
22
H A . 357358. T u r k e s t a n . 175179.
23
I t a h r i , 333.
24
K . C z e g l d y : Die K a r t e der D o n a u l a n d s c h a f t g r u p p e nach al-Ruwrizmi. A c t a
Orient. H u n g . 1 (I960) 51.
25
H A . 119. 3 5 7 - 3 5 8 . E I . I V . 720.
28
C. A. N a l l i n o : Al-Ruwrizmi e il suo refacimento della Geografia di Tolomeo.
R a c e o l t a di Scritti editi inoditi. Vol. V. 514.
27
1 k n o w t h i s on the basis of G. U r a y ' s verbal communication.
28
H A . 93.
29
I b n R u s t a , 97.

330 M. MARTH
t o Syrian mediation. 3 0 Thomson regards this point in Ptolemy as a fabulous

element, 3 1 b u t Berthelot identifies it w i t h t h e Chinese Sung-Pan. 3 2
(407) jeJ J~r J* jl longitude : 14915' and latitude : 39. I t can also
be f o u n d in Suhrb, b u t t h e longitude given b y him is 14950'. Unfortunately,
al-Huwrizmi did n o t give any name, a n d t h e city is situated outside the
territories well-known b y t h e Arabic itineraries. We tried to identify t h e city
on t h e basis of its place occupied by i t on our map. I n t h e H A we can find
a city, 3 3 about which we can read : It is . . . close to the mountains. I n his
e x p l a n a t i o n V. Minorsky adds t h a t t h e c i t y was situated on the bank of a
small river, southeast of this. I t can be identified very likely with this city.
(467) f j j J - longitude : 9150' a n d latitude : 4210'. I n Suhrb the
longitude is 914'. J u s t as above, the c i t y bearing the name of the province
m e a n s t h e capital also here. On the basis of t h e itineraries it seems t h a t the
capital of Hwrizm was Kth. 3 4 This view is supported also b y Barthold. 3 5
T o d a y only the remainders of the city can be found, a n d t h e y are known
u n d e r t h e name Sh ' A b b a s Wall. Ibn R u s t a places this city into Climate V. 36
(468) j > J ' longitude : 93 a n d l a t i t u d e : 45. Suhrb's d a t a agree
w i t h those of al-Huwrizmi. The reports on t h e city of the K h a z a r s can also
be f o u n d in the Arabic itineraries. I h n H u r r a d d b e h calls the K h a z a r city
Khamlig. 3 7 I t was s i t u a t e d at the lower course of the Volga. V. Minorsky
proposes the reading Khamlih. 3 8
(469) L-TJ l o n g i t u d e : 104 and l a t i t u d e : 44. Suhrb gives 42 as
l a t i t u d e . I t is one of t h e important stations of the route leading towards
China, according to Q u d m a it was a big city. 39 The route led from here to
B a r s h n to be found a t t h e border of China, on the northern hank of the
Issik-Kl. 4 0 The n a m e of the city was pronounced Nawikat, al-Huwrizmi
a n d t h e Arab geographers transcribed t h e long with alif mamdda. 4 1
(470) o > l o n g i t u d e : 110 and l a t i t u d e : 4445'. The city is charac-
terized with these d a t a also in Suhrb. I t was situated between N a w i k a t and
30
H . M2ik : A f r i k a n a c h der a r a b i s c h e n B e a r b e i t u n g der ] v<prjyrj<n
des Claudius P t o l e m a e u s v o n M u h a m m a d ibn M s a l - f j v r i z r n i . W i e n 1917. I X .
31
Thomson: 310311.
32
A . B e r t h e l o t , C a r t e de l'Asie Centrale. L'Asie Centrale Ancienne. 224225.
33
H A . 85. 232.
34
I b n R a w q a l , 351. Maqdisi, 287. I s t a h r i , 300.
35
T u r k e s t a n . 144145, 150. E I . H . 974.
36
I b n R u s t a , 98.
37
I b n H u r r a d d b e h , 124. Against t h i s I s t a h r , 220 : Jl , I b n H a w q a l , 365 : Jjl ,
I b n R u s t a , 139 : J J L .
38
HA. 450-454.
39
Q u d m a , 206.
40
H A . 289.
41
H A . 290.

B a r s h n , t h u s it was situated on t h e n o r t h e r n route of t h e three routes leading

t o Barshn, n o r t h of t h e IssikKol. This is shown also b y the circumstance
t h a t on our m a p it can be found b y 245' more to t h e n o r t h t h a n B a r s h n
s i t u a t e d on its southern shore. R e a d i n g t h e itineraries we f i n d t h a t it is t h e
most i m p o r t a n t s t a t i o n of the n o r t h e r n r o u t e t h a t a p p e a r s in every e n u m e r a -
tion, it was t h e city of JIT. 4 2 Very l i k e l y , a t this point this c i t y was also repre-
sented b y al-fjuwrizml. 4 3
(471) L l o n g i t u d e : 11245' a n d l a t i t u d e : 42. T h e same d a t a are
given also by Suhrb, b u t u n f o r t u n a t e l y he does not m e n t i o n the n a m e of
t h e city either. E x a m i n i n g the itineraries of Ibn H u r r a d d b e h and Q u d m a ,
however, we can be sure t h a t this c i t y was OUip 44 a n d its n a m e has t o be
read correctly Barshn. 4 5 This city was situated a t t h e border of China, 4 0
a n d was t h e final point of the itineraries. From the t e r r i t o r y reviewed here
one could get here on three routes. O n e of the routes led f r o m N a w l k a t on
t h e n o r t h e r n b a n k along the Issik-Kl. The second r o u t e s t a r t i n g o u t f r o m
T u m k a t ran on t h e southern bank of t h e Issik-Kl, while t h e third one led
t h r o u g h t h e cities Qb, Os and U z k a n d from the south t o t h e north. 4 7 This
is t h e final station of each route, f r o m here the route leads already in foreign
territory. The city of Barshn was s i t u a t e d on the b a n k of t h e Issik-Kl in
t h e vicinity of present-day Przevalsk. 4 8 F r o m here one could get f u r t h e r t o
China, to t h e l a n d of the Toguzguz, t o Slnistn. 49
( 4 9 7 l o n g i t u d e : 14810' a n d l a t i t u d e : 4645'. Suhrb's d a t a
accurately tally with these. The city does not occur in P t o l e m y . I t c a n be
found in t h e land of the Toguzguz, n e a r the border of China. On t h e basis
of the different d a t a and of the c o m p a r i s o n of names it c a n be identified w i t h
inngkth of t h e H A , the winter c a p i t a l of the Toguzguz. 5 0 The city is men-
tioned b y other sources by other n a m e s . I t is known by t h e Chinese u n d e r
t h e n a m e K a o - c h ' a n g , and by t h e T u r k s under the n a m e Qoo. T o d a y beside
t h e place of t h e city the ruins of I d i q u t - s a h r i are situated.
(536) J ^ ^ j - l l l o n g i t u d e : 131 a n d l a t i t u d e : 5810'. Suhrb's d a t a
tally with those of al-Huwrizmi. I n w r i t i n g Suhrb gives t h e form j - U t o
t h e n a m e of t h e city, thus to some e x t e n t he differs f r o m al-Huwrizm. I n
42
Q u d m a , 206. I b n H u r r a d d b e h , 29.
43
Identification see H A . 291.
44
I b n H u r r a d d b e h , 29. Q u d m a , 206.
45
. 292, n o t e 3.
48
Q u d m a , 206.
47
. 289. I b n H u r r a d d b e h , 30. Q u d m a , 206.
48
Q u d m a , 262. J . d e G o e j e : D e M u u r v a n Gog en Magog, reviewed b y -
s c h e k in W Z K M 1889. H A . 293.
49
I b n H u r r a d d b e h . 30.
50
H A . 271. M. P . P e l l i o t : K a o - t c h ' a n g , Qoo, Houo-tcheou et Q a r - K h o d j a .
J A (1912) 6 7 9 - 6 0 3 . A . S t e i n : I n n e r m o s t A s i a . O x f o r d 1928. 6 6 6 - 6 0 9 .

332 M. MAHTH
absence of p u n c t u a t i o n , however, the n a m e is illegible w i t h both of them, it

is also difficult to identify it. On the m a p this city can be found in the region
of L^-I , country of t h e 7 . I n the itineraries we can read t h a t the people
n a m e d dLT was living there. 51 According to the HA, there was a city in t h e
K m k region, in agreement with our t e x t beyond Climate VII, its n a m e
was 4L or JJ. Comparing these d a t a we find t h a t the two cities
correspond to each other. 5 2
(537) longitude : 14130' a n d latitude : 5120'. Suhrb's d a t a
t a l l y with those of al-Huwrizm, b u t he gives the n a m e in the form ^.LL.
On account of t h e absence of p u n c t u a t i o n , the reading of the name is un-
certain. I t is difficult to identify, because it has been preserved in a very
distorted form. T h e difference is significant even between the forms of al-
Huwrizm and Suhrb. On the basis of our m a p t h a t m u c h is certain t h a t
t h e city was situated in the Toguzguz territory. Pangikth, t h e summer capital
was the second most important city a f t e r Oinngkth. According to the H A
these two cities were separated from each other by a mountain. 5 3 The repre-
sentation of al-Huwrizm meets this requirement. The Chinese name of t h e
t r i b e living in t h e vicinity was Pa-si-mi ( Basuzil). I t appears t h a t this
n a m e can also be restored from the city n a m e of al-Huwrizmi that differs
considerably from P a n g i k t h mentioned in t h e HA as well as from the Turkish
n a m e of the city, viz. from Bisbaliq. 54
II. 2. Mountains
(817 -818) longitude: 88 a n d l a t i t u d e : 36 longitude: 106

a n d latitude : 3920'. The direction of t h e peaks is North. On the basis of its
site, the mountain can be identified with t h e Ptolemaic and t h e
Eoupa orj. On t h e basis of H e r m a n n ' s identification, the present-day
H i n d u k u s h m o u n t a i n corresponds to t h e one-time Paropanisos, while t h e
K a i t u mountain a n d t h e Safid-kh correspond to the Sariphos. 1
(819 820) J ^ longitude: 106 a n d l a t i t u d e : 3920' - longitude:
12435' and l a t i t u d e : 3945'. The direction of the peaks is north. On t h e
basis of its site, t h e m o u n t a i n can be identified with the Ptolemaic Kavxaiov
OQO. According to H e r m a n n 2 and Berthelot 3 the mountain between the Amu-
D a r y a and the I n d u s can be regarded as t h e KavxoLov ooz. As compared
51
Ibn HurrudfidVjch, 31.
52
H A . 94. 469. 4 7 1 474.
53
HA. 309.
54
H A . 272.
1
RE 18/3. 1 7 7 8 - 1779, 11/1.2514-2515.
2
R E . 11. 60.
3
A. B e r t h e l o t : 265.

with the reality, the mountain to be found on the m a p of al-juwrizm j a

disproportionately elongated. The reason of this will be discussed in t h e
Conclusions.
(821 822) J^- longitude: 12750' and l a t i t u d e : 3950' l o n g i t u d e :
13930' a n d latitude : 38. The direction of the peaks is North. Suhrb gives
also the name of the mountain, viz. : ^ j U I . Taking into consideration this
name and the site of the mountain, we find t h a t it can be identified w i t h
"I/iaov OQO of Ptolemy. The name of t h e mountain also occurs in Syrian geo-
graphical literature. The spelling Y to be found there is the accurate
transliteration of the Greek form. 4 I t seems t h a t the Arabic variant of t h e n a m e
of the mountain originates from the misreading of the Syrian variant.
(823 - 8 2 4 ) J~r longitude : 137 a n d latitude : 72 longitude : 13920'
and latitude : 38. The direction of t h e peaks is West. Suhrb's co-ordinates
differ from the d a t a of al-Huwrizmi. Suhrb gives the name of the m o u n t a i n
also in this case, viz. : Lil. Thus, on the basis of its site and its name, also
this mountain can be identified with Ptolemy's "Ipaov OQO, exactly with its
long branch of N-S direction.
(825 826) 0>. J->- longitude: 14030' and l a t i t u d e : 37 l o n g i t u d e :
14920' a n d latitude : 38. The direction of the peaks is North. W i t h S u h r b
the name appears in the form o ^ . On the basis of t h e better form of t h e
name and on the basis of the site, the mountain can be identified with t h e
mountain 'Hpco OQY] occurring with Ptolemy. The last three mountains in
reality can be identified with the mountain system of the Himalayas. 5
( 8 6 5 - 8 6 6 ) jU ^ longitude: 10155' and l a t i t u d e : 415' longitude:
10155' a n d latitude : 4215'. The direction of the peaks is E a s t . I t does n o t
occur in Suhrb. I t can be identified with Ptolemy's Ai&tvo , although
according to the co-ordinates of al-Huwrizml it is situated much more t o
the West t h a n t h a t . Earlier it was regarded by the researchers as the equivalent
of Taskurgan, 8 b u t as a result of t h e more recent investigations, Treidler, 7
Markwart 8 and Thomson 9 represent t h e unifie view t h a t it can be found on
the Alai plateau, in the vicinity of present-day D a r a u t k u r g a n . The m o u n t a i n
was of great significance from the viewpoint of travels, because the r o u t e
to China led through this.
( 8 6 7 - 8 6 8 ) Jer longitude: 11750' and latitude : 4530' l o n g i t u d e :
12220' and latitude : 4150'. The direction of the peaks is South. Suhrb's
4
A. H j e l t : t u d e s sur l'Hoxameron de Jacques d'Edesse. Holsingfors 1892.
P p . X X X V I I I . , XL1.
5
R E . 10. 2261 2263. 5. 2502 2604. 9. 5214. 9. 2542.
6
T h o m s o n : 309.
7
R E . 24. 3 3 - 4 6 .
8
J . M a r k w a r t : Wehret und Araiig. Leiden 1938. 39 65.
8
T h o m s o n : 308.

334 M. MARTH
d a t a tally with those of al-fjuwrizmi, b u t neither of them gives the name.

On t h e basis of its site, it can be identified with Ptolemy's Kcoprjdwv ogeivr].
I n t h e form jT, t h e name also occurs in with I b n R u s t a . 1 0 I n reality t h e
Alai mountain system corresponds to this mountain. Kretschmer 1 1 a n d
Markwart 1 2 share this opinion.
( 8 6 9 - 8 7 0 ) li Ua-I J*. longitude : 12455' and latitude : 4950' - longi-
t u d e : 13540' and l a t i t u d e : 43. The direction of the peaks is North. Suhrb's
d a t a tally with those of al-fjuwrizml, there is a small difference only in t h e
f o r m of name given b y him : it appears in the spelling U IL U-. On the basis
of its site and n a m e t h e mountain can be identified with Ptolemy's ';
ogrj. The identification has been done b y Tomaschek. 1 3 According to him t h e
m o u n t a i n range consisting of the U s u n - t a u , t h e K o g a r t - t a u and the Karakol-
t a u corresponds to this mountain, indeed.
( 8 7 1 - 8 7 2 ) L-1 J ^ longitude: 141 a n d l a t i t u d e : 4110' longitude:
14745' and l a t i t u d e : 47. The direction of the peaks is North. I n Suhrb
t h e n a m e of the m o u n t a i n appears in t h e form L-L. The co-ordinates at two
points do not agree w i t h the co-ordinates of al-Ifuwrizm. The mountain, on
t h e basis of its n a m e a n d site, corresponds to t h e Ptolemaic mountain. The direct
predecessor of t h e Arabic name is very likely t h e Syrian spelling f .fi? Jo . 14
I t can be identified w i t h the present-day K a r a k o r u m mountain, otherwise
called Mus-tag mountain. 1 5
(885886) J->- longitude: 96 a n d l a t i t u d e : 4555' longitude:
10050' and l a t i t u d e : 4950'. The direction of the peaks is North. Suhrb
slightly differs f r o m t h e co-ordinates of al-IJuwrizm. On the basis of its site
t h e mountain can be identified with t h e " f e t a gr/ known from Ptolemy.
According to H e r m a n n today the Hissar m o u n t a i n corresponds to this. 16
( 8 8 7 - 8 8 8 ) J ^ longitude: 102 a n d l a t i t u d e : 47 longitude:
10720' and l a t i t u d e : 47. The direction of the peaks is North. There is a
smaller, insignificant difference between t h e co-ordinates of Suhrb and those
of al-Huwrizml. On t h e basis of its site t h e mountain can be identified with
t h e Eyia gt] t o be found also in P t o l e m y . In reality the mountainous
district situated between river Baskatis a n d river Demos corresponds to this
mountain. 1 7
10
H A . 361 363.
11
R E . 2/3. 788791.
12
M a r k w a r t : op. cit. 59 65.
13
R E . 2. 16141615. W . T o m a s c h e k : K r i t i k der ltesten Nachrichten ber
d e n skythischen N o r d e n . 11. 51.
14
Jacob of E d e s s a , 38.
15
H A . 199.
16
R E . 18. 2003.
17
R E . 263. 789.

AL-HUWRIZMI'S D E S C R I P T I O N OF CENTRAL ASIA 335
(927 928) -jj J-T- longitude: 88 and l a t i t u d e : 57 l o n g i t u d e :

9325' and latitude : 55. The direction of the peaks is N o r t h . Suhrb's d a t a
slightly differ from the d a t a of al-JJuwrizmi. On the basis of its site and n a m e
the mountain can be identified with t h e Ptolemaic 'Pvptfitx . According
to Eckman, 1 8 Berthelot 1 9 and Hermann 2 0 equally t h e Ural corresponds to
the IS-yj J--.
( 9 2 9 - 9 3 0 ) J ^ longitude: 8936' and l a t i t u d e : 61 l o n g i t u d e :
9450' and latitude : 61. The direction of the peaks is N o r t h . In the co-ordi-
nates there is only one smaller difference between Suhrb a n d al-Huwrizmi.
Neither of them mentions the n a m e of the mountain. On t h e basis of its site
it can be identified with the 'Ynegogeia of Ptolemy, and in reality t h e
Ural corresponds to it. 2 1
(931 932) U Ujl longitude: 9540' and l a t i t u d e : 5510' longi-
t u d e : 10050' and l a t i t u d e : 5920'. The direction of t h e peaks is N o r t h .
The d a t a of Suhrb and al-IJuwrizmi completely tally with each other. On t h e
basis of the site and the similarity of name, the mountain can be identified
with t h e Ptolemaic Ngoaaov OQO. This corresponds to t h e hilly c o u n t r y
joining t h e Ural from the south. I n this respect there is a complete identity of
views among t h e researchers. 22
( 9 3 3 - 9 3 4 ) J l o n g i t u d e : 102 a n d l a t i t u d e : 6020' l o n g i t u d e :
10820' and latitude : 5930'. The direction of the peaks is North. The moun-
tain has no n a m e either in al-IJuwrizmi or in Suhrb. T h e mountain c a n n o t
be found in Ptolemy, and it has no trace in the Syrian geographical literature
either. This is a new material on the m a p of al-fjuwrizmi. I t is s i t u a t e d
between the Arsn and Alan mountains. The former mountain represents
the hilly country connecting with t h e Ural. About t h e Alan mountains we
can give the same statement below. Accordingly, t h e mountain s i t u a t e d
between the two can also be regarded only as a hilly country of the U r a l .
(935936) Jer longitude : 9720' and latitude : 6730' longitude :
10260' and latitude : 6030'. T h e direction of the peaks is North. In Suhrb
t h e direction of t h e range of t h e mountain differs from t h a t represented b y
al-IJuwrizmi. This mountain also represents a new material just like t h e
former one. The sites of both are also similar, thus it can also be regarded
only as t h e southern hilly country of the Ural. Accordingly, just like t h e
'Ynegogeia OQT), described by Ptolemy as one mountain, was regarded by al-
IJuwrizmi as three nameless mountains, similarly the NQOOOOV OQO, con-
18
J . E c k m a n n : op. cit. 99.
19
B e r t h e l o t : 224.
20
R E . 2/1. 1884-1886.
21
J . E c k m a n n : op. cit. 99.
22
Op. c. 98. B e r t h e l o t : 224. R E 17. 1048.
Acta Antqiua Academiae Scientiarum Hungaricae 28, 1980

336 M. M A l l T H
sisting in Ptolemy of one range, was represented by al-juwrizm as a system

consisting of three mountains.
( 9 3 7 - 9 3 8 ) L- U-l J~>- longitude: 10425' and l a t i t u d e : 5640'
longitude : 10840' a n d l a t i t u d e : 5220'. T h e direction of t h e peaks is N o r t h .
Suhrb's d a t a tally w i t h those given above. On the basis of its name and
site t h e mountain can be identified with t h e 'Aon'iaia ogrj of Ptolemy. T h e
f o r m of the Arabic n a m e clearly shows t h a t the direct source was not Greek,
because the errors of t h e transliteration point to Syrian mediation. I n Jacob
of Edessa this name can be found in the f o r m ? j . 23 I n t h e variants of the
manuscript the spelling i'( also occurs. Thus, it is easily understood t h a t
the was read as ( = J). And the transliteration of t h e long is in Arabic
alif mamdda. 2 4 F r o m a m o n g all scholars who dealt w i t h t h e identification
of t h e mountain, H e r m a n n ' s solution is most likely. He sees its real equivalent
in t h e Karatau. 2 5
( 9 3 9 - 9 4 0 ) VI longitude : 10745' a n d latitude : 6225' - longitude :
11440' and latitude : 6220'. The direction of the peaks is North. Suhrb's
d a t a completely tally with the data of al-IJuwrizm. On the basis of t h e
i d e n t i t y of the name a n d the site it can be identified with the mountain
'AXav or] of Ptolemy. According to Berthelot 2 " and Tomaschek 2 7 this moun-
tain also corresponded to t h e hilly country connecting w i t h the southern p a r t
of t h e Ural.
(941942) \jjHe J-t- longitude: 10920' and l a t i t u d e : 5545' longi-
t u d e : 11540' and l a t i t u d e : 5120'. T h e direction of t h e peaks is North.
Suhrb's d a t a completely tally with the d a t a of al-Huwrizml. On the basis
of its name and site t h e mountain can be identified with t h e Tnovga of
P t o l e m y . According to H e r m a n n ' s identification in reality this corresponds to
t h e Alexander range. 2 8
( 9 4 3 - 9 4 4 ) L i j - J r v longitude: 11415' and latitude : 5820' longi-
t u d e : 124 and latitude : 63. The direction of the peaks is North. I n Suhrb
t h e n a m e is given erroneously as ^ . T h e d a t a of the longitude are different.
On t h e basis of the site a n d the similarity of the n a m e of the mountain, i t
can be identified with t h e Eovfia ogrj of Ptolemy. The Arabic spelling of the
n a m e of the mountain points to Syrian mediation. 2 9 The distortion of the n a m e
results from the erroneous punctuation of t h e correctly written letter forms.
23
Jacob of Edessa, 38.
24
C. B r o c k e l m a n n : Syrische G r a m m a t i k . Leipzig 1960. 6 7.
23
R E . 2/4. 2272 2273. R E . 2. 1 7 3 6 - 1 7 3 7 . Kritik, I I . 51.
26
B e r t h e l o t : 224.
2
' R E . 1. 1281.
28
R E . 2/4. 2272 2273.
23
About this in greater detail H . MZik : Afrika, II. V I V I I I .

AL-HUWARIZMI'S DESCRIPTION OF CENTRAL ASIA 337
( 9 4 5 - 9 4 6 ) J*- longitude: 118 and l a t i t u d e : 6745' longitude:

125 and l a t i t u d e : 63. The direction of the peaks is N o r t h . The mountain
is nameless both in Suhrb and in al-Huwrizmi. According to both of t h e m
i t connects with the former mountain. Very likely, it forms a unit with t h a t
one, and therefore it did not receive a separate name either. T h e two m o u n t a i n
ranges form one unit, just as we have already seen an example for this in t h e
case of the Ngoaaov got; and the 'Ynegdgeia . The complex consisting of
these two mountains corresponds to the ranges of the Ala-tau. Their common
n a m e correctly written is : L j . 3 0
( 9 4 7 - 9 4 8 ) h I j l J - - longitude : 12210' and latitude : 5910' - longi-
t u d e : 12840' and l a t i t u d e : 5425'. The direction of t h e peaks is N o r t h .
I n Suhrb the d a t a agree with the above ones, b u t the form of the name has
f u r t h e r deteriorated, viz.: LULL On the basis of its site a n d n a m e the mountain
corresponds to Ptolemy's 'Avgea . The Arabic transliteration shows t h a t
the Greek name was t a k e n over with Syrian mediation. 3 1 According to Toma-
schek the mountain can be identified with the western foothills of t h e
Tien-Shan. 3 2
(949-950) longitude: 12645' and l a t i t u d e : 6130' longitude:
13520' and l a t i t u d e : 57. The direction of the peaks is N o r t h . There is a
difference between Suhrb and al-Huwrizmi in the l a t i t u d i n a l data. This
mountain is also nameless in both of them. No Ptolemaic antecedents can be
traced and it is not known by the Syrians either. We can f i n d also this first
in al-Huwrizmi. I n all probability, it can be regarded as a more easterly
range of the 'Avgea . Thus, together with t h a t , it can be identified with p a r t
of the Tien-shan. This is also shown by the fact t h a t it is situated between
the 'Avgea and the "Avvia , and as we shall see, t h e "Avvia
are identified by the research with the eastern range of t h e Tien-shan.
(951 952) ojJjjl - longitude : 14140' and latitude : 5120' longi-
t u d e : 14850' and latitude : 58. The direction of the peaks is North. I n
Suhrb the name, a p a r t from the a t the end of the word, is not p u n c t u a t e d ,
while in al-Huwrizmi the dot is missing only from the j . On the basis of its
name and site the mountain can be identified with P t o l e m y ' s Avijaxa .
I t stretches from the K a r a k o r u m mountain to the eastern foothills of t h e
Tien-shan, and thus it corresponds to the present-day Altyn-tag.
( 9 5 3 - 9 5 4 ) longitude : 14740' and latitude : 65 longitude :
15555' and latitude : 6010'. The direction of the peaks is North. Suhrb's
co-ordinates and those to be found in al-Ruwrizmi slightly differ from each
30
J a c o b of Edessa, 37 : it s t a n d s in t h e f o n n j Oil?
31
H . M2ik : A f r i k a , loc. cit.
32
R E . 1. 2063.
33
R E . 1. 2 2 6 8 - 2 2 5 9 .

338 M. M A R T H
other. I n Suhrb we f i n d the n a m e of t h e mountain in t h e form JA Ii I.

On t h e basis of its site it can be identified with the "Avna of P t o l e m y .
Tomaschek (in R E ) identifies it with t h e eastern p a r t of t h e Tien-shan, men-
tioning especially t h e Qarasar and T u r f a n mountains.
I I . 3. Names of countries
(1585) -UJJ Ik j jiliJI ^ longitude : 98 and latitude : 42. The name can-
n o t be found in Ptolemy, it belongs to t h e new acquirements of al-IJuwrizm.
(1600) Lî-I l o n g i t u d e : 114 a n d l a t i t u d e : 5730'. I n Ptolemy t h e
t e r r i t o r y corresponds to t h e Uxv&ta xrj vx 'luov OQOV oi. The Syrian
v a r i a n t of the word is (- 2. OPT } The Arabic form c a m e into existence on
t h e basis of the l a t t e r one.
(1601) l&U l o n g i t u d e : 143 and l a t i t u d e : 5730'. The territory cor-
responds to P t o l e m y ' s Hxv&ia fj xx /IOV OQOV at. This word through
t h e Syrian transliteration (_ 2. onVi? can be traced b a c k to the Greek word
Uxv&la.
II. 4. Rivers
River t ^ . (2196 2201) Its beginning is a lake a t t h e longitude : 10030'

a n d latitude : 48. I t reaches the corner of a mountain a t the longitude : 102
a n d l a t i t u d e : 47. I t t u r n s back to t h e longitude: 9640' and l a t i t u d e :
4430'. I t flows to t h e longitude : 9630' and latitude : 42. Crossing the
city of Hwrizm it arrives a t longitude : 9130' a n d latitude : 3940'. I t
flows farther in the vicinity of Balh, until a t l o n g i t u d e : 88 and l a t i t u d e :
3910' i t flows into L a k e Aral. Looking a t the m a p we see t h a t of the two big
river systems the b r a n c h described now represents t h e main course of t h e
southern river system. T h e southern river system clearly corresponds to river
r
Qo to be found in P t o l e m y and to its tributaries, while t h e northern water
system is nothing else t h a n the Ptolemaic 'Iagxrj a n d its branches. Today
t h e TQo corresponds to t h e Amu-Darya, and the 'Iagxrj to the Syr-Darya.
Comparing t h e Oxus-course described by al-Huwrizm with the descrip-
tion to be found in P t o l e m y , we can f i n d several differences. One of the dif-
ferences is t h a t in al-Huwrizm the ^ ^ flows into L a k e Aral, while inPtolemy
it flows into the Caspian Sea. I n this respect, in accordance with reality, al-
Huwrizm has corrected Ptolemy. 1 The other difference is t h a t in Ptolemy the
1
L. Jacob of Edessa, 38.
1
On the mouth of t h e Oxus seo W . B a r t h o l d : Nachrichten ber don Aral-See
u n d den unteren Lauf des A m u - d a r j a von den ltesten Zeiten bis z u m X V I I . J a h r h u n d e r t .
Leipzig 1910.

direction of the course is South-North, while in al-IJuwrizm it is North-South,

or exactly the reverse of it. E x a m i n i n g the actual circumstances, we can see
t h a t t h e river-heads of the A m u - D a r y a (Oxus) and the I n d u s f l a n k the moun-
tainous district of the Hindukush, by its ancient name Kavxaiov OQO, as refer-
red to by us already while discussing t h e mountains. 2 E x a m i n i n g the m a p of
al-Huwrizm, we find t h a t also in him there is a river t h a t with the river-head
of t h e Indus flanks the Kavxaiov OQO. H e describes this river as follows.
(2211 2213) I t comes out f r o m t h e mountains a t l o n g i t u d e : 99 a n d
latitude : 39, it flows on along the City of merchants u p to the point with t h e
co-ordinates longitude : 100 and l a t i t u d e : 4130', then it flows into the
a t the longitude : 4640' and l a t i t u d e : 4430'. According to t h e conditions
mentioned above this river corresponds to the main course of the Ptolemaic
Oxus, the present-day Amu-Darya. This view is supported also by the fact t h a t
also another river flows into this river n o t much after its source.
(2214 2215) I t is the t r i b u t a r y of the former river. I t s source can be
found a t the longitude : 101 40' a n d l a t i t u d e : 3920'. I t flows beside the City
of merchants and a t the longitude : 100 and latitude : 41 30' it flows into t h e
Amu-Darya. This river can be identified with the Murgb to be found a t t h e
river-head of the Amu-Darya. The M u r g b in reality takes its source from t h o
P a m i r mountain, on our m a p from t h e Paropanisos mountain. This is t h e m o s t
i m p o r t a n t t r i b u t a r y of the upper course of the Amu-Darya, to such an e x t e n t
t h a t Istahri takes this as the main b r a n c h , and regards t h e main course as a
t r i b u t a r y of the Murgb. 3
After these it has to be clarified t h a t if these two rivers represent the m a i n
branch of the Amu-Darya, then to w h a t does the river correspond t h a t is shown
by al-Huwrizm as the Amu-Darya ? We get the answer, if we t a k e t h r e e
factors into consideration. The first is t h e North-South direction of the course
of t h e river. The second is t h a t the source of the river is a lake (2196). The t h i r d
is t h a t in al-Huwrizm one river is missing t h a t can be found in P t o l e m y .
This missing river is the /.vTt'prjro which can be identified with t h e Zarafsan. 4
The direction of the course of the river in question in general outlines corre-
sponds to t h a t of the Zarafsan, and on t h e other hand, in t h e Middle Ages it w a s
believed t h a t the Zarafsan took its source in a lake, in f a c t in Lake I s k a n d a r .
Thus, the lake described under point 2196 corresponds to the Iskandar-kl, 5
and t h e river in question to the Z a r a f s a n . Thus, on the one hand, a t this p o i n t
al-Huwrizm corrected Ptolemy, who represented the IJoXvTprjro so t h a t i t
flew into t h e Caspian Sea, and on the o t h e r hand, in the course of t h e correction
2
B e r t h e l o t : 265.
3
I f a h r i , 296.
4
W . ToMasChek : Centralasiatisehe Studien. I . Sogdiana. Wien 1877. 14 22.
5
. 73. 211. J . M a r k w a r t : W e h r o t u n d Arang. 84.

340 M. M A R T H
he also fell victim to an error, because he mixed up the Zarafsan with the Amu-
D a r y a . According to t h e sources, however, this was a traditional error, it cannot
b e p u t down to his account. 6 Besides, i t is also an evidence in favour of the view
j u s t expounded t h a t in Ptolemy the co-ordinates of the river-head of the Oxus
a r e longitude : 11930' and latitude : 39. Comparing these d a t a with the d a t a
of t h e river of al-Huwrizm taking its source in the Kaukasion oros, we can see
t h a t the latitudinal d a t a exactly tally with each other, while in the longitude
we find a difference of 20. This, however, can be a t t r i b u t e d to the general
displacement of t h e longitudinal degrees. This question will be discussed in
t h e Conclusions.
The A m u - D a r y a received its present n a m e from the city of mul situated
on its bank. 7 On t h e basis of a similar consideration, al-Huwrizm called it
jdj , because in f a c t , t h e river before flowing into L a k e Aral, flew along b y
t h e city of Balh. Sulirab mentions in his book t h a t its most widespread name
in Arabic scientific literature is . I n P a h l a v i literature the river was also
called Weh-rt. 8 The Greek variant of its n a m e is This n a m e was derived
f r o m the one-time n a m e of the river to he discussed below.
(2216 2217) A t r i b u t a r y from the right bank flows into the Amu-Darya.
I t s source can be found a t the longitude : 9720' and latitude : 4450', and its
m o u t h at the longitude : 10430' and latitude : 46. I t flows into the Amu-
D a r y a . This can be identified with present-day river W a h s , the biggest a n d
a t t h e same time most important t r i b u t a r y of the Amu-Darya. 9 At its middle
course it is called S u r h b and at its u p p e r course Kizil-su. The rQo is derived
f r o m the name WahS, and this name was applied later on not to this t r i b u t a r y ,
b u t to the main river, t h e Amu-Darya. 1 0
(2206 2207) I t is a tributary of t h e river j J j ^ ' o n t h e left bank. I t s co-
ordinates are longitude : 9130' and l a t i t u d e : 3940'. I n Ptolemy we find here
t h e river named Margos, this flows through t h e province of Margiana. The prov-
ince, a t the same time, received its n a m e from it. The present-day Murghab 1 1
corresponds to this river, b u t in reality, in contradiction to al-Huwrizm's m a p ,
i t does not reach t h e Amu-Darya. Al-Huwrizm does not give the name of this
river, therefore t h e river can be identified w i t h the Margos-Murgb only on t h e
basis of Ptolemy. I n this area geographical tradition takes note also of another
river, the name of t h a t is r Qo in Greek sources, t o d a y it can be found under
t h e n a m e Herirud-Tagand. In reality this river does not reach the Amu-Darya
6
J . M a r k w a r t : E r n a h r nach der Geographie des Ps. Moses Xoronac'i. Berlin
1901. 150. Wehrot. 77.
' B a r t h o l d : op. cit. 35. This n a m e occurs also in Suhrb.
8
M a r k w a r t : op. cit. 34.
9
. 209. I s t a h r 295. Ihn Hawqal, 398.
10
11
M a r k w a r t : loc. cit.

either. 1 2 I n Ptolemy we can also f i n d a river named rQtjo. This flows, however,
according to him first into the Aagyapvrj, and then t h e Aagya/ivrj falls i n t o
the Amu-Darya. The rQi;o and t h e Aagya/uvijg to be found in Ptolemy can b e
identified with t h e Qunduz and its tributary, 1 3 b u t these tributaries were n o t
represented by al-JJuwrizm.
(2218 2222) J J > . The river has its source a t t h e longitude : 12930'
and l a t i t u d e : 46, finally at the longitude : 905' and l a t i t u d e : 4130' it flows
into L a k e Aral. I t can be identified with Ptolemy's 'Iagxrj. The name is of
I r a n i a n origin, the original form Hasartu can be found in t h e HA. There h a v e
already been a t t e m p t s to explain t h e origin of the name. 1 5 I n Suhrb we find it,
beside the J i _ > b t o he found in al-Huwrizm, also under the name
widespread in Arabic scientific literature. Sometimes we f i n d it under the n a m e
(jiliflrfl, thus also in I h n Hawqal. T o d a y it is called Syr-Darya. This section of
the river t h a t stretches from point 2218 to point 2246, can be identified w i t h
present-day River Narin. This appears in the HA under the name Hatlam.16
Barthold also mentions this b r a n c h as the biggest among the sources of t h e
Syr-Darya. 1 7 In harmony with al-Huwrizm, Suhrb also regards this b r a n c h
as t h e main stream. As we shall see below, this view was n o t of general validity.
(2245 2246) The source of t h e river is longitude : 11940' and l a t i t u d e :
4430'. I t flows into the Long R i v e r a t the longitude : 1185' and l a t i t u d e :
465'. This river can he identified w i t h the present-day K a r a - D a r y a . I t occurs
in the H A under the name Uzgand. I n the Middle Ages the geographers, as well
as t h e m a j o r part of the population even today, regard this as the main course.
The H A is also of this opinion. 18
(2247 2248) Nameless river, it flows into the Syr-Darya. Its source is
the longitude : 12430' and l a t i t u d e : 495'. Its mouth is a t longitude : 123
and latitude : 4620'. On the basis of its site the river can be identified w i t h
the K a r a - s u .
Three rivers flow into the Syr-Darya from the T a p u r a mountain. All t h e
three are nameless both in al-Huwrizm and Ptolemy.
(2239 2240) longitude: 112 and l a t i t u d e : 5310'. I t flows into t h e
Syr-Darya at the longitude : 10920' and latitude : 505'. I t can be found also
12
B e r t h e l o t : 176. M a r k w a r t : op. cit. 4-4. I n P t o l e m y t h e Margos is described
in V I . 10-1.
13
14
H A . 116. T o m a s c h e k : K r i t i k , 260. I I . 50.
15
H A . 210.
16
H A . 73. 210.
17
T u r k e s t a n , 155.
18
T u r k e s t a n , 155. H A . 211. This view is rejected b y ibn H a w q a l , 392 : j ^liJI j + j
# , , - Li L-ljli- . . . IfcJl jyljo J^J jA

342 M. M A R T H
in Suhrb, b u t with him t h e data of the river-head are different. On the basis of
its site, it can be identified with the river Bossu. 19
(2241 2242) T h e river-head is a t t h e longitude: 1145' and l a t i t u d e :
4930'. I t can also be f o u n d in Suhrb, a n d t h e longitudinal a n d latitudinal d a t a
given there completely tally with those of al-Huwrizm. On t h e basis of its site
it can be identified w i t h t h e river P a r a k mentioned in the HA, 2 0 as well as with
the river Cirik according to Barthold. 2 1 T h e province of Ss is situated in the
valley of the river.
(2243 2244) T h e co-ordinates of t h e river-head are longitude : 11530'
and l a t i t u d e : 5130', a n d those of the m o u t h are longitude : 1145' and lati-
t u d e : 4830'. The river is also mentioned b y Barthold. 2 2 I t can be identified
with t h e present-day Angren. The province of Uq wa3 situated in the valley
of this river.
F r o m Mount L- li-l similarly three rivers flow into t h e Syr-Darya. All
t h e three are nameless b o t h in al-TJuwrizmi and Ptolemy.
(2231 2232) T h e co-ordinates of t h e source are longitude : 1085' and
l a t i t u d e : 5230', a n d those of the m o u t h longitude : 10720' and latitude :
5030'. The river can also be found in Suhrb, and the co-ordinates given by
him completely tally w i t h those given here. I t can be identified with the present-
d a y Aris. The course of this was described b y Barthold. 2 3
(2229-2230) T h e river-head is a t longitude : 10630' a n d 545', and the
m o u t h a t longitude : 10430' and latitude : 51. The latitudes of the river-head
a n d t h e mouth of t h e river are equally different in Suhrb. On the map of al-
Huwrizm, from t h e L-15-1 mountain, t h r e e rivers flow into the Syr-Darya.
I n reality from t h e mountain corresponding to the L-li-l similarly three
m a j o r rivers are heading for the main river. Of these the first one was the Aris
a n d t h e middle one was t h e Bugun. The river in question can be identified with
this.
(2227 2228) T h e co-ordinates of t h e river-head are a t longitude : 10440'
a n d latitude : 565'. Those of the m o u t h are a t longitude : 10030' and lati-
t u d e : 515'. In S u h r b we can find exactly those co-ordinate d a t a as in al-
Huwrizml. On the basis of the consideration expounded in t h e previous point,
this can be identified with the present-day river Inicke.
F r o m the Norosson mountain two rivers are heading for the Syr-Darya.
I n al-IJuwrizml b o t h are nameless.
(22232224) T h e river takes its source in the liLyl mountain at longi-
t u d e : 99 and l a t i t u d e : 5730'. It flows into the Syr-Darya a t longitude:
" T u r k e s t a n , 173.
20
H A . 73. 211.
21
Turkestan, 169.
22
Loc. cit.
23
Turkestan, 176.

925' and l a t i t u d e : 455'. I n al-TJuwrizmi, differing f r o m Ptolemy, t h e D a i x

takes its source in the Rhymmika mountains, thus in t h e Norosson oros there
remains place only for the 'Iarrj. On the basis of these, this river can be identi-
fied with the Ptolemaic 'Iarrj, a n d t h a t is identical with the present-day
Emba. 2 4
(22252226) I t takes its source in the 6Ly1 mountain a t l o n g i t u d e :
100 and l a t i t u d e : 595', and it flows into the Syr-Darya a t the longitude :
945' and latitude : 465'. This river has no equivalent in Ptolemy. E x a m i n i n g
the m a p carefully, we see t h a t beside t h e Emba-Iastes, only one river flows f r o m
the hilly country connecting with t h e Ural corresponding to the Norosson oros
towards the Syr-Darya and this is t h e Sari-su. In reality this does not reach t h e
catchment river either, just like t h e E m b a , or the Murghab t h a t does not r e a c h
t h e Amu-Darya. North of the S y r - D a r y a we can still f i n d another significant
river, this is t h e u. The site of this, however, completely differs from t h a t
represented in al-fjuwrizmi. Thus, this river cannot be anything else t h a n t h e
Sari-su. 25
(2233 2234) The source of t h e river can be found a t longitude : 103 a n d
latitude : 4830'. Its mouth is a t t h e longitude : 10310' a n d latitude : 5130'.
I n al-IJuwrizmi the Syr-Darya has three tributaries on t h e left side. This river
is the most westerly of them. In P t o l e m y , the Iaxartes has only two tributaries
on the left side, these are the Afjfto a n d the Bactxari. All t h e three tributaries
mentioned b y al-TJuwrizmi are nameless. Similarly, in the HA we can f i n d
three rivers here, t h e names of all t h e three can also be found. The river charac-
terized by al-IJuwrizmi with t h e above co-ordinates corresponds to the Qb,
because this is t h e most westerly of t h e three. According to Barthold's descrip-
tion, we do n o t know with which of t h e small brooks flowing in the territory in
question it can be identified, with t h e Aravan, the Absura, or the Isfayram. 2 6
I n Suhrb this river does not occur.
(2235 2236) The source of t h e river is at longitude : 10330' and lati-
t u d e : 4730', and its mouth a t longitude: 10430' and l a t i t u d e : 5040'.
Of the rivers on the left side the middle one corresponds to the s indicated in
the HA. This was identified by B a r t h o l d with the Aqbura. 2 7 This river does not
appear in Suhrb.
(2237 2238) The source of t h e river is at longitude : 10445' and lati-
t u d e : 485', and its mouth is a t longitude : 1065' and latitude : 5020'. I t is
t h e most easterly tributary on the l e f t side. I t corresponds to t h e f j u r a b in t h e
24
B e r t h e l o t : 224.
25
R E . 9. 1178-1179.
28
T u r k e s t a n . 159. H A . 72. 211.
27
HA. loc. cit. Turkestan, loc. cit.

344 M. MARTII
H A . This is the n a m e of t h e river even t o d a y . W . Tomaschek identified it with

P t o l e m y ' s Arjyo.26 This river does not a p p e a r in Suhrb.
(2046 2048) y b l y i . Its source is longitude: 146 and l a t i t u d e :
3910'. Through longitude : 149 and l a t i t u d e : 4130' it flows to the longitude :
153 a n d latitude : 4010'. On the basis of its name and site it can be identified
w i t h Ptolemy's Bavxtao. Tomaschek identified it with the present-day Hoang-
ho. 29
(2050 2051) I t is t h e tributary of t h e ^-LL^J. Its source is at longi-
t u d e : 14720' and l a t i t u d e : 4520'. I t s m o u t h is a t longitude : 14930' and
l a t i t u d e : 3840'. Tomaschek identified i t w i t h the Hoang-sui. 3 0
(20522053) Similarly, it is a t r i b u t a r y of the ^-LL y . I t s source is a t
l o n g i t u d e : 14930' a n d l a t i t u d e : 3840', a n d its mouth a t longitude: 153
a n d l a t i t u d e : 40. Tomaschek identified it with the Tar-ho. 3 1 T h e entire system
of river ^j-hL, unlike t h e kL> y and the , completely agrees with the
Ptolemaic picture. T h e tributaries of the can be identified with the tribu-
taries of the Bavxtoo.
(22492258) J. Its beginning is at l o n g i t u d e : 14530' and
l a t i t u d e : 47. F r o m here it flows to t h e longitude : 15230' and latitude :
5130'. On the basis of its name and site it corresponds to Ptolemy's Olyorj.
A. H e r m a n n identified i t with the river Tarim. 3 2
T h e beginning of its tributary is a t longitude: 14730' and l a t i t u d e :
5230'. I t is the t r i b u t a r y of the Tarim. H e r m a n n identified i t with the Tsung-
ling. 3 3 The system of t h i s river is identical b o t h in al-Huwrizm and Ptolemy.
T h e rivers t a k i n g their source in Mount R m i q have already been dis-
cussed b y J . E c k m a n n .
CONCLUSIONS
I n the i n t e r p r e t a t i o n of al-Huwrizm's t e x t and in t h e drawing of his

m a p t h e first d i f f i c u l t y is the interpretation of the Arabic letter-numerals.
H . Mzik has already corrected the misinterpretations and a great part of t h e
errors, when he published the work of al-Huwrizm entitled Kitb Srat
al-Ard, b u t of course, h e was not immersed in the investigation of the territories
28
R E . 5. 154. R E I I / 3 . 789.
29
R E . II. 175176.
30
Loc. cit.
31
Loc. cit.
32
R E . 17. 2101 2102. I n Ptolemy t h e river flows into the E a s t e r n Sea. According
to H e r m a n n this r e p r e s e n t s Lake Lop Nor. H e identifies its branch taking its source
in t h e Auzakia ore w i t h t h e Tsung-ling a n d t h a t taking its source in t h e Kasia with the
Yu-tien (Khotan).
33
Cf. note 32.

of Central Asia represented here. Thus he has left some questions open, and he
did not notice a few hidden mistakes. Although t h e n u m b e r of these errors is
insignificant, the work h a d still to be s t a r t e d with t h e clarification of these
mistakes, exactly on t h e basis of the hints of Mzik himself. 1
I n the case of t h e letter-numerals one of the possible sources of misunder-
standing is t h a t in t h e case of several Arabic letters t h e tracing is identical, t h e
differentiating mark is indicated by the dots applied. If t h e dots are o m i t t e d ,
t h e n sometimes several interpretations are possible. I t is t r u e t h a t t h e meaning
of t h e numeral is u n a m b i g u o u s and clear in m a n y cases even in absence of d o t s ,
exactly because one of t h e two possible interpretations is a f i g u r e of place value
one, a n d the other is a figure of place value ten, a n d t h e u n p u n c t u a t e d form in
question is on place value one. W e have to do with such a case in point 399,
where ( J ^ ) can be found. This there can only represent 2, because v = 2,
= 50.
I could establish t h e correct value with the help of o t h e r criteria, there,
where there was no such obvious evidence. I should like t o mention only a
few cases by way of example. Most of t h e difficulties were connected with t h e
writing of the ^ and t h e The separation of t h e a n d j was similarly
difficult, if they stood in t h e middle or in t h e beginning of t h e group of letters
signifying a figure (e.g. points 865, 866, 471, etc.). T h e 3 also presented a n
o p p o r t u n i t y for misunderstanding, if two dots were missing f r o m it. This h a p -
pened in the case of 469, where t h e form given can equally m e a n 84 a n d ' 1 0 4 .
I t did not create misunderstanding, however, in the case of p o i n t 470, where it is
clear even without dots t h a t t h e J can only be interpreted as 110.
I n t h e case of t h e rivers it could not be decided, w h e t h e r t h e mark is n o t
a clerical error in place of t h e 1 ? Suhrb gave the same form in each case, when
in al-IJuwrizm it stood like this, b u t a t t h e same t i m e t h e r e was no Ptolemaic
evidence available because of t h e m a j o r changes t a k i n g place in t h e representa-
tion of the rivers. I deemed it correct t h a t because of t h e m e r e assumption of
a clerical error it is still n o t w o r t h while t o correct the t e x t in reality. I n these
cases I have left t h e 5 ' unchanged everywhere.
I have also found such hidden errors which resulted f r o m the erroneous
writing of two letter-numerals similar in form. I n point 821 j stood instead of j ,
a n d in point 872 j stood instead of j .
E x a m i n i n g the d r a w n m a p itself, we can m a k e some interesting observa-
tions.
If we examine t h e cities represented on the m a p , we f i n d t h a t besides t h e
eastern Ottorokorra we do n o t find any other Ptolemaic p o i n t . I t has also been
established by Mzik t h a t t h e most changing layer of P t o l e m y ' s rewriting b y
1
These general hints are contained in the preface a t t a c h e d to t h e publication
of the t e x t .

346 M. M A R T H
al-Huwrizm is represented by the cities. 2 The great changes in Central Asia

h a v e two reasons. One of them is t h a t during the about one thousand years
t h a t have elapsed between Ptolemy and al-Huwrizm certain peoples disap-
peared and new ones came in their place. The other reason is t h a t the Arabs
conquered a great p a r t of these territories, and could gather more accurate
knowledge. 3 The question of the cities has already been clarified by K . Czeg-
ldy. 4 The cities represented on the m a p mean the main points of the routes
given by the itineraries. Viewing the map, we can also discover three routes.
I n the North there is the route U s r s a n a t - H g a n d a - B a n k a t h Traband
Isbgb Tarz-Nawkath Barshn, in the south the route Usrsanat
Hoganda Ahskath al-Tubbat, and towards the North the route Huw-
rizm City of theKhazars. I t strikes the eye at the first glance t h a t the cities
situated in reality on the banks of the Syr-Darya, are on the map south of the
place due to them. The reason of this shift is very likely t h a t as from Antiquity
t h e Yaxartes was mixed up with the Tanais. 5 I t also points to this t h a t on our
m a p the City of the Khazars is on the bank of the Syr-Darya, although, as
generally known, it was situated on the bank of the Volga. Of course, this could
only happen as a result of a further error, in fact so t h a t the Arabs sometimes
mistook the Don for the Volga. Ibn Hurraddbeh writes t h a t H a m l i h is situated
on the bank of the river t h a t comes from the land of the Slavs and flows into
t h e sea of Gurgni. 6 (This is the present-day Caspian Sea). Comparing with the
H A , it can be established without any doubt that the river coming from the
l a n d of the Slavs is the Don. 7 The mistakes made in the identification of the
rivers resulted, of course, in new errors. Not only the City of the Khazars b u t all
cities have shifted in southern direction. A result of this was t h a t the ->- ,
a n important place from the view-point of the itineraries, has also shifted. This
is a serious mistake, because it disturbes the system of mountains, although it
was taken over by al-Huwrizm from Ptolemy almost without any change. As
regards the number of degrees, the significant shifting of the h U has with-
o u t doubt also other reasons, because such a great displacement can only be
explained by the joint effect of several reasons. The other reason will be men-
tioned later on. At any rate, in order to preserve the harmony with the itinera-
2
H . M2ik : Parageographische E l e m e n t e . 176.
3
The considerable change of t h e n a m e s of cities is also explained by t h e fact t h a t
p a r t of t h e points represented b y a l - B u w r i z m did n o t y e t exist in t h e age of P t o l e m y .
I t is written by B a r t h o l d in t h e E I . I I . 04 69, under t h e e n t r y F e r g a n a t h a t this
n a m e is mentioned for t h e first t i m e in t h e 5 t h century: The n a m e of Tarz occurs also
in a Greek source, in 568 A. D.
4
K . C z e g l d y : Die K a r t e , 51.
8
J . J u n g e : Saka-Studien. Leipzig 1939.
6
I h n I J u r r a d d b e h , 1 24.
' H A . 75. 216.
Acta Antiqua Academiue Scientiurum Hungaricae 28, 1980

A L - H U W R I Z b l l ' S D E S C R I P T I O N O F CENTRAL ASIA 347
ries, the ^ had to occupy a different place on al-Huwrizm's m a p in

contrast to the Ptolemaic map. This meant such an insoluble problem and the
observation of different view-points necessary in such a degree (its situation as
compared with the itineraries its situation as compared with the other
mountains, etc.) t h a t Suhrb found it better to disregard its representation.
The mountains represented are without exception of Ptolemaic origin.
The mountains tally with the Ptolemaic system also in regard to their sites.
We can still find a few minor differences. Al-Huwrizm differs from the Ptole-
maic pattern on three occasions, in fact in the case of the Arsn, Syeba and
Anarea mountains. He represented all the three in a more differentiated form.
I t is, however, important to observe t h a t all the amplifications were made so
t h a t they do not disturb the Ptolemaic internal order already formed, each
change modifies the edge of the map.
Very likely, the knowledge regarding the mountains of Central Asia did
not come in possession of the Arabs directly from the Greeks, but with Syrian
mediation. The Arabic writing of the names of mountains shows t h a t we must
presume Syrian variants in the course of the transmission. The intermediate
Syrian forms can be inferred from two phenomena. One of them is the transliter-
ation of the original Greek name into Arabic according to Syrian orthog-
raphy. In this case I tried to point to the Syrian orthographic rules concealed
behind the Arabic script on the basis of the hints of Mzik. The transliteration of
the names of the Anarea and Syeba mountains belongs to this group. The other
phenomenon is the distortion of the names. The name transliterated into Syrian,
as a result of the similarity of the Syrian letters, was misread when the Arabic
rewriting was prepared, and the Arabic form, as a result of its distortion charac-
teristic of Syrian, betrays its origin. The following names belong to this group :
Imaos, Emoda ore, Askatagkas, Aspisia ore, etc. Although Al-Huwrizm follows
Ptolemy almost accurately in the representation of the system of mountains,
he does not take over namas of mountains from the Geographike Hyphegesis to-
gether with the mountains in several cases. Such mountains taken over without
names are the Paropanisos, the Saripha ore, the Kaukasion oros, the Oxeia ore
and the Sogdia ore. In connection with this it is interesting to note t h a t these
mountains do not occur in the geographical chapter of the Hexameron of Jacob
of Edessa either. Thus, very likely this also confirms the Greek-Syrian-Arab
route of the names of the former mountains.
If we take a glance a t the map drawn on the basis of al- Huwrizm's d a t a ,
it immediately strikes the eye t h a t in the East, the rivers situated in Serike
completely correspond to the Ptolemaic pattern, while the two rivers to be
found in the West, the Amu-Darya and the Syr-Darya, slightly differ from it.
The most important change and this reflects a t the same t i m e the
changed circumstances, as well as the fundamental principle of al-Huwarizmi's
rewriting (where he has more accurate knowledge, there he does not t a k e over

348 M. MARTH
P t o l e m y ' s d a t a without criticism) is t h a t w i t h him these two rivers do not

flow into the Caspian Sea, b u t into Lake Aral. T h e course of the Zarafsan simi-
larly bears witness to a b e t t e r knowledge. T h e Polytimetos corresponding to t h e
Z a r a f s a n flows into t h e Caspian Sea. In al-Huwrizmi this river flows already
i n t o t h e Amu-Darya. A similarly striking change is the E W direction of t h e
course of the Syr-Darya. Ptolemy's Iaxartes flows first from the South to t h e
N o r t h , then later on it t u r n s to the west. P t o l e m y ' s error is traced back by A.
H e r m a n n to two erroneous conceptions. According to the one the Greeks be-
lieved t h a t the I a x a r t e s separates the Sogdians from the Sakas, while according
t o t h e other conception t h e Sogdians lived a t t h e border of India. As a result of
t h e linking of these two ideas, Ptolemy placed t h e source of the Iaxartes as far
t o t h e South as he could.
Al-Huwrizmi represented the system of the Oxus in m a n y respects dif-
ferently from Ptolemy. H e did not represent t h e system of tributaries on t h e
l e f t side consisting of t h e rivers Artamis a n d Zariaspis as well as the Ochos a n d
t h e Dargamanes, b u t we can find with him t h e course of the Murgb, mentioned
b y P t o l e m y under t h e n a m e Margos. In P t o l e m y we do not find the river Wahs,
this belongs to al-Huwrizmi's new knowledge. A t this point he further devel-
oped t h e inherited knowledge. We know this river also from the H A . There,
however, several tributaries of the A m u - D a r y a are also mentioned, b u t very
likely al-Huwrizmi held these insignificant, therefore, he did not mention
t h e m in his work. I n this respect he included t h e Ptolemaic material into his
work a f t e r a certain sceening, just like t h e material of the Arabic itineraries
t h a t was recorded by him only after a very strict selection. This was also charac-
teristic of his working method.
I n the representation of the system of t h e Syr-Darya, al-Huwrizm! is
m o r e accurate t h a n P t o l e m y , who mentions e.g. t h a t from the Norosson moun-
t a i n several rivers flow into t h e Iaxartes. Al-Huwrizmi gives also the accurate
co-ordinates of these rivers. There is a difference also in the question of the left
side tributaries between t h e two of them. I n t h e Geographike Hyphegesis we can
read only about two tributaries, while in al-Huwrizmi we can read about three.
T h e H A also discusses three tributaries. This renders likely t h a t al-Huwrizmi
a n d t h e author of t h e H A used common sources. The three tributaries on the
l e f t side on al-Huwrizmi's m a p do not flow into the K a r a - D a r y a b u t into t h e
Syr-Darya. The reason for this representation is very likely t h a t these rivers
were also important f r o m t h e view-point of t h e itineraries (one of the routes
leading to Barshn went along here), and w i t h the displacement of the cities
these also shifted towards t h e West.
On the m a p t h e general shifting of t h e degrees can also be observed.
A t t e n t i o n was drawn to this on several occasions also by H. Mzik, who studied
8
H . M Z I K : op. cit. 176.

A L - H U W R I Z b l l ' S D E S C R I P T I O N OF CENTRAL ASIA 349
t h e eastern territories described by al-JJuwrizm equally condensed the Ptole-

maic material in order to ensure place to the new material in the territories of
Central Asia. H e also had to make place for the Syr-Darya, because the direc-
tion of the course of this in al-JJuwrizm is more elongated in the E W direc-
tion t h a n in Ptolemy. This disputable procedure has brought about several
detrimental consequences. Of these we can mention first t h a t the Kaukasion
oros and the Paropanisos were unproportionately elongated, and the river-heads
of the Amu-Darya a n d t h e Indus situated very close to each other, differently
from reality, shifted very f a r from each other. Add to this al-JJuwrizml's error
mentioned above resulting in an even more westerly shifting of the cities which
h a d been pressed to the West even without this. Thus, a large empty space has
come about in the middle. These two reasons jointly explain t h a t the jr j-.
in al-JJuwrizm lias shifted with 34 as compared with t h e Ptolemaic localiza-
tion.
However, it has to be noted t h a t for the filling of t h e e m p t y space in t h e
middle al-IJuwrizm did not elongate all surface forms. I n his work, e.g.,
VI J~r is shorter t h a n in Ptolemy. The reason for this is t h a t the Norosson
and Syeba mountains situated on its two sides were represented by him in a
more differentiated form and on a larger area. In the southern territories, on t h e
other hand, the shifting as compared with the Ptolemaic d a t a is greater t h a n
in the northern territories.
I t is not sure, however, whether all these errors h a v e to be a t t r i b u t e d to
al-JJuwrizm's activity. Studying his work, we can observe t h a t he used several
kinds of sources. H e had such sources (e.g. Ptolemy) t h a t were not known to
him first hand. H e has t a k e n over not only the positive knowledge b u t also t h e
errors of these sources, and the linking up and a d j u s t i n g of these sources were
not always successful either.
As it could be seen already in the Introduction, on t h e basis of different
references we can presume the existence of even three P t o l e m y translations.
I n the beginning of his work, I h n IJurraddbeh mentions 9 t h a t he translated
Ptolemy's work in question. H e wrote this book in 846, t h u s he had completed
the translation earlier. We know t h a t al-IJuwrizm's work came into existence
in these years. Thus it can be imagined t h a t , eventually, he could already use
this translation. According to the evidence of the Fihrist 1 0 a Ptolemy transla-
tion is also connected with the name of al-Kindi, this, however, was bad. I t is
not known exactly when was the translation prepared, b u t in all probability
it could already be known also by al-IJuwrizm. T h e third translation was
prepared by Thbit ihn Qurra. He was born in 836, and t h u s it is not likely t h a t
9
I b n Hurradlbeh, 3.
10
A 1 - N a d i m : K i t b al-Fihrist. Leipzig 1871 1872. See in t h e chapter entitled
Geography.

350 M. 1IARTH
his translation could h a v e come to t h e hands of al-Huwrizm before the writing

of his work. Thus, a b a d Arabic translation and eventually another Arabic
t r a n s l a t i o n was a t al-Huwrizm's disposal, b u t we are not informed about the
v a l u e of the latter. Similarly, according to the Fihrist, P t o l e m y ' s work also had
a Syrian translation.
The Syrian translation of Ptolemy's work was in all probability known by
al-Huwrizm, since as i t was already pointed out above, we saw in connection
w i t h t h e names of t h e mountains t h a t f r o m the writing of t h e Arabic forms of
n a m e s we can frequently make conclusions to an intermediary Syrian v a r i a n t .
T h e question arises spontaneously t h a t there, where al-Huwrizm does not give
t h e names of the mountains and we do not find any reference to t h e names either
in J a c o b of Edessa or in Suhrb, w h a t brings about this deficiency as compared
w i t h t h e other cases ? The names of t h e Kaukasion oros a n d t h e Paropanisos
s i t u a t e d in the South (where the shifting is greater than in t h e North) are known
e.g. only from Ptolemy's Greek t e x t . If we examine these mountains more thor-
oughly, we find t h a t in their representation al-Huwrizm differs considerably
f r o m the Ptolemaic traditions. I n t h e co-ordinates of these, t h e difference of
degrees is even greater t h a n in the case of the others. I n the description of these
mountains, where al-Huwrizm also otherwise relied upon Ptolemy in the
highest degree, he probably drew first of all on the Syrian translation, while in
t h e elaboration of t h e mountains separating India and Central Asia he could
use another source. This could be either one of the two Arabic sources j u s t
mentioned, or an earlier Arabic map. Unfortunately, we do n o t know, whether
t h e two Arabic translations were prepared as literal translations, or they were
a l r e a d y rewritings.
About the representation of the rivers it can be established t h a t the rivers
t o be found in Serike tally with the picture to be found in P t o l e m y . In Central
Asia, however, the hydrographie m a p differs from the known patterns. In the
representation of the rivers flowing in Serike, al-Huwrizm very likely used the
S y r i a n translation, because in fact, t h e orthography used in t h e transliteration
of t h e names of rivers does not contradict to this assumption. I n the drawing of
t h e waters of Central Asia, however, his main source could already be an Arabic
rewriting. Besides, very likely, he also used the d a t a originating from the Arabic
itineraries, because on t h e m a p we do not find only corrected Ptolemaic ele-
m e n t s , b u t also several new rivers. The description of these rivers can also be
r e a d in the books belonging to the Maslik wa Mamlik literature of geographi-
cal writers living in the period a f t e r al-Huwrizm.
The description of the cities completely differs from t h e representation
t o b e found on Ptolemy's map. With t h e exception of the d i s t a n t city of Ottoro-
k o r r a , all elements are new. All the cities of Central Asia without exception can
be found in the enumerations of the itineraries, which were t h e main sources of
al-Huwrizm.

AL-HUWRIZbll'S DESCRIPTION OF CENTRAL ASIA 351
I n the case of t h e names of countries we also find a duality. The n a m e s

byiM and Ljj-I are clearly of Greek origin, they can be traced back to Scythia
extra Imaum a n d Scythia infra Imaum. On the basis of the correct writing
we can conclude t h a t this n a m e also got into the Arabic with Syrian media-
tion, where on account of the error in t h e punctuation t h e form of the n a m e
f u r t h e r deteriorated. For the time being it cannot be decided, whether this error
in t h e puctuation was made by al-Huwrizm himself, or he took it over already
in such a defective form. The country n a m e AJ jh> j ^LILL ; is similarly of
Arabic origin.
Suhrb also prepared his own world m a p using the work of al-Huwrizm.
Examining this m a p , it immediately strikes the eye t h a t it is much more sche-
m a t i c than t h a t of al-Huwrizm. This can be observed in t h e representation of
t h e mountain separating India and Central Asia, as well as in t h e drawing of t h e
system of the Indus. The description of t h e rivers is more or less identical in
b o t h works. The greatest difference can be observed in t h e case of the cities.
Beside the material to be found in al-Huwrizm, Suhrb included also new
elements in his work. The -uJI a n d t h e v j i - i l ! also belong here. 11 T h e
insertion of the new material of knowledge was not successful. Comparing t h e
localization of t h e -UAII OX. with the cities represented, we find t h a t it is too
f a r to the North, while the y.^ l s too f a r to the E a s t , beyond the P a m i r .
The new material did not become an organic p a r t of t h e map, it remained a
foreign body. The fiasco of the insertion, however, is not due to Suhrb alone.
T h e m a p of al-Huwrizm is also concerned in this in a great degree. A f t e r all,
t h e AAJI J-L drawn in a situation comparatively well a d j u s t e d to the surface
forms is not in h a r m o n y with the system of cities taken over in a defective form,
without correction.
If we make some statistics, it will become clear how Suhrb took over
al-Huwrizm's d a t a . On our m a p al-Huwrizm represented 17 cities. Of these
12 can easily be identified, the preserved t e x t is unambiguous and clear. I n t h e
case of 5 cities, on account of the absence of names or uncertain reading, t h e
identification is difficult. 1 2 In the first group the given co-ordinates in the ma-
jority of the cases slightly differ in Suhrb from those of al-Huwrizmi (in 7
cases they differ, while in 5 cases they are identical). As regards the cities be-
longing to the other group, their situation was probably not clear to Suhrb
himself. H e took over al-Huwrizm's d a t a without hesitation, in the form he
found them. This is shown by the f a c t t h a t he does not deviate from al-Huw-
rizm's d a t a even in a single case.
11
Both cities occur in the H A , p p . 561 652.
12
The cities belonging in the first group o r e : Ursanat, R o u a n d a , B a n k a t h ,
Ahskath, Traband, Isbgb, al-Tarz, al-Tubbat, Atrqr, Rwrizm, Madnat a l - R a z a r ,
N a w k a t h . The second group contains : Madlnat (470) m a d n a t (471) Snistn, t y G
L.

352 . m a r o t h : a l - h u w r i z m i ' s d e s c r i p t i o n of c e n t r a l a s i a
A significant difference between t h e m is that in Suhrb we cannot find

the y j , as well as the Ptolemaic Demos and Baskatis.
I t can be mentioned as an influence of al-ljuwrizm t h a t a city named
L-L- c a n also b e f o u n d in Idris.
Budapest.
APPENDIX
Beside the mountains drawn on t h e map, Suhrb in his work indicates

t h e following :
Mountain. Longitude : 1 2 8 5 ' a n d latitude : 5830' longitude : 1285'
a n d l a t i t u d e : 525'.
Mountain. Longitude : 1325' a n d latitude : 605' longitude : 1325'
a n d l a t i t u d e : 585', here it meets another one : at longitude: 1295'and lati-
t u d e : 5830' this is m e t by a m o u n t a i n , a t longitude: 141 and l a t i t u d e :
5530' it ends.
These mountains were described b y Suhrb somewhat confusedly, b u t
t h a t m u c h is still sure t h a t on the m a p t h e y cannot be represented. Mzik also
f o r m e d the same opinion. According to him Suhrb did not t a k e over these d a t a
f r o m Ptolemy, b u t from some other work. Disregarding the inconsistency of his
d i f f e r e n t sources, he wrote down everything. At the drawing of Suhrb's m a p
t h e above d a t a had to be disregarded b y me.
ABBREVIATIONS
W . BARTHOLD : T u r k e s t a n down t o t h e Mongol Invasion. L o n d o n 1958. (Turkestan)

A . BERTHELOT : L'Asie Ancienne Centrale et Sud-Orientale d'aprs P t o l m e .
P a r i s 1 9 3 0 . (BERTHELOT)
V. MINORSKY : H u d d al-'lam. Oxford 1937. (HA)
J . O. THOMSON : H i s t o r y of Ancient Geography. Cambridge 1948. (THOMSON)
A l - I s t a h r i : K i t b al-maslik wa '1-mamlik. E d . M. J . DE GOEJE. Lugd. B a t a -
v o r u m 187.
I b n H a w q a l : K i t b al-maslik wa '1-mamlik. E d . M. J . DE GOEJE. Lugd. B a t a -
v o r m n 1873.
Al-Maqdis : A h s a n al-taqsim fi m a ' r i f a t al-aqlim. E d . M. J . DE GOEZE. L u g d .
B a t a v o r u m 1877.
I b n R u s t a : Al-I'lq al-naflsa. L u g d . B a t a v o r u m 1877.
A l - J a ' q b : K i t b al-buldn. E d . M. J . DE GOEJE. L u g d . B a t a v o r u m 1892.
I b n K h u r r a d d b o h : K i t b al-maslik wa'1-mamlik. E d . M. J . DE GOEJE. L u g d u n i
B a t a v o r u m 1889.
Q u d m a ibn G a ' f a r : K i t b a l - R a r g . E d . M. J . DE GOEJE. Lugduni B a t a v o r u m
1889.

I. F R H L I C H
HISTORIOGRAPHIE ET AGGADA DANS LE LIBER

ANTIQUITATUM BIBLICARUM DU PSEUDO-PHILON
1. La gense de l'uvre et les problmes de genre
Le L A B dont la matire s'tendant de la cration jusqu' la m o r t de

Saiil (2 Samuel 1) est puise la Bible f u t compos aprs 70 de n. ., (voir
l'allusion faite en X I X . 7 visant la destruction du second temple de Jrusalem).
D'aprs la prcision de M. R. J a m e s il serait le p r o d u i t de la mme cole que le
4 e livre d ' E s d r a s et l'Apocalypse de Baruch, conserve en langue syriaque. 1
L'uvre elle-mme passe pour un des derniers produits de cette poque
littraire allant du temps de la destruction du premier temple jusqu' la dis-
persion sous Tite et Vespasien. Le dbut de cette activit littraire est m a r q u
par la canonisation des livres bibliques en Palestine. 2 C'est au cours de cette
priode que naquirent la plupart des uvres pseudpigraphiques, les parties
ultrieures du Testament des douze patriarches, 3 le Livre des Jubils, 4 le Livre
de Hnoch sous sa forme dfinitive, 5 le 4 e livre d ' E s d r a s 6 et l'Apocalypse syria-
que de Baruch dont le texte subit peut-tre l'influence du LAB. 7
L'histoire politique de la priode de la gense d u L A B comprend l'poque
de l'tat national juif suivant la rvolte des Macchabes, ensuite la domination
de Rome et le temps de la guerre juive. Le LAB f u t compos aprs la destruc-
tion du second temple, t o u t probablement sur territoire palestin, fait d o n t
1
M. R . J a m e s : T h e Biblical Antiquities of Philo. N o w first translated f r o m t h e
Old Latin Version. L o n d o n 1918. 37.
2
L ' t a p e suivant de la canonisation de la Tora ( E s d r a s 7 et suiv., Nhmie 8 et
suiv.) est m a r q u e p a r l'tablissement d u canon de la Tora et des prophtes. Processus
qui f u t termin j u s q u ' 117 av. n . . ; Josphe Flavius, lui, m e n t i o n n e 22 livres canoniss
(Contra Apionem, I . 8) ; cf. O. E i s s f e l d t : Einleitung in d a s Alte Testament. T b i n g e n
1964 3 ; W . F e l l : Der Bibelkanon des Flavius Josephus. BZ 7 (1909) 1 16, 113122,
235244; G. W. A n d e r s o n : Canonical a n d non-canonical. The Cambridge H i s t o r y
of t h e Bible. Cambridge 1970. 1 1 3 - 1 5 9 .
3
0 . E i s s f e l d t 1964. 8 5 5 - 8 6 2 .
4
O . E i s s f e l d t 1964. 8 2 3 - 8 2 4 .
5
O . E i s s f e l d t 1964. 8 3 8 - 8 4 3 .
6
O. E i s s f e l d t 1964. 848 - 8 4 9 .
' O. E i s s f e l d t 1964. 852 853. Sur le rapport, e n t r e le L A B et l'apocalypse syrai-
que de Baruch v. M. R . J a m e s 1917. 46 54 (avec collationnemont dos textes) ; L. G r y :
L a d a t e de la fin des temps, selon des rvlations ou les calculs d u Pseudo-Philon et d e
B a r u c h (Apocalypse syriaque). R B 48 (1939) 3 3 7 - 3 5 6 .

354 I. FRHLICH
t m o i g n e entre autres le t e x t e originel crit en hbreu. 8 Au cours de la guerre

juive, e t prcisment aprs la chute de M a s a d a (73 de notre re) le sort de
P a l e s t i n e se voit comme rgl, un peu plus t a r d le territoire lui-mme choit
Vespasien. La rsidence d u gouverneur imprial de J u d e se t r a n s m e t
Csare, Jrusalem, ruin ne p o u r r a plus j a m a i s reprendre son ancienne im-
p o r t a n c e ; la domination romaine m e t l'accent sans cela sur le dveloppement
des villes de fondation ou de reconstitution romaines. 9 L a p o p u l a t i o n , elle, se
t r o u v e puise et dpouille p a r la guerre. Officiellement le culte est supprim
e t malgr une survie s p o r a d i q u e m e n t atteste 1 0 le sanctuaire de Jrusalem,
en t a n t que centre e t r e p r s e n t a n t officiel d u culte, n'existe plus.
U n e uvre littraire bouleversante e t qui nous renseigne sur la guerre
c ' e s t le Bellum J u d a i c u m de Josphe Flavius, crit de p r o p a g a n d e e t apologie
la fois, avertissement adress a u x J u i f s v i v a n t hors la Palestine, et prcision
d u rle e t de la responsabilit des J u i f s palestiniens du point d e vue des vne-
m e n t s lis l'clatement de la guerre. 1 1 Si l ' u v r e de Josplie est le seul expos
h i s t o r i q u e dtaill sur l ' p o q u e en question, la p l u p a r t des u v r e s littraires
contemporaines ne r e s t e n t p a s elles non plus sans toucher l'histoire, d o n t celles
d e P h i l o n d'Alexandrie, I n Flaccum et Legatio a d Gaium, ouvrages polmiques
c o n t r e les offenses concrtes faites a u x J u i f s de la Dispersion. 1 2 C'est dans ce
m m e genre que se r a n g e le Contra Apionem de Josephus. 1 3
L e LAB, lui n ' e s t p a s une rponse des attaques aussi directes, ni une
a n a l y s e des vnements contemporains d u p o i n t de vue des J u i f s ; il n'en
r e s t e p a s moins une u v r e apologtique. C'est en effet cette intention qui se
f a i t valoir dans la manire de voir d o m i n a n t l ' u v r e entire,dans l'arrangement
e t d a n s le remaniement de la matire lgendaire de l'histoire. L e L A B est une
apologie indirecte crite d a n s le dessein de dmontrer que le j u d a s m e n'est
d u p o i n t de vue ni culturel ni historique infrieur aux grands peuples culturels
d e l'Antiquit. Sous ce r a p p o r t le L A B est comme un parallle historique
d e l ' u v r e de Philon d'Alexandrie, 1 4 ce dernier explique d e s notions de la
8
L. C o h n : A n A p o c r y p h a l W o r k Ascribed to Philo of Alexandria. J Q R 10
(1898) 311.
9
E . S c H R R E R : Geschic htede sjdischen Volkes im Zeitalter J e s u Christi. I .
L e i p z i g 1890. 537545.
10
E . S c h r e r 1890. 546 647 ; A. B c h l e r : Economie Conditions of J u d a i s m
a f t e r t h e Destruction of t h e Second Temple. L o n d o n 1912. Sur la signification du Temple
et sui le culte v. G. F . M o o r e : J u d a i s m in t h e F i r s t Centuries of t h e Christian E r a . I .
C a m b r i d g e 1950. 2 8 1 - 3 0 7 .
11
S. Z e i t l i n : A s u r v e y of Jewish historiography from t h e Biblical Books to t h e
Sefer h a - K a b b a l a h with special emphasis on J o s e p h u s . J Q R 49/3 (1969) 171 214.
12
L e premier doit sa naissance la perscution des Juifs d'Alexandrie, le second,
l ' a p o t h o s e et l'introduction d u culte de Caligula.
13
Contra Apionem t o u t comme les Antiquits judaques e n t r e n t dans la
c a t g o r i e de l'apologie indirecte. L ' a u t e u r y e x p r i m e ses ides sous f o r m e de trait
h i s t o r i q u e . Sur les genres apologtiques v. J . B e r g m a n : Jdische Apologetik im neu-
t e s t a m e n t l i c h e n Zeitalter. Berlin 1908.
14
Le L A B reprsente u n e analogie indirecte d ' u n e part de la m t h o d e exgetico-
p h i l o s o p h i q u e des Questiones, d ' a u t r e p a r t de l'apologie directe.

H I S T O R I O G R A P H I E E T AGGADA 355
religion en d m o n t r a n t la priorit des notions bibliques sur celles de la religion

grecque, e t cela p a r le moyen de la philosophie grecque; le Ps.-Philon, lui
intgre les passages bibliques qu'il traite, l'ide de l'alliance, de l'lection.
L a conception n ' e s t pas nouvelle, la composition de la Bible elle-mme se
f o n d a n t sur l'ide de l'alliance reparaissant sans cesse d a n s le t e x t e biblique
le procd du L A B est p o u r t a n t t o u t nouveau. P a r m i les histoires bibliques il
ne retient que les passages insistants sur l'importance de l'alliance, e t les
complments qu'il y apporte s ' a j u s t e n t eux aussi ce p o i n t de vue. Le t e x t e
d u L A B ne va-t-on p a s donc le considrer comme un e x t r a i t de la Bible le
Livre des Jubils t r a i t a n t les histoires de la Gense n'en est lui non plus u n
simple abrg 15 il s'agit d ' u n t e x t e remani suivant un seul point de vue,
d ' u n e u v r e autonome, de voix individuelle, compose d ' u n e technique t o u t e
spciale.
L a mthode de composition d u L A B et son procd de r e m a n i e m e n t
relvent c e p e n d a n t d ' u n e tendance b e a u c o u p plus ancienne, de celle commence
p a r les premiers remaniements des cycles de tradition bibliques. L a matire d e
la Bible, la mise au point canonique de ses composants d i f f r a n t p a r leur carac-
tre et leur genre, se t r o u v e n t ds le d b u t domines p a r une ide runissant le
t o u t de l ' u v r e e t p o r t a n t sur le r a p p o r t entre vie humaine, socit et s u r n a t u -
rel ; loin d'tre un simple rcit d'vnements, la matire de la Bible est u n e
sorte de philosophie de l'histoire, dtermine toutes les fois p a r la vision reli-
gieuse du monde de chacun des remanieurs. D a n s la conception de ces derniers
il y a ceci de c o m m u n que la directrice suprme des vnements c'est p a r t o u t la
justice de Y a h v ('emet) la manire de la vision du m o n d e des Msopota-
miens o c'est la justice (kittum) qui soutient l'ordre cosmique, le roi n ' e n
t a n t que le serviteur, la garantie. 1 6 D a n s la Bible les souverains terrestres ne
sont jamais e x e m p t s de faute au lieu d'tre les incarnations d u 'emet ils
en sont les serviteurs. L a conception historique de la Bible a pour deux p o i n t s
focaux l'alliance d u peuple lu avec Y a h v et la dlivrance d u peuple p a r
Yahv. Chacun des livres bibliques se fonde sur ces prmisses; et, la conception
des bases de l'alliance, l'explication de la dlivrance, l'espoir e t le pessimisme
y relatifs a u t a n t de moments qui varient suivant l'poque p o u r caractriser
chacune des parties de la Bible. 17
15
M. R . J a m e s 1917. 4 6 ; S. Z e i t l i n : The Book of Jubilees, its Character a n d
its Significance. J Q R 30 (1939 40) 1 32 ; S. Z e i t l i n : The Book of Jubilees. J Q R 36
(1944-46) 1 2 - 1 6 .
16
E . A. S p e i s e r : The Biblical I d e a of H i s t o r y in its Common Near E a s t e r n Set-
ting. I E J 7 (1967) 203.
" M. B u r r o w s : Ancient Israel. The I d e a of H i s t o r y in t h e Ancient Near E a s t .
N e w H a v e n - L o n d o n 1966 3 . 1 1 1 - 1 1 7 . G. F . M o o r e I960, I I . 48 ; E . A. S p e i s e r : I E J
7 (1967) 269 ; R . B u l t m a n n : Primitive Christianity in its c o n t e m p o r a r y setting. E d i n -
burgh 1960. 4 0 - 6 1 .

356 I. F R H L I C H
L e L A B a p o u r modles bibliques le Livre des Rois et celui des Chroni-

ques. 1 8 L ' u n comme l ' a u t r e r e m o n t e n t des sources diffrentes et d a t e n t d e
l ' p o q u e no-babylonienne et achmnide (le Livre des Rois o b t i n t sa f o r m e
d f i n i t i v e au milieu d u 6 e sicle, les Chroniques, de mme que le Livre d ' E s d r a s
e t celui d e Nhmie f o r m a n t , trois, une unit, crits aprs le retour de l'exil de
Babylone). 1 9 Ces deux livres sont les premires tentatives d ' u n aperu gnral
de l'histoire des tribus d'Isral ; il s'agit d'anthologies historiques composes
d a n s u n ordre chronologique bien dfini, p u i s a n t leur matire d a n s des sources
d i f f r e n t e s et p o r t a n t ainsi l'empreinte d e s ides nationales e t religieuses de
leur rdacteur. L e u r gense semble avoir t influence p a r les chroniques
synchronistes assyriennes, puis no-babyloniennes, d u 7 e sicle. 20 Le L A B est
m a r q u s u r t o u t p a r l'influence d u Livre d e s Chroniques qui, postrieur l ' a u t r e
en m e t lui-mme en u v r e quelques passages. L a similitude a p p a r a t au premier
c o u p d ' i l : les Chroniques, le Livre d ' E s d r a s et celui de Nhmie formrent
p r i m i t i v e m e n t une unit contenant l'histoire de la r o y a u t de J u d a 2 1 c'est
dire, d a n s leur conception l'empire d e Yahv, t o u t en o m e t t a n t l'histoire
de la r o y a u t d u Nord impie e t en p a s s a n t en revue, en ordre chronologique
les v n e m e n t s p a r t i r d ' A d a m j u s q u ' a u x temps t o u t rcents, c'est--dire
j u s q u ' a u retour et la rorganisation d u culte. E s d r a s et Nhmie m e t t e n t au
centre, eux, le culte et le Temple plus t a r d , dans la conception religieuse plus
a b s t r a i t e d u Ps.-Philon cette position sera rserve la loi. D a n s les Livres des
Chroniques et des Rois l'histoire a p p a r a t comme l'enchanement d u dveloppe-
m e n t organique des v n e m e n t s et ces vnements les a u t e u r s a j o u t e n t des
explications politiques, religieuses et morales.
D e p a r leur origine leur matire t a n t prise d a n s les livres plus anciens
les livres en question sont de caractre midrashique, c'est dire que la
m a t i r e remanie y est suivie d'explications et de complments. Le caractre
m i d r a s h i q u e d u Livre des Chroniques est d d ' u n e p a r t a u x extraits qu'il
c o m p o r t e du Livre de la Gense et de celui des Rois, d ' a u t r e p a r t a u x additions
(pour la p l u p a r t des prires, des discours). 22
18
O. EISSFELDT 1964. 654 656. Sur le L A B et le Livre de la Chronique v. M. R .
JAMES 1917. 33.
19
O . EISSFELDT 1964. 749.
20
O . EISSFELDT 1 9 6 4 . 3 3 8 - 3 7 9 ; A . C. WELCH: T h e W o r k of t h e Chronicler.
I t s P u r p o s e and its Date. London 1939 ; A. NORDTZIJ : Les intentions du Chroniste.
R B 49 (1940) 1 6 1 - 1 6 8 .
21
O. EISSFELDT 1964. 654-656.
22
L ' a u t e u r de la Chronique se servit, e n t r e autres, du midrash du Livre des Rois,
cf. O. EISSFELDT 1964. 721 723. Sur le caractre et l'origine d u midrash, sur les pas-
sages midrashiques des livres canoniss v. K . BUDDE : Midrasch des Buches der Knige.
Z A W 12 (1892) 37 51 ; R . BLOCH: Ezechiel X V I : Exemple parfait du procd mid-
rashique dans la Bible. CahSi 91 (1955) 193 223; R. BLOCH: Midrash, Dictionnaire
de la Bible. Suppl. 5. P a r i s 1957 ; R . BLOCH : Ecriture et traditon dans le judasme.
Aperus sur l'origine du Midrash. CahSi 8 (1954) 9 34 ; S. ZEITLIN : Midrash. A Histori-
cal S t u d y . J Q R 44 (1953) 21 36.

Le Ps.-Philon emprunte au Livre des Chroniques non seulement des

noms, mais aussi un procd, leur conformit consistant principalement d a n s
la propagande qu'ils font : les Chroniques idalisent la thocratie ralise p a r
la royaut lgitime de David et runissant les Juifs retourns de l'exil; l'idal
du Ps.-Philon est, son tour la thocratie d a vidique qui sera tablie comme p o u r
accomplir les calculs eschatologiques.
P a r m i les uvres pseudpigraphiques de date postrieure c'est le Livre
des Jubils qui met en uvre, la manire du LAB une matire biblique, en
appliquant une mthode historique. Le Livre des Jubils remaine la matire
de la Gense en la rpartissant en priodes de ybl e t en y a p p o r t a n t des
complments explicatifs aggadiques et halachiques ; son b u t consiste en effet
illustrer, travers la matire de la Gense la voie de l'observance de la loi.
P a r sa conception et sa mthode il servit, sur plus d'un point de modle p o u r
le LAB. 2 3
2. Alliance et histoire universelle eschatologique
L a matire mise en uvre d a n s le L A B finit sur la m o r t de Saiil, mais

cette fin n'est pas celle envisage p a r l'auteur. C'est que le t e x t e s'arrte brus-
quement, en plein milieu d'une phrase du Saiil mourant, ce qui rend probl-
matiques la fin et aussi l'intention de l'auteur quant l'tendue et l'achve-
ment de la matire historique traite par lui. Malgr la forme inacheve on
devine quand mme une structure prcise de l'uvre qui p e r m e t d'en reconsti-
tuer d u moins hypothtiquement l'achvement vraisemblable, la conclu-
sion de l'histoire universelle prsente par le LAB.
L e LAB s'ouvre non sur la Gense, mais sur l'numration de la famille
d'Adam. 2 4 L ' a u t e u r n'indique que brivement la priode qui s'coula depuis
le premier couple d'hommes jusqu' No, sous forme de tableaux gnalogiques,
en compltant le texte d'une histoire sur l'invasion du pch (I. 8). La mthode
que l'auteur du L A B applique ici est celle du livre des Chroniques, en matire
la fois de la mise en uvre et des noms. Il parat sr que l'auteur a trs bien
connu la Bible et qu'il en f i t emploi, d'une faon consciente, de tels passages,
tels cycles de tradition. 2 5
L'histoire de No suivant les tableaux gnalogiques (III. 1 12) ne
diffre pas, pour l'essentiel de celle de la Bible. Cependant la description
donne par le Ps.-Philon de l'alliance conclue avec No est d'une importance
23
L. F i n k e l s t e i n : The Book of Jubilees and t h e R a b b i n i c H a l a k a h . H T R 1
(1923) 3 9 - 6 1 ; S. K l e i n : P a l s t i n i s c h e s in J u b i l e n b u c h . Z D P V 57 (1934) 727.
Sur les relations aggadiques du L A B et du Livre des Jubils cf. M. R . J a m e s : 1917. 45.
24
Cf. Chronique 1 : 1 s ' o u v r a n t galemont sur la descendance d ' A d a m .
25
P . ex. l'omission consquente ou transformation des passages relatifs a u x S a m a -
ritains, d a n s la matiro b i b l i q u e : v. A. S p i r o : Samaritans, T o b i a d s a n d J u d a h i t e s in
Pseudo-Philo. P A A J R 20 (1951) 2 7 9 - 3 5 5 .

particulire. Aprs le dluge No prsente un sacrifice d'action de grce qui

sert de base pour l'alliance effective : Et dixit Deus : Non adiciam iam male-
dicere terram pro homine, quoniam figura cordis hominis desiit a iuventute
sua, et ideo non adiciam simul disperdere omnes viventes sicut feci. Erit autem
c u m peccaverunt h a b i t a n t e s terram, diiudicabo eos sive in gladio, sive in igne,
sive in morte, et erit terremotus et dispergentur in inhabitabilia. Terram autem
non adiciam corrumpere a q u a diluvii (III. 9). Cum a u t e m completi fuerint
anni seculi, tune quiescet lumen et extinguentur tenebre, et vivificabo mortuos,
e t erigam dormientes de t e r r a . . . E t erit terra alia, et celum aliud habitaculum
sempiternum (III. 10).26
C'est dire que la description de l'alliance, t o u t en compltant l'ide
formule dans l'Ancien Testament sur l'alliance conclue avec No, la divise en
d e u x parties : la premire c'est en conformit avec la Bible cette promesse
d e Y a h v qu'il ne d t r u i r a plus jamais la Terre par un dluge. Le Ps.-Philon,
lui y ajoute une autre ide, de caractre eschatologique, servant de base pour
une seconde alliance : il y aura d'autres sortes de chtiments, suivant la mesure
des pchs des mortels. C'est ce complment qu'achve la conclusion eschato-
logique encadrant, elle, l'alliance conclue avec les mortels. Du point de vue de la
s t r u c t u r e ce passage n ' a point son pareil d a n s l'uvre et y a un rle de pilier.
C'est que l'uvre entire f u t tout probablement une sorte d'apologie juive
encadre d'une histoire universelle conception eschatologique, et l'auteur
chercha runir les ides de l'eschatologie cosmique avec celles de l'histoire
nationale juive, en imposant, la fin de l'uvre une conclusion commune
a u x passages situs sur des plans t o u t diffrents. 2 7
Chez le Ps.-Philon l'ide du peuple lu apparat la premire fois dans
l'histoire d'Abraham (VI. 1 18). C'est l'histoire de Babel q u ' a t t a c h e l'auteur
le motif explicatif et unique dans son temps de l'lection d'Abraham
(L'analyse de G. Vermes a dmontr que sur ce point le t e x t e du Ps.-Philon
servit de base pour les midrasli les plus anciens aussi bien que les rcents). 28
L a situation de d p a r t est la mme que d a n s le cas de No : parmi les mortels
coupables et corrompus Y a h v ayant t r o u v un seul qui soit juste pour conclure
u n e alliance avec lui. Mais tandis que, selon le Ps.-Philon du temps de No
sur la terre adimpleta e r a n t opera malignitatum (III. 3 en pleine confor-
m i t avec les paroles de l'Ancien Testament, v. Gn 6 : 5, mais en ne se servant
q u e de ce seul passage), l'poque d ' A b r a h a m le pch des hommes s'est mani-
26
Sur la consommation des temps v. Test. R u b e n 6 : reAelai ynovojv ; Test.
L e v i 10 : awreia rdjv alvcov; Test. Zebuion 9 : xaigg ovvzeAeia Sur le commence-
m e n t d u nouvel on : l'apocalypse syriaque de Baruch 44 : 12 : m u n d u s ille ; 48 : 5,
51 : 10 : m u n d u s novus ; 51 : 3 : m u n d u s qui non moritur (les citations sont t r a d u i t e s
p a r O. F . F r i t z s c h e : Libri Apocryphi Veteris Testamonti. 1871). Avec Baruch concorde
E s d r a s IV. 7 : 50, 4 : 2, 27, 6 : 9, 7 : 12, 8 : 1, 6 : 55, 7 : 113, etc.
27
R . B u l t m a n n 1960. 9 4 - 1 0 2 .
28
G. V e r m e s : Scripture and Tradition in Judaisme. H a g g a d i c Studies. Leiden
1961. 88.

feste d a n s ce qu'ils s'opposrent la rvlation divine : Tune hi qui divisi

e r a n t omnes habitantes terram, postea congregati h a b i t a v e r u n t simul. E t
profecti a b oriente invenerunt c a m p u m in terra Babilonis, et habitantes ibi
dixerunt singuli quique ad p r o x i m u m suum : Ecce futurum est ut dispergamur
unusquisque a fratre suo, et in novissimis diebus a l t e r u t r u m erimus expu-
gnantes nos. Nunc ergo venite et edificemus nobis ipsis turrim, cuius c a p u t erit
usque ad celum, et faciamus nobis nomen et gloriam super terram (VI. 1).
C'est dire que le b u t de la construction de la tour n'est-il pas ce que suggre
Gn 1 1 : 4 faisons-nous un nom afin que nous ne soyons pas disperss sur la
face de toute la terre ; ils construisent la tour tout en s a c h a n t bien qu'ils seront
disperss sur la terre et qu'une lutte aura lieu entre les hommes aussi se
mettent-ils, comme par vengeance contre Yahv, construire une tour afin
de perptuer leur nom.
Le passage analys est donc une interprtation vaticinante de la prci-
sion du Gn 1 1 : 4 , interprtation conforme l'esprit du L A B et qui, f i g u r a n t
l'endroit donn se voit intgre au cadre eschatologique de l'histoire de No.
l ' i n t e r p r t a t i o n d e l ' h i s t o i r e d'abraham dans l e lab"*

L a rvolte contre Yahv et le monothisme est figure par une image
qui, intrieur du systme de symboles du LAB, figure ailleurs en symbole
de l'idoltrie (motif des noms gravs sur les pierres, X X V I ) . Dans l'histoire
d ' A b r a h a m les constructeurs de la tour de Babel gravent leur propre nom sur
les pierres qu'ils vont cuire aux fins de la construction de la tour. L a signifi-
cation du motif se trouve par l modifie : l'audace des constructeurs de la
t o u r devient le symbole de la rvolte contre l'avenir rvl. 29 Chez le Ps.-Philon
les constructeurs de la tour de Babel se rvoltent contre la direction divine d u
processus de l'histoire voil le plus qu'apporte le L A B cette histoire qui
d a n s l'Ancien Testament et ailleurs raconte la rvolte de l'homme contre le
ciel. 30 E n dehors de ce qui y est effectivement rapport l'importance de l'his-
toire de Babel prsente par le Ps.-Philon est encore augmente par ceci que
de l'pisode de Yar et les sept dvots pisode parallle celui du salut mira-
culeux d ' A b r a h a m et ses compagnons le motif des pierres est absent
( X X X V I I I . 1 4). Le motif, on le voit bien, n'est pas sans fonction dans le
texte; comme ceci, Yar y figure en simple idoltre: . . . ipse edificavit
sacrarium B a a l . . . ( X X X V I I I . 1).
23a
L'article de P. M. B o g a e r t Sur la figure d ' A b r a h a m d a n s le L A B tait pour
moi inaccessible jusqu' l'apparition du prsent article (P. M. B o g a e r t L. D e q u e k e r
et al.: A b r a h a m dans la Bible et dans la tradition juive. Bruxelles 1978.
29
D'aprs J u b . 10 : 1 c'est encore aprs le temps de No que commencent se
propager, sous l'effet des dmons malins, les pchs le t e x t e interprte, cette fois
encore, le passage comme e x p r i m a n t le dtachement de Yahv.
30
L a suite d'ides commence en J u b . 10 : 11 ne se poursuit non plus. D ' a p r s
J u b . 11 : 1019 le b u t des hommes consiste galement m o n t e r a u x cieux.

Aprs ces vnements Y a h v conclue l'alliance avec A b r a h a m puisque

celui-ci ne s'est par rvolt contre la providence divine, contre la direction
divine d u destin des gnrations futures et, qui plus est, il a prouv par son
comportement la fermet de sa foi. Au lieu d'accepter le secours de Yectan, au
lieu d'chapper, la manire de ses compagnons la punition des rvolts il se
confie Yahv, la providence de celui-ci t a n t garantie et s'avrant la seule pos-
sibilit de se sauver : Ecce ego fugio hodie in montana, et si evasero ignem,
exient de montibus fere bestie et venientes comedent nos, a u t esce nobis defi-
cient et moriemur fame, et inveniemur fugientes ante populum terre, et caden-
tes in peccatis nostris. E t nunc vivit in quo confido, quia non movebor de
loco meo in quo posuerunt me. E t si fuerit aliquod peccatum meum consumens
u t consumar, fiat voluntas Dei (XI. 11).
L'alliance conclue ensuite avec A b r a h a m est d ' u n caractre autre que
celle conclue avec No. A b r a h a m , lui a mrit d'tre lu, a v a n t mme de conclure
l'alliance il en a r p o n d u aux exigences. 31 L'lection d ' A b r a h a m se double,
comme d'une ligne secondaire plus vague, de l'lection des fils de Sem, celle-ci
a y a n t toutefois une signification autre que pour la descendance d'Abraham, et
se rglant sur l'enseignement de l'pisode de Yectan et de l'histoire des com-
pagnons d'Abraham. Les deux exemples reprsentent les deux degrs de
l'observance de la loi, de la crainte de dieu ; Yectan, le chef des fils de Sem
cherche sauver A b r a h a m et ses compagnons le rle qu'il assume est celui
d u personnage qui p r t e aide sans participer activement d a n s l'tablissement
de l'alliance. Ayant reconnu que l ' a t t i t u d e d'Abraham est la seule qui soit
juste, il lui porte aide sans pour a u t a n t s'associer l'opposition d'Abraham.
L e cercle plus exclusif des douze dvots (symbole des tribus d'Isral auprs
de Y e c t a n symbolisant, lui les peuples smitiques) a t t e i n t un plus haut point
de l'observance de la loi. C'est de concert avec Abraham que ceux-ci se refusent
l'obissance signifiant l'adhsion aux rvolts. Tout en persistant dans leur
intention ils s'avrent incapables de suivre la seule voie juste. Leur erreur ne
signifie quand mme p a s une chute. L a passage final de l'histoire annonce leur
solidarit avec A b r a h a m dlivr de la fournaise ardente : Et surrexit Abraham
de camino, et concidit caminus ignis. E t salvatus est Abraham, et abiit ad unde-
cim viros qui erant absconsi in montanis, et renunciavit eis omnia que contin-
gerunt illi. E t descenderunt cum eo de montanis gaudentes in nomine Domini
et nemo eis obvians t e r r u i t die illa (VI. 18).
L a gradualit d u mrite se fait valoir, d'une manire analogue, aussi
d a n s la conclusion de l'alliance. Y a h v conclue alliance uniquement avec
A b r a h a m : Ibi ergo faciam inhabitare puerum meum A b r a h a m , et disponam
t e s t a m e n t u m cum eo, et semen eius benedicam, et nominabor ei Deus in eter-
31
D'aprs J u b . 11 : 16 24 c'est l'ge de 14 ans q u ' A b r a h a m p a r v i n t la con-
naissance parfaite de dieu. J u b . 13 : 26 donne une enumeration dtaille des lois et de
l'ordre des sacrifices q u ' A b r a h a m aurait observs.

num (VII. 5). Les trois degrs le plus h a u t en t a n t l ' a t t i t u d e et l'lection

d ' A b r a h a m r e n d e n t ncessaire une sorte de distinction de la providence
divine au profit des tribus de Sem et d'Isral, l'oppos d u reste coupable des
mortels.
L ' i n t e r p r t a t i o n que donne le Ps.-Philon de l'histoire d ' A b r a h a m a d e u x
m o m e n t s remarquables : le premier c'est la position que prend l'histoire elle-
m m e l'intrieur de la matire traite, le rle de l'alliance l'endroit d o n n .
C'est que, en conformit avec la conception du codex sacerdotal, 3 2 l'auteur d u
L A B met au centre, au lieu de l'alliance sinatique celle conclue avec A b r a h a m ,
en l'insrant d a n s une srie d'alliances, d o n t les tapes sont reprsentes p a r la
gense, le dluge et l'alliance de Y a h v avec Abraham, de la mme faon que
l ' a f f i r m a t i o n de Gn 17 : 8 : Et je serai leur dieu v a u t , d a n s la Bible encore,
p o u r la descendance d ' A b r a h a m . E t voil nous sommes arrivs au deuxime
m o m e n t , a u x degrs de l'lection. D ' a p r s le passage cit d e la Gense l'lec-
tion v a u t pour les descendants d ' A b r a h a m l'alliance se m o n t r a n t donc p l u s
complte que les antrieures ensuite la rdaction cite r d u i t l'alliance gale-
m e n t aux douze tribus, descendant de Jacob. 3 3
l ' h i s t o i r e des douze t r i b u s

P a r l'insertion de l'pisode d ' A b r a h a m et p a r l ' i n t e r p r t a t i o n de celui-ci
le Ps.-Philon fait suggrer l'lection collective des t r i b u s d'Isral et m m e
sans le dire expressment il tend l'alliance de Y a h v conclue avec A b r a h a m
l'Isral entier. A p a r t i r de ce m o m e n t la Bible remanie p a r le Ps.-Philon
se poursuit en histoire exclusivement juive, ses v n e m e n t s restant l'int-
rieur des ples Y a h v peuple lu, d e u x cts de l'alliance.
L'ide et le motif priodiquement rpt de l'alliance apparaissent p o u r
la premire fois d a n s l'histoire de la Gense pour r a p p a r a t r e dans les a u t r e s
livres bibliques, en r e f l t a n t une conception qui varie avec la d a t e de la compo-
sition des livres et en m e t t a n t en relief des aspects t o u j o u r s nouveaux d e
l'alliance. 3 4 L a tradition ancienne contient dj les bases de la vision historique
des poques postrieures : alliance, promesse de l'occupation de la terre d e
Canaan et du triomphe sur les peuples hostiles. 35 D a n s ces histoires le rdacteur
insiste t o u t le t e m p s sur la faiblesse humaine, sur le besoin de l'homme d ' t r e
aid la clmence divine succdant a u x pchs des mortels s'appuie de c e t t e
manire exclusivement sur la bienveillance du pouvoir surnaturel. Aux yeux des
prophtes l'alliance devient la hase d ' u n e suite analogique Isae, p. ex. rap-
32
E . A. S p e i s e r : I E J 7 (1957) 210. Sans quo le Livre des Jubiles ait mis en relief
l'alliance conclue avec A b r a h a m , le passage suivant la mort de celui-ci n'en semble p a s
moins, ici encore, m a r q u e r une limite (23 : 1).
33
Test. Simeon insiste galement sur l'lection dos fils de Sem l'oppos de l'assu-
jetissement des fils de H a m (6).
31
M. B u r r o w s 1966. 102.
33
Un exemple classique en est D e u t . 26 : 9.

pelle, hors la sortie d ' g y p t e , autres cas de la dlivrance. (Is 1 : 9). E n mme
t e m p s l'alliance s'exprime dsormais d a n s les notions de la promesse et de
l'accomplissement. A l'effet des v n e m e n t s des poques assyrienne et nobaby-
lonienne, des guerres e t de la captivit, la conception de l'alliance dans la tra-
dition biblique sera domine p a r l'autocritique. Les p r o p h t e s accusent le
p e u p l e des pchs anciens et rcents la fois. 36 L'esprance en l'accomplisse-
m e n t a p p a r a t pour la premire fois sous u n aspect nouveau. Qu'on pense
la juridiction divine (c'est dire p a s une dlivrance collective, mais u n
j u g e m e n t plus individuel, supposant une sorte de responsabilit individuelle
des pchs) figurant d a n s la 17 e parabole d'zchiel. L a conception de la vision
d'histoire deutronomique insiste d ' u n e faon encore plus marque sur les
pchs p o u v a n t dcouler des r a p p o r t s tablis avec les peuples trangers
( J g 2 : 1 3). C'est ici que s ' a j o u t e l'ide de l'lection celle de l'isolation d ' a v e c
les a u t r e s peuples. L ' a t m o s p h r e des livres composs aprs la captivit exprime
la joie d'tre sauv ce qui s'exprime chez Is. 5 2 : 4 p a r l'analogie entre la
dlivrance de l ' E g y p t e et celle de B a b y l o n e s ' o f f r a n t d'elle-mme. Auprs de
la dlivrance un des lments constitutifs de l'alliance une a u t r e ide
celle d u service du Dieu a u r a dsormais u n rle plus m a r q u d a n s la pense d u
j u d a s m e post-exilique (p. ex. Is. 52 53). L'histoire ancienne sera rapprcie,
la conception antrieure, qui considre le pass comme l'enchanement des
peines cde la place celle qui tient les souffrances, les dsastres comme pr-
p a r a t i o n d ' u n e mission spciale du peuple lu remplir d a n s le monde. La rdac-
tion sacerdotale (P), crite au 6 e sicle, r e m e t t a n t la m a t i r e d u texte de la
Bible p r e n d l'alliance p o u r constante e t indissoluble; c'est cette source qui
d ' u n e faon analogue l'histoire d ' A b r a h a m cite plus h a u t met au centre
non l'alliance sinatique, mais celle conclue avec Abraham. 3 7
C'est partir des lments de conception des priodes antrieures que
le L A B de mme que la conception contemporaine d'alliance cre son
ide d'alliance lui qui, pareillement la rdaction sacerdotale domine l ' u v r e
entire et en assure m m e la cohrence. Au lieu des ides religieuses plus gn-
rales d o n t la relation d u monde et de l ' h o m m e avec la sphre divine le
Pseudo-Philon met en relief celles qui sont spcifiquement juives ; chez lui
la premire place revient au r a p p o r t d u peuple juif avec Yahv, dieu d'Isral.
Ce rapport-ci est en e f f e t la base laquelle t o u t se laisse comparer, autour de
laquelle t o u t se laisse regrouper. U n e conception qui d'ailleurs n ' e s t pas unique-
m e n t celle du L A B : d a n s cette priode t a r d i v e du judasme elle prend une
i m p o r t a n c e remarquable. 3 8 Chez le Pseudo-Philon elle f a i t dominer l ' u v r e
entire p a r l'interprtation de l'ide de l'alliance. D a n s la relation Yahv
36
M. B u r r o w s 1966. 11126.
37
E . A. S p e i s e r : I E J 7 (1957) 210 ; M. B u r r o w s 1966. 124.
38
G. F . M o o r e 1950. I. 219234 ; R . B u l t m a n n 1960. 72.

peuple lu la notion de dieu se voit, elle aussi comme largie : Ayant q u i t t

le rle de dieu des tribus israliens J a h v apparait comme un Deus, dieu
omnipotent qui a lu les tribus d'Isral afin de pratiquer la charit exclusive-
ment avec elles. Le ple de Y lection se complte cependant de celui de la loi :
il leur fit encore don de la loi pour que, en observant celle-ci elles puissent
mriter la providence divine. Cette ide est toute courante dans les uvres
littraires de l'poque de la composition du LAB, 39 alors que la notion de la loi
se voit, elle aussi, largie pour devenir une notion surtemporelle, valable donc
pour l'ternit. P a r consquance et en partie par suite du contenu chang
des notions la conscience religieuse concentre autour des ides de l'lection
et de la loi mme encore plus loin p o u r projeter l'lection sur l'avenir. Voil
le b u t auquel seront rattaches toutes les ides de cet ordre et qui signifie en-
core la base de la conception tlologique qui se trouve chez le Ps.-Philon. L a
projection de l'lection sur l'avenir voil en quoi consiste en gnral la foi
des Juifs d'aprs la guerre ayant, pour moyen de raliser la foi, l'observance
de la Loi. Les deux ples de la conception primitive se restituent sur un autre
plan runissant, lui, les ples de Y observance de la loi et de Y attente de la rcom-
pense. La valeur des notions de loi et d'observance de la loi se trouve, elle
aussi modifie. La loi ne signifie plus les recueils de loi bibliques : la notion
va se concentrer sur quelque chose d'origine divine ordonne par dieu et trans-
mise par Mose (Esd 7 : 6). La formation de cette conception de la loi s'attache
la tendance nomistique de la priode qui succda la captivit de Babylone. 4 0
Le changement notionnel des catgories de l'observance et du respect de la
loi se traduit par la canonisation d u respect de la loi, p a r la reconnaissance
officielle de son origine divine, cette dernire tant d'ailleurs justifier partir
des sources bibliques galement (p. ex. Esd 7 : 6, Ne 8 10). L'interprtation
de la loi fournie par le Livre des Jubils reprsente une tape antrieure mais
qui fait dj penser au LAB. L a loi est ici aussi intemporelle (les patriarches
observent eux-mmes les prescriptions de la loi) 41 et malgr les explications
concrtes de lois que comporte le t e x t e on sera enclin voir, dans le choix
des lois expliquer un signe de l'abstraction, ces lois se rattachant toutes,
l'office divin et ayant par consquent un caractre p l u t t thologique que moral.
Chez le Ps.-Philon la loi a p p a r a t sous une forme t o u t abstraite. U n m o m e n t
caractristique et qui reprsente fidlement la conception religieuse de l'poque
c'est que, en r a p p o r t a n t la transmission de la loi le Ps.-Philon complte de
maints pisodes le passage narratif de l'histoire (XI. 1 15) alors que, en expo-
39
D ' a p r s J u b . 13 : 26 la mrite do l'lection se fonde galement sur l'observance
de la loi. C'est justement p a r leur citation pormanente que la loi et l'alliance deviennent
intemjxnelles et peuvent p a r consquent t r e remontes a u p a s s ; cf. R . B u l t m n a n
1960. 72.
40
H . S t r a c k : Einleitung in T a l m u d und MidraS. Leipzig 1921. 195.
41
J u b . 13: 26.

364 I. FRHLICH
s a n t le contenu de la loi il se rduit ces prcisions les plus courtes possibles et

ne d o n n e qu'un rsum bien sommaire des commandements et de l'ordre des
sacrifices. La base idelle de cette nouvelle conception de la loi est fournie p a r
D t 6, o il est dit que l ' a m o u r de Yahv et l'observance de ses lois la crainte
d u dieu sont insparables ; aimer Y a h v et suivre son chemin, c'est la
m m e chose. Or, d a n s la situation d ' a p r s la guerre des J u i f s lorsque le temple
est dtruit, le culte, supprim, la seule force de cohsion c'est en effet la loi,
ce que semblent indiquer telles manifestations religieuses de l'poque. 4 2 D'aprs
la nouvelle conception de loi du judasme c'est justement le respect de la loi
qui p e u t tre le moyen d'accomplissement des esprences eschatologiques des
J u i f s et c'est ce q u ' e n t e n d viser le respect officiel de la loi. Le passage o le
Ps.-Philon dcrit la transmission des lois de Mose exprime la mme ide :
Et adiecit Dominus : A d h u c ostendi vias paradysi... ( X I I I . 9).
Le dpart mme d u motif de l'alliance tmoigne d a n s le LAB, de la
nouvelle conception de la relation acte foi (observance de la loi a t t e n t e
de l'avenir) e t d e la nouvelle interprtation base sur celle-ci de l'alliance.
L a raison pour laquelle Y a h v lit pour peuple favori la tribu de Sem, c'est l'at-
t a c h e m e n t d'Abraham la loi. Dans l'esprit de la pense biblique la mention
de la loi propos d ' A b r a h a m serait un anachronisme. Bien sr, le Ps.-Philon,
lui, parle aussi en d ' a u t r e s termes, mais la pit d'Abraham n'en signifie p a s
moins l'observance d u premier commandement de la loi de Mose.
A la couple Y a h v peuple lu, force motrice de l'histoire des douze
t r i b u s s'ajoute un troisime lment, compltant et runissant les deux pre-
miers : il s'agit du territoire accorder Isral qui figure d a n s l'alliance comme
un lment de promesse. Chez le Ps.-Philon cet a t t a c h e m e n t territorial est
b e a u c o u p plus troit qu'il ne l'est dans l'Ancien Testament. L e LAB ne rap-
pelle que d'une faon vague les passages succdant l'migration d'Abraham
(l'arrive d'Abraham en Canaan, la sparation de Lot, la strilit de Sara, t o u t
cela n'est que vaguement abord, le sjour en E g y p t e et l'histoire de Hagar
t a n t entirement absents V I I I . 1 3). D a n s le passage en question l'im-
p o r t a n c e de l'alliance est la seule tre souligne par ceci q u ' a v a n t la concep-
tion et la naissance d ' I s a a c le texte en fait mention une nouvelle fois l'ide
o b t e n a n t par l une place de choix dans la suite des vnements et la naissance
d ' I s a a c devenant t o u t n e t t e m e n t le sceau de l'alliance. (VIII. 3). Ceci revient
dire que, par la concentration du texte le Ps.-Philon t r a n s m e t l'accent sur le
r a p p o r t rciproque entre la naissance d ' I s a a c et l'alliance, m e t t a n t ainsi en
relief la famille d ' A b r a h a m et avanant, p a r la concentration le motif de la
terre promise.
P o u r ce qui est de l'histoire de J a c o b l'auteur n'en entre non plus dans
les dtails ayant d u reste la tendance remanier ou o m e t t r e tout simple-
42
G. F. M o o r e : 1950. 267 280.

m e n t les parties narratives. Or, l'omission de l'histoire de J a c o b ne reste p a s

sans se faire sentir dans une partie postrieure du texte, et prcisment d a n s la
description de la lutte entre David e t Goliath : parmi les noms crits par D a v i d
sur les lance-pierres se retrouvent en effet ceux d'Isaac et de J a c o b (XI. 5).
Les noms crits sur les pierres sont les symboles de l'aide divine offerte p a r
l'alliance ce qui revient dire que dans l'alliance Isaac et J a c o b o n t le
mme rle prfrentiel que p. ex. A b r a h a m ; si leur histoire est quand m m e
omise par l'auteur, c'est pour que l'unit notionnelle de peuple lu terre
promise soit par l assure et que l'accent soit dsormais mis uniquement sur
l'ordre d'ides de l'observation de la loi providence divine, soit l'espoir
fond sur l'avenir.
Quant la priode qui s'coula depuis la naissance d'Isaac jusqu' la
captivit en g y p t e le Ps.-Philon ne l'indique que sous forme de tableaux
gnalogiques et largissant ceux d ' s a et des fils de J a c o b en r a t t a c h a n t
l'histoire de Dina le rcit du mariage de Dina avec J o b (Gn 34) ; de cette faon
la descendance de Jacob se voit enrichie des enfants de Dina (VIII. 7 8). 43
Le passage servant d'introduction l'occupation de la terre promise e t
qui contient une description assez dtaille des vnements s'tend de la priode
d ' a v a n t la naissance de Mose j u s q u ' Josu et la construction du sanctuaire
de Gilgal (IX. 1 X X I . 7). Le L A B fait prcder ce passage d ' u n dtail inter-
cal et qui parat trs important d u point de vue de l'alliance : il s'agit du dis-
cours que prononce Amram, pre de Mose, au moment o le pharaon prescrit
le meurtre des enfants des Juifs ( I X . 3 6). Amram se refuse luder le dcret
de cette faon que le lui conseillent les vieux (que chacun vive sparment
de sa femme : ainsi il n'y aura p a s d ' e n f a n t s tuer). A m r a m est le seul avoir
confiance en la providence divine : Et respondit A m r a m et dixit : Celerius
est u t in victoria minuatur seculum a u t in immensurabile m u n d u s indicat a u t
cor abissi astra contingat, quam genus filiorum Israel minuatur (IX. 4).
L a foi d'Amram est elle-mme fonde sur des analogies ; il rappelle le cas de
T a m a r qui se fit aider elle aussi p a r le dieu pour sauver sa famille (l'histoire
de T a m a r , elle, Gn 38 ne figure p a s dans le LAB) : Et salvavit earn con-
silium eius ab omni periculo. Nunc ergo et nos faciamus sic. E t erit cum com-
pletum fuerit tempus parturitionis, si potuerimus, non proiciemus f r u c t u m
ventris nostri. E t quis seit si pro hoc zelabitur Deus, ut liberaret nos de humilia-
tione nostra? (IX. 5 6). Dans le discours d'Amram on aperoit la mme
conception tlologique que chez les insurgs de l'histoire de Babel (VI. 1)
cette diffrence prs que, cette fois elle sert exciter accomplir l'avenir pr-
dit d a n s la rvlation divine. A m r a m mrite l'lection t o u t aussi bien qu'Abra-
ham encore que l'accomplissement n'en ait lieu que sous son fils Mose.
43
Le passage intercal reflte u n e tendance d'hostilit envers les Samaritains, cf.
A. SPIRO : l ' A A J R 20 (1951) 319.

L'accomplissement succdant l'acte reprsente, d a n s la structure du

L A B la premire analogie avec l'histoire d ' A b r a h a m , avec l'exemple de celui-ci
persistant dans sa foi e t se trouvant sauv. A leur tour l'histoire d'Abraham
et celle d'Amram serviront, l'intrieur du L A B de base p o u t toute une srie
d'vnements analogique afin d'exprimer la valeur et la force permanentes de
l'alliance. Signalons q u e l'histoire d ' A b r a h a m et celle d ' A m r a m illustrant dans
le L A B le caractre d e l'alliance la force de l'alliance sauvant des mains de
l'ennemi ceux qui persistent dans leur foi ne figurent pas dans la Bible.
L'accomplissement a lieu par le moyen d'une forme intermdiaire par
u n e srie d'pisodes : il s'agit du rve de Miriam sur la naissance de Mose et
ses proprits extraordinaires, ensuite de l'activit entire de Mose, l'exode, la
lgislation le cours des accomplissements y t a n t interrompu par l'histoire
d u veau d'or. Chez le Ps.-Philon chacun des deux lments l'importance de
la lgislation et l'idoltrie a une porte plus grande que d a n s l'Ancien Testa-
m e n t . Le rcit de Sina comporte d a n s le L A B plusieurs lments nouveaux :
les phnomnes n a t u r e l s accompagnant la transmission de la loi, l'apocalypse
de Mose sur le m o n t , ensuite la splendeur du visage de Mose rentr du mont et
le changement de son extrieur (XI. 4, X I I . 1) soit la punition divine prc-
d a n t le motif de l'crasement du veau d ' o r et celui de l'ingurgitation de l'eau
c o n t e n a n t ses m o r c e a u x : la disparition de l'criture des tables de pierre
( X I I . 5). La punition divine, au lieu d ' t r e un simple lment de supplment
pose cette fois sur de nouvelles bases le contenu de l'alliance entre le dieu et son
peuple lu. La disparition de l'criture signifie la cessation de l'alliance et la
rvocation de la loi q u e Yahv va lui-mme renouveler sur les instances de
Mose (XII. 8). Le caractre du discours de Mose diffre lui aussi et d'une
faon remarquable d e celui qui figure d a n s l'Ancien T e s t a m e n t : ici Mose
reconnat le pch d u peuple et en d e m a n d e pardon ou p l u t t il fait l'offre de
l'expier en personne p o u r obtenir ensuite la promesse concernant l'occupation
de la terre de Canaan (Dt 32 : 31 35, 33 : 1 2) ; chez le Ps.-Philon cepen-
d a n t Mose, dans la parabole sur le cep commence p a r parler de la providence
divine pour en venir l'autre aspect de l'alliance : Y a h v a lui-mme besoin
de ceux qui le glorifient aussi bien que le peuple a besoin de lui : Tu es enim
qui omne lumen es, e t domum tuam ornasti lapidibus preciosis et auro et aro-
m a t i b u s . . . decorastis domum t u u m , et in diversis escis, et suavitate diversi
p o t u s saturasti c r e a t u r a m tuam. Si ergo misertus fueris vinee tue, omnia Domi-
ne in vano facta s u n t , e t non habebis qui te glorificaret. N a m etiam aliam vine-
a m plantaveris, nec hec tibi eredet, eo quod priorem dissipasti. Si enim relin-
ques, relinqueris in seculum ( X I I . 9). L'alliance de Y a h v conclue avee
Mose, casse donc chez le Ps.-Philon d ' t r e base sur la bienveillance de la
puissance surnaturelle e t sur la faiblesse humaine et a formellement le carac-
tre d'un trait. L'ide d u trait entre dieu et homme n'est pas un phnomne
t o u t neuf l'Orient a n t i q u e : elle remonte en effet la formule des traits poli-

H I S T O R I O G R A P H I E ET AGGADA 367
tiques officiels. D a n s la Bible l'alliance conclue avec Yahv se rgle aussi chaque
fois sur cette formule : la conclusion effective de l'alliance est introduite p a r
un prlude et un aperu historique. 44 D a n s le LAB l'ide d u commencement de
l'alliance implique d'elle-mme le caractre de trait sans l'expliquer nette-
m e n t : ce n'est que plus tard, travers des cas analogues priodiquement rp-
ts (l'acte de mriter l'alliance signes assistance) que devient manifeste
l'ide de trait pour devenir en mme t e m p s le propre de la technique de com-
position de l'uvre.
Le moment o Mose mrite l'alliance (par la suite nous allons e n t e n d r e
p a r mriter l'accentuation de la force de l'analogie en t a n t qu'antcdent d u
renouvellement de l'alliance) est suivi p a r l'accomplissement, par la transmis-
sion de la loi. L a rptition priodique de la suite alliance (aussi sous la forme
de l'vocation de l'alliance, donc d ' u n e srie analogique) signes intercala-
tion ventuelle des motifs du pch et de la punition accomplissement de la
promesse, divise en des priodes systmatiquement revenant l'histoire des douze
tribus. 4 5 Le texte du Pseudo-Philon ne s'loigne nulle p a r t outre mesure de
celui de la Bible : il ne fait que runir les rcits bibliques ou les regrouper de
faon que la tendance en question soit plus f r a p p a n t e (p. ex. la rvolte de K o r a h
se trouve chez lui troitement lie a u x vnements accompagnant la transmis-
sion de la loi XVI. 1).
Cette vision priodique de l'histoire dont les bases sont bibliques, mais
l'laboration est toute spciale et particulire au LAB annonce la vision
historique eschatologique dominant le t e m p s de composition de l'uvre, vision
d'aprs laquelle c'est l'apoge du processus priodiquement rpt de l'histoire
que signifieraient la venue du Messie et l'avnement d e la thocratie, soit le
jugement dernier, suivi par le renouvellement du monde c'est--dire p a r la
rptition de la priode passe, au f o n d statique, u n niveau plus haut. 4 8
Sous cet aspect c'est le Livre des Jubils qui est le plus rapproch de la concep-
tion du L A B : dans ce livre-l l'auteur, en remaniant la matire de la Gense
recre l'histoire p a r le moyen des priodes Ybl d'origine traditionnelle. ( E n t r e
les deux crits il y a maintes autres concidences ; dont la forme de chronique
44
J . B e g r i c h : Berit. ZAW 60 (1944) 1 11 ; J . M u i l e n b e r g : The F o r m a n d
the Structure of t h e Covenant Formulation. V T 9 (1969) 347 365.
45
La tendance de priodisation, c o m m e n a n t au 2e sicle a v . notre re n ' e s t p a s
sans laisser son empreinte sur presque t o u t e s les uvres pseudpigraphiques et a p o -
cryphes. L'histoire sora diviso on 12, 10 ou 7 priodes. Division en douze : 4" F s d r a s
14 : 11 et suiv. ; Baruch, syriaque 5 3 7 2 ; l'apocalypse d ' A b r a h a m 29. Division en
d i x : Henoch 9 3 : 1 1 0 ; 9 1 : 1 2 1 7 ; Oracula Syb'illina 4 : 4 7 et s u i v . ; 4 E s d r a s
14 : 11. Division en sept : Hnoch (slave) : 33 : 1.
46 e
4 Esdras 7 : 50 s'attend le t o u r n a n t sparant les deux poques dfinies lors
de la cration d u m o n d e et qui va survenir d a n s un proc he avenir ; d'aprs le B a r u c h
syriaque 4 9 : 51 la nouvelle poque p o r t e r a de la gloiie. De la mme faon B a b .
Berakot 17a.
24Arta Antiqua Academiae Scientiarum Hungaricae 28, 1980

368 I. FRHLICH
qui, d a n s les deux cas est largie de tableaux gnalogiques au dbut de chacune
des uvres.) 47
Le processus historique priodiquement rpt d a n s la conception d u
Ps.-Philon, l'histoire des douze tribus est dirige par l'omnipotence divine
celle-ci aidant chaque fois les tribus d'Isral contre les ennemis. L a conclusion
de l'alliance avec Mose signifie non seulement le renouvellement des alliances
antrieures (XII. 8 X I I I . 10) mais aussi la garantie de l'lection pour les tri-
b u s . Selon les paroles de Yahv : Sciens a u t e m scio quoniam corrumpent vias
suas e t relinquam eos et obliviscentur t e s t a m e n t a que disposui patribus eorum,
et ego tamen non in sempiternum obliviscar eos ( X I I I . 10). Ce n'est pas d ' u n e
f a o n directe que la providence divine rgit le cours de l'histoire des tribus il
y a u r a des pchs et des punitions il n ' e n reste pas moins qu'il est dirig p a r
le secours divin systmatiquement apparaissant.
L'histoire de Mose et l'histoire nationale y succdant sont embrasses
p a r les motifs relevant de cette vision, motifs qui leur assurent en mme t e m p s
u n e isolation formelle d u reste de la matire d u LAB. L a providence divine se
ralise surtout d a n s les qualits charismatiques du chef d u peuple. Cette nou-
velle relation a ses nouvelles expressions symboliques : il s'agit de ces motifs
q u i dominent l ' u v r e entire ou s'y r p t e n t (gnralement trois fois). Ayant,
c h a c u n le mme sens ils symbolisent la force divine attribue au chef, la force
d e l'alliance voil pourquoi il se trouve que, au cours de l'histoire ils se
s u b s t i t u e n t l'un l ' a u t r e ou qu'ils s'assimilent l'un l'autre alors que, l'int-
rieur de chacun des passages du texte, l ' a u t e u r emploie, pour la plupart du
t e m p s les mmes motifs.
U n motif qui reste dominant t o u t le long du texte, c'est le sceptre de l'al-
liance. Son acte de donation tant pass sous silence il figure la premire fois
lors d u passage de la mer Rouge. (X. 5). Ici le sceptre pareillement en ceci
la conception de la Bible (Gn 14 : 16) symbolise la puissance miraculeuse,
d'origine divine de Mose : c'est grce au sceptre que celui-ci russit fendre
en d e u x les eaux de la mer. Le sceptre de Mose, en t a n t que symbole constant
d e l'alliance est a t t e s t une deuxime fois d a n s l'histoire de Sina du L A B :
Et nunc virga t u a in qua facta sunt signa, erit in testimonium inter me et
populum meum, e t erit cum peccaverint irascar eis, et memorabor virge mee, et
p a r c a m eis iuxta misericordiam meam. E t erit virga t u a in conspectu meo in
commemorationem omnium diebus, et similabitur arcui in quo disposui testa-
m e n t u m ad Noe cum exiret de arca ( X I X . 11).
La force divine symbolisant la matrise sur le peuple continue se trans-
m e t t r e par la personne de Mose. Aprs la mort de celui-ci Josu sera lu chef
47
M. r . J a m e s v e u t que l'auteur d u L A B ait connu le Livre des Jubils, les
t a b l e a u x gnalogiques de celui-l t a n t , p a r leurs s t r u c t u r e s troitement a t t a c h s
c e u x d u Livre des J u b i l s . Les donnes fictives de ce dernier s ' y trouvent compltes
d ' a u t r e s donnes d u m m e t y p e .

historiographie et aggada 369
( X X . 1). A la diffrence de la Bible le Ps.-Philon n'en ajoute pas moins l'in-

stallation de Josu le motif de l'obtention de la force et sans mentionner, il est
vrai, le sceptre, il y parle quand mme de l'hritage de Mose, du pouvoir divin
assur par le sceptre : . . . et dixit Deus ad I h e s u m : U t quid luges e t u t
quid speras in vanum, quod Moyses adhuc vivat? E t ideo superflue sustines,
quoniam defunctus est Moyses. Sed accipe vestimenta sapientie istius et indue te et
zona scientie eius precinge lumbos tuos, e t i n m u t a b e r i s e t cris in virum alium
Nonne pro te locutum sum servo meo Moysi dicens : I s t e ducabit populum post
te. I n manum eius t r a d a m reges Amorreorum ( X X . 2). Les symboles de l'hri-
tage de Mose transmis Josu, les vestimenta sapientiae et la zona scientiae
sont, dans la conception religieuse contemporaine des Juifs, ceux du savoir,
non pas du savoir abstrait, mais bien de la morale pratique, de la connaissance
des voies, des choses de monde, de la distinction d u vrai d'avec le faux bref,
de l'emploi pratique de la loi que le texte exprime d ' u n e faon qui est celle du
symbolisme essnien. 48 Aussi le terme de vestimenta sapientiae est-il par-
faitement compatible avec les qualits du chef du peuple, qualits qui, d a n s la
pense du Ps.-Philon relvent sans cela de la direction spirituelle, de la direction
de l'observance de la loi. 49
La force divine accorde aux fins de la direction et semblable celle des
prophtes a p p a r a t une nouvelle fois propos de la personne de K e n a z (en
latin : Cenaz) (XXV. 2 X X V I I I . 10). Sa prsence d a n s le texte p a r a t en
elle-mme assez trange : la Bible n'en fait q u ' u n e vague mention, en conne-
xion de son fils (Jg 1 : 13), or, dans le LAB il a un rle de poids. Le rcit sur
Kenaz figure chez le Ps.-Philon dans la tradition d u Livre des Juges, aprs
le rcit sur la m o r t de Josu. L ' a u t e u r du LAB fait figurer la personne de K e n a z ,
un chef nouveau, le successeur d u chef dfunt, qui a les qualits et la popularit
de Josu. 50 Voil pourquoi largit-il l'histoire de K e n a z par des pisodes lgen-
daires, dont le motif de la force divine : a v a n t d ' e n t r e r en lutte contre les
Canaanites Kenaz se trouve possd par une force divine : Et factum est u t
audivit Cenaz verba eorum indutus est spiritus virtutis et transmutatus est in
virum alium, descendit in castra Amorreorum et cepit percutare eos ( X X V I I .
10).
Un autre exemple de la force divine attribue au chef du peuple est fourni
par l'histoire de Gdon qui reprsente, elle, une dernire tape de la transfor-
mation successive du motif. Gdon est, lui aussi, lu : Et nunc veni, m i t t a m
te et liberabis Israel de manu Madianitum. . . N u n c ergo vade, et cinge lumbos
tuos, et Dominus erit tecum ( X X X V . 45). C ' e s t s e u l e m e n t sur sa p r o p r e d e -
48
D'aprs Ps. Sal. 22 le Messie de David arrivera cercl de force.
49
La Mishna, elle, divise en deux le motif f i g u r a n t avec deux sons chez le Ps.-
Philon ; Josu d u t son savoir au sceptre de Mose d o n t il tait en possession; J e r .
Nedarim 9, 91B.
50
La m a t i r e d u Livre des J u g e s ne concordant pas, pour la plupart, a v e c la
conception postrieure, il resta sans tre propag, cf. A. S p i r o : P A A J R 20 (1951) 298.
24 Arta Antiqua Academiae Scientiarum Hungaricae 28, 1980

m a n d e que Gdon obtient les signes de son lection : Ecce Moyses primus
o m n i u m prophetarum peciit a Domino signum, et datum est ei. Ego autem
quis s u m , nisi forte se elegit me Dominus, d e t mihi signum, ut sciam, quoniam
dirigor (XXXV. 6).
P o u r ce qui est de J e p h t c'est le peuple mme qui est d o n a t e u r du pou-
voir d e chef que celui-ci se refuse d ' a b o r d accepter : Et f a c t u m est dum
e x p u g n a r e t u r Israel, venerunt in terram Tobi ad Ieptan, et d i x e r u n t ad eum :
Veni principare populo ( X X X I X . 3). Que J e p h t se charge d ' t r e le chef de
l ' a r m e dirige contre les Ammonites, cela est d uniquement la d e m a n d e du
p e u p l e lui-mme a y a n t peur de la colre de Y a h v cause des pchs d u peuple.
Le p e u p l e rappelle J e p h t la vocation d'Isral ce qui ne l'empche pas
d'insister, dans son argumentation, au mpris de l'importance de la personne
du chef sur la force de l'observance de la loi, force qui assure la plus grande
puissance : Nunc ergo ponite cor vestrum in lege Domini Dei vestri, et depre-
cemur unanimes et sic pugnabimus dversus inimicos nostras, confidentes et
s p e r a n t e s in Domino, quoniam non t r a d e t nos usque in finem ( X X X I X . 6).
D a n s la suite, la force divine du chef n ' a p p a r a t que priodiquement dans
les rcits sur les juges : Et penituit D e u s ire sue, et confortavit spiritum
lepte ( X X X I X . 8). Sur la personne d ' A d d a , ce juge succdant J e p h t , le
Ps.-Philon ne fait aucune observation et, q u a n t la lgende de naissance de
Samson il se contente de remarquer propos de sa personne : Et e r a t Dominus
cum eo (XLIII. 1).
L'apparition du secours divin travers la personne du chef va donc en
se s u p p r i m a n t alors que les pchs se propagent; les juges c o m m e t t e n t eux-
mmes des fautes contre l'alliance, tel J e p h t dont le pch le v u auda-
cieux e t le sacrifice de sa propre fille se trouve, chez le Ps.-Philon particu-
lirement mis en r e l i e f e t comdamn ( X X X I X . 11).
A u cours de la priode allant j u s q u ' D a v i d Israel est sans chef : Et in
diebus illis non erat d u x in Israel, sed faciebant unusquisque que placita erant
a n t e conspectum eorum (XL1V. 1). L'exigence d'avoir un chef n'en reste pas
moins vive et pourra m m e prluder, d ' u n e faon efficace l'histoire de David,
la confrontation de D a v i d avec la royaut de Sal : Et in t e m p o r e illo cepe-
r u n t filii Israel requirere a Domino, et d i x e r u n t : Sorciamur omnes nos et
v i d e m u s quis sit qui possit principari nobis sicut Cenez (XLIX. 1).
Aprs la mort de Samuel c'est D a v i d qui sera lu chef: en connexion
avec l'apparition de sa personne rapparaissent dans l'uvre les signes miracu-
leux particuliers au chef politique du peuple mais, cette fois sous la forme d'un
motif nouveau qui, jusque-l servit exprimer le secours divin direct. Avant
d ' e n t r e r en lutte avec Goliath David, par le moyen des noms crits sur les
lance-pierres demande secours Yahv t o u t en lui rappelant la force de l'al-
liance ( L X I . 5). Au cours du combat il sera aid par un ange, force divine per-
sonnifie : Angelus a u t e m erexit faciem D a v i d , et nemo agnoscebat eum. E t

H I S T O R I O G R A P H I E ET aggada 371
videns Saul David, interrogavit eum quid esset, et non erat qui cognosceret
cum (LXI. 1). Le fait que la figure de David devient indiscernable des mortels
est cependant un analogue do l'pisode sinatique de la transmission de la loi :
ici c'est le visage de Mose qui, aprs l'acceptation des tables reut une splen-
deur d'origine divine telle que le peuple lui-mme ne p t plus le reconnatre
(XXV. 1 et E x . 3 4 : 2 9 - 3 5 ) . 5 1
Auprs des principaux symboles de la force divine d u chef et qui sont
attests tout le long de l'uvre apparaissent d'autres motifs, un peu moins
marqus, avec des sens ventuellement diffrents et qui, au lieu de dominer
l'uvre entire n ' y sont prsents qu'en un nombre dfini (gnralement trois
fois); indiquons, en outre ces parties qui sont la rptition ou la paraphrase
de tels passages antrieurs (des discours par exemple) et o n t pour fonction de
souligner la force de l'analogie.
Cependant la royaut de Saiil qui prcda celle de D a v i d apparat, dans
le LAB, sous un aspect t o u t autre. Ce portrait de Saiil diffre essentiellement
non seulement de l'idal du chef dessin dans le LAB, mais aussi de celui de
la Bible. D'aprs le Ps.-Philon la priode prcdant la royaut de David signi-
fierait l'chec de l'alliance d aux pchs du peuple. (Pour faire dcor la
situation extraordinaire voil la description du double chtiment pour l'ido-
ltrie de Mika soit pour le meurtre de la concubine du lvite X L V I I . 712).
E n mme temps le peuple commence sentir le besoin d u retour du secours
divin en la personne du chef. Or, l'lection d'un nouveau chef, le peuple ne l'a
p a s encore mrite : c'est ce qui signalent l'insuccs du tirage au sort, et aussi
ce fait que le choix se p o r t a n t sur E l q a n a (dans la t e x t e latin : Helcana)
celui-ci, tout en connaissant les intentions de Yahv se refuse tre le chef,
justement cause des pchs du peuple : Ego non possum dux esse super
populum istum, estimare quis possit esse vobis in ducem. Sed si peccata mea
comprehenderunt me u t supervolaret me sors, ego me interficiam, u t non me
contaminetis. I u s t u m est enim me pro peccatis meis t a n t u m m o d o mori, magis
quam sustinere pondus populi ( X L I X . 5). Samuel, fils d ' E l q a n a dont la
naissance est antrieurement rvle par Yahv est le seul avoir le privilge
de sacrer le f u t u r roi. Aprs la description de l'enfance de Samuel, de la lutte
contre les Philistins, de la m o r t des fils d'li et de la victoire dcisive le pro-
blme de la royaut se trouve de nouveau au premier plan : Et in illo tempore
concupierunt filii Israel petentes regem, et congregati s u n t ad Saulem et dixe-
r u n t : Ecce nunc tu senuisti, et filii tui non a m b u l a n t in viis tuis. E t nunc
constitue super nos regem, qui nos diiudicet, quoniam completum est verbum
quod dixit Moyses patribus nostris in heremo dicens : Constituendo constitue
de fratribus tuis super te principem (LVI. 1).
51
Pour los parallles v. H . S t r a c k P . B i l l e r b e c k : K o m m e n t a r z u m N e u e n
T e s t a m e n t . IV. 2. M n c h e n 1922. 298.

372 I. FRHLICH
Mais, l'intrieur de l'histoire de la royaut le Ps.-Philon fait opposer

d ' u n e faon marque le rgne de D a v i d celui de Saiil. Le p o r t r a i t ngatif que
d o n n e le Ps.-Philon d u rgne de Saiil, est loin d'tre une sorte de propagande
contre la royaut; il sert p l u t t m e t t r e en relief l'lection de David. D'aprs
les paroles de Samuel : Ecce nunc video, quoniam non est adhuc tempus
regnandi nobis in sempiterno et edificare domum Domini Dei, petentibus regem
a n t e tempus (LVI. 2). L a fin tragique du rgne de Saul est pralablement
dtermine par Y a h v : Mittam enim eis regem qui eos exterminet, et ipse
postea exterminabitur (LVI. 3). Aprs quoi Samuel lui-mme se dsolidarise
d'avec le nouveau roi.
Quant la figure de Saiil le Ps.-Philon y apporte des modifications con-
sidrables par r a p p o r t d u texte de la Bible. E n r a p p o r t a n t sa naissance il
omet tous les passages qui rappellent l'origine benjamite de Saiil pour lui impro-
viser l'origine phramite : . . . Saul filius Cis veniebat de m o n t e E f f r a i m . . .
(LVI. 4). Du point de vue de la conception historique du L A B cette indication
de l'origine de Saiil a en elle-mme une grande importance. C'est qu'en la figure
de Saiil le Ps.-Philon nous dpeint le Messie ben p h r a m d o n t la venue est,
d a n s les ides messianistiques des J u i f s antrieure celle d u Messie davidique. 52
P a r sa personne, se voit justifie cette conception toute reue d a n s les ides d'his-
toire escliatologiques d ' a p r s laquelle le temps de la rdemption sera prcd
d ' u n e priode pire que t o u t e autre et la dernire tape de celle-ci sera marque
des pertes et des guerres. 5 3 L'existence de cette conception chez le Ps.-Philon
est bien mise en vue p a r un passage antrieur, lorsque Mose aperoit son
peuple en train d ' a d o r e r le veau d'or : aperte sunt m a n u s eius, et factus est
similis mulieri parturiendi in primitivis suis, que cum t e n e a t u r in doloribus et
m a n u s eius super pectus illius, et virtus non erit que adiuvet partus eius
( X I I . 5). Dans la l i t t r a t u r e de l'poque les maux d ' e n f a n t du Messie signi-
fient, selon la conception eschatologique de l'histoire ce soudain redressement
succdant la dernire priode la plus grave ainsi que la voie qui conduit
la rdemption. 5 4
Le Ps.-Philon prsente la royaut de Sal commme coupable et en agran-
d i t mme les pchs; p a r exemple lorsque, au cours d u combat contre les
Philistins l'arche t o m b e d a n s les mains de ceux-ci Saiil, lui prend la fuite (LIV.
4), cependant, la diffrence d u premier livre de Samuel (p. ex. 1 S 15 : 11)
52
C. C . T o r r e y : T h e Messiah Son of E p h r a i m . J B L CO (1947) 253277 ; J . L i v e r :
T h e Doctrine of t h e T w o Messiahs in t h e Sectarian L i t e r a t u r e in t h e T i m e of t h e Second
Commonwealth. H T R 52 (1959) 149 185 ; A. S p i r o : Pseudo-Philo's Saul and R a b b i s '
Messiah ben E p h r a i m . P A A J R 21 (1952) 136.
53
Henoch 99 : 4 ; J u b . 23 : 22 ; 4F E s d r a s 4 : 5 1 - 5 : 13 ; 6 : 1 8 ; 8 : 6 3 9 : 6 ;
Baruch 25: 1 - 2 9 .
54
1 . H a h n : F l a v i u s J o s e p h u s s a Bellum J u d a i c u m eschatolgiai httere.
(Josphe Flavius et l'arrire-plan eschatologique du Bellum J u d a i c u m . ) A n t . T a n . 8
(1961) 199.

HISTORIOGRAPHIE ET aggada 373
il ne le blme jamais pour ses pchs. A u x yeux de l ' a u t e u r du LAB l ' t a t de

culpabilit de Saiil est t o u t naturel ; c'est en effet la force divine qui lui im-
pose ce rle. Aussi sa figure n'est-elle p a s conforme celle que doit avoir le
chef, suivant la conception du LAB. A vrai dire, il n'est m m e pas roi oint (c'est
que le texte ne fait mention nulle p a r t de ce que Samuel l'aurait oint roi cf.
1 Sam. 10 : 1 pour David, il le fait bien : L I X . 2). Saul n'ignore pas que, le
spiritus sanctus l ' a y a n t quitt, il n'est plus souverain (LX. 1). Aprs la con-
versation qu'il poursuivit avec la sorcire d ' E n d o r c'est encore lui-mme qui,
la femme n ' a y a n t pas reconnu, constate ceci : Ecce nunc scio quoniam species
mea m u t a t a est et gloria regni mei t r a n s i v i t a me (XLIV. 4). La destine de
Saul sera bientt accomplie cause d u pch commis contre Agag ( L V I I I .
2 - 3 LXV. 4).
Cependant les signes reprsentatifs de la personne de David, leur confron-
tation avec ceux de la personne de Saiil d u r a n t le rgne de celui-ci, d o n n e n t
se faire des ides sur le caractre de la royaut de D a v i d conu par le Ps.-
Philon, et ceci malgr que, le texte t a n t fragmentaire, la dscription des vne-
m e n t s de la royaut de David elle-mme en est absente. D a v i d est un vritable
lu de Yahv. C'est ce qui ressort des paroles de Samuel adresses David lors
de leur premire rencontre : Ecce n u n c sanctus christus Domini (LIX. 2).
L'onction de David a lieu sous l'effet d ' u n ordre divin : Et dixit Dominus a d
Samuel : Vade ungue quem tibi dixero, quoniam completum est tempus in quo
adveniret regnum eius (LIX. 1).
C'est dans les deux psaumes c h a n t s par David que l'opposition d u
Messie ben phram au Messie ben David apparat la plus marque ( L I X .
4, L X 23). 55 C'est en pleine conscience d e son tat d'lu et avant d'entrer en
lutte avec Goliath que David chante les psaumes. Dans chacun d'eux il m e t en
parallle, par le moyen des rcits de la Gense la personne de Saiil avec la sienne.
D a n s le premier, en voquant l'histoire de Can et Abel il f a i t le rcit de la haine
de ses frres et celui de sa propre lection : . . . eustodivit me Deus et. . . ange-
lis suis t r a d e t me et custodibus suis u t me custodiant. . . (LIX. 4). D a n s le
second il est dj question de la p r i m a u t de David sur Saiil et de la victoire
qu'il va remporter sur celui-ci, tout cela partir de l'histoire de la cration
o comme le veut le psaume 9 l'me de Saul figure p a r m i les choses cres d u
chaos. Quant sa propre personne D a v i d la prsente comme la perfection l'em-
p o r t a n t sur le chaos (ce dernier ayant cette fois le sens de force finie, dsordre),
comme une force suprieure : 5 6 Aut inmemor es quoniam de resultatione in
chaomato n a t u s est vestra creatura. Arguet autem te m e t r a nova unde n a t u s
85
r . B u l t m a n n 1960. 96.
86
Ps. Sal. 17 : 37 a f f i r m e que le Messie l'emporte sur ses adversaires; en T e s t .
Lvi 1 8, Test. Zebulon 9 il triomphe sur Beliar ; Ass. Mosis a t t e n d la victoire de l'arme
cleste. Sur le caractre dualiste essnien de l'hymne v. M. I ' h i l o l e n k o : Origine
essnienne des psaumes de David. Semitica 9 (1959) 36 38.

s u m , de qua nascetur post tempus de lateribus meis qui vos dominabit (LX. 3). 57
L a personne de D a v i d est entoure de signes miraculeux : dans sa l u t t e
contre Goliath c'est conformment la prophtie qu'il a donne dans son
h y m n e en effet u n ange qui le fait triompher (LXI. 2). C'est David qui re-
oit l'attrihut christus (en hbreu rTPD L I X . 2) a t t e s t une seule fois
d a n s le texte du LAD. Cependant, d ' a p r s la conception d u LAB le Messie
r d e m p t e u r figurant d a n s l'histoire des douze tribus n'est p a s David lui-mme :
ses calculs eschatologiques, le Ps.-Philon les fait aboutir sa propre poque,
u n e date proche de la naissance de l'uvre. 5 8 La confrontation de la r o y a u t
de D a v i d avec celle de Saiil n'est qu'une illustration de ceci que la force divine
favorable aux lus triomphe de ceux qui s'opposent elle une illustration
des esprances de sa propre poque.
L'auteur du L A B croit savoir que la fin des temps arrivera dans des
cadres apocalyptiques ; c'est ce qui ressort des passages apocalyptiques qui
reparaissent dans l ' u v r e entire. On y trouve le schma des apocalypses
juives, la marche de ces lments pour ainsi dire runissant le processus des
vnements rapports p a r la narration historique. Une p a r t i e des descriptions
apocalyptiques n'est q u ' u n e addition g r a t u i t e au texte de la Bible, 59 une a u t r e
p a r t i e en sert de supplment, d'largissement de la description de tel signe ou
d e tel phnomne surnaturels, par des lments apocalyptiques (p. ex. lors d u
passage de la mer R o u g e X. 5 ou lors de la rvlation de la Tora X I .
4 5). Dans l ' a d a p t a t i o n du LAB deviennent apocalyptiques ces parties qui
suivent fidlement les passages bibliques comportant des lments miraculeux
d a n s ces cas-l l'importance des lments se voit-elle augmente par leur
mise en relief, leur intgration un t e x t e raccourci. Deux fragments m a j e u r s
de l'apocalypse sont fournis par la vision implique dans le rcit de Mose : la
prsentation, par le moyen des lments cosmiques, de la terre promise Mose,
e t le mouvement cosmique lors de l'pisode de combat renferm dans l'histoire
d e D b o r a (XIX. 1016 soit X X X I . 1).
L'eschatologie apocalyptique et l'historiographie nationale se trouvent,
chez le Ps.-Philon, d a n s un rapport spcial : alors que, d a n s les apocalypses
57
La priode initiale de l'eschatologie juive est marque de tendance nationaliste
et de l'esprence en la r o y a u t davidienne, celleci consistant a t t e n d r e le rtablisse-
m e n t de la royaut d a v i d i q u e sous la conduite d ' u n roi provenant de la maison de D a v i d .
Ce messie serait, d'api s Ps. Sal. 17 une figure humaine qui va occuper le trne de
J r u s a l e m , les Juifs de la Dispersion, eux, vont retourner et la domination universelle
d e l'Isral se verra p a r l ralise.
58
Sur le t e r m e final des calculs eschatologiques du LAB et sur le rapport e n t r e
los calculs de celui-ci avec ceux de l'apocalypse syriaque de B a r u c h v. L. G r y : K B
(1939) 3 3 7 - 3 5 6 .
59
Les libres additions des lments apocalyptiques du L A B ont beaucoup do
ressemblances avec l'apocalypse de 4e Esdras 5 : 4 12 ; les parallles de l'arbre sanglant,
d u e feu jaillissant de la terre, des pierres criantes, des fleuves de sang, du soleil brillant
p e n d a n t la nuit sont, chez le Ps.-Philon, les s u i v a n t s : feu, dlivrance du feu, destruc-
tion de l'ennemi par t r e m b l e m e n t de terre, le soleil qui, l'effet des sorcelleries d ' A o d
brille pendant la nuit, prodige (eau) offert Gdon lequel se transforme en sang et feu.

rgulires la conception de l'eschatologie cosmique d'origine trangre se

mle avec les esprances nationales juives, avec les prophties a n n o n a n t d'une
p a r t les esprences glorieuses d'Isral, son futur magnifique, d ' a u t r e p a r t la
perte des puissances adversaires, 60 chez le Pseudo-Philon le r a p p o r t des l-
ments constitutifs est juste le contraire : ici c'est dans l'historiographie natio-
nale que se mlent des lments cosmiques eschatologiques et apocalyptiques.
L a structure et la conception analyses ci-dessus nous a m m e n t suppo-
ser que le texte primitif du L A B s'est tout probablement termin p a r un juge-
m e n t de Dieu a y a n t lieu dans des cadres apocalyptiques et par la description
de la thocratie de l'on suivant. D a n s la thocratie du nouvel on comme
donne y conclure, l'intrieur de l'uvre, la conception du r a p p o r t peuple
Dieu les descendants des douze tribus ont un rle privilgi : Dabo lumen
mundo, et illuminabo habitabilia, et disponam testamentum cum filiis hominum
et glorificabo populum meum super omnes gentes; in quo disposui excelsa
sempiterna, que ei erunt in lumine, impiis vero in punitionem (XI. 1).
3. Sphre humaine et sphre divine
La reprsentation du processus historique n'est qu'un des aspects de la

vision du monde du LAB. Un a u t r e lment constitutif en est la conception du
r a p p o r t entre individu et dieu, individu et univers, donc un lment statique
par rapport au dynamisme du processus historique.
homme e t d i e u
Pareillement aux tendances spiritualisantes des prophtes le L A B trans-
forme certaines conceptions bibliques pour les rendre plus abstraites. Son
image de dieu est exempte de t o u t t r a i t anthropomorphe. Yahv n ' y est plus
comme un dieu de tribu plus puissant que celui de t o u t autre peuple, mais bien
le dieu tout-puissant, qui, cause de sa persistance d a n s sa foi a lu A b r a h a m
pour conclure l'alliance avec lui. Dieu ternellement subsistant et qui sait tout,
qui connat l'intention des gnrations futures a v a n t mme que celles-ci ne
soient nes : . . . tu sois sensum omnium generationum antequam nascantur
( X X I . 2).
P a r sa loi il dirige l'univers, l'histoire des douze tribus. Ce sens abstrait
de la loi a p p a r a t d a n s l'uvre p a r des symboles qui rappellent le symbolisme
essnien des rouleaux de Qumran : 61 et nunc si ambulaverimus in viis eius
erimus filii sui. . . (XVI. 5).
80
Telles les visions de 4, E s d r a s ou bien la srie do vaticinia ex eventu de l'Oracula
Sibyllina.
61
Les uvres littraires voisines d u L A B tmoignent, elles aussi, d ' u n e influence
essnienne (le Livre des Jubils, H e n o c h , le Testament des douze patriarches), v. A.
D u p o n t - S o m m e r : Los crits essniens d c o u v e r t s prs de la Mer Morte. Paris 1959. 308.
Acta Antiqua Academiae Scientiarum Hungaricae 28, 11/80

376 I. FRHLICH
L'omnipotence divine dirige cependant les actes des hommes non seule-
m e n t sur la voie de la loi antrieurement prescrite, mais assez souvent par une
intervention directe. E n apprenant la promesse de J e p h t Yahv se met en
colre et dcide lui-mme que J e p h t aille rencontrer le premier sa propre fille
p o u r sacrifier j u s t e m e n t celle-ci, la plus chre ( X X X I X . 11).
C'est encore d ' u n e faon directe que Y a h v dirige les douze tribus d'Is-
ral, lorsqu'il leur conseille d ' a t t a q u e r la tribu coupable de Benjamin pour que
celle-ci expie par la l u t t e le meurtre de la concubine du lvite : Ascendite,
quoniam tradam eos in manus vestras. Ipse a u t e m seduxit eos u t compleret
v e r b a sua (XLVI. 1).
Kenaz ayant achev le combat contre les Amorites n'arrive pas se
debarrasser de son pe miraculeuse dont il eut t u t a n t d'ennemis. L'Amorite
c a p t u r par Kenaz lui conseille de se laver les mains du sang d ' u n Hbreu. Or,
d ' a p r s le Pseudo-Philon l'ennemi donne ce conseil mchant pour se dceler et
p o u r faire par cela t r i o m p h e r le peuple lu de Y a h v ( X X V I I . 11).
L a plus belle et la plus complte illustration de la force de l'omnipotence
divine qui dirige les actes humains est fournie p a r un pisode part, l'histoire
de B a l a a m (XVIII. 1 14). G. Vermes en a y a n t fait l'analyse 62 a dmontr
que c e t t e conception de la figure de Balaam remonte sur N u m . 31. E n cons-
quence Balaam a p p a r a t chez le Ps.-Philon non comme un homme mchant,
mais comme un hros tragique. Cette conception de la figure de Balaam est-
elle t o u t isole les a u t r e s commentateurs considrant Balaam, tous, d'aprs
la tradition de Num. 22 24 comme un coupable p a r excellence qui personnifie
l'avidit, l'orgueil et qui, a y a n t corrompu son peuple est en plus le moyen de la
c h u t e d'Isral. C'est p a r avidit qu'il eut accept l'appel de Balak, c'est par
f a n a t i s m e qu'il n'eut p o i n t observ l'avertissement divin, et lorsque Yahv le
c o n t r a i n t bnir Isral, il donne tout de suite un conseil mchant Balak en
se rvoltant par cela contre l'ordre divin. 63
Le Balaam du Ps.-Philon est en revanche u n hros tragique qui ne h a t
p o i n t Isral. Il assume son rle non par avidit, mais par ceci que, accomplir la
volont de Jahve, c'est l son unique ambition. Il n'ignore par que les Juifs
sont peuple lu, lui-mme adore leur dieu. Son activit entire est dtermine
p a r u n e premire erreur : il demande la permission de se mettre en route avec
Balak t o u t en sachant que Isral, en t a n t que peuple lu, ne peut tre bless.
Si p o u r la seconde fois il va chez les Moabites, c'est q u ' a y a n t pris piti de Balak
il v e u t offrir un sacrifice pour lui. Yahv, t o u t en le lui p e r m e t t a n t le prvient
que sa mission fera d u mal chacun d'eux ( X V I I I . 8). C'est pour cela qu'il
lui enlve le talent de prophte. Dsormais Balaam accepte son sort consciem-
m e n t et se soumet la destine ordonne p a r Yahv. Aussi, le dernier conseil
32
G. Vermes 1961. 140.
63
G. V e r m e s 1961. 174.

qu'il donne Balak n ' e s t pas, d ' a p r s le Ps.-Philon u n f r u i t de sa mchancet,

p l u t t le moyen de son suicide physique. 6 4
Cette conception de la figure de Balaam est base, d ' a p r s G. Vermes non
p a s sur l'ignorance de la tradition palestinienne, mais p l u t t sur le refus con-
scient de celle-ci 65 (il ressort d'autres passages galement que l'auteur du L A B
a trs bien connu la tradition des livres bibliques et que t o u t ce qu'il en insra
d a n s son uvre, f u t filtr d ' u n e faon consciente e t conforme ses propres
ides.) Au lieu de la rdaction rabbinique de la Tora e t la tradition rabbinique
se modelant sur celle-l le Ps.-Philon a ici recours u n e tradition d'origine dif-
frente. Dans son u v r e la figure de B a l a a m est destine illustrer la concep-
tion entire du LAB, suivant laquelle l'activit et la puissance du dieu d ' I s r a l
o r i e n t a n t les v n e m e n t s du monde n'en agit pas moins sur les actes de l'ennemi
e t fait ainsi t r i o m p h e r son peuple lu.
U n e preuve de la puissance directrice immdiate de Y a h v c'est retarde-
m e n t des punitions. Chaque crime commis doit tre puni, mais pour la p l u p a r t
d u t e m p s la peine n'accable le coupable qu'ultrieurement : non pro hoc
peccato quod nunc locuti sumus moriemur, sed in illo priori de quo comprehensi
confessi sunt viri, qui arserunt in peccatis suis. ( X X V I I . 15). L'infcondit de
la femme de Manoah est d u e des pchs a n t r i e u r e m e n t commis. ( X L I I . 2).
A b r a h a m a d m e t la menace de la peine de m o r t pour u n e punition de ses pchs
antrieurs commis malgr soi (VI. 11). C'est pour des pchs anciens que K e n a z
se fait tuer au cours d u combat (XVII. 7), et la m o r t de la concubine d u lvite
s'explique de la m m e faon (XIV. 3). E l q a n a , le pre de Samuel qui craint
dieu et qui est le seul du peuple comprendre la volont divine hsite accep-
t e r le pouvoir suprme offert p a r le peuple parce qu'il n'ignore pas que ceci
contreviendrait la volont de Yahv. Il pense qu'un ventuel accs au pouvoir
passerait pour une punition de ses anciens pchs : Sed si peccata mea compre-
h e n d e r u n t me u t supervolaret super me sors, ego me interficiam, u t non me con-
taminetis. I u s t u m est enim me pro peccatis meis t a n t u m m o d o mori,magis q u a m
sustinere p o n d u s populi ( X L I X . 5).
Nous avons p u constater non seulement dans la manire de penser d ' E l -
q a n a mais aussi propos d'autres chtiments que les anciens pchs s'expient
p o u r la p l u p a r t p a r la mort. Pour avoir fait l'idoltrie Mika et sa mre v o n t
mourir dans les t o u r m e n t s ; ici encore la punition a bien l'air d ' u n j u g e m e n t
(XLIV. 10) et en plus le t e x t e insiste sur la gnralit de la punition : Et sic
non hoc solummodo Miche faciam, sed omnibus qui p e c c a n t in me (XLIV. 10).
64
Le dualisme du Test. Aser 1 montre de la simiilitude avec la conception d u
Pseudo-I'hilon ; il y est question des deux voies ordonnes l'homme ; de celui qui
suit le mal, chaque acte tourne en mal alors qu'il veut faire du bien, cf. 1 QS,
IV. 1 5 - 1 9 .
65
G. V e r m e s 1961. 175.

Les punitions retardes t m o i g n e n t d ' u n e p a r t d u c h a n g e m e n t de la con-

c e p t i o n du pch. L e L A B pose la question de la punition non partir de la
responsabilit morale, mais bien d ' u n p o i n t de vue dterministe : l'expiation
d u c o u p a b l e est invitable en ce qu'elle d p e n d de la volont d u dieu tout-puis-
s a n t e t non de l'acte d u coupable. C'est ce qui est soulign aussi d a n s les pas-
sages o il s'agit de la punition d ' a p r s la m o r t : Gdon, de son v i v a n t n'est
p a s p u n i pour avoir f a i t le culte des idoles : Una via posita est u t non redar-
g u a m Gedeonem in v i t a sua, eo q u o d dissipaverit sacrarium Baal, quia dixe-
r u n t t u n e omnes : Vindicet se Baal ( X X X V I . 4). Gdon v a donc mourir
d a n s sa vieillesse, m a i s aprs sa m o r t , comme d i t Y a h v : . . . castigabo e u m
semel ( X X X V I . 4). L a spcialit d e la punition consiste non seulement d a n s
ceci q u e le Ps.-Philon trace une ligne d e dmarcation entre la manire paenne
d e la punition et celle signale d a n s le L A B , mais aussi qu'il place la punition
d a n s u n temps irrel, a p r s la mort. C e t t e conception, celle de la punition aprs
la m o r t est d m o n t r e r aussi p a r t i r d e certains passages d u L A B : Quando
senuerimus, tune penitebimur ( L I I . 4), disent les fils d ' l i . L ' a f f i r m a t i o n
v a u t t o u t p r o b a b l e m e n t pour la p u n i t i o n aprs la mort.
Les uvres pseudpigraphiques rapproches du L A B e t qui sont les
p r o d u i t s de la m m e cole reprsentent des ides toutes pareilles. D a n s le
4 e livre d'Esdras a p p a r a t une d u a l i t t o u t e pareille celle d u LAB, propos
d e l'expiation et d e l'lection individuelles. Chacun doit expier ses propres
p c h s , mais dans l'on prochain les t r i b u s d'Isral, finissent, p a r l'intervention
d u Messie par tre sauves. 6 6 D a n s le livre syriaque de B a r u c h le pch n'est
p a s non plus suivi d e l'expiation f a i t qui p e r m e t que les peuples paens qui,
a u commencement e u r e n t une sort f a v o r a b l e ne soient punis qu'ultrieurement
e t la situation de c e u x qui sont p u n i s d a n s le prsent sera meilleure dans l'on
q u i arrivera trs prochainement. 6 7
L e problme d e la punition a p r s la m o r t soulve celui de la foi dans l'autre
m o n d e laquelle n ' e s t peut-tre p a s a b s e n t e du LAB. Au sujet de cette question
la conception du L A B n'est pas c o m p l t e m e n t dveloppe. Le Ps.-Philon, lui
e s t s r de l'tre a p r s la mort et de la rsurrection, le t e m p s de cette dernire
se r a t t a c h a n t l'accomplissement des calculs eschatologiques ; lorsque les
a n n e s du monde s e r o n t consommes, Y a h v , lui va faire n a t r e un tremble-
m e n t de terre p o u r ressusciter ceux qui d o r m e n t . Shol et A b a d d o n (deux
n o m s de l'enfer d a n s la croyance de l ' a u t r e monde du judasme) payeront leurs
d e t t e s 6 8 et jugeront c h a q u e hommes selon leurs actes. Le j u g e m e n t sera suivi
d e la recration d e la terre celle-ci s e r v a n t dsormais de demeure ternelle
p o u r les justes. 69 ( I I I . 10).
66
Cf. Esdras I V . 7 : 50.
67
Cf. l'apocalypse syriaque de B a r u c h , 13 20.
68
De la mme faon 4e Esdras 7 : 32, 4 : 42 ; Henoch 56 : 8 ; apocalypse syriaque
de B a r u c h 21 : 23.
68 e
4 Esdras 7 : 50 ; apocalypse syriaque de Baruch 49 : 51.

Les mes a t t e n d e n t l'arrive d u t e m p s du j u g e m e n t en leur d e m e u r e

d ' a p r s la m o r t ;70 c'est ce qui ressort d u passage selon lequel Mose ne p e u t
entrer sur la terre promise cette poque-l, mais il d o r m i r a avec les p a t r i a r -
ches et restera ici bas j u s q u ' ce que Yahv ne visite la terre pour les ressusciter
c'est ce moment-l qu'ils regagneront leur demeure ternelle ( X I X . 7).
Malgr que le jugement n ' a u r a lieu q u ' a v e c la consommation des temps l'au-
teur du L A B fait q u a n d mme distinction entre la demeure aprs la mort, des
justes et celle des coupables ; distinction qui ressort le plus n e t t e m e n t de c e t t e
r e m a r q u e que la rcompense des justes c'est la vie ternelle, leurs m e s se
reposant en paix, j u s q u ' ce que les t e m p s du monde ne soient consomms
( X X I I I . 13). Les mes des patriarches se reposent elles aussi j u s q u ' ce m o m e n t -
l isoles ( X X I . 9, X X I V . 6., X X X I I . 13). 71 Pour ce qui est c e p e n d a n t des m e s
des paens et des coupables, c'est la punition ou du moins une demeure pareille
l'enfer qui les a t t e n d juste aprs leur mort. Sisera, une fois mort p a r v i e n d r a
l'enfer ( X X X I . 7), K o r a h et ses compagnons rvolts a t t e n d r o n t le j u g e m e n t
sur le lieu des tnbres et de la dsolation (XVI. 3). Les coupables l'op-
pos des justes seront jets p a r Y a h v aux tnbres (LI. 5).72
Toutefois, les ides sur la distinction des coupables et des justes d ' a p r s
la m o r t ne sont pas loignes de la conception biblique de l'au-del, c'est--dire
de l'ide do l'tre sans t e m p s ni pense aprs la mort. Aprs la mort il n ' y a
pas grce immdiate, les patriarches, aprs tre morts, ne p e u v e n t eux non plus
intervenir d a n s l'intrt d'Isral ( X X X I I I . 2 5). Dbora du L A B raisonne de
la mme faon lorsqu'on la prie d'intervenir cause des pchs du peuple
( X X X I I I . 5). 73
Le rle d'intermdiaire entre la sphre des hommes e t celle du dieu est
rempli p a r les anges qui tous figurent chaque fois c o m m e personnificateurs
de la volont divine. C'est ce que semble suggrer le fait que, d ' a p r s la con-
ception du L A B les anges n'interviendront pas dans l'intrt d u peuple d a n s le
cas o celui-ci c o m m e t r a des pchs (XV. 5). Les anges d u Ps.-Philon s ' a v r e n t
donc diffrents de ceux de l'Ancien T e s t a m e n t . Au lieu d ' t r e des envoys,
des messagers ils sont comme les expressions autonomes des notions abstraites.
Leur figure ne remonte pas une tradition antrieure, ni l'influence perse,
mais elle annonce une volution intrieure. Au niveau de l'histoire des reli-
,0
J u s q u ' a u jugement qui aura lieu d a n s le nouvel on le Shol servira de lieu de
sjour transitoire pour les mes, et ceci ds l'apocalypse syriaque de Barueh, 30 : 2 ; de
la mme faon Henoch 51 : 1 et suiv.
,l
La rsurrection des justes est p a r t mentionne en 2 M 7 : 9 ; 14 : 23 ; 12 : 43
etc., Henoch 91 : 10 ; 92 : 3 ; 100: 5. Solon Baruch 1 1 : 4 los justes d o n n e n t en paix.
D'aprs Test. Benjamin 10 les justes seront spars des mchants lors do la rsurrection.
72
De la mme faon Baruch 34 : 4 et 4 e . Esdras 7 : 67 ; d a n s la conception phari-
senne dans le Shol aussi il y a peine et rcompense (Josephe Flavius, Ant. 18: I, 3).
73
D'aprs J o b 24: 20 et Ps. 88: 13 le Shol serait le lieu de l'oubli. Selon Qoh.
9 : 10, Ps. 49 : 16 il n'y a ici qu'existence sans temps ni pense. P a s mme d'intervention,
selon Baruch 85 : 9.

gions l'poque est sans cela marque d ' u n e sorte d ' a u t o n o m i e des notions reli-
gieuses telle la shekina, prsence divine d o n t l'autonomie se t r a d u i t aussi
p a r le changement d u caractre du culte. 74 L'anglologie d u L A B est le fruit
d ' u n e volution pareille. L ' a n g e y est c h a q u e fois la force surnaturelle personni-
fie s ' i n c a r n a n t d a n s u n e forme qui convient sa fonction. L a force divine
incarne est exprime p a r le nom de l'ange : l'ange qui figure d a n s l'histoire de
D a v i d s'appelle Zervihel. L e nom rappelle la fonction : il drive du nom hb-
reu J?i1T qui veut dire e n t r e autres force, bravoure. L a dnomination est
justifie p a r l'explication insre dans le t e x t e latin aussi, explication qui, d a n s
l'origine hbraque doit avoir eu une valeur tymologique : Zervihel angelum
p r e p o s i t u m super v i r t u t e m . L a figure d e l'ange de l'histoire de Samson qui,
chez le Pseudo-Philon s'appelle Fadahel ( X L I I . 10) m a r q u e d j une nouvelle
t a p e de la personnification sans pour a u t a n t dpasser l'incarnation de la force
divine. L'tymologie d u nom drive d u verbe hbreu sauver, dlibrer
e t le nom de l'ange rappelle t o u t p r o b a b l e m e n t les actes f u t u r s de Samson qui
n a t r a . C'est que la naissance de Samson est prdite j u s t e m e n t par cet ange.
P l u t t que messager il est cette fois encore personnificateur de la force divine
bien que sa figure ait t certainement influence p a r l'ange de l'histoire de
Samson dans l'Ancient T e s t a m e n t (Jg 13); toutefois la p a r t i e finale ne laisse
p o i n t conclure sur la personne a u t o n o m e de l'ange. D ' a p r s les paroles de
Manoah : Morte moriemur quia vidimus dominum facie a d faciem (XVII. 9)
e t c'est ce qu'indique aussi le nom de l'angle qui est une dnomination fonc-
tionnelle aussi bien que celui de l'autre ange.
P a r m i les anges qui sont nomms d a n s le t e x t e du L A B citons Gethel
(Ingethel), ange des choses caches ( X X V I I . 10) et Nathaniel, ange d u feu
( X X X V I I I . 3). Ce dernier figure dans l'histoire de Y a r et les sept dvots, his-
toire qui est l'analogue de celle d ' A b r a h a m ( X X X V I I I . ) . D a n s l'histoire de
Y a r c'est aussi un ange qui dlivre les h o m m e s jets d a n s la fournaise. L ' a n g e
a ici le mme rle que, d a n s l'histoire d ' A b r a h a m le t r e m b l e m e n t de terre qui
d t r u i t les impies (VI. 17). U n collationnement des deux t e x t e s rend t o u t vi-
d e n t que les expressions ange et tremblement de terre sont interprter
c o m m e notions e x p r i m a n t le secours divin (C'est cette m m e analogie d o n t se
sert l'endroit parallle la chronique de Yerahmeel o l'ange s'appelle N a t h a -
nael). L'histoire de Y a r rappelle l'ange de p a r sa fonction : . . .anglus qui
p r e e s t igni ( X X X V I I I . 3).
Quelquefois la d n o m i n a t i o n de l'ange dsigne seulement la fonction de
la force divine en question (c'est ce q u ' a conserv du moins la version latine) :
P r e m i t t a m angelum ire mee in ipsos p r o m e t Y a h v J o s u (XV. 5).
D e mme, l'ange figure comme l'lment intermdiaire de la volont divine
74
H . R i n g g r e n : Israelitische Religion. Die Religionen der Menschheit. S t u t t g a r t
1963. 283.

dans le passage d u t e x t e o c'est lui qui rpond d i r e c t e m e n t la question d u

peuple pose J a h v e : Et dixit eis anglus Domini : Preponite ( X X V . 2).
L a fonction des anges rappele plus h a u t se m a n i f e s t e encore d a n s le fait
que dans le t e x t e d u Ps.-Philon d a n s la dscription d u mme vnement figu-
rent la p l u p a r t d u t e m p s plusieurs anges la fois, leur nombre correspondant
aux lments constitutifs de l'vnement. Le plus bel exemple en est fourni
p a r l'histoire de K e n a z o dans le rcit sur les prodiges prsents Kenaz a p p a -
raissent trois anges : Et post lie venit anglus et combussit eos (i.e. libros).
Alius autem anglus accipit lapides preciosos. E t alius anglus iens a t t u l i t
duodecim lapides. . . ( X X V I . 8). D a n s l'histoire d e Samuel il y a galement
deux anges (LXIV. 6).
Le secours anglique est la ralisation priodique d u secours divin per-
m a n e n t accord au peuple lu. Sans tre t o u t aussi direct le motif des pierres
a juste le mme sens d a n s l ' u v r e parfois il se r a t t a c h e mme celui d u
secours anglique. D a n s l'histoire de Kenaz les d o u z e pierres prcieuses, de
sortes diffrentes signifient, avec les noms de tribus g r a v s sur elles l'assurance
du secours divin p e r m e t t a n t l'occupation de la terre d e C a n a a n (XXVI. 1011).
Les sept pierres f i g u r a n t d a n s l'histoire de D a v i d symbolisent galement
l'assistance divine; a v a n t d ' e n t r e r en lutte avec Goliath David grave sur les
pierres ramasses p o u r sa f r o n d e les noms de A b r a h a m , Isaac, J a c o b Mose,
ensuite son propre nom et celui de dieu, de l'autre p a r t i e de l'alliance (LX1. 5).
Le motif en question indique donc chaque fois un recours une puissance, en
fonction du nom g r a v sur les pierres (c'est ce sens qu'il p r e n d dans l'histoire de
Babel analyse ci-dessus, en dsignant la rvolte des hommes. Les pierres e t
les noms gravs sur elles sont rattachs, dans l'histoire d e Kenaz et de D a v i d au
motif des anges. D a n s la premire l'ange a p p o r t e les pierres du ciel, d a n s la
dernire il donne assistance D a v i d dans sa lutte.
D a n s la conception de l ' a u t e u r du LAB les p l a n t e s sont galement des
lments mdiateurs, porteurs de la volont divine. C ' e s t grce une interven-
tion divine que les tribus d'Isral sortent victorieux d u combat contre Sisera.
D ' a p r s l'Ancien T e s t a m e n t Yahv mit le trouble en Sisera (Jg 4 : 15)
chez le Ps.-Philon un succinct discours adress au p e u p l e est suivi de la prfi-
guration de l'assistance divine, e t ceci sous la f o r m e d ' u n e action cosmique,
dcrite p a r l ' a u t e u r d a n s le cadre de la vision de D b o r a . D a n s les paroles de la
prophtesse p r f i g u r a n t le c o m b a t les participants s o n t reprsents p a r des
tres cosmiques, des p l a n t e s : . . . video astra c o n t u r b a r i in dispositione
sua et parari in p u g n a m vobiscum. . . ( X X X I . I). 75
Bien que la fonction des anges et des autres l m e n t s mdiateurs d e l'as-
sistance divine soit pareille au systme d'hypostases d e Philon et de la Gnose,
les figures mdiatrices du L A B ne seront quand mme p a s considres c o m m e
75
Cf. Henoch 14 : 8. Sur la similitude et sur l'influence p r o b a b l e du Livre de H e n o c h
ce passage du L A B v. M. R . J a m k s 1917. 43 44.

r s u l t a n t d ' u n effet pareil. L a vision d u m o n d e du LAB reste, j u s q u ' a u bout

de f o r m u l a t i o n mythologique, sans s'lever en un systme philosophique
c o m m e qui est le cas chez Philon. C'est ce que prouve aussi ce f a i t que parmi
les m d i a t e u r s il y a plusieurs sortes d ' l m e n t s qui, l'intrieur d u systme
p e u v e n t se remplacer l'un l ' a u t r e avec la m m e valeur (ange, pierres, phno-
m n e s cosmiques).
D ' a u t r e p a r t des notions telles que la sagesse mdiatrice (cf. la Sophia de
P h i l o n ) tiennent leur place, en t a n t q u ' i n v e n t i o n s autonomes d a n s l'exgse rab-
b i n i q u e galement, et parfois mme d a n s u n e conception assez rapproche de
celle d e Philon. 76
L e s anges du L A B tablissent u n r a p p o r t permanent e n t r e sphre divine
e t s p h r e humaine, r a p p o r t qu'exigent la conception de dieu et celle de l'alliance
p l u s abstraites, plus spirituelles du L A B , e t ceci conformment la nouvelle
i m a g e d e dieu et d a n s une conception qui p o u r diffrente qu'elle soit de celle
des livres bibliques antrieurs en est q u a n d m m e la suite.
l e s puissances opposes l a s p h r e divine: idoles e t dmons
L e s reprsentations de l'idoltrie sont d a n s le LAB les symboles des reli-

gions trangres, opposes la religion d e Yahv. La description des idoles de
M i k a rappelle les s t a t u e s d ' u n sanctuaire d e Mithras ( X L I V . 5). 77 U n autre
p a s s a g e comporte la description des cultes cananens formes diffrentes de
l ' i d o l t r i e pratique p a r les tribus ( X X V . 9 13). Les pchs d e s tribus sont
d ' o r d r e moral : vnration des dieux trangers, railleries de la providence
divine. Cependant, p r o p o s des idoles de la tribu d'Issachar le t e x t e a j o u t e
ce q u i suit : Nos voluimus interrogare p e r demones idolorum si manifeste
a n n u n c i a r e n t (XXV. 9). L e passage est conforme l'esprit d e l'histoire de
M i k a ou les idoles servaient galement les b u t s de la divination (Mika a en effet
install un oracle, comme ceci devient clair d u texte) et o n t u n pouvoir parti-
culier : facis tibi idola, e t e r u n t tibi in deos e t t u eris eis sacerdos (XLIV. 2).
A u p r s de la divination faite avec le concours des puissances s'opposant
Y a h v le L A B cite encore les sorcelleries d ' A o d ( X X X I V . 1 5). Sous l'effet
d e la magie d'Aod le soleil brille p e n d a n t la nuit; Aod tient sa force magique
d e s anges chus : . . . e t fecit magicis sui precipiens angelis, qui preerant
maleficiis, quoniam m u l t o t e m p o r e i m m o l a b a t eis ( X X X I V . 2). Plus t a r d le
t e x t e explique que ces anges sont e x e m p t s d e pouvoir : Et q u i a transgressi
s u n t , f a c t u m est u t angeli in p o t e s t a t e non e s s e n t . . . E t e n s u i t e : Et in his
70
D'aprs Prov. 3 : 17 le symbole de la sagosso intermdiaire serait l'arbre de vie,
d ' a p r s Philon Leg. all. I . 64 65, le jardin d e l ' d e n . Selon Sap. Sal. 7 : 15 la sagesse
m a n e d e Yahv. D ' a p r s H n o c h 1 : 42 elle h a b i t e dans le ciel, avec les anges. P . M.
G r a n t : Les tres i n t e r m d i a i r e s d a n s le j u d a s m e tardif. Studi in onore di Alberto
P i n c h e r l e . I. P o m e 1967. 246 t r o u v e qu'il ne s'y agit pas d ' i n t e r p r t a t i o n gnostique,
m a i s bien de l'volution de la m t h o d e de l'exgse rabbinique qui, elle eut des rsultats
semblables.
" M . R . J a m e s 1917. 39.

operantur qui ministrant hominibus in maleficiis, usque quo veniat immen-

surabile seculum ( X X X I V . 3).
Des passages cits ci-dessus se dgage une conception suivant laquelle
il existe des puissances s'opposant la puissance divine, en la personne des
anges chus dont le pouvoir limit n'est valable qu' cette poque-l du monde,
jusqu' l'arrive du nouvel on. 78 Les forces s'opposant Y a h v aspirent
s'emparer du savoir possd par la sphre divine (c'est dire la connaissance
de l'avenir).
Le caractre fragmentaire des informations rend difficile de prciser la
dmonologie du L A B c'est dire le caractre des puissances extrieures la
sphre divine. De ce point de vue l'uvre n'est pas domine d ' u n e conception
unique, elle a recours aux lments d ' u n e tradition atteste ailleurs aussi p o u r
les insrer dans son propre systme.
U n aspect de la dmonologie d u L A B apparat d a n s la conception domi-
n a n t la premire majeure partie de l'uvre, conception qui n'est p a s
d'origine gnostique. D a n s la conception de l'auteur du L A B les forces opposes
la sphre divine et apparaissant sous la figure d'anges dchus ne sont p a s
gales au divin, elles n'exercent leur pouvoir que dans la mesure permise p a r
celui-ci, elles sont, somme toute, sa disposition, et servent excuter les puni-
tions. L a conception du LAB reprsente ici la tradition se rencontrant d a n s
Gn 6 : 1 14 (sur la chute des anges) ou p l u t t une forme plus dveloppe de
celle-l. 79
Auprs de l'opposition cependant la subordination la puissance divine
se trouve galement mise en relief par le texte : Et dixit Dominus ad ante-
ciminum : Vides quemadmodum conturbatus populus insipiens in terra. . .
(XLV. 6). Ensuite Yahv dcide de punir des hommes de leurs pchs, c'est
dire de l'idoltrie de Mika et de la participation du peuple celle-ci. Le terme
anteciminus employ cet endroit du texte (transcription du grec vTixei/uevo)
signifie adversaire, qui s'oppose et p a r son sens est rapprocher de la notion
Mastema figurant dans le Livre des Jubils. L'adversaire n'est, dans aucun
des deux cas le principe du mal pris d a n s le sens du dualisme gnostique, mais
bien celui qui s'oppose l'homme, qui est porteur de la tentation t o u t comme
par exemple la figure du Satan dans le Livre de J o b o c'est de par sa position
subordonne au dieu et avec la permission de celui-ci qu'il t e n t e Job, ou bien
comme d a n s le Livre des Jubils ou Mastema est le chef des esprits s'opposant
au dieu et protecteur de l'idoltrie. C'est lui qui demande au dieu le sacrifice
d ' A b r a h a m , c'est lui qui a t t a q u e Mose alors qu'il est lui-mme une force
subordonne au divin. 8 0
78
Cf. 2 Cor. 4 : 4 o la tte des anges inalins n'est le dieu que do cette poque-l.
79
Do la mme faon lo Hnoch grec 6 : 11 ; Henoch thiopique 19 ; Test. J u d a 6 ;
J u b . 5 : 1 14; 10: 1, 5.
80
J o b 1 : 6 - 1 2 ; J u b . 17: 16.

D a n s les a u t r e s crits pseudpigraphiques et certains endroits d u Nou-

v e a u T e s t a m e n t on rencontre de nouvelles variantes de la ligure de l'adver-
saire. 8 1 Mais le dualisme qui s'exprime d a n s ces passages n'est jamais de con-
ception gnostique, d'existence a priori. Aucun des t e x t e s en question n ' a f f i r m e
q u e le mal serait lui-mme principe crateur, son origine s'y explique l
o il en est question p a r la chute des anges.
Les lments littraires de la dernire partie d u L A B concernant la per-
s o n n e de Sal sont c e p e n d a n t remarquables d u point de vue d u dualisme de
l ' u v r e . Saiil d o n t le rle se rduit prparer, en t a n t que Messie ben E p h r a i m
le rgne de D a v i d se t r o u v e obsqu p a r l'esprit malin. L ' e s p r i t sera exorc p a r
D a v i d . Le second p s a u m e chant p a r D a v i d (LX. 2 3) est au fond une for-
m u l e d'exorcisation, p a r le moyen de laquelle il chasse le dmon qui possde
Sal. C'est que, a p r s la rvlation de l'lection de D a v i d Sal est abandonn
p a r la force divine : Et in illo tempore a b l a t u s est spiritus sanctus a S a u l e . . .
( L X . 1), et suit le second psaume de D a v i d .
Les deux p s a u m e s de David c o n s t i t u e n t une unit organique avec le
t e x t e du LAB, e t il n'est par probable qu'ils y aient t postrieurement
insrs. Le second qui, p a r sa conception p a r a t p e u t - t r e plus tranger sert
souligner, les diffrences qui s p a r e n t les deux Messies et en expliquer les
ides fondamentales. C'est en effet ce t e x t e qui nous fait comprendre la con-
c e p t i o n messianistique d u LAB, ce p o i n t culminant d u t e x t e conserv.
Le psaume de D a v i d redit l'histoire de la cration, non la manire de la
Gense, mais d a n s une conception dualiste. L'opposition des expressions
spiritus sanctus e t spiritus pessimus propos de la personne de Sal sug-
gre dj, il est vrai, la conception dualiste d u Pseudo-Philon, p o u r t a n t le sens
d e s expressions ne nous est clairci que p a r le p s a u m e en question. D a n s le
rcit de conception gnostique de l'histoire de la cration le spiritus pessimus
d e Sal figure aprs l'achvement d u premier acte de la cration, c'est dire
p a r m i les choses secondairement cres : Et post hec f a c t a est tribus spiri-
t u u m vestrorum ( L X . 2). Le t e x t e insiste sur le caractre secondaire d u spi-
r i t u s pessimus : de resultatione in c h a o m a t o n a t a est vestra creatura (LX.
3). L e spiritus pessimus a y a n t un pouvoir secondaire est donc oblig le
s o u m e t t r e la puissance d'origine divine de David.
A cet endroit d u L A B l'opposition de D a v i d e t de Sal reflte pour ainsi
dire celle de d e u x puissances cosmiques. D ' a p r s M. Philolenko le psaume en
question serait-il u n e incantation d'origine essnienne 82 employe par l ' a u t e u r
d u L A B d ' u n e faon indirecte et conforme ses ides messianistiques.
81
L'adversaire est souvent rappel d a n s le livre des Jubils (p. ex. 17 : 16) ;
Tost. Dan. 3 : 5, 6 ; Test. Gad 4 : aarav Test. Naphtali 3 ; Test. Aser 4 : tdoXog
T e s t . R u b e n 2 : 4, 6 etc. : 1 Cor 10 : 10.
82
L'analyse d u p s a u m e v. M. P h i l o l e n k o : Semitica 9 (1959) 35 48.

3. Les origines de l' gada
1. L E LIVRE DE DANIEL ET D E U X PISODES DU PSEUDO-PHILON
Les rcits indiqus dans le titre et qui vont tre analyss dans ce qui suit
constituent avec les supplments explicatifs la matire aggadique d u LAB.
Les pisodes ont ceci de commun que chacun d'eux sert de parallle non ex-
pliqu mais suggr, pour d ' a u t r e s pisodes bibliques, apocryphes ou antiques,
et n o t a m m e n t grce la base spirituelle commune des rcits.
D a n s le 3 e chapitre du Livre de Daniel on lit l'histoire de Nabuchodonosor
et des trois Juifs dvots, Shadrak, Meshak et Abed Nego. L a suite des vne-
m e n t s y est la suivante :
1. Nabuchodonosor fait excuter une statue d'or de dieu, et appelle le
peuple la vnrer, tout en m e n a a n t ceux qui s'y opposent de les jeter d a n s
une fournaise ardente (3: 1 6).
2. Malgr la menace rpte trois fois, les trois jeunes gens refusent l'ido-
ltrie en se rfrant l'assistance d u dieu des Juifs (exposition du b u t parn-
tique, 3 : 7, 18).
3. Les trois jeunes gens seront jets dans la fournaise ardente, mais au
lieu de mourir ceux qui les y avaient jets doivent prir : le feu de la fournaise
les t u e (3: 1 9 - 2 2 ) .
4. Les trois dvots, accompagns d ' u n ange sortent de la fournaise sains
et saufs (3: 2227).
Le b u t parntique du Livre de Daniel est rendu vident non seulement
par l'argumentation des trois Juifs, qui parlent de la russite probable de leur
rsistance, mais aussi par le contexte t o u t entier. L'histoire fait au fond p a r t i e
d ' u n e srie de rcits parntiques. Le premier lment en serait le songe de
Nabuchodonosor expliqu par D a v i d (sur les quatre empires du monde, D n 2);
le passage suivant raconte la persuasion de Nabuchodonosor par un miracle. Le
rcit qui est son tour suivi p a r un nouveau songe de Nabuchodonosor et la
prophtie de Daniel. Originairement ces rcits taient sans former une unit
cohrente, ce qui les unit ce sont le b u t parntique et la scne commune, soit
la figure de Nabuchodonosor. 8 3
Or, le Pseudo-Philon a recours au schma du rcit expos ci-dessus du
Livre de Daniel, et mme deux fois : d'abord en racontant la construction de la
tour de Babel, deuximement d a n s l'histoire de Yar et des sept dvots (VI.
1 8 soit X X X V I I I . 1 4). Le schma de l'histoire d ' A b r a h a m est le suivant :
1. L'idoltrie se rpend parmi les gens. Le pch le plus grave et qui met
le comble aux sacrilges c'est la construction de la tour de Babel (IV. 16, VI. 1).
83
O. E i s s f e l d t 1964. 716. A u x pisodes du Livre de Daniel v. A. B e n t z e n :
Daniel 6. Ein Versuch zur Vorgeschichte der Mrtyrerlegende. Bertholet-Festschrift 1950.
58 6 4 ; J . B. A l e x a n d e r : New Light on the Fiery F u r n a c e . Theology 58 (1955)
340 345.
24 Ar ta Antiqua Academiae Scientiarum Hungaricae 28, 1980

2. Abraham e t ses 14 compagnons se r e f u s e n t cuire les pierres signales

de leurs noms (notons la prfiguration d u motif du feu; V I . 13).
3. Abraham e t les a u t r e s exposent p a r - d e v a n t les chefs d u peuple le motif
de leur rsistance : U n u m Deum novimus e t ipsum adoravimus (Le carac-
t r e compos de l'histoire d ' A b r a h a m d u L A B et la multiplicit des sources
sont voir n e t t e m e n t d e ce que l ' a r g u m e n t a t i o n d ' A b r a h a m e t les autres an-
nonce cette fois le c o n t r a s t e de polythisme et de monothisme alors que la
construction de la t o u r symbolise, d ' a p r s le sens originaire de l'histoire la
r v o l t e contre le destin divin VI. 4).
4. Les chefs d c i d e n t de punir A b r a h a m et les a u t r e s puisque ceux-ci
a v a i e n t rvolt contre leur ordre. Ils veulent les faire brler eux-mmes avec
les pierres offertes p a r les constructeurs de la t o u r (le motif des pierres t a n t
ici le symbole, d ' a p r s le systme de symboles de l ' u v r e entire aussi, de
l'idoltrie - VI. 5).
5. Apparition d ' u n auxiliateur en la personne d n chef des fils de Sem; il
d e m a n d e d'accorder s e p t jours de dlai a u x coupables en a t t e n a n t qu'ils aban-
d o n n e n t leur p r o j e t son intention effective consistant sauver Abraham e t
les a u t r e s (VI. 6).
6. Les compagnons d ' A b r a h a m a c c e p t e n t le p r o j e t de l'auxiliateur et
s ' e n f u i e n t dans les m o n t a g n e s . Abraham qui est le seul avoir confiance en
la justesse du j u g e m e n t divin reste, t o u t en p r t a n t aide ses compagnons
rfugis (VI. 7 - 1 4 ) .
7. Abraham d a n s la fournaise a r d e n t e et sa dlivrance. Le feu jaillissant
b r l e ceux qui l ' e n t o u r e n t . L a dlivrance des compagnons d ' A b r a h a m (VI.
5-18).
Voyons la m a r c h e de l'histoire de Y a r e t des sept d v o t s :
1. Yar tablit u n a u t e l Baal et m e n a c e de m o r t ceux qui se refusent
lui offrir un sacrifice ( X X X V I I I . 1).
2. Sept h o m m e s d u peuple s'y opposent e t se r f r a n t la loi refusent le
sacrifice Baal ( X X X V I I I . 1 - 2 ) .
3. Suivant l'ordre de Y a r on les j e t t e au feu mais, voil un ange (ang-
lus q u i preest igni) l ' t i e n t et, le feu brle ensuite les serviteurs de Y a r aprs
t r e averti il tue Y a r qui s'approche du feu et le feu f i n i t p a r dtruire la
s t a t u e de Baal et ceux qui lui offrent un sacrifice ( X X X V I I I . 23).
E n t r e les d e u x histoires racontes p a r le Ps.-Philon c'est celle de Y a r
qui, p a r ses motifs c o n s t i t u a n t s est plus rapproche d u rcit d u Livre de Daniel.
Voyons la srie des m o t i f s qui constituent la structure c o m m u n e des deux his-
toires : appel l'idoltrie (opposition d u dieu local e t d u dieu tranger) e t
perspectives d ' u n e sanction immdiate; les opposants se rfrent leur
dieu (rfrence la loi); punition a p p a r i t i o n de l'assistance divine
la destruction des perscuteurs p a r le feu.

L'histoire de Y a r reprsente une variante plus simple de cette srie de

motifs. L a rptition triple connue d u Livre de Daniel, est absente des passages
r a c o n t a n t l'appel l'idoltrie et la menace de supplice. L e discours p a r n t i q u e
de ceux qui refusent l'idoltrie, discours qui figure c e p e n d a n t dans le Livre d e
Daniel est galement absent d a n s ce rcit dans l'histoire de Y a r l'efficacit
de la providence divine est justifie p a r l'histoire elles-mme.
L'histoire d ' A b r a h a m c o m p o r t e plusieurs passages supplmentaires, sans
reprsenter en elle-mme aucune vision unitaire. Le rcit de Babel, s e r v a n t d e
base pour celle-ci p e r m e t lui-mme beaucoup moins d ' t a b l i r l'analogie avec le
passage cit du Livre de Daniel que l'autre passage d u Ps.-Philon. Ici l'histoire
place l'poque des juges suggre elle-mme la comparaison du polythisme
chananen, une sorte d'idoltrie, avec le monothisme des Juifs.
Cependant l'histoire de Babel reste sans former u n e unit voire m m e
avec son propre contexte. L'histoire, elle, raconte la rvolte contre le destin
divin rvolte invitable qui s'exprime dans la construction de la t o u r les
antcdents, les pchs antrieurs des mortels t a n t effectivement considrs
comme idoltrie : Tune ceperunt hi qui i n h a b i t a n t terram, inspicere in
astris, et inchoaverunt imaginari, et ex his divinationes facere, et filios e t filias
suas traicere p e r ignem (IV. 16).
Le motif de la peine inflige ceux qui se refusrent l'ordre est gale-
m e n t absent de l'histoire de Babel d u LAB. Mme le commencement de l'his-
toire description des pchs de l ' h u m a n i t et de la rvolte des dvots est
d i f f r e n t de celui de l'autre rcit d u Livre de Daniel. Chez le Pseudo-Philon
ce passage est plus rapproch d u moins p a r ses m o t i f s de l'histoire bib-
lique de Babel. Le changement survient dans la p a r t i e dcrivant le refus de
Tordre par A b r a h a m : dans les paroles de celui-ci le pch des constructeurs
de la tour a p p a r a t comme s'il s'agissait d'idoltres. (VI. 4). T o u j o u r s les
m m e s paroles suggrent aussi la mode du supplice : et si nos m i t t a t i s in
ignem cum lapidibus vestris, non consentiemus vobis (VI. 4). Ce n'est q u ' a p r s
cela que les chefs des peuples dclarent la mode du supplice d ' A b r a h a m et ses
compagnons : Et irati duces d i x e r u n t : Sicut locuti s u n t , sic eis facite. E t erit
nisi consenserint vobiscum m i t t e r e lapides, consumetis eos igni cum lapidibus
vestris (VI. 5).
Les motifs que le Ps.-Philon fait succder ce passage (le rle d'auxilia-
t e u r s des chefs des fils de Sem, l'hsitation des compagnons d ' A b r a h a m e t leur
fuite) sont rendus ncessaires p a r la conception d'alliance de l'uvre, p u i s q u e
l'alliance d ' A b r a h a m et de Y a h v est comme la base de celle des douze tribus,
ce qui explique que l'histoire d ' A b r a h a m est aussi le p o i n t de d p a r t du symbo-
lisme des nombres relatif aux t r i b u s et qui reste d o m i n a n t dans les passages
suivants.
L'histoire de Babel semble encore diffrer des a u t r e s dans la troisime
t a p e de la srie de motifs aussi. Si A b r a h a m est dlivr d e la fournaise a r d e n t e

p a r aide divine, la figure de l'ange n ' y est p o u r t a n t p a s prsente. F a i t qui

t r o u v e son explication encore une fois d a n s la conception d'alliance de l'uvre.
t a n t donn que d a n s la manire de penser d u Ps.-Philon l'ange a p p a r a t
p a r t o u t comme personnificateur de la puissance divine, t o u t en indiquant, p a r
son apparition le caractre immdiat de l'aide divine, l ' a p p a r i t i o n de l'ange
a v a n t la naissance de Mose ne pourrait correspondre la g r a d u a l i t temporelle
de la conception de l'alliance. Une formule moins personnelle qu'est le motif
d u t r e m b l e m e n t de t e r r e exprime beaucoup mieux la signification de l'alliance
d ' A b r a h a m l'intrieur de la structure d e l'uvre.
Les motifs de l'idoltrie et de la mise au feu (dans le passage dcrivant
les antcdents, IV. 16, les deux sont r a t t a c h s l'un l'autre) sont retrouver
en d ' a u t r e endroits d e l ' u v r e . D a n s le passage cit l'idoltrie e t la mise au feu
f i g u r e n t ensemble c'est dire non en supplice et rappellent t o u t probable-
m e n t le sacrifice d ' e n f a n t s des cultes cananens. D a n s la t r a d i t i o n juive
p r o b a b l e m e n t en conformit avec] la m e n t a l i t du talio le supplice qu'en-
t r a n e l'idoltrie c'est ce qu'on le voit d ' u n passage pareil celui cit plus
h a u t le supplice d u feu. (L'histoire de Babel doit reprsenter chez le Ps.-
P h i l o n une tape de c e t t e suite d'ides : ici, c'est en plein cours des vnements
que le supplice du feu f r a p p e soudainement Abraham, en t a n t que renverse-
m e n t des pchs de l ' h u m a n i t . )
Mika, le fils de Samson et de Dlila expie l'idoltrie galement par le sup-
plice d u feu. Le Ps.-Philon dcrit l'idoltrie de Mika avec u n e minutiosit qui
v o q u e celle des m i d r a s h : Et fecit sibi effigies trs p u e r o r u m et vitulorum
trs, e t leones a t q u e a q u i l a m , et draconem e t columbam ( X L I V . 5). L'Ancien
T e s t a m e n t rappelle galement l'idoltrie d e Mika (dans certains passage du
L i v r e des Juges) sans nous informer sur son destin ultrieur. Le Ps.-Philon
c o n n a t lui aussi, le genre de m o r t qui a t t e n d Mika, et l'avertissement, la pr-
d i c t i o n de la punition n ' e s t p a s non plus absente du L A B : Micham autem
t r a d a m igni (XLIV. 9), ensuite : In illo t e m p o r e reddidit D o m i n u s Miche et
m a t r i eius omnia que locutus est. E t e r a t Micha dissolutus igne, et mater eius
marcescens, sicut locutus f u e r i t Dominus de eis (XLVII. 12). Ce qui revient
dire que la dscription des circonstances de la mort de Mika dans l'uvre
d u Ps.-Philon est loin d ' t r e le produit de la fantaisie; ce d t a i l est bien p l u t t
la p r e u v e d'un symbolique s y s t m a t i q u e et consquant d u L A B et il est en
m m e temps un exemple illustrant la naissance d ' u n e t r a d i t i o n de motifs.
L a combinaison d e la personne d ' A b r a h a m avec la fournaise ardente est
assez courante dans la t r a d i t i o n aggadique. 8 4 D ' a p r s G. Vermes cette tradition
r e m o n t e r a i t l ' u v r e d u Ps.-Philon, celle-ci t a n t la plus ancienne parmi les
sources comportant le motif en question. 8 5
84
P o u r les lgendes d ' A b r a h a m v. L. G i n z b e r g : The Legends of tho Jews. Phila-
d e l p h i a 1938. I. 202, V. 215.
85
G. V e r m e s 1961. 88.
Acta Antiqua Academiae Scientiarum Hunqaricae 2i, 1980

Cependant l ' a u t r e ligne de tradition, c'est--dire le motif de la p u n i t i o n

de l'idoltrie p a r supplice du feu n'est pas elle non p l u s sans exemples d a n s la
l i t t r a t u r e aggadique. L a tradition v e u t que le pre d ' A b r a h a m ait brl t o u t e s
ses idoles; son frre H a r a n essayant de les sauver f u i t lui aussi la proie des
flammes. 8 6
Le supplice d u feu est a t t e s t encore chez H i e r o n y m u s , mais cette fois
combin avec un motif contraire : H a r a n est brl p o u r s'tre refus adorer
le feu lequel les Chaldens vnrent en dieu. Cette histoire reprsente donc u n
mlange spcial des d e u x types. 8 7
P o u r le pch et le supplice de H a r a n les m i d r a s h conservent de diff-
rentes traditions nous nous sommes cette fois contents de ne citer que celles
qui o f f r e n t aucun parallle avec les histoires exposes.
Sans bien sr p e r m e t t r e de suivre la voie de la t r a d i t i o n les deux exemples
m d i v a u x que nous allons citer tmoignent p a r les m o t i f s qu'ils comportent,
d ' u n e sorte de p a r e n t avec les exemples dj cits. D ' a p r s le premier u n gar-
on juif fait a p p r e n d r e son pre son intention de se faire chrtien. Le pre
d a n s sa f u r e u r j e t t e son fils d a n s la fournaise mais celui-ci en sera dlivr grce
Marie. 8 8 D ' a p r s l ' a u t r e lgende un dvot reoit chez lui un homme p a u v r e
d o n t la figure cle en effet le diable. Le mchant saisit le fils du dvot e t le
j e t t e d a n s une fournaise ardente, mais, voil deux anges qui, en la figure d e
plerins en dlivrent l'enfant. 8 9 L a possibilit du conflit dcoule ici d ' u n e con-
ception dualiste. Seule l'apparition des anges rappelle d e quelque faon l'his-
toire primitive.
2. LE LIVRE DE JUDITH ET PS.-PHILON
P a r son genre littraire, le Livre de J u d i t h est un rcit parntique, genre

littraire trs la mode de son poque. Malgr son s u j e t a p p a r e m m e n t histori-
que l'vnement qu'il traite ne l'est pas. 9 0 Il ne v e u t p a s faire de l'historio-
graphie son caractre a p p a r e m m e n t historique ne s e r v a n t qu' a u g m e n t e r
l'authenticit de l'crit, mais il veut donner le bon exemple, un enseignement
moral. U n v n e m e n t cette fois fictif d u pass n a t i o n a l est expos p o u r
prouver que le dieu p r t e son aide son peuple lu, le sauve de sa situation
difficile. Le caractre p r o p a g a t e u r d u livre est d e n t r e autres aux circon-
stances de sa gense, c'est dire l'poque de la rvolte des Macchabes.* 1
88
H a r a n subit le supplice d u feu p o u r avoir sauv les i d o l e s : J u b . 1 2 : 1 1 2 ;
E p h r a e m le Syrien I . 166 D-167 A (d'aprs le J u b . ) .
87
H i e r o n y m u s : Quaestiones 11 : 28.
88
F . C. T u b a c h : I n d e x exemplorum. A H a n d b o o k of Medieval Religious Tales.
F F Communications 204. Helsinki 1969. 204.
88
F . C. T u b a c h 1969. H . 662.
90
E . S c h r e r 1890. I I . 601.
91
. E i s s f e l d t 1964. 796.
Acta Antiqua Academiae SeierUiarum Hungaricae 28, 1980

L a mise en parallle de l'histoire de Yal et celle de J u d i t h , soit l'inser-

t i o n d e certains m o t i f s de l'histoire de J u d i t h dans celle de Y a l sont, chez le
P s . - P h i l o n , tout videntes. Il p a r a t certain que le Livre de J u d i t h f t inspir
j u s t e m e n t par l'histoire biblique de Y a l (Jg 4 : 17 23, 5 : 2429). L a mise
en u v r e du rcit p a r le Ps.-Philon continue, pour ce qui est de l'histoire du
m o t i f , justement c e t t e tendance. Chez lui, c'est l'histoire de J u d i t h qui se trouve
r e p l a c e dans son c o n t e x t e primitif. L a comparaison des t e x t e s f a i t voir maints
t r a i t s spciaux et conformes la conception de chacun des d e u x auteurs : 92
D'aprs le rcit biblique la m a r c h e de l'histoire est c o m m e ceci :
1. Baraq se m e t la poursuite de Sisera ; celui-ci le f u i t t o u t seul, pied,

son arme entire s ' t a n t faite tuer (3 : 1516).
2. Recontre avec Yal, p a r - d e v a n t la t e n t e de celle-ci (4 : 17).
3. Yal offre son aide Sisera et le cache ( 4 : 18) Sisera demande de
l ' e a u de la femme Y a l lui donne d u lait boire. A la d e m a n d e de Sisera
Y a l se tient l'entre de la t e n t e (4 : 19 20).
4. Yal t u e Sisera endormi ( 4 : 21).
5. Yal m o n t r e B a r a q l'ennemi vaincu (4: 22).
L ' a u t r e rcit s u r Yal et Sisera d a n s le Livre des J u g e s (5: 24 28)

r a c o n t e les lments d e l'histoire cites sous le 3 e point de l'analyse, avec plus
d e dtails que le passage prcdent, d a n s le cadre du c h a n t de triomphe de
D h o r a et de B a r a q . Les passages cits sont des pisodes qui, a p p a r e m m e n t
i n t e r r o m p e n t la suite des vnements p o u r les enrichir de nuances, p a r leur
c o n t e n u cependant l'accueil a p p a r e n t d e Yal et la t o u r n u r e inattendue qui
le s u i t , c'est dire la m o r t de Sisera m a r q u e n t le point c u l m i n a n t de l'histoire.
5 : 2428 du Livre des Juges souligne aussi justement ces d e u x pisodes :
il d e m a n d a de l'eau, elle a donn d u lait, ainsi que la m o r t de Sisera subie
s a n s rsistance aucune, d ' u n e faon impuissante.
Ce sont j u s t e m e n t les passages cits d u Livre des J u g e s qui devaient ser-
vir d e base la conception du Livre de J u d i t h . L a marche des vnements est
d a n s ce dernier, la s u i v a n t e :
1. Prsentation d e la situation difficile de Bthulie au cours du sige, dif-
f r e n d s entre Ozias e t le peuple (les diffrends rsultant des conceptions dif-
f r e n t e s de la confidence en la providence divine 7 : 16).
2. Judith, j e u n e veuve, vnre p a r tous pour son d v o t i s m e et sa puret,
p a r la ville entire expose son p r o j e t ; elle se rend dans le c a m p de l'ennemi,
preparatives : la prire de J u d i t h et le changement de ses habits quoi
v i e n t s'ajouter une b e a u t extraordinaire, d'origine divine ( !) (9 : 1 : 15).
92
Sur le t y p e i d e n t i q u e des d e u x pisodes v. E . J . B u r n s : J u d i t h or J a e l ? CBQ
16 ( 1 9 6 4 ) 1 2 - 1 4 ; E. J. B u r n s : T h e G e n e a l o g y of J u d i t h . C B Q 18 ( 1 9 5 6 ) 19-22.

3. A force de manuvres J u d i t h russit parvenir dans le camp de

l'ennemi (10: 616).
4. Rencontre avec Holopherne qui, lui est bloui par la beaut et la sa-
gesse de J u d i t h : Non est talis mulier super t e r r a m in aspectu et pulchri-
tudine, et in sensu verborum (Vulg. 11 : 19). (Cette scne n'est pas sans voquer
la rencontre d ' E s t h e r avec Assurus, Est. 5 : 2). J u d i t h refuse Holopherne
qui l'invite u n festin (12 : 2). J u d i t h . . .exibat noctibus in vallem Behu-
liae, et b a p t i z e b a t se in fonte aquae conservation et renforcement de la
force de l'assistance divine (12 : 7).
5. La deuxime invitation de la p a r t de Holopherne, cette fois accepte
aprs avoir prononc une courte prire J u d i t h , l'aide de sa servante t u e
l'ivre Holopherne (la dscription rappelle vivement la m o r t de Sisera).
6. J u d i t h rentre dans la ville assige l'arme assigeante, la vue
de son chef mort, prend la fuite (1415).
Pour finir voyons la marche de l'histoire de Y a l chez le Ps.-Philon :
(les divergeances d'avec l'Ancien Testament seront soulignes)
1. La vision cosmique de Debora fait prfigurer le cours du combat et f a i t

apprendre les intentions de Sisera qui, lui veut acqurir des captives pour en
faire des concubines. D'aprs la prophtie de Dbora: in manus mulieris inci-
d a t etiam ipse ( X X X I . 1).
2. Bataille Sisera s'enfuit ( X X X I . 2).
3. Yal . . .ornavit se ornamentis suis et exivit obviam eis ( X X X I . 3).
4. Rencontre description de la beaut de Y'al ( X X X I . 3).
5. Yal offre a Sisera de le sauver: Et ingressus Sisera ut vidit rosas
sparsas in lecto dixit : Si liberabo ibo ad m a t r e m m e a m , et Iahel mihi erit
mulier (le motif figure la place de la premire invitation du Livre de J u d i t h
et a peu prs la mme fonction que celle-l, seuls les rles sont inverses
X X X I . 3), Sisera d e m a n d e de l'eau Yal donne une rponse vasive,
ensuite : cum obdormisset Sisera Iahel perrexit ad pecora et mulsit ex eis lac
ensuite dans u n e prire elle d e m a n d e un signe divin qui puisse prouver son
pouvoir sur Sisera ( X X X I . 5). Sisera demande de nouveau boire voil
le signe que vient de demander Yal.
6. Yal lui donne boire, en mlant le lait avec d u vin. Sisera s'endort
et Yal, a y a n t reu un nouveau signe divin lui enfonce le clou dans la t t e
( X X X I . 7). H y m n e de Yal les dernires paroles de Sisera qui vient de
reprendre ses sens, ensuite la m o r t de celui-ci ( X X X I . 7).
7. Les paroles de la mre de Sisera (dans l'Ancien Testament rappeles
par le cantique de Dbora) Yal montre Baraq l'ennemie tu ( X X X I . 8 9).
Les trois histoires examines comportent de t r a i t s essentiellement com-
muns et qui dfinissent en grandes lignes leur structure. Au cours du c o m b a t
de deux peuples se t r o u v a n t en relations hostiles c'est u n e femme qui, du milieu

392 I. FRHLICH
d u p e u p l e a t t a q u fait triompher son p e u p l e en soumettant sa puissance (en

s d u i s a n t p a r son a m o u r ) le chef ou u n hros minent de l'ennemi.
A partir des t r a i t s mentionns plusieurs histoires bibliques seraient
i n c l u r e d a n s ce groupe, en dehors de celles que nous venons de citer. Telle
l ' h i s t o i r e de Samson e t Dlila o les v a i n q u e u r s sont du moins d a n s l'his-
t o i r e les Philistins. L'histoire de Samson comporte aussi ce motif caract-
r i s t i q u e qui bien que omis dans l ' n u m r a t i o n prcdente n ' e n figure pas
m o i n s d a n s les autres histoires que la hrone livre sa famille l'ennemi
v a i n c u (Yal montre B a r a q le cadavre de Sisera, J u d i t h p r s e n t e la tte de
H o l o p h e r n e Dlila, elle aussi livre a u x h o m m e s de son p e u p l e Samson aux
c h e v e u x coups, J g 16 : 2021). E t j u s t e m e n t le motif de la remise marque
p a r t o u t le t o u r n a n t : d a n s l'histoire de Y a l bien que l'arme de Sisera se soit
f a i t e b a t t r e antrieurement, la victoire ne sera complte que p a r la m o r t du
chef ou plutt p a r la prsentation collective de Baraq victorieuse et de Sisera
t u . D a n s le Livre de J u d i t h aussi, les assigeants poursuivent le sige et ne
p r e n n e n t la fuite q u ' a y a n t appris la m o r t de Holopherne. D a n s l'histoire de
S a m s o n la captivit e t la m o r t de Samson p o r t e n t malheur au peuple entier.
E n dehors des t r a i t s gnraux c e p e n d a n t le LAB c o m p o r t e des motifs
ou d e s analogies spciaux qui figurent u n i q u e m e n t dans le L i v r e de J u d i t h .
D ' u n e analyse formelle des deux t e x t e s v o n t se dgager les parallles suivants
( l ' a n a l y s e ne p o r t a n t que sur les v n e m e n t s suivant la rencontre) :
A la diffrence d u rcit de l'Ancien T e s t a m e n t Yal v a elle-mme la
r e n c o n t r e de Sisera. Qu'elle ait des i n t e n t i o n s pareilles celles de J u d i t h a
se v o i t bien de ce qu'elle se m e t des p a r u r e s et qu'elle dispose la maison pour
l'accueil de Sisera. Y a l t o u t comme J u d i t h blouit de sa b e a u t Sisera,
t e l p o i n t que celui-ci lui f a i t des propositions de mariage. Y a l offre Sisera
d e l a s a u v e r fait qui m o n t r e fort bien son rle d'initiatrice qui est aussi celui
de J u d i t h .
P o u r une deuxime t a p e des m o t i f s parallles voil le passage qu'on
p o u r r a i t peut-tre r s u m e r sous le t i t r e de prparation de l ' i n s t r u m e n t magi-
que. L'hrone, a y a n t bloui de sa b e a u t le chef de l'ennemi e t l ' a y a n t soumis
s a puissance a recours de nouveaux m o y e n s d a n s le b u t d ' u n t r i o m p h e dci-
sif. Voil pourquoi J u d i t h se rend-elle B t h u l i e o elle baptizabat se in fonte
a q u a e (12 : 7), et que, au cours d u festin elle se met boire avec Holopherne
p o u r pouvoir ensuite t u e r l ' h o m m e e n t i r e m e n t ivre. Chez le Ps.-Philon l'acte
m a g i q u e est-il mis en relief avec plus de n e t t e t que dans le L i v r e de J u d i t h ;
d a n s ce dernier il est i n t i m e m e n t li l'ide d u secours divin. Yal, elle, donne
Sisera son propre lait ml avec d u vin. E n faisant ceci elle gagne le pouvoir
s u r lui d ' u n e faon consciente, et finit p a r le tuer.
Malgr les concidences et les parallles des vnements il y a u n e certaine
d i f f r e n c e de conception e n t r e le Livre de J u d i t h et l'histoire de Yal rappor-
t e p a r le Ps.-Philon. Ceci ressort clairement surtout de la description des

antcdents. D a n s le Livre de J u d i t h l'action de l'hrone a lieu afin de justi-

fier l'ide de la providence divine. L a situation est, elle-mme ds le d b u t a u t r e
que chez Ps.-Philon qui, ce s u j e t ne s'carte point de la tradition de l'Ancien
T e s t a m e n t si bien que, dans la vision de Dbora il fait prfigurer le t r i o m p h e
de la providence divine, voire m m e le mode de la triomphe : in m a n u s
mulieris incidat ipse ( X X X I . 1). A l'intrieur de l'histoire les prodiges o f f e r t s
Yal renforcent eux-mmes cette conception tlologique sans pour a u t a n t
tre de valeur dmonstrative. Ce qui a p o u r effet que le caractre p a r n t i q u e
de l'histoire entire s'efface un peu et, au lieu de stimuler la crainte de dieu e t
l ' a t t a c h e m e n t la loi elle sert la foi, en pleine conformit avec la conception
historique d u Ps.-Philon, justifier l'lection divine, comme le font n e t t e m e n t
voir les paroles que prononce Dbora lors de sa vision : Nam etsi populus m e u s
peccavit, ego tarnen memorabor ei. E t his dictis, p r o f e c t a sunt astra sicut
p r e c e p t u m f u e r a t eis et incenderunt inimicos eorum ( X X X I . 2). L'ide d u
triomphe se t r o u v e donc chez le Ps.-Philon ds le d b u t comme garantie, l'his-
toire de Y a l ne servant qu' la raliser et ceci p a r le moyen de la s t r u c t u r e
e t de certains lments de l'histoire de J u d i t h , de caractre parntique e t
dmonstratif. 9 3
L'histoire de Yal est richement atteste aussi d a n s la tradition rabbi-
nique, propos de Yal, y apparaissent plus d ' u n e fois les motifs f i g u r a n t chez
Ps.-Philon ou pareils ceux-ci. On y lit p a r exemple de l'ivresse de Sisera, ce
qui fait supposer que Yal le f i t boire non pas d u lait pur. 9 4 D'aprs une a u t r e
information des midrash palestiniens Sisera est fascin p a r la beaut de Yal,
mais celle-ci rsiste la tentation. 9 5
D a n s le T a l m u d babylonien l'histoire de Yal est t o u t e pareille celle
d e J u d i t h ; Y a l se soumet la passion de Sisera en y v o y a n t la seule m a n i r e
d e le tuer. 9 6
U n e seule aggada contient cette variante du Ps.-Philon suivant laquelle
Y a l fit boire Sisera de son propre lait; pour deux a u t r e s passages c e p e n d a n t
il y a lieu supposer la connaissance d u motif. 9 7
L ' a u t r e ligne de la tradition postrieure de l'histoire de Yal et Sisera
r e m o n t e au t e x t e biblique. E n se f o n d a n t sur la bndiction de D b o r a ( J g
93
D ' a p r s E . J . B u r n s l'essence d u Livre de J u d i t h serait ne l'intrieur d e la
c o m m u n a u t d ' l o p h a n t i n e , sous l'effet des ralits politiques, en imitation consciente
d e l'histoire do Yal p o u r devenir ce qu'elle devint, c e t t e histoire connue, p a r m i les J u i f s
d e L o n t o p o l i s : E . J . B u r n s : C B Q (1954) 1 2 - 1 4 ; E . J . B u r n s : C B Q (1956) 19 2 2 .
L e parallle subsistant chez Pseudo-Philon laisse cependant conclure un emploi p l u s
gnral du m o t i f , son caractre de s u j e t folklorique.
M
M H G I . p . 3 3 6 ; d'aprs le T a l m u d babylonien a u s s i : J e b a m o t 103a103 b ;
Nazir 23b ; H o r a j o t 10b ; Niddah 55b et T o s e f t a S h a b b a t 8 : 24.
95
Midrash ha-Gedol I . 336 ; Midrash wa-Yiqra R a b b a 23 : 10.
96
Y e b a m o t 103a103b; Nazir 2 3 b ; H o r a j o t 10b.
97
L ' a g g a d a cite p a r Rimzo H a f t a r o t et Niddah 55b ; v. L. G i n z b e r g 1938. V I .
198. n. 85.

5 : 24) le T a l m u d babylonien rappelle, p a r m i les femmes bnies, Yal avec Sara,

R b e c c a , Rachel et La. 9 8
D e u x sources postrieures m e n t i o n n e n t la vision de la mre de Sisera 99
d ' a p r s laquelle Sisera devait rentrer victorieux. Cette vision ne figure p a s
chez le Ps.-Philon d'ailleurs, l'endroit cit le motif semble remonter J g
5 : 28 31 la description de la vision rappelle cependant l'histoire connue
d u Ps.-Philon. L a m r e d e Sisera espre le triomphe de son fils aprs l'avoir vu
coucher d a n s le lit d ' u n e femme h b r a q u e ce qu'elle explique d e faon qu'il v a
r e n t r e r chez lui avec beaucoup de captives. Pareillement ceci le texte d u
P s . - P h i l o n fait mention des intentions d e Sisera a v a n t le c o m b a t , lesquelles
consistent apporter d e la guerre des captives. Le motif se f o n d e encore une
fois s u r les vers cits d u Livre des J u g e s t o u t comme les passages du Zohar et
de l ' a u t r e source, mais sous une forme anticipe le motif de l'acquisition des
c a p t i v e s figure aussi chez le Pseudo-Philon. U n r a p p o r t ventuel entre les sour-
ces rabbiniques et le Ps.-Philon est encore supposer de ce f a i t que le nom de
la m r e d e Sisera est d a n s chacun d ' e u x T e m a h un lment qui est cependant
a b s e n t des autres sources.
3. l e m o t i f a n t i q u e d u l i b e r a n t i q u i t a t u m b i b l i c a r u m :
l e c h a n t f u n e b r e de l a f i l l e de jepht
D a n s la Bible la personne de J e p h t figure dans le 11 e chapitre du Livre

des J u g e s . E n 11: 1 28 on lit de J e p h t lui-mme, tandis que les vers 29 40
r a c o n t e n t son voeu e t le sacrifice de sa fille. Chacun des deux passages compor-
t e n t d e s lments f a b u l e u x a y a n t b e a u c o u p de parallles e t qui sont gnrale-
m e n t connus aussi d u folklore mondial d o n t la mise en f u i t e de J e p h t de
chez soi, ensuite les h o n n e u r s qu'il obtient. 1 0 0 L'histoire d u sacrifice de la fille
de J e p h t contient elle aussi un motif f a b u l e u x , celui du v u f a i t sur une chose
p r a l a b l e m e n t inconnue. D a n s les contes parallles antiques e t europens le
hros f a i t v u que, u n e fois sauv de sa situation difficile il v a consacrer la
p r e m i r e chose (ou le premier homme) qu'il rencontre celui qui lui a p r t
aide. 1 0 1
98
L e m m e fait r a p p a r a t , sous u n e f o r m e raccourcie en J e r a h m e e l 5 8 : 172.
L e T a l m u d rappelle Y a l avec Sara et a u t r e s , en se b a s a n t sur Gn 24 : 67 et 41 : 33 ; B R
48 : 15, N a z i r 23b ; cf. T a r g u m J u d . 5 : 24.
99
Z o h a r I I I . 119a et S a b b a t wa-Jeze 27d. A v e c le Pseudo-Philon concorde J e r a h -
meel 58 : 184.
100
S. T h o m p s o n : Motif-Index of F o l k - L i t e r a t u r e . F F Communications 106.
Helsinki 1932. L. 10.
101
S. T h o m p s o n 1932. S 241. Les v a r i a n t e s antiques d u m o t i f : Idomne, roi
de C r t e fait Posidon le v u de sacrifier la premire personne q u ' i l rencontrerait,
lorsqu'il serait rentr sain et sauf de la guerre d e Troie : Servius ad Virg. Aen. I I I . 121
et X I . 264, ensuite le p r e m i e r et le second m y t h o g r a p h e s du Vatican : Scriptores r e r u m
m y t h i c a r u m Latini, I . (ed. G . H . B o d e , 1834). 195, 210. Lors de la guerre phrygienne
M a e a n d r o s fait le v u d e sacrifier la Magna M a t e r , pour la victoire la premire personne
qui, chez lui, vienne le saluer c'est son p r o p r e fils qui le saluera le premier, en com-
pagnie d e sa mre et sa s u r , cf. P l u t a r q u e De fluviis I X . 1.

L'histoire biblique de J e p h t comporte, hors les motifs de conte des a u t r e s

motifs qui rappellent les cultes cananens. Tel le passage sur lequel s'achve
l'histoire d u sacrifice de la fille : Et de l vient cette coutume en Isral : t o u s
les ans les filles d'Isral s'en vont se lamenter q u a t r e jours par an sur la
fille de J e p h t le Galaadite (Jg 11 : 39 40). L'histoire elle-mme ou p l u t t
le vu de J e p h t sert expliquer un culte cananen en r a t t a c h a n t le motif
de conte l'histoire tiologique du culte de vgtation. C'est que le deuil de
q u a t r e jours des jeunes femmes dans les montagnes fait effectivement p a r t i e
du rite du culte de Mip et d ' a u t r e s c o u t u m e s sont encore citer d o n t le deuil
d ' A n a t sur le Baal m o r t dans l'pope ougaritique. 1 0 2
L a premire p a r t i e de l'histoire de J e p h t ne figure, dans l'uvre d u Ps.-
Philon que brivement, titre d'indication. D ' a u t a n t plus dtaille est la
seconde partie, la description du sacrifice de la fille de J e p h t .
Du point de vue de la tradition d u t e x t e latin d u LAB, le passage a ceci
d'intressant qu'il a t d'abord s p a r m e n t publi. Sa premire publication
est d u e J . B . P i t r a (Analecta sacra); or dans la srie Texts and Studies
de M. R . J a m e s il a t publi suivi des t e x t e s de Oratio Moysi, Visio Zenez e t
Citharismus regis David, c'est dire d ' a u t r e s passages de la chronique d u
Ps.-Philon. 1 0 3
D a n s le t e x t e du Ps.-Philon l'histoire est relate non seulement sous une
forme largie en comparaison du t e x t e biblique, mais aussi dans une conception
diffrente de l'originale. Le premier motif de conte, le rcit sur l'origine d e
J e p h t y est absent, et le vu de J e p h t p r e n d lui-mme un autre sens d a n s la
conception d u Ps.-Philon que dans l'Ancien T e s t a m e n t . Le motif garde son
caractre de conte en ceci que le Ps.-Philon souligne qu'il s'agit de la fille u n i q u e
de J e p h t , mais le vu, lui obtient une interprtation thologique. Que J e p h t
soit un chef lu p a r le peuple et qu'il soit lu chacun des deux m o m e n t s sont
bien mis en relief d a n s le t e x t e du Ps.-Philon : Vehi principare populo. Quis
enim seit si p r o p t e r e a salvatus es in dies istis, a u t propterea liberatus es de
manibus f r a t r i s t u o r u m , u t principeris in tempore hoc populo tuo ( X X X I X . 3).
L a conscience de l ' t a t d'lu de J e p h t fait voir le v u sous un t o u t a u t r e
aspect le t e x t e du Ps.-Philon insiste nergiquement sur l'absurdit et la
vanit du vu contre la justice divine : Et iratus est Dominus ira et dixit :
Ecce devovit I e p t a n ut offerat mihi o m n e quod primum obviaverit ei. E c c e
nunc si canis p r i m u s obviaverit l e p t e n u m q u i d canis offeretur mihi. E t n u n c
f i a t l e p t e oratio eius in primogenitum f r u e t u m ventris sui, et peticio eius in
unigenita ipsius. Ego autem liberans lihero populum meum in eo tempore, non
prod eo, sed p r o oratione quam oravit Israel ( X X X I X . 11).
101
H . R i n g r r e n 1963. 45.
103
M. R . J a m e s 1893.

396 I. FRHLICH
L a fille de J e p h t appele d a n s le t e x t e d u L A B Sheila devient p a r

l c o m m e destine au sacrifice. C'est-ce que souligne aussil'tymologie h b r a q u e
de son n o m : Iuste v o c a t u m est nomen t u u m Seila, u t offereris in sacrificium
(XL. 1). L'tymologie d u t e x t e h b r a q u e d o i t se baser sur le radical ^XIT
a y a n t le sens de demander, exiger, interroger, sa forme qui en drive e t qui
est le plus rapproche d u nom de Sheila s ' e m p l o y a n t dans le sens de demande
Aprs le v u que f i t son pre Sheila, suivie de ses amies se retire d a n s les
m o n t a g n e s p o u r y pleurer. Le rite de deuil est mentionn d a n s la Bible aussi :
Elle sien alla avec ses compagnes e t elle p l e u r a sa virginit sur les montagnes
(Jg 11 : 38). L a Bible ne fait que'effleurer le c h a n t funbre, le Ps.-Philon, lui
en p r s e n t e le t e x t e entier. Le c h a n t f u n b r e de la fille de J e p h t n ' a p a s son
p a i r p a r m i les variantes de l'histoire de J e p h t . Les manuscrits, eux aussi le
f o n t ressortir l'intrieur d u texte (dans le manuscrit de B u d a p e s t le c h a n t est
i n t r o d u i t p a r l'intitulation Planctus Seile Filie lepte) :
L e c h a n t f u n b r e de Sheila s ' u v r e sur u n tableau naturel : c'est qu'elle
p r e n d tmoin la n a t u r e pour sa m o r t (les passages cits seront articuls con-
f o r m m e n t la s t r u c t u r e intrieure d u texte) :
Audi t e montes trenum m e u m

intendite colles lacrimas oculorum m e o r u m
testes estote petres in planctu anime mee ( X L . 5)
D a n s la suite il s'agit du motif de la m o r t de Sheila, ensuite le t e x t e re-

p r e n d la structure liturgique :
non sum saturata thalamo meo

nec r e p l e t a sum coronis n u p t i a r u m mearum
non enim vestita sum splendore s e c u n d u m ingenuam m e a m
non sum usa Moysi odoris mei (dans l'dition de M. R . J a m e s :
preciosi o d o r a m e n t i mei)
nec f r o n d u i t a n i m a m e a oleo unctionis quod p r e p a r a t u m est mihi (XL. 6)
Voil la troisime partie du t e x t e q u i prcise le vritable sens du c h a n t

entier :
m a t e r , in v a n o peperisti unigenitam t u a u i
(ngation de l'existence terrestre)
quoniam f a c t u s est internus t h a l a m u s m e u s
(combinaison de la m o r t avec les noces)
. . . e t confectio omnis olei quam p r e p a r a s t i mihi e f f u n d e t u r
(la ngation des noces terrestres p a r la ngation des lments de la p a r t i e
prcdente, de caractre numratif )

E t referentes me convirgines mee in genitum per dies p l a n g a n t me (XL. 6)

(adieux l'existence terrestre)
Le chant termine sur le tableau naturel initial :
Inclinate arbores ramos vestros et plangite i u v e n t u t e m m e a m
Venite fere silvae et conculcate supra virgines meas
(James : supra virginitatem meam)
quoniam abscisi sunt anni mei, e t t e m p u s vite mee in tenebris inveterabit (XL. 7 )
Le chant f u n b r e est suivi de la description du sacrifice et des funrailles

de Sheila, ensuite le Ps.-Philon reprend le texte de la Bible : Et f e c e r u n t
filie Israel p l a n c t u m eius, et constituerunt in mense eo X I I I I die mensis u t
convenientes per singulos annos plangerent filiam l e p t e per dies quatuor. E t
fecerunt nomen sepulchri eius i u x t a nomen eius Seila ( X L . 8). L a t r a d i t i o n
biblique se t r o u v e donc largie, chez le Ps.-Philon p a r l'indication d u t e m p s
d u culte et le motif tiologique relatif au lieu du culte.
L'Ancien T e s t a m e n t e t le t e x t e d u Pseudo-Philon d o n n e n t ensemble u n e
ide exacte du culte de vgtation des Cananens, les d e u x textes se c o m p l t a n t
mutuellement. A travers le t e x t e du Ps.-Philon s'impose aussi un parallle
littraire grec : p a r l'interprtation du sacrifice, p a r le contenu et la s t r u c t u r e
d u t e x t e le c h a n t funbre de Sheila est r e m a r q u a b l e m e n t analogue la com-
plainte d'Antigone d a n s la tragdie de Sophocle. 104
Le chant de Sheila se fonde sur la tradition biblique, il reprsente u n e
variante du motif connu de la Bible sa mise en u v r e se modelant sur u n e
forme littraire grecque. Les d e u x histoires, le Ps.-Philon les runit p a r t i r d e
ce qu'elles ont de c o m m u n . Or, l'lment commun de la m o r t des deux filles
Sheila et Antigone c'est le m y t h e de Persphone, le rite du deuil sur la
K o r d o n t le rite de vgtation des Cananens est un parallle. Dans la tragdie
de Sophocle la figure d'Antigone et celle de Persphone se trouvent c o m m e
compntres, ou p l u t t ce n ' e s t q u ' u n aspect de Persphone que Sophocle m e t
en relief. Dans le m y t h e de Persphone c'est encore la m o r t et le mariage qui
se r a t t a c h a n t l'une l'autre. 1 0 5 Leur liaison est vidente, la mort a y a n t p o u r
oppos dialectique la renaissance, et le mariage, soit le rite d'initiation des
jeunes filles sont galement concevoir comme une sorte de m o r t qui est renais-
sance. C'est que d a n s la langue des mythes le mariage achve une existence
p o u r indiquer en mme t e m p s le commencement d ' u n e nouvelle existence.
104
Ce n'est q u ' a p r s l'achvement d u prsent article q u e m e sont devenus acces-
sibles : M. A l e x i o u P . D r o n k e : Tho lament of J o p h t h a ' s d a u g h t e r : themes, t r a d i -
tions, originality. Studi medievali 12 (1971) 819 863, ainsi q u e M. P h i l o l e n k o : I p h i -
genie et Sheila. I n : Les Syncrtismes dans les religions grecque et romaine. Paris 1973.
165177. Ces tudes sont consacres l'analyse du passage d u Pseudo-Philon e x a m i n
aussi dans le prsent article, ensuito ses parallles littraires grecs dont le passage
s u r Antigone et ceci sous un aspect diffrent d u mien.
105
Sur les diffrents aspects do P e r s p h o n e : C. C. J u n g . e r n y i : D a s
gttliche Mdchen. A m s t e r d a m Leipzig 1941. 3 9 - 50.
Ada Antiqua Academiae Scientiarum Hunqaricae 28, 1980

398 I. FRHLICH
Cette conception se manifeste nettement d a n s les rites de noces qui, eux signi-
f i e n t p a r t o u t dans le m o n d e une fte la fois de joie et de deuil.
D a n s la tragdie de Sophocle, la figure d'Antigone se ralise, au cours du
t r a i t e m e n t du m y t h e d e vgtation, travers certains intermdiaires. Pers-
p h o n e m e u r t par ceci qu'elle se marie p o u r Antigone de m m e que pour Shei-
la, le destin est de mourir au lieu de se marier. C'est toujours la m o r t et les noces
qui se trouvent identifies dans ces m y t h e s d'Antigone o celle-ci doit tre
enterre par Hmon.10 Il se peut qu'il ne s'agit pas tellement d ' u n e modifica-
tion d u mythe de Persphone, de son a j u s t e m e n t la figure d'Antigone mais
p l u t t de ceci que le m y t h e d'Antigone est en contact avec un rite qui fait
revivre continuellement le mythe de Persphone, rite d a n s lequel le person-
nage principal la jeune fille destine tre sacrifie a p p a r a t en la figure
d'Antigone. On ignore la tradition littraire dans laquelle Sophocle prit la
m a t i r e de sa tragdie, la figure et le problme mme de l'hrone t a n t absent
des cercles mythiques antrieurs. 1 0 7
Les lments c o m m u n s du sacrifice de jeune fille sont retrouver dans
chacune des deux uvres. Sheila est e n f a n t unique, Antigone, elle, premire-ne.
L e u r m o r t est reprsente sous forme de claustration. Le Ps.-Philon, il est vrai,
n'en f a i t aucune mention, il ne parle que du tombeau de Sheila, qui deviendra
le centre du culte, toutefois le motif sera retrouver chez les auteurs juifs pos-
trieurs. D'aprs D. K i m h i J e p h t aurait b t i une maison afin que, enferme,
sa fille y passt ses jours. 1 0 8 Ce moment rappelant le m y t h e de Dana rappa-
rat d a n s la tragdie de Sophocle :
xka xai Aava ovgvtov tpw

AAfat fia v xa^xoxoi ava.
xgvnxopva 'v xvpr\gEi aX/ y.axecpevxdrj-
(944-946)
Certains passages de la complainte d'Antigone sont galement comparer

avec le chant funbre de Sheila :
rv nayxoirrjv ci' gc dhifiov

xr/v' 'AvxiyvTjV vvxovaav
(804-805)
106
Euripide, f r a g m e n t d'Antigone et H y g i n e : 72. fab. 72. Sur les d e u x : J .
C s e n g e r y : Euripidsz elveszett drminak tredkei (Les f r a g m e n t s des drames perdus
d ' E u i i p i d e ) . Budapest 1926. 261 264.
107
L'hrone et les problmes t o u c h a n t sa personne sont a b s e n t s d u cercle de
lgendes d'Eschyle. G. M l l e r : Sophokles' Antigone. Heidelberg 1967, affirme qu'il
s'y agirait d'une libre invention de Sophocle. 21.
108
L'exgse mdivale de D. K i m h i sur J g 11 : 39, v. L . G i n z b e r g 1938. VI.
203. n . 109.

(identification de la mort avec les noces)
. . . offl' v/tevaiv yxXrjgov, ovx'mvvfMpei

ji> [t ri vpvo; v/uvrjasv, XX' 'AyoovTi
vvficpevaoi (ngation des noces terrestres)
(812-816)
Les vers cits identification de la mort avec des noces et numration

des lments de la fte de noces constituent justement l'essentiel de l'ana-
logie entre les deux chants. P a r leur forme et leurs motifs l'un et l'autre s o n t
au fond des pithalames sous une forme ngative. 109 L e motif de l'identifi-
cation de la mort avec les noces passe pour un lieu commun de la l i t t r a t u r e
grecque. L a mme identification se fait noter dans la figure de l'Hymne per-
sonnifi en t a n t que gnie qui meurt jeune et qui est en r a p p o r t avec les noces. 110
L'identification est atteste dans les genres trs divers : drame, pitaphe, po-
sie, roman, etc. 1 1 1 P a r m i les identifications littraires qu'il nous soit permis de
citer les paroles d'Iphignie sur sa p r o p r e mort sa figure t a n t aprs Anti-
gone la plus rapproche de celle de Sheila :
Ti)v 'ai rXaivav nagvov ri nag&vov;

"AlTJ VLV W EOiXE VV[KpEVOEt x"/(l
(Eur. I p h . Aul. 4 6 0 - 4 6 1 )
Antigone prend tmoin de sa m o r t galement la nature il se p e u t

que ce motif, attest d a n s chacun de deux chants, ait son origine dans le culte
de vgtation :
l Atgxalai xgfjvat
Orjag 'evagpxov aXao, /jjia v/tfioxvga vfift mxzwpai,
oa tpiXcov xXavxo, oloi vfioi
ngo gyfia rv/ioxcoarov xcpov noxaivov
(843 848)
Antigone donne galement sa maldiction sa propre naissance la

maldiction passant, d a n s la tragdie grecque pour l'accomplissement du destin
de famille, se transforme, chez Pseudo-Philon en une acceptation force de la
volont divine :
109
P a r m i ses a n t c d e n t s littraires les mieux connus s o n t les hymnes m e n t i o n -
ns en Horn. Iliade 18 : 493 et Euripide, Iphignie on Aulide 1036.
110
P i n d a r e , frg. 139 ; Servius ad A e n . I . 661.
111
M. A l e x i o u P . D r o n k e : Studi Medievali 12 (1971) 830.

l) / Xxzomv
axai xoifirjpaz z'avzoyvvrjz' p> nazal vct/iogov fiazq
ov y) '' xaXaiepwv cpvv.
(861-865)
L a c o m p l a i n t e d ' A n t i g o n e s ' a c h v e p a r la c o m b i n a s i o n de l'image d u

t o m b e a u e t de la c h a m b r e nuptiale, en r s u m a n t p a r l les m o t i f s rappels :
ai zvfiog, ai vVfMpeov, < xaxaaxacpi)

oxr]cn eicpQOVQO, ol nogevopai
ngo xov pavzrj, wv pid/iov v vsxgol
nXecrxov exzat ' SXcozv
(891-894)
E n c o m p a r a n t les d e u x c h a n t s f u n b r e s on sera a m e n constater q u e

celui d e Sheila est c o m p o s avec p l u s d e c o n c e p t i o n e t f a i t plus n e t t e m e n t
valoir le sens p r i m i t i f d u rite. L a f i g u r e d e Sheila est plus rapproche, d a n s le
r c i t lui-mme, d u r i t e , en t a n t c o m m e la personnificatrice, t a n d i s que, d a n s
le c a s d ' A n t i g o n e les m o t i f s d u m y t h e d e P e r s p h o n e ne s o n t q u e des l m e n t s
accessoires i l l u s t r a n t la m o r t de la j e u n e fille. L a s t r u c t u r e cadres d u c h a n t
f u n b r e d e Sheila, les t a b l e a u x n a t u r e l s i n i t i a u x e t f i n a u x i n d i q u e n t eux-
m m e s le r a p p o r t d e l'histoire avec le m y t h e d e vgtation e t , l ' i n t r i e u r
d u c a d r e la suite d e s m o t i f s et leur r a p p o r t e n t r e eux r e p r s e n t e n t les t a p e s
du rite de vgtation.
L e s motifs grecs q u e P s e u d o - P h i l o n e m p r u n t a Sophocle ne sont p a s
s a n s parallle n o u s s o m m e s p o u r t a n t c o n d u i t s croire q u e d a n s le L A B il
n e s ' a g i t p a s d ' u n l i b r e e m p r u n t d e m o t i f s grecs, mais p l u t t d ' u n e x e m p l e
l i t t r a i r e c o n c i e m m e n t choisi. L ' i m i t a t i o n consciente est t r s p r o b a b l e p a r cela
m m e q u e d a n s le u v r e s p s e u d p i g r a p h i q u e s e t a p o c r y p h e s (de m m e q u e
d a n s la l i t t r a t u r e a g g a d i q u e tardive) 1 1 2 l ' i m i t a t i o n consciente des motifs an-
t i q u e s , e t plus s p c i a l e m e n t des d r a m e s a n t i q u e s est g n r a l e m e n t rpandue. 1 1 3
L e t e x t e d u t e s t a m e n t d e J o s e p h d a n s le T e s t a m e n t des D o u z e Patriarches 1 1 4
s e m b l e avoir subi l ' i n f l u e n c e d u rcit grec d e P h d r e bien qu'il s'agisse d ' u n
s u j e t biblique (Gn 39 : 7 20) qui t a i t t r s connu et r p a n d u d a n s l ' O r i e n t
a n t i q u e entier. C'est encore une fois u n e x e m p l e grec sur lequel se modle le
112
A. Scheiber : Die Parabel vom Schatz des Gelehrten. Acta Ant. Hung. 10
(1962) 233 236; A. S c h e i b e r : Zu den antiken Zusammenhngen der Aggada. Acta
Ant. Hung. 13 (1963) 267 272 ; A. S c h e i b e r : Antikes und Aggada. Acta Ant. Hung.
17 (1969) 449 457; A. S c h e i b e r : Neue Bemerkungen zu den antiken Zusammen-
hngen der Aggada. Acta Ant. Hung. 14 (1966) 225 229.
113
111
M. HADAS: Hellenistic Culture. Fusion and Diffusion. New York 1959. 300.
Test. Joseph 3.

d r a m e intitul Exode d u juif Ezchiel mais d o n t n o u s ne connaissons que

quelques fragments 1 1 5 et c'est aussi le cas du Christus p a t i e n s de Grgoire de
Nazianze, d a t du 4 e sicle de notre re. P a r leur forme les uvres en question
sont des imitations de d r a m e , p a r leur contenu, des t r a i t s philosophiques e t
morales.
L a tendance d'intgrer d a n s les uvres des genres littraires non piques,
des paraphrases h y m n i q u e s est d ' a b o r d illustre p a r Pseudo-Philon lui-mme
qui, lui introduisit d a n s le L A B cette incantation rappele plus haut, d'origine
essnienne. Le c h a n t f u n b r e de Sheila, on le voit bien, tient-il sa place de t o u t e
vidence p a r m i les h y m n e s d u LAB.
4. l e s c o m p l m e n t s a g g a d i q u e s d u l i b e r a n t i q u i t a t u m b i b l i c a r u m .
P a r complments aggadiques nous entendons ces passages du L A B qui

p o r t a n t des additions plus ou moins courtes et d o n t les informations, sans
dpendre directement de la composition structurale e t d e la conception idelle
de l'crit, y a p p o r t e n t p l u t t des menus complments, d ' u n point de vue qui
est en pleine conformit avec l'esprit de l'uvre m m e . Leur fonction n ' e s t
pas de formuler une nouvelle conception, mais p l u t t d e m e t t r e en relief l'im-
portance des motifs subsistants et qui forment des parallles avec d ' a u t r e s mo-
tifs l'intrieur de l ' u v r e ou avec ceux d ' a u t r e s u v r e s , bref, ils sont c o m m e
des chos des motifs principaux.
Parfois, il ne s'agit mme pas de motifs, mais des additions de quelques
m o t s ajoutes au t e x t e et lui p e r m e t t a n t de fournir u n e explication des vne-
m e n t s plus dtaille que ne le fait la Bible. Les c o m p l m e n t s en question ser-
vent expliquer s u r t o u t ces passages qui p r t e n t q u i v o q u e ou qui, p o u r tre
p a r f a i t e m e n t compris, ncessitent une opration, u n supplment logiques.
U n complment explicatif de cette sorte est chez le Pseudo-Philon ce passage
de l'histoire de Mose o l'ordre donne p a r le P h a r a o n d e massacrer les garons
des Hbreux et de laisser la vie a u x filles est motiv c o m m e ceci : . . . feminas
eorum reservabimus, u t eas nostris demus domus in u x o r e s ; et erit qui n a t u s
fuerit ex eis servus et serviebat nobis. E t hoc est q u o d pessimum visum est
coram Domino (IX. 1). Les midrash tardifs expliquent ce mme passage p a r
le caractre sensuel des gyptiens. 1 1 6
Le L A B fournit des explications non seulement l o la comprhension
d u texte l'exige absolument. Les nombreux c o m p l m e n t s ajouts a u x d a t e s
historiques et aux noms des personnes e t qui ont u n e fonction lgalisatrice,
sont exigs p a r l'intention apologtique. Tels les t a b l e a u x gnalogiques con-
115
1. T r e n c s n y i - W a l d a p f e l : Uno tragdie g r e c q u e sujet biblique. AOr 2
(1952) 1 4 3 - 1 6 3 .
116
E R 7 : 40 et G i n z b e r g 1938 I. 222.
24 Ar ta Antiqua Academiae Scientiarum Hungaricae 28, 1980

402 I. FRHLICH
s t i t u a n t la premire p a r t i e presque entire du L A B et qui d o i v e n t leur conti-

n u i t l'numration s y s t m a t i q u e e t ininterrompue des donnes. D a n s
ceux-ci l'auteur ne m a n q u e d ' n u m r e r le nom des m e m b r e s des races. Les
d o n n e s numriques sont rapproches de celles de la version des Septante, y
s o n t presqu'identiques. Il est seulement regretter que la f o r m e trs corrom-
p u e des noms ne p e r m e t t e p a s chaque fois une analyse.
A propos de l'histoire de Can et d ' A b e l le Pseudo-Philon ne m a n q u e d a s
d ' i n d i q u e r la date d u sacrifice e t l'ge des d e u x frres : au t e m p s du sacrifice
C a n a u r a i t eu 40 ans, Abel, lui 50. (II. 3). L a mme explication avec des
d o n n e s numriques u n peu diffrentes rapparat dans d ' a u t r e s sources. 117
Des complments de cette sorte s ' a j o u t e n t aussi aux n o m s . D a n s le L A B
la f e m m e de Can s'appelle T e m a c h la mre de Sisera a y a n t le mme nom
( I I . 3 e t X X X I . 8). L e Ps.-Philon, lui, n'ignore pas le nom des p a r e n t s de Mose :
A m r a m et Yokabed. ( I X . 9). Sur la m m e tradition semble avoir t fond ce
p a s s a g e du Testament des douze patriarches o il est dit que A m r a m et Yoka-
b e d n a q u i r e n t le m m e jour. Le nom d ' A m r a m y est i n t e r p r t comme le
redressement du peuple puisque c'est lui qui fait sortir le p e u p l e d'Egypte. 1 1 8
L ' l e c t i o n d ' A m r a m est rappele dans plusieurs midrash, chez J o s p h e Flavius
d e m m e que dans la biographie de Mose crite par Philon. 1 1 9 De pareilles
explications tymologiques s ' a t t a c h e n t au nom de Yokabed e t on voit en elle
l ' u n e d e ces dvotes femmes-sages qui s a u v e n t les nouveaux-ns t o u t en s'oppo-
s a n t l'ordre du pharaon. 1 2 0
Chez le Ps.-Philon Mose est appel p a r sa mre Melkiel (IX. 6). Les
a g g a d a s postrieures m e n t i o n n e n t plusieurs noms de Mose d o n n s celui-ci
p a r ses parents, les a u t r e s m e m b r e s de sa famille ou son peuple. 1 2 1
Les parents de Samson s'appellent Manoah (dans le t e x t e latin Manoe) et
E l u m a , filia Remac ( X L I I . 1). L e Ps.-Philon ne m a n q u e p a s de raconter
q u ' i l s se disputaient souvent en s'accusant mutuellement de la strilit de leur
m a r i a g e ( X L I I . 2). D a n s les autres midrash les parents de Samson figurent aussi
m a i n t e s fois. Dans les autres sources la mre de Samson s'appelle Hazlelponit
o u Zelalponit. L a A g g a d a t Bereshit semble savoir de sept f e m m e s striles pour
q u i u n miracle se produisit : Sara, Rbecca, La, Rachel, H a n n a , Hazlelponit
e t Cion. U n e autre source v e u t que Hazlelponit soit l'une des 22 femmes chastes
m e n t i o n n e s dans le dernier chapitre des Prov. 1 2 2 Manoah, le p r e de Samson
f i g u r e , lui aussi, assez souvent. A p a r t i r de J g 15 : 8 le T a l m u d identifie E t a n o
117
Tan. Bereshit 9.
118
Test. Lvi 12.
119
Josphe Flavius : A n t i q u i t a t e s I I . 8 : 6 ; Philon : Vita Mosis 1.2.
120
L. G i n z b e r g 1938. V. 393. n . 16.
121
Yashar Shemot 131 a ; W R 1.3 ; Dibr ha-Yamim 3 ; P R E 48 ; Megilla 13a ;
T a r g u m sur Chronique I . 4 : 18 ; avec les dnominations du Pseudo-Philon concordent
celles d e Clment d'Alexandrie : S t r o m a t a I . 21.
122
Aggada Bereshit 62, 106 107 ; sur Hazlelponit MHG I . 337 ; sur les parents
de S a m s o n : Tan. . IV. 160 ; B a b a B a t r a 91a ; B a R 10 : 6.
Acta Antiqua Academiae ScietUiarum Hungaricae 28, 1980

rappel dans le Livre des Chroniques avec le pre de Samson. D ' a p r s J o s p h e

Manoah aurait t le plus grand h o m m e de son poque. 1 2 3
L e Ps.-Philon nous fait a p p r e n d r e le nom de ces d e u x espions qu'envoie
J o s u sur la terre occuper : Et vocavit Cenez et Semeniam f r a t r e m eius,
duos filios C a l e p h . . . ( X X . 6). Selon d ' a u t r e s sources les espions sont P e r e t s
et Zerah, fils de Juda. 1 2 4
L e Ps.-Philon omet les histoires de J a c o b mais a j o u t e des complments
celle de Dina. P l u s t a r d la jeune fille f u t donne en mariage un h o m m e n o m m
J o b qui, ce qu'on voit d u t e x t e serait identique au principal personnage d u
Livre de J o b et de qui D i n a aurait eu sept fils et trois filles, et ensuite, aprs la
m o r t de ceux-ci encore sept fils et trois filles (VIII. 8).
L a naissance de Mose est p r v u e d a n s une vision de sa sur Miriam. Celle-
ci-voit d a n s son rve un h o m m e v t u en blanc qui lui p r d i t la naissance d e
Mose e t le salut du peuple (une scne pareille a lieu dans le t e s t a m e n t de L vi ;
d a n s son rve Lvi voit galement u n h o m m e p o r t a n t des vtements blancs
de lin e t p a r des a t t r i b u t s sacerdotaux). 1 2 5
L e rcit de la sortie d ' g y p t e est interrompu chez le Ps.-Philon p a r u n
court pisode : a v a n t le passage de la m e r Rouge, lorsque les soldats d u p h a r a o n
sont sur le point de r a t t r a p e r le peuple guid p a r Mose. A la vue d u d a n g e r
l'opinion de la foule est partage : Les t r i b u s de Ruben, Issachar, Zabulon e t
Simon p r f r e n t mourir dans les eaux au lieu de se laisser t u e r p a r l'ennemi.
Celles de Gad, Asher, D a n et Nephtali veulent rentrer avec les perscuteurs,
puisqu'il est mieux de servir v i v a n t que de mourir. Seules les tribus de Lvi,
J u d a , J o s e p h et B e n j a m i n restent a t t a c h e s leur dcision, ils prfrent p r e n d r e
les armes et l u t t e r (X. 3). Le systme d e la division des t r i b u s est t o u t de suite
vident. Les descendants de J a c o b ns des captives choisissent la servitude,
les tribus de La, la m o r t et seules les t r i b u s de Rachel et les deux tribus lues
de La ont confiance en l'assistance divine (l'apprciation des tribus est t o u t e
pareille dans le Testament des Douze Patriarches : Malgr leurs pchs com-
mis J u d a e t Lvi seron lus, l'lection des t r i b u s de Rachel allant de soi d a n s
cet crit la division des tribus se fait encore une fois en considration de leur
origine).
L'opinion p a r t a g e des tribus reflte chez le Ps.-Philon cette mme con-
ception qui, dans l'histoire d ' A b r a h a m a p o u r effet la diffrence de comporte-
m e n t e n t r e A b r a h a m et les autres. L'pisode du passage de la mer Rouge sert,
chez le Ps.-Philon expliquer l'lection des tribus de Lvi, J u d a , J o s e p h e t
B e n j a m i n t o u t comme explique l ' a u t r e pisode celle des descendants d ' A b r a h a m
e t des douze tribus (VI. 1 18).
123
Joseph. A n t i q u i t a t e s V. 8 : 2 3.
124
Tan. . IV. 62 ; Tan. Shela 1 ; B a R 16 : 3.
125
Test. Lvi 3.

L e motif de la division des tribus r a p p a r a t ailleurs. U n e partie des

sources semble savoir en conformit avec le Ps.-Philon de trois groupes
d'opinion. 1 2 6 Quelque p a r t d a n s le Y a s h a r Shemot on t r o u v e une division
quatre. 1 2 7 Ici la division des tribus ne semble pas logique, au lieu de quelque
conception a u t o n o m e il s'agit plutt de l ' e m p r u n t d ' u n e tradition corrompue.
A l'endroit en question ce sont les t r i b u s de R u b e n , Simon, Issachar
Zabulon, B e n j a m i n , N e p h t a l i J u d a , D a n , Joseph Lvi, Gad, Asher, qui
se s p a r e n t les unes des autres.
L. Ginzberg fait remonter cette division P s 68 : 28; d'aprs lui elle
reprsenterait donc u n e tradition a u t r e que celle du Ps.-Philon. 1 2 8 Le passage
d c r i v a n t les idoles de Mika passerait lui aussi, pour u n e prcision aggadique
( X L I V . 5).
L e passage d u L A B t r a i t a n t le Livre de Samuel est particulirement riche
en dtails nuancs. Les dtails servent ici prsenter Samuel sous un aspect
favorable, et ceci en conformit avec la conception de la r o y a u t de Saiil et de
D a v i d , soit avec celle d u rle que joua Samuel dans la royaut. Sans largir
la suite des vnements p r o p r e m e n t dite, de l'histoire de Samuel, le Ps.-Philon
ne fait q u ' a j o u t e r des explications c h a c u n des vnements. Il prcise p a r
exemple les raisons p o u r lesquelles A n n a prononce une prire h a u t e voix
a v a n t la naissance de Samuel (I. 5). Lorsque, suivant le v u d'Anna, Samuel
est p o r t dans le temple, il y est accompagn p a r le peuple entier, celui-ci
j o u a n t des i n s t r u m e n t s de musique voil c o m m e n t l'offre de Samuel de-
vient-elle chez le Ps.-Philon, p a r t i r d u v u d'Anna, l'offre du peuple entier,
e t la voie de Samuel, elle, la marche solennelle de l'lu (LI. 7). Le Ps.-Philon
n o u s fait aussi a p p r e n d r e que Samuel a v a i t huit ans q u a n d il eut la premire
vision pareillement en ceci Mose. E n conformit avec le rcit biblique,
S a m u e l ignore d ' a b o r d qui lui avait parl. C'est li qui renseigne le garon :
si la voix se fait entendre deux fois, m i n u i t e t midi, c'est l'esprit malin, si elle
se f a i t entendre une troisime fois, c'est la voix de l'ange. P o u r la troisime fois,
Y a h v lui parle p a r la voix d ' E l q a n a , le pre de Samuel. Si Samuel l'aperoit
d e l'oreille gauche, c'est d ' a p r s l'explication d'li, la voix de l'esprit malin.
S'il e n t e n d l'oreille droit, c'est celle de l'esprit bon ( L I I I . 3 6).
Parfois, les lments aggadiques servent intgrer les passages en ques-
tion d a n s un cercle de lgendes. Tel le passage suivant de l'histoire de Samson :
S a m s o n aprs avoir sorti la porte de ville de Gaza, enlve u n vantail, comme u n
bouclier au-dessus de l'paule, et se sert de l'autre c o m m e d ' u n pe, p o u r
t u e r 25 000 Philistins ( X L I I I . 3). Sur la force, sur la g r a n d e u r physique, sur
126
Pirqe R a b b e n u ha-Qedo (d. G r n h o t ) L i k k u t i m I I I Jerusalem 18981902,
4 3 ; Wa-Yosa 5 1 - 5 2 .
127
Yashar Shemot 146 a (Ruben, Simeon, Issachar, Zabulon, B e n j a m i n ,
N a p h t a l i J u d a , Dan, J o s e p h Lvi, Gad, Aser).
128
L. G i n z b e r g 1938. V I . 4. n. 23.

HISTORIOGRAPHIE ET AGGADA 405
les pieds mutils de Samson, il y a nombre de traditions aggadiques : p o u r

celles-ci, la manire de la transmission est de second ordre, puisque le rcit
biblique lui-mme, de p a r son caractre de lgende pique, a p u trs facilement
inspirer des lgendes postrieures, et il est aussi supposer que l'histoire bib-
lique de Samson ait t largie des lments lgendaires, prts d u folklore
mondiale.
L ' a u t r e lment de conte se r a t t a c h e la naissance d'Isaac. Chez le Ps.-
Philon, la naissance d ' I s a a c est mentionne d a n s le discours de Josu o il e s t
indiqu que celui-l n a q u i t sept mois aprs sa conception et avait p a r cons-
q u e n t des qualits extraordinaires : Et dedi ei Ysaac et plasmavi e u m in
m e t r a eius que eum genuit, et precepi ei u t cicius restituens reddere eum mihi in
mense septimo. E t propterea omnis mulier qui peperit septimo mense v i v i t
filius eius, quoniam super e u m vocavi gloriam meam ( X X I I I . 8). Le motif
de l'enfant n au septime mois fait gnralement p a r t i e de la srie s u i v a n t e :
une femme strile supplie pour avoir u n e n f a n t sa d e m a n d e est accomplie :
elle accouche d ' u n fils a y a n t des qualits extraordinaires (dons de prophte).
Les midrash postrieurs nous informent sur les mmes circonstances p r o p o s
d ' a u t r e s personnages. Auprs d'Isaac, Samuel vit aussi le jour au septime
mois. 129 (Il est remarquable que le Ps.-Philon reste sans le mentionner alors
que d a n s son rcit l'histoire se trouve galement largie ce point : aprs la
prire d ' A n n a , li a p p r e n d p a r inspiration divine q u ' A n n a sera accouche d ' u n
garon mais t o u t ce qu'il en fait savoir la femme c'est que sa prire f u t
exauce (I. 8). D ' a p r s Josphe Flavius, Ichabod, le petit-fils d'li n a q u i t au
septime mois (Antt. V. 2 : 4). A propos de Mose, un midrash suppose qu'il
ait t n dans des circonstances pareilles 130 e t en y lit encore que chacun des
prophtes n a q u i t comme l'affirme aussi le Ps.-Philon au septime mois.
Le mme motif a p p a r a t dans le roman arabe d ' A n t a r o la force extraordinaire
d u fils le plus b r u t a l d ' A n t a r , K h u d z r u f s'explique p a r cette croyance gnrale-
m e n t r p a n d u e que si une femme enceinte vivant sparment de son m a r i
m e t au jour son e n f a n t au septime mois, celui-ci sera beaucoup plus dvelopp
e t plus fort que de coutume. 1 3 1
Auprs des additions objectives, des courts pisodes et lments lgen-
daires qui illustrent la conception de l ' a u t e u r sans pour a u t a n t s'intgrer d a n s
la s t r u c t u r e de l'uvre, une nouvelle catgorie des complments aggadiques est
forme p a r ces renseignements dans lesquels le Ps.-Philon se rfre une his-
toire fort bien connue de la Bible, comme p a r exemple d a n s le discours d ' A m r a m
l'histoire de T a m a r sans cette fois raconter l'histoire elle-mme (IX. 5). U n
,29
R o s h h a - S h a n a l i a , Samol 3 : 52. Sur I s a a c : R o s h h a - S h a n a le. Sur I k a b o d :
J o s e p h e F l a v i u s : A n t i q u i t a t e s V. 2 4 ; M H G I I . 13 croit savoir que chacun des p r o p h -
tes serait n de c e t t e faon.
no Pirgo R a b b i Eliezer, W a r s z a w a 1852, 48.
131
B. H e l l e r : Az a r a b Antar-regny (Le r o m a n a r a b e d ' A n t a r ) . B u d a p e s t
1918. 198.

406 I. FRHLICH
passage d e ce t y p e p e u t tre cit p r o p o s de Samuel : en dcrivant l'orgueil

de F e n e n n a envers A n n a , strile, l ' a u t e u r d u L A B cite l'exemple de Rachel que
J a c o b a v a i t aime malgr sa strilit mais ceci sans raconter l'histoire elle-
m m e (I. 2). Cette m t h o d e est une caractristique gnrale de l'aggada primi-
tive. Pareillement au Ps.-Philon, certains endroits du Livre des Chroniques,
c o m p o r t e n t galement des allusions d ' a u t r e s passages (13 : 22 et 24 : 27), au
Livre d u p r o p h t e I d d o et au midrash d u Livre des Rois.
D a n s un a u t r e groupe des c o m p l m e n t s aggadiques se rangent les rcits
o le t e x t e d u Ps.-Philon m e t en relief u n pisode de l'histoire biblique alors
que p o u r les autres il se rduit les m e n t i o n n e r (tel le passage relatif Dlila
d a n s l'histoire de Samson, X L I I I . 5 7). Lorsque l'auteur laisse inexpliqus
certains passages, il le fait dessein, a y a n t n o t a m m e n t le b u t de prsenter
t r a v e r s le motif mis en relief l ' v n e m e n t entier de l'histoire, sous forme
d ' e n s e i g n e m e n t moral. D a n s l'histoire d e Samson, la cause de la chute de Sam-
son n ' e s t p a s sa passion p o u r Dlila c o m m e le veulent les midrash tardifs, 1 3 2
mais bien le pch de la commixtio, c'est--dire celui de s'tre d t o u r n d u
Dieu de son peuple. C'est pour cela que Samson perd sa force et d e v i e n t pri-
sonnier, e t non p o u r s'tre laiss coup les cheveux. Notons d'ailleurs que c'est
ce t y p e d'explications aggadiques qui est le plus rapproch de l'interprtation
morale des exgses postrieures.
Les additions aggadiques causent leur tour, un problme, celui d u rap-
p o r t e n t r e l ' u v r e d u Pseudo-Philon e t le genre aggadique. Les motifs de l'ag-
g a d a prise d a n s son sens traditionnel, se sont forms dans des circonstances
a u t r e s que le t e x t e d u LAB. L ' a g g a d a f u t cre au cours de la dcomposition
d u t e x t e biblique et l'explication de certains passages, de certaines propositions
de celui-l. Ces explications doivent leur naissance des raisons diffrentes :
tel passage crit comporte un mot d o n t le sens est prciser (corruptions ven-
tuelles d u texte), ou bien le texte est lui-mme dfectueux, c'est--dire qu'il
n'est p a s en lui-mme suffisant pour tre compris ou qu'il se t r o u v e en contra-
diction a p p a r e n t e avec d ' a u t r e s passages bibliques et enfin, le t e x t e ou son
sens a p p a r e n t semblent inadmissibles. Les t y p e s numrs, G. Vermes les
qualifie de pure exegesis. 133 Cependant l'exgse pure ne se rencontre que
d a n s le cas des textes officiellement admis, canoniss et admis p a r la t r a d i t i o n
aussi, e t fixs avec exactitude, c'est--dire l, o c'est j u s t e m e n t le r a t t a c h e -
m e n t littral au t e x t e qui rend ncessaire les explications aggadiques. Or, il
n'en est p a s ainsi des t e x t e s du Ps.-Philon. Ici, il s'agit p l u t t d ' u n e libre inter-
p r t a t i o n de la Bible, de la formation de la matire, p a r t i r d ' u n propre point
de vue. L e mobile de l ' u v r e t a n t t o u t f a i t autre, le texte, lui, n'en comporte
pas moins des passages analogues p a r leur m t h o d e des explications des textes
132
L . G i n z b e r g 1938. V I . 208. n. 121.
133
G. V e r m e s : Bible a n d Midrash : E a r l y Old T e s t a m e n t Exegesis. T h e Cam-
bridge H i s t o r y of t h e Bible I . Cambridge 1970. 203.

historiographie e t agoada 407
postrieurs, voire m m e et c'est ce q u e p r o u v e la t r a d i t i o n p o s t r i e u r e

d a n s plus d ' u n cas, c ' e s t j u s t e m e n t le L A B q u i s e r v i t d e base p o u r les a g g a d a s
postrieures.
Va-t-on d o n c a f f i r m e r q u e le L A B soit u n e u v r e de caractre m i d r a -
s h i q u e ? S'agit-il d ' u n e a g g a d a ? Ce p r o b l m e d u L A B touche bon n o m b r e
d ' a u t r e s u v r e s . L e m i d r a s h est d ' a p r s la conception de R . Bloch 1 3 4
u n e u v r e qui a p o u r p o i n t de d p a r t la Bible, en est d o n c une sorte d ' c h o ,
u n e m d i t a t i o n s u r la Bible. P r a t i q u e m e n t , il d r i v e d e la lecture et d e l'expli-
c a t i o n liturgiques d e la Tra. D ' o r s u l t e n t son b u t e t sa spcialit d e g e n r e
c o n s i s t a n t celle-ci claircir les passages o b s c u r s e t chercher des analogies
au t e x t e lu. Bien e n t e n d u , a u cours des explications, s'y t r o u v e n t i n t r o d u i t s
des l m e n t s f a i s a n t d e p u i s l o n g t e m p s p a r t i e d e la t r a d i t i o n orale ou qui com-
p o r t e n t des a c t u a l i t s ( v n e m e n t s historiques). P a r la suite des diffrences d e
genre s u b s i s t a n t l ' i n t r i e u r de la Bible, la l i t t r a t u r e m i d r a s h i q u e se d i v i s e
en d e u x p a r t i e s : la h a l a k a est l'explication d e s passages d ' o r d r e juridique, d e s
articles de la loi, t a n d i s que l ' a g g a d a sert clairer les passages piques (histo-
riques, n a r r a t i f s ) .
L a f o r m a t i o n d e s u v r e s m i d r a s h i q u e s e s t t r o i t e m e n t lie la f o r m u -
lation d u t e x t e b i b l i q u e ce qui ne r e v i e n t p o u r t a n t p a s dire q u ' a n t r i e u r e -
m e n t , il n ' a i t t c r i t de midrash, le L i v r e des Chroniques canonis en t a n t
lui-mme u n . D e m m e , les crits des p r o p h t e s d ' a p r s la c a p t i v i t c o m p o r t a n t
m a i n t s passages q u i s o n t , eux, l ' i n t e r p r t a t i o n d ' a u t r e s passages bibliques. Ce
processus a y a n t p r i s naissance d a n s le c o n t e x t e biblique, se p o u r s u i t d a n s les
u v r e s postrieures q u i ne sont p l u s canonises : d a n s la l i t t r a t u r e d e sagesse,
d a n s la l i t t r a t u r e p s e u d p i g r a p h i q u e ( d o n t n o u s nous p e r m e t t o n s d e m e t t r e
en relief p a r sa p a r e n t m t h o d i q u e e t t h m a t i q u e avec le P s e u d o - P h i l o n ,
Le Livre des Jubils), d a n s les passages bibliques d e la ' I o v a x f ] 'Ag/aioXoyia
d e J o s p h e F l a v i u s , d a n s les t e x t e s des r o u l e a u x d e la Mer Morte, et aussi d a n s
la version des S e p t a n t e et, f i n a l e m e n t , d a n s le N o u v e a u T e s t a m e n t o, d ' a p r s
G. Vermes, c h a q u e f o r m e midrashique serait atteste. 1 3 5 Les u v r e s cite ne
s o n t p a s i d e n t i q u e s a v e c la notion des m i d r a s h a m o r e n s e t t a n n a t i q u e s pos-
trieurs. Elles s ' i n s p i r e n t d e la Bible sans en e x p l i q u e r le texte, leur s t r u c t u r e
ne s u i v a n t que v a g u e m e n t celle d e l'Ancien T e s t a m e n t . Mais, malgr la d i f f r -
ence d e genre p a r r a p p o r t au m i d r a s h postrieur, la m a t i r e relve d u m i d r a s h
ces t e x t e s sont, en effet, des formes ralises d e l ' a g g a d a d ' a v a n t l'exgse
p r t a n n a t i q u e e t c h r t i e n n e . A la lumire des t e x t e s cits, la dfinition d u g e n r e
134
R. B l o c h : N o t e mthodologique pour l'tude de la littrature rabbinique.
R S R 48 (1965) 194 2 2 7 .
135
G. V e r m e s 1970. 8. Pour les midrash p r i m i t i f s : R . B l o c h : CahSi 8 (1954)
9 34. E . J . E i k e r m a n n : The Septuagint as a Translation. P A A J R 28 (1969) 1 3 9 ;
W. B a c h e r : Die Promien der alten jdischen Homilie. Leipzig 1913; G. V e r m e s :
Haggadah in the Onkelos T a r g u m . J S S 8 (1963) 159 169; G. W i d e n g r e n : Tradit ion
and Literature in E a r l y J u d a i s m and t h e Early Church. N u m e n 10 (1963) 42 83.
Acta Antiqua Academiae Scientiarum Hunqarirar 28, 1980

408 I. FRHLICH
de F a g g a d a et du midrash, sera, elle-mme, modifie, pour comprendre dsor-

m a i s n o n seulement les explications de t e x t e proprement dites, mais aussi ces
oeuvres d'inspiration biblique t r a i t a n t e t largissant la matire de la Bible sous
un a s p e c t spcial, u v r e s qui f u r e n t cres a u cours du processus a c c o m p a g n a n t
l'tablissement prcis et la formation d u canon dfinitif des Livres de l'Ancien
T e s t a m e n t . C'est ce processus qui reprsente l'histoire de genre de l'aggada,
e n s u i t e sa division en exgse chrtienne e t aggada, ces dernires p a s s a n t p o u r
des f o r m e s figes d u genre.
Cependant, les exgses d ' a v a n t les explications t a n n a t i q u e s et chr-
tiennes, comportent d j mme si ce n ' e s t sous la forme postrieure d u mid-
rash les caractristiques essentielles des midrash. Si, en se b a s a n t sur le
L A B , n o u s cherchons dfinir la mesure d a n s laquelle les i n t e r p r t a t i o n s de
m o t i f s d u texte obissent-elles aux rgles d u Hillel, celles-ci prescrivant les
caractristiques de la h a l a k a et aussi celles de l'aggada, pareille p a r ses mtho-
des celle-l, nous observons plusieurs t r a i t s communs. Le Ps.-Philon a souvent
recours a u x analogies (comme p a r exemple, d a n s le discours cit ci-dessus d ' A m -
r a m o est rappel l'exemple de T a m a r , I X . 5). On y retrouve des passages
semblables d'autres passages bibliques et c'est l une caractristique fon-
d a m e n t a l e de la composition du texte. Cette ressemblance n'est bien e n t e n d u
p a s telle qu'elle a p p a r a t dans les exgses : le Ps.-Philon ne fait que rendre
reconnaissables certaines ressemblances, le t e x t e ne les prcisant nulle p a r t ,
bref, c e t t e mthode se t r o u v e mi-chemin entre le texte biblique et l'explica-
tion aggadique.
E n mme temps, la mthode du Ps.-Philon laisse dj entrevoir l'exgse
c h r t i e n n e dont les d b u t s apparaissent en effet dans l'activit des aptres
e x p l i q u a n t , eux, les doctrines de J s u s t r a v e r s le texte de l'Ancien T e s t a m e n t .
C'est d a n s les interprtations d'Origne que prend une forme dfinitive la
m t h o d e explicative, allgorique des exgses chrtiennes, 136 l'explication litt-
raire, morale et m y s t i q u e de la Bible se p o u r s u i v a n t dans l'activit d'Ambroise
e t d ' A u g u s t i n . Le c o m b a t de David e t Goliath qu'on lit chez Augustin e t qui
est interprter symboliquement (dans le sens d u combat de J s u s avec l'es-
p r i t malin) 1 3 7 p e u t avoir pour modle l'histoire analogue relate p a r le Ps.-
P h i l o n , o on devine les d b u t s de l'interprtation symbolique d a n s la descrip-
tion d e la lutte de D a v i d aid p a r l'ange.
Somme toute, le L A B sera rang p a r m i les aggadas primitives. L ' a n a l y s e
de s a s t r u c t u r e et de son t e x t e nous p e r m e t d ' a f f i r m e r que le t e x t e comporte
non seulement les d b u t s des rgles logiques des midrash postrieurs, mais aussi
les spcialits de genre de ceux-ci. Cette manire de composition rappele au
cours d e l'analyse de structure, et qui consiste tablir des analogies e t ins-
136
H . d e L u b a c : Exgse mdivale. P a r i s 1959. I . 171 219.
13
' Augustin : Sermo X X X I I . 6 - 6 .

H I S T O R I O G R A P H I E ET AGGADA 409
rer des motifs de rptition et ceci avec la composition d'analogies ou de sries

analogiqes, sera au fond considrer comme le d b u t de genre des m i d r a s h
homiltiques. Cependant, la m a t i r e est ici plus riche que d a n s les m i d r a s h
rappels. Non seulement elle comporte les passages destins lecture solennelle
d u Deutronome et des Livres des prophtes et leurs parallles, mais elle re-
prsente le p e n d a n t de l'entire matire biblique, comprise et traite p a r le
LAB. 1 3 8
Les analogies d u L A B sont d j marques de la caractristique postrieure
des midrash homiltiques. C'est que les passages mis en parallle, d r i v e n t ,
pour la p l u p a r t , non d u P e n t a t e u q u e , mais des livres hagiographiques.
Les additions qualifies d a n s notre analyse, de complments aggadiques,
m a r q u e n t non p a s en elles-mmes les passages aggadiques d u LAB, mais elles
reprsentent l ' a u t r e groupe de genre monumental de la t r a d i t i o n midrasliique,
celui d u midrash explicatif (Auslegungsmidrasim) 1 3 9 qui, lui, explique le t e x t e
suivant l'ordre des vers en y a j o u t a n t des complments, des pisodes. Les mid-
rash explicatifs sont, pour leur p l u p a r t , de caractre halachique e t s a n s
tre eux-mmes les traits caractristiques de l'uvre d u Ps.-Philon confrent
p a r leurs complments, une couleur spciale au texte, et p o u r ce qui est de leur
popularit, la transmission postrieure est l en tmoigner.
Budapest.
138
H . STRACK 1 9 2 1 . 1 9 7 .
139
H . STRACK 1 9 2 1 . 1 9 7 .
Arta Antigua Academiae Scientiarum Hungaricae 28, 1980

T. NAGY
DIE ZWEI L. NERATII PRISCI

UND NERATIUS PRISCUS DER JURIST
Eine sechszeilige Bauinschrift aus Saepinum, die mit Th. Mommsen's

Ergnzungen in das Corpus Inscriptionum Latinarum aufgenommen w u r d e 1
u n d die seither als die zuverlssige epigraphische Grundlage f r die Beurteilung
der L a u f b a h n e n der in dieser Inschrift erwhnten zwei L. Neratii Prisci gegolten
hat, schien es unzweideutig bezeugen, da von diesen zwei Senatoren der
ltere noch die ungeteilte Provinz Pannnia, der Jngere dagegen nach-
einander schon die Provinzen Pannnia inferior u n d superior verwaltet h a b e n .
Mit besonderer Rcksicht auf diese Angaben unserer Inschrift wurde die
pannonische Statthalterschaft des lteren L. Neratius Priscus in der F a c h -
literatur entweder in die J a h r e zwischen 94 97,2 oder in jene zwischen
102 106, 3 bzw. 102 1054 whrend die Legation des jngeren homonymen
Senators in Unterpannonien in die S p t j a h r e der Regierung Traians, 5 oder in
die Anfangsjahre der Regierung Hadrians, u n d jene in Oberpannonien ent-
weder in die frhere, 7 oder in die sptere 8 Regierungsjahre Hadrians versetzt.
1
CIL I X 2466. Danach D. 1034, mit Weglassung der ergnzten E n d f o r m e l ,
[. . .s.p.f.):
L. Neratius L. f . [Vol. Priscus] | praef. aer. Sat., cos., l[eg. pr. pr. prov.] | Pannnia.
L. Neratius L. f . Vol. Pr[iscus ? f., cos.) \ VII vir. epul., leg. Aug. pr. pr. P[annonia] | infe-
riore et Pannnia [superior.).
1
W . REIDINGER, Die Statthalter des ungeteilten Pannonin und Oberpannoniens,
Antiquitas, R . 1., B. 2., Bonn 1956, 58 f. Vgl. A. MCSY . v. Pannnia, R E Suppl.
I X (1962), 589.
3
So schon E . RITTERLING, Die S t a t t h a l t e r der pannonischen Provinzen, AEM. 20
(1897), 14 f. hnlich A. DOB, Die Verwaltung der r m . Provinz Pannonin von A u g u s t u s
bis Diocletianus, B u d a p e s t 1968, 41 f.
8
R . SYME, Governors of Pannnia, Gnomon, 29 (1957), 515 ff. = Danubian P a p e r s ,
Bucharest 1971, 177 ff., bes. 185 f. (zit. im spteren nach diesem Sammelband), der n a c h
d e m Vorgang E. GROAG'S, s . v . Metilius, R E X V (1935), 1401, n. 19 den P. Metilius
Nepos (//. 103), als den immittelbaren Nachfolger des L. Neratius Priscus d. . f r
die J a h r e 105/106 vorgeschlagen h a t . I h m folgt W . ECK, Senatoren von Vespasian bis
H a d r i a n , Vestigia, . 13., Mnchen 1970, 160 f.
S
E. GROAG, S.V. Neratius, R E X V I (1936), 2547 f., n. 13. Dafr war a u c h
W . REIDINGER, . O . 70.
8
E. RITTERLING, Die legati pro praetore von Pannnia inf. seit Traian, A r c h r t .
41 (1927), 283 f. (um 119 ff.). J . FITZ, Legati Augusti pro praetore Pannoniae inferioris,
ActaAnt. 11 (1963), 253 f. (zwischen 1 1 8 - 1 2 0 ) , vgl. auch AlbaR. 11 (1970/1971), 146.
R . SYME, DP, 227 und 234 (from 1 1 9 - ) . W. ECK, . O. 186 f. stimmte der Datierung v o n
J . FITZ ZU. . DOB, . . 48 f. lt die Datierungsfrage offen.
7
W . REIDINGER, . O . 69 f. ( 1 2 0 ff.). h n l i c h . DOB, . . 125.
8
E. RITTERLING, Statthalter (Anm. 3.) 18 f. (zweite H l f t e der hadrianischen Zeit).
R . SYME, D P 186 (before 133). J . FITZ, . O. 254 (before 130). W . ECK, . O. 230 legt
sich nicht fest.

412 T. NAGY
I m letzteren J a h r z e h n t e kndigte man jedoch wichtiges neues Material an,

das geeignet ist, die bisherige teils unsichere Zeitordnung der L a u f b a h n dieser
zwei Senatoren vom Grund aus zu modifizieren.
Zuerst wurde von einer neuen I n s c h r i f t aus Saepinum benachrichtigt,
die den cursus bonorum des jngeren L. Neratius Priscus enthlt. 9 Aus dieser
Inschrift konnte dann B. Thomasson durch das freundliche Entgegenkommen
italienischer Kollegen unlngst den nachfolgenden Textteil mitzuteilen : 1 0
l[eg. Aug. pr. pr. in ]a inferiore et Pannnia.
Schon e t w a frher, wurde auch die anllich dieser neuen Inschrift dringende
ntig gewordene Revision der eingangs erwhnten Bauinschrift aus Saepinum
von G. Camodeca durchgefhrt 1 1 und der revidierte Text, wie folgt, vorgelegt :
L. NeratiusL. f . [Vol(tinia) Priscus] | praef(ectus) aer(arii) Sat(urni), co(n)s(ul),

l[eg(atus) pr(o) pr(aetore) in provincia)] \ Pannnia.
L. Neratius L. f . Volftinia) Pr[iscus, q(uaestor), tr(ibunus) pl(ebis), pr(aetor),
co(n)s(ul),] I Vllvir epul(onum), leg(atus) Aug(usti) pr(o) pr(aetore) i[n pro-
vincia) Germania] | inferiore et Pannnia.
D a s wesentliche Neue, worber uns die obigen zwei Inschriften belehren,

b e t r i f f t vor allem die konsulare L a u f b a h n des jngeren L. Neratius Priscus.
Seine Legation vor 107 in der noch ungeteilten Provinz P a n n n i a bezeugen
die beiden Inschriften in vollem E i n k l a n g miteinander. Vor dieser Legation
war unser Senator S t a t t h a l t e r in einer provincia . . . inferiore. Der Vollname
dieser Provinz fehlt in beiden Inschriften. Vor der administrativen Teilung
P a n n o n i e n s etwa im W i n t e r 106/712 existierten jedoch im rmischen Reich
bekanntlich nur zwei Provinzen, die in ihren Namen das komparative Adjektiv,
inferior, getragen haben u n d zwar Germania inf. und Moesia inf. G. Camodeca
entscheidet f r die erstere Provinz. I n der T a t ist kaum eine andere Lsung
9
S. die k n a p p e Mitteilung von V. C i a n f a r a n i , F a s t i A r c h . 22. 1967 (1971),
3 4 4 , . 5 0 3 0 .
10
. THOMAE, L P I . 2 (1979), 18, p . 7, wo b e m e r k t er noch u . a . : Torellii
mecum communicaverunt, post verba 'et Pannnia' nihil incisum esse.
11
G. C a m o d e c a , L a carriera del g i u r i s t a L . Neratius Priscus, A t t i dell'Accademia
di scienze morali e politiche Napoli, 87 (1976), 19 ff., bes. 24.
12
I n der neueren L i t e r a t u r wird die Zweiteilung der P r o v i n z im allgemeinen in
die Zeit u n m i t t e l b a r n a c h d e m E n d e des 2. Dakerkriegs verlegt. J . F i t z , . O. 246 f.
R . S y m b , D P 227 (u. a.), G. A x f l d y , Die Legionslegaten der rmischen R h e i n a r m e e n ,
E S 3, 1967, 23 f. W . E c k , Senatoren, 9. G. C a m o d e c a , . O. 28. - An d a s J . 103 h l t
sich A . M c s y nach d e m Vorgang W . R e i d i n g e r ' s (. O. 18 f.) fest: P a n n n i a a n d U p p e r
Moesia, L o n d o n and B o s t o n 1974, 92 w a s a b e r n a c h der revidierten Lesung der saepi-
n a t i s c h e n B a u i n s c h r i f t der Neratier (D 1034) k a u m a u f r e c h t e r h a l t e n werden k a n n .
E i n e N e u b e h a n d l u n g der prtorischen L a u f b a h n H a d r i a n ' s [s. H i s t r i a , 30 (1981), i m
D r u c k ] h a t mich zur obigen, im T e x t a n g e g e b e n e n Datierimg, E n d e 106/Anfang 107
geleitet.

D I E Z W E I L. NERATII PRISCI 413
mglich, insofern es zu besttigen lt, da der jngere L. Neratius Priscus

mit dem gleichnamigen Suffektkonsul des J . 97. identisch ist. In diesem Fall
konnte nmlich dieser Senator nach seinem K o n s u l a t u n d vor seiner panno-
nischen Legation, die er wohl schon vor Ende 102 angetreten hat, 1 3 nur die
Provinz Germania inferior verwalten, wo in den Provinzialfasten nach 90 u n d
vor 101/102 eine grere Lcke sich zeigt, 14 nicht aber die Provinz Moesia
inferior, wo alle S t a t t h a l t e r aus dem Zeitraum zwischen 97 und 102 b e k a n n t
sind. 15 Die angenommene I d e n t i t t des jngeren L. Neratius Priscus mit dem
homonymen Suffektkonsul des J . 97. unterliege jedoch nur dann keinem Zwei-
fel, wenn aus der senatorischen Familie der Neratier von Saepinum das hchste
Amt unter Domitian u n d Nerva von L. Neratius Marcellus abgesehen
nur allein die zwei Neratii der revidierten Bauinschrift bekleidet h t t e n .
Die auf diese J a h r z e h n t e bezglichen Konsulfasten bezeugen nmlich nur
zweimal den Konsulat je eines L. Neratius Priscus u n d zwar in den J a h r e n
87. 16 und 97. 17 E s scheint nach diesem Tatbestand wirklich auf der H a n d
liegen, da der ltere L. Neratius Priscus mit dem homonymen Suffektkonsul
des J . 87., whrend der Jngere mit jenem des J . 97. identische Persnlich-
keiten sind und demgem der letztgenannte Senator in den ersten Regierungs-
jahren Traians die Provinz Germania inferior, d a n a c h aber die ungeteilte
Provinz P a n n n i a verwaltet h a t .
Diese einfache Lsung b e t r b t aber ein d r i t t e r Neratius Priscus der
bekannte Jurist, der einmal ebenfalls Suffektkonsul gewesen war 18 u n d zwar
k a u m bedeutend f r h e r oder spter als sein B r u d e r L. Neratius Marcellus
( s u f f . 95).19 F r diese drei Neratii Prisci, die alle drei einmal in den letzten
Jahrzehnten des 1. J h . die Fasces gefhrt haben, sind jedoch, wie es im Obigen
schon angemerkt wurde, in den erhaltenen Konsulfasten nur zwei P l t z e
13
Sein Vorgnger, Q. Glitius Atilius Agricola, ist a m 19. November 102 n o c h in
der Provinz bezeugt (CIL X V I 47), a m I d u s J a n u a r 103 (belegt ist zwar erst a m 19.
J a n u a r 103: A. D e g r a s s i , FC 31) h a t aber Agricola seinen zweiten Suffektkonsulat wohl
schon in R o m und n i c h t in absentia in der Provinz angetreten.
14
Vgl. W. E c k , Senatoren, 141 ff., 242, der aber n o c h m i t der mglichen S t a t t -
h a l t e r s c h a f t des L. Licinius Sura zwischen 97/98 und 100/101 gerechnet hat. Diese Mg-
lichkeit h a t aber inzwischen nach der v o m M. S p e i d e l vorgelegten Inschrift aus G r a m -
moni [ J R S 60 (1970), 76 ff. = A E 1969/1970, 583] ihre Berechtigimg eingebt. F r die
Einzelheiten s. die A r g u m e n t i e r u n g von G. C a m o d e c a , . O. 25 f.
15
So bezeugt sind: im J a n u a r 97, (L.) Iul. M a r [ i n u s i ] (CIL X V I 41); a m 14.
August 99, Q. P o m p o n i u s R u f u s (CIL X V I 44 und 45); a m 25. O k t o b e r 100, M ' L a b e r i u s
Maximus ( P I R 2 L 9), den wohl noch vor E n d e 102 Q. F a b i u s P o s t u m i n u s abgelst h a t .
P I R 2 F 54. Vgl. W. E c k , . O. 160. G. C a m o d e c a , . O. 25, A n m . 44.
16
N. A l f i e r i , I fasti consolari di Potentia, A t h e n a e u m , 26 (1948), 110 ff. = A E
1949, 23. A. D e g r a s s i , FC 26.
17
G. B a r b i e r i , Nuovi f r a m m e n t i dei F a s t i Ostiensi, S t u d i Romani, I (1953),
366 = A E 1954, 220. S. auch L. V i d m a n , F a s t i Ostienses, P r a h a 1957. 17 und zuletzt F .
Zevi, Consoli del 97 d. Cr. in d u e f r a m m e n t i gi editi dei FO., L i s t y Filologick, 96 ( 1973),
1 2 5 ff., bes. 1 3 7 = A E 1 9 7 6 , 1 3 2 .
18
Dig. 1.2.2, 53.
19
Einige Angaben ber die Intervalle der K o n s u l a t e n zweier Brder findet der
Loser dieser Zeilen unton, S. 461, A n m . 68.

414 T. NAGY
v o r h a n d e n , ein im J . 87 u n d ein anderer im J . 97. Diese Schwierigkeit wurde

v o n G. Camodeca mit der nicht nher begrndeten Annahme umgegangen, da
der J u r i s t mit dem jngeren L. Neratius Priscus identisch ist u n d so der gleich-
n a m i g e Suffektkonsul des J . 87. nur mit dem lteren, whrend jener des J . 97.
n u r m i t dem jngeren L . Neratius Priscus, d. h. mit dem J u r i s t e n identisch
sein k a n n .
Diese angenommene I d e n t i t t des J u r i s t e n mit dem jngeren L. Neratius
P r i s c u s ist aber mehr als fraglich. Zunchst, wir kennen n i c h t einmal den
V o r n a m e n des Juristen. I n den gewi auf ihn bezglichen schriftlichen Quellen,
v o r allem in den Digesten, wird er nie mit tria nomina g e n a n n t , sondern nur
als Neratius Priscus, bzw. Priscus Neratius, oder schlicht n u r als Neratius. 2 0
U n s e r Rechtsgelehrte knnte also hchstens, mte aber nicht unbedingt den
Lucius-Yorname f h r e n . Dieser U m s t a n d m a c h t schon an u n d f r sich berech-
t i g t e s Bedenken gegen seine Identifizierung mit dem jngeren Lucius Neratius
Priscus.
Eine weitere B e t r a c h t u n g verstrkt nur die Schwierigkeiten einer
solchen Identifizierung. Nach der Verffentlichung der Cursusinschrift des
jngeren L. Neratius Priscus aus L a r i n u m 2 1 ist es nmlich unzweifelhaft
geworden, da dieser Senator, wie auch sein gleichnamiger V a t e r zur plebeji-
schen Schicht des Senatorenstandes gehrt haben. Ebenso s t e h t es aber von
jeher fest, da der B r u d e r des Juristen dagegen ein Neupatrizier gewesen war. 22
W e n n nun jetzt n i m m t j e m a n d die I d e n t i t t des Juristen m i t d e m jngeren
L . Neratius Priscus an, der mu notgedrungen irgendeine Adoptionsart postu-
lieren, um die Zugehrigkeit des Rechtsgelehrten zur plebejischen Nobilitt,
w h r e n d jene seines Bruders zum P a t r i z i a t verstndlich zu machen. Dies
E r f o r d e r n i s erkennend war es also ganz logisch, da n a c h d e m die frhere
F a s s u n g einer Adoption 2 3 in der Beleuchtung des neuen Inschriftbestandes sich
als berwundene herausgestellt hat, zur weiteren Aufrechthaltung der Iden-
t i t t eine neue Version der Adoption vorgelegt wurde. 24 Diese lt sich, wie
folgt, kurz zusammenzufassen :
Der Jurist u n d sein Bruder waren die Shne des lteren L. Neratius
P r i s c u s (suff. 87). Der Erstgeborene L. Neratius Marcellus w u r d e dann von
seinem vermeintlichen Onkel, M. Hirrius F r o n t o Neratius P a n s a ( s u f f . 73?)
a d o p t i e r t und bald darauf vom Kaiser Vespasian im J . 73 oder 74 unter die
P a t r i z i e r adlegiert. D e r jngere Sohn, der J u r i s t , und sein V a t e r blieben dage-
gen weiterhin Plebejer.
20
S. die zusammengestellte N e r a t i u s - Z i t a t e n in der Textausgabe v o n P . K r u e g e r
T h . Mommsen, p. 938. S. noch, V. H a d r . 18,1: N e r a t i u s Priscus.
21
O. F r e d a , E p i g r a f i inedite di L a r i n o , Contributi d e l l ' I s t i t . di Filol. Class.
Milano, I 1963, 238 f. = A E 1969/1970, 152. Die letzten Zeilen e m e n d i e r t jetzt von G.
C a m o d e c a , . O. 30, A n m . 75.
22
D. 1032.
23
R . Syme, T h e J u r i s t N e r a t i u s Priscus, Hermes, 85 (1957), 480 ff., bes. 491 f.
24
G. C a m o d e c a , . O. 24, 32 f.

D I E Z W E I L. N E R A T I I PR I SCI 415
Die Adoption des Marcellus durch Hirrius Fronto lt sich jedoch nicht
beweisen, ja das meiste spricht dagegen. In jeder F o r m einer Adoption 2 5
m u t e nmlich der Adoptivsohn nach der eingebrgerten Sitte den Vollnamen
oder zumindest einen Namensteil seines Adoptivvaters aufnehmen, 2 mit
mglicher Beibehaltung seines vorigen Namens, bzw. Namensteilen. W e n n also
der Adoptivvater M. Hirrius F r o n t o Neratius Pansa hie, dann wre es zu
erwarten, da der Adoptivsohn ebenfalls ein Polyonymer gewesen war, 27
dessen Vollname etwa als M (oder mit mglichen Beibehaltung des vorigen
Vornamens, L) Hirrius Fronto Neratius Marcellus gelutet hat. In den auf
unserem Marcellus bezglichen schriftlichen Quellen, in den Konsulfasten u n d
dem Militrdiplom von Malpas, 28 t r g t aber dieser Senator nur die schlichten
tria nomina: L. Neratius Marcellus. I n den Konsullisten u n d Militrdiplomen
kommen zwar die Konsuln, bzw. die S t a t t h a l t e r meist nicht mit ihren Voll-
namen vor. 29 Eben d a r u m sind die Angaben dieser D o k u m e n t e f r die Beur-
teilung der evtl. Vollnamen der in diesen Quellen benannten Amtstrger nicht
immer richtungsgebend. Wie man es auch nimmt, im Fall des Marcellus sind
nur die tria nomina bezeugt, die aber keinen Rckschlu auf seiner vermeint-
lichen Adoption durch Hirrius P a n s a zulassen. Das beste, von dieser ganz
hypothetischen Annahme ein f r alle Mal abzusehen.
In dem ganzen komplizierten Fragekomplex der Familienverhltnisse
der Neratier wre zurzeit die meist beruhigende Lsung mit zwei verschiedenen
Zweigen innerhalb dieser senatorischen Familie aus Saepinum zu rechnen,
von denen die eine, zu der Marcellus u n d sein Bruder, der J u r i s t angehrt
haben, von Vespasian u n d Titus im J . 73 oder 74 unter die Patrizier aufgenom-
25
Vgl. Dig. 1.7,1 : quod adoptionis eat quidem generale, in duos autem species
dividitur, quarum altera adoptio similiter dicitur, altera adrogatio. adoptantur filii jamilias,
adrogantur qui sui iuris sunt. hnlich schon Gai., I n s t . 1,99.
26
Zur F r a g e m i t k o n k r e t e n Beispielen s. B r . D o e r , Die rmische N a m e n g e b u n g ,
S t u t t g a r t 1937 (Reprint, A r n o Press New Y o r k 1976), 74 ff.
27
I n der f r h e r e n L i t e r a t u r t a u c h t e a u f g r u n d eines I n s c h r i f t f r a g m e n t s a u s Saepi-
n u m (CIL I X 2613) in der T a t schon einmal jene k h n e Gedanke auf, d a Marcellus w a r
ein Polyonymer u n d hiess vielleicht L. N e r a t i u s [Marcellus] I u n i u s Macer. G r o a g , R E
X V I , 2642. Die Zusammengehrigkeit dieses F r a g m e n t s und der seit B. Borghesi's Vor-
gang allgemeinen d e m L. N e r a t i u s Marcellus zugewiesenen akephalen I n s c h r i f t a u s
Saepinum (D 1032) h a t sich aber bisjetzt n i c h t besttigt.
28
Cf. A E 1940, 92, CIL V I 527, bzw. X V I 48.
29
Es ist unntig, diese w o h l b e k a n n t e T a t s a c h e bezglich d e r Konsullisten m i t
Belegen zu bezeugen. Die von den Militrdiplomen befolgte P r a x i s k a n n m a n a b e r hin-
reichend m i t einem einzigen, doch m. E . sehr lehrreichen Beispiele beleuchten.
Cn. Pinarius Aemilius Cicatricula P o m p e i u s Longinus h a t in den A m t s j a h r e n
85/86 u n d vielleicht noch 86/87 die Provinz I u d a e a , zwischen 93/94 imd 96/97 Moesia
superior u n d d a n a c h u n v e r m i t t e l t die Provinz P a n n n i a etwa bis zur Mitte 98 v e r w a l t e t .
Aus seinen Amtszeiten kennen wir 4, auf die Auxiliarheere dieser Provinzen bezglichen
Militrdiplome. I n den f r h e s t e n von diesen, welches das Auxiliarheer der P r o v i n z
I u d a e a b e t r i f f t (CIL X V I 33), f h r t unser S e n a t o r n u r die schlichten tria nomina : Cn.
P o m p e i u s Longinus. Die f r das obermoosische Auxiliarheer ausgestellten zwei D i p l o m e
(CIL X V I 39 u n d M M. R o x a n , R o m a n Military Diplomas, L o n d o n 1978, 6) lassen a u s
d e m Vollnamen dieses Senators das nomen P i n a r i u s weg. Allein in d e m p a n n o n i s c h e n
Diplom von Felsnna (CIL X V I 42) lesen wir den Vollname dieses Polyonymers.

416 T. NAGY
men wurde, whrend der andere Zweig der Familie, deren Representanten
im ffentlichen Leben u n t e r den Flaviern u n d dem Kaiser Traian der ltere
u n d der jngere L . Neratius Priscus gewesen waren, weiterhin Plebejer
geblieben sind. Diese ganz offensichtliche Lsung, die, wie ich es jetzt sehe,
schon vor einigen J a h r e n aufgeworfen wurde, 3 0 macht die verschiedene u n d
unbeweisbare Identifizierungs- und Adoptionstheorien berflssig.
Gegen die I d e n t i t t spricht zuletzt auch eine in dieser Zusammenhang bes.
wichtige Angabe aus der Lebensgeschichte u n d Ttigkeit des Juristen Neratius
Priscus. Es ist die E r w h n u n g von der aktiven Teilname unseres Rechtsgelehr-
t e n gemeinsam mit einem anderen Juristen, namens Titius Aristo, an einem Kog-
nitionsverfahren, das von dem obersten Gerichtshof Traians (consilium Traiani)
im Beisein des Kaisers, wohl im Rom beurteilt wurde. 3 1
M. Ulpius T r a i a n u s traf nun nach der b e r n a h m e des Prinzipats erst im
H e r b s t 99 in R o m ein, wo er bis 25. Mrz 101 blieb, als er gegen die Daker auf-
brach. 3 2 Sein R c k k e h r von dem Kriegsschauplatz in die H a u p t s t a d t erfolgte
e t w a im Herbst 102. Der Princeps verweilte d a n n von kurzweiligen Reisen in-
n e r h a l b Italiens abgesehen bis 4. J u n i 105 in Rom, 3 3 da zog er zum zweiten
Mal gegen die D a k e r ins Feld. Der siegreiche Herrscher kehrte im Sommer 107
wieder nach R o m zurck. Die spteren D a t e n f r die Aufenthaltsorten Traians
hier anzufhren, ist unntig, da die im obigen wachgerufenen liefern schon einen
gengenden chronologischen Rahmen f r die Nchstfolgenden.
Die spteste Nachricht ber den Lebenslauf des Titius Aristo, Mitglieds
des von Traian prsidierten kaiserlichen Gerichtshofs s t a m m t nmlich aus
einem im J . 105 an Aristo gerichteten Briefe des Plinius. 3 4 In diesem Brief
ersucht Plinius die Begutachtung des Aristo ber seine Rolle und Stellung-
n a h m e in jener Untersuchung, die f r die A u f k l r u n g der Todesumstnde des
w h r e n d seiner Amtszeit, am 24. J u n i 105 in seiner Wohnung leblos gefundenen
Suffektkonsuls, Cn. Afranius Dexter 3 5 im Senat durchgefhrt wurde. Wenn
a u c h das 8. Buch der Epistularum, das dieses Brief des Plinius enthlt, mit
einigen J a h r e n s p t e r verffentlicht wurde, 3 6 ist es k a u m denkbar, da Plinius
die B e g u t a c h t u n g des von ihm hochgeschtzten Juristen erst mit J a h r e n
s p t e r nach dem obigen Ereignis u n d nicht gleich nach der Beendigung der
senatorischen U n t e r s u c h u n g einzuholen bestrebt hat. Der Lebenslauf des
30
L. S c h u m a c h e r , Prosopographische U n t e r s u c h u n g e n zur Besetzung der vier
h h e n rmischen Priesterkollegien im Zeitalter der Antonine u n d der Severer, Mainz
1973 222 f.
31
D i g . 37. 12,6. C f . T H . MoMMSEN, G . S e h r . 2.22, A n m . 6.
33
Fl - diese u n d die nchtsfolgenden D a t e n s. R . H a n s l i k , M. Ulpius Traianus,
R E Suppl. X (1968), 1050 ff.
33
R . H a n s l i k , . O. 1075 gibt irrig den 6. J u n i 105 an. I m Frg. X I X . Z. 5 der F O
( l . v i d m a n , . o . 18) liest m a n aber: Pr. nn. In.
34
Plin., . 8,14.
35
Das D a t u m gibt das Frg. X I X , Z. 6 f. der F O an.
36
A. N. S h e r w i n - W h i t e , The L e t t e r s of Pliny, Oxford 1966, 41. Cf. 461.

D I E Z W E I L. N E R A T I I PRISCI 417
Titius Aristo lt sich demnach wohl hchstens nur bis zum dritten Viertel
105 verfolgen. Als Titius Aristo noch die Vorlesungen des C. Cassius Longinus
( s u f f . 30), des einflureichen Rechtsgelehrten unter Claudius und bis 65 u n t e r
Nero gehrt hat, 3 7 war er im J . 105 gewi schon b e j a h r t und stand allem
Anschein nach schon a m Ende seiner Laufbahn. 3 8 Seine mit dem Juristen
Neratius Priscus gemeinsame Ttigkeit am kaiserlichen Gerichtshof u n t e r
Traian lt sich allenfalls quellenmig nur bis 105, f r die J a h r e 99101
u n d (oder) 103105 unterbauen. Der jngere L. Neratius Priscus verwaltete
aber eben in diesen J a h r e n nacheinander zwei kaiserliche konsulare Provinzen
und konnte folglich k a u m whrend seiner hiesigen Amtszeiten als consiliarius
am obersten Gerichtshof ttig sein. E r ist also nicht der fast gleichnamige
Jurist, den wir neben ihm und dem lteren L. Neratius Priscus als den d r i t t e n
fast homonymen Senator aus dieser Zeit in Evidenz halten mssen.
F r die Konsulaten der obigen drei Neratii Prisci sind aber bis jetzt
nur zwei Daten in den Konsulfasten berliefert. E b e n d a r u m wre es zurzeit
noch verfrht, den lteren L. Neratius Priscus bestimmt mit dem gleichnamigen
Suffektkonsul des J . 87., den Jngeren aber mit jenem des J . 97. ohne Weiteres
identifizieren. Man m u also zuerst noch genauer umsehen, wann der d r i t t e
Neratius Priscus das hchste Amt bekleiden konnte. U n d wenn diese Nach-
forschung zur Erfolge f h r e und demzufolge sich herausstelle, da der J u r i s t
weder im J . 87, noch i m J . 97 Konsul gewesen sein konnte, dann werden die
im obigen noch mit Fragezeichen versehenen Behauptungen in der Tat zwin-
gend, da nmlich die zwei L. Neratii Prisci der revidierten Bauinschrift aus
Saepinum den Konsulat im J . 87, bzw. 97 bekleideten u n d der Jngere von
ihnen nach seinem K o n s u l a t zwischen 97/98 u n d 100/101 die Provinz Germania
inferior, danach aber zwischen 102/103 und 105/106, bzw. 104/105 die noch
ungeteilte Pannnia verwaltete. 3 9
II
F a s t vor achzig J a h r e n hat O. Benndorf von einem 68 cm hohen u n d

25 cm breiten Quaderstein, der auf dem Gebiet der antiken lykischen S t a d t
Tlos in ein neuzeitliches Gebude sekundr erbaut wurde, den Rest einer vier-
zeiligen Inschrift verffentlicht, 4 0 die nachdem in das Berliner Corpus der
37
Dig. 4.8,40: nam et Cassium audisse se dicentem Aristo ait etc. ber C. Cassius
Longinus s. P I R 2 501. W. K u n k e l , H e r k u n f t u n d soziale Stellung der rmischen
Juristen 2 , Graz Wien Khl 1967, 130f. D . L i e b s , Rechtsschulen u n d R e c h t s u n t e r r i c h t
im P r i n z i p a t , A N R W , I I . 15, B e r l i n - N e w Y o r k 1976, 212.
38
T h . Mommsen, G. Sehr. 2,21. W. K u n k e l , . O. 141 f. . M a y e r - M a l y , s . v .
Titius Aristo, R E Suppl. I X (1962), 1395 f.
39
W e n n die maxima provincia, deren V e r w a l t u n g zu b e r n e h m e n P. Metilius N e p o s
in der ersten H l f t e des J . 105 schon in Begriff w a r (Plin., E p . 4.26,2) beziehe sich auf seine
Legation in P a n n n i a (R. S y m e , D P 185 f.), d a n n blieb L. N e r a t i u s Priscus n u r etwa bis z u r
Mitte 105 in dieser- Provinz. Gegen diese letztere Ansicht sprechen a b e r andere Grnde, die
in der a m E n d e dieses A u f s a t z e s angezeigten Arbeit eingehender e r r t e r t werden.
40
O. Benndorf, W S 24 (1902), 250, A n m . 1.

418 T. NAGY
griechischen Inschriften, 4 1 wie auch in die von R. Cagnat besorgte Sammlung

der auf die Rmerherrschaft bezglichen griechischen Inschriften 4 2 A u f n a h m e
f a n d , zuletzt aber in der magebenden Wiener Ausgabe der klein asiatischer
I n s c h r i f t e n in dem nachfolgenden revidierten Textgestalt 4 3 abgedruckt wurde :
. . . [, vlov xov I ixaio[xov Aovxov | Neoax\ov Qiaxov, | TXcocov / [ovXr/

xal r) ye I Qova[a.
Von allen Herausgebern wurde der Neratius der obigen Inschrift, wenn auch
meist mit gewissem Vorbehalt, mit dem L . Neratius Priscus pater, dieser aber
m i t dem fast gleichnamigen J u r i s t identifiziert und dessen lykische S t a t t -
h a l t e r s c h a f t in einer nher nicht festgelegten Zeit in Evidenz gehalten. hnlich
beurteilte unsere Inschrift die zeitgenssische prosopographische L i t e r a t u r
e t w a bis zu den fnfzigeren Jahren. 4 4 W . Reidinger hat d a n n mit einem Schritt
noch weiter gegangen u n d versuchte die lykische Legation dieses Neratius,
d e n sonst auch er als mit dem J u r i s t e n u n d dem lteren L. Neratius Priscus
identische Persnlichkeit beurteilte, auf die Jahre zwischen 84 u n d 87 zu
bestimmen. 4 5 Die U n h a l t b a r k e i t dieser Datierung, wie auch die Grundlosigkeit
der Identifizierung des Neratius der Inschrift aus Tlos entweder mit dem lte-
ren, oder mit dem jngeren L. Neratius Priscus der Bauinschrift aus Saepinum
h a t d a n n R. Syme erwiesen, der seine kritische Bemerkungen letzten E n d e s
darin zusammengefat h a t , da es das Beste wre, bezglich des Juristen von
d e m Neratius der lykischen Inschrift k n f t i g ganz abzusehen. 4 6 Diesem ableh-
n e n d e n S t a n d p u n k t schienen zwei zeitlich dicht darauf folgenden Mitteilungen
vllig R e c h t zu geben. Zuerst wurde aus einer unpublizierten Inschrift von
X a n t h u s der Name des [M. 4Q\QLOV 0gvxvo Negaxo[v Ilava) bekannt-
gegeben, 4 7 wodurch die lykische Legation dieses Neratius einen einwandfreien
Beleg erhielt. Aufgrund dieser neuen Belehrung wurde d a n n in einer zweiten
Mitteilung die besagte Inschrift von Tlos ebenfalls auf den N a m e n des \M. 'Iggiov
0QVXCVO] I Negax[ov Ilava] ergnzt 4 8 u n d diese Ergnzung h a t eine allge-
meine Zustimmung gefunden. 4 9 So h t t e also die Provinz Lycia u n d P a m p h y l i a
u n t e r den Flaviern aus der senatorischen Familie der N e r a t i e r von Saepinum
einzig nur M. Neratius P a n s a ( s u f f . 73 ?) verwaltet.
41
CIG I I I 4240b.
42
I G R I I I 1511.
43
I I 568.
44
P I R N 46. E . G r o a g , s . v . N e r a t i u s , R E X V I (1935), 2547, n. 13. D . Magie,
R o m a n R u l e in Asia Minor, I I . P r i n c e t o n 1950 (Reprint, A r n o P r e s s N e w Y o r k 1975),
1599.
45
W . R e i d i n g e r , . O. (Anm. 2), 58.
46
R . S y m e , H e r m e s , 1957, 486 f. Vgl. S E G 19 (1963), n o 868.
47
J . et L. R o b e r t , Bull. 1958, n o 294.
48
J . et L. R o b e r t , Bull. 1959, n o 112.
49
W . E c k , Die L e g a t e n v o n L y k i e n u n d P a m p h y l i e n u n t e r Vespasian, Z P E 6
(1970), 65 ff. I d e m , s. v. Neratius, R E S u p p l . X I V (1974), 283 f., n o 10. . ,
S P Q R , Gothoburgi 1975, 43 f. I d e m , L P I I . 2 (1977), no 30, p. 4. . K r e i l e r , Die S t a t t -
h a l t e r Kleinasiens u n t e r d e n F l a v i e r n , M n c h e n 1975, 106.

Doch stellt sich die epigraphische Grundlage dieser Folgerung bei einer
N a c h p r f u n g als unzulnglich heraus. Von der Neratius-Inschrift von Tlos
sind nur die nachfolgenden vier Zeilenreste erhalten geblieben :
ixaio\ . .
Negari[ . . .
TXcoecovyl . . .
Qovaia
Die diesem Inschriftfragment im Wiener Corpus beigefgte schriftliche u n d

zeichnerische Dokumentation lt d a r b e r kaum einen Zweifel aufsteigen,
da der regelmige Quaderstein, der dieses Inschriftfragment a u f b e w a h r t
h a t , als der linksseitige untere Eckteil des ursprnglichen Steindenkmals a u f -
gefat werden soll und ferner, da auf diesem Eckteil eigentlich die vier letzten
Zeilenanfnge der ursprnglichen Inschrift auf uns gekommen sind. Links-
wrts fehlen also keine Buchstaben u n d f r die Ergnzung bleibt so nur die
rechtswrts fehlende Schriftflche brig. Die einzelne Zeilenlnge der ursprng-
lichen Inschrift ergeben sich nun aus Z. 3 und 4, die k a u m anders, als r/[ovXr]
xal rj ye] I Qovdi[a\ sich ergnzen lassen. In Z. 3 s t a n d e n also ursprnglich
gewi 18 Buchstaben, von denen 6 (und ein halbes) auf die erhaltene, whrend
11 (und ein halbes) auf die verlorengegangene Schriftflche eingemeielt wurden.
Als die Hhe der Buchstaben wie auch die Schreibweise in allen 4 Zeilenresten
durchwegs die Gleiche sind, die zur Z. 3 bemerkten gelten auch f r die Z. 1,
wo sonst der Quaderstein ebenfalls nur 6 (und ein halbes) Buchstaben a u f -
bewahrt hat. Nach der gesicherten Ergnzung, txato[rov bleibe R a u m
also auch in dieser Zeile hchstens n u r f r einen siebenbuchstblichen T e x t
brig. Die zuletzt vorgeschlagene Ergnzung, M. 'IQQOV tovo, mit ihren
16 Buchstaben lt sich jedoch eben deswegen kaum in der verfgbaren Schrift-
flche unterbringen, auch nicht, wenn man mit kleineren Schwankungen in
den Buchstabenzahlen der einzelnen Zeilen Rechnung trage. In dieser Zeile,
nach dem Wort, ixaio[orov, m t e also entweder n u r der ausgeschriebene
Vorname, oder aber dieser verkrzt u n d noch ein Familienname dieses Neratius
die Schriftflche ausfllen. Auer der von 0 . Benndorf vorgeschlagene u n d
auch von den spteren Herausgebern beibehaltenen Ergnzung Aovxiov
s t a t t dessen wohl auch ein anderer Vorname, wie z. B. Mgxov, oder hn-
licher stehen konnte wre eine Ergnzung auf M. | IQQOV, die auf die Abstam-
mung unseres Senators von M. Hirrius Pansa hindeute, epigraphisch noch
ebenfalls mglich. F r ein weiteres, dreisilbiges Kognomen, bleibt aber in
dieser Zeile kein Platz mehr brig.
In der Z. 2 die erhaltene Schriftflche hat nicht mehr, als nur 6 Buch-
staben a u f b e w a h r t , was d a f r sprechen kann, da der N a m e des Statthalters
mit etwa lockerer Buchstabenverbindung eingemeielt wurde, als die anderen
Textteilen. Mit Rcksicht darauf, drfen wir auf der fehlenden Schriftflche

420 T. NAGY
der Z. 2, nach der Ergnzung des Familiennamens, NegaT[ov f r das Kog-

n o m e n im Genitiv zumindest und gleich hchstens mit 7 8 Buchstaben
rechnen. Zum Weiterkommen in dieser F r a g e ist auch die Zeitstellung unserer
I n s c h r i f t vom Belang. Den Spteren teils vorgreifend, sind wir auch zu dem
E r g e b n i s gekommen, d a unsere Inschrift in die Flavierzeit gehrt. 5 0 Nun,
f r die mnnlichen Mitglieder der senatorischen Familie der Neratier aus
Saepinum, zu denen wir zweifellos auch den Neratius unserer Inschrift zurech-
nen mssen, sind aus diesem Zeitabschnitt nur die cognomina Priscus (drei-
mal), 5 1 ferner Marcellus, 52 Fronto u n d Pansa, 5 3 je einmal bezeugt. Nimmt man
diese Namengebung-Praxis bei den ersten Generationen der senatorischen
F a m i l i e der Neratier in Betracht, so scheint es berechtigt zu sein, auch das
K o g n o m e n unseres Neratius, dessen Zugehrigkeit zu diesen Generationen
der Neratier aus Saepinum kaum bezweifelt werden darf, auf den einen von
den obigen Zunamen zu ergnzen. Von diesen p a t nun allein das Kognomen
im Genitiv, IIQCXOV, mit seinem 7 Buchstaben einwandfrei in die fehlende
Schriftflche der Z. 2 ein. Das Kognomen im Genitiv Tlvaa flle dagegen die
zur Verfgung stehende Schriftflche nicht aus. Das Kognomen, MagxXXov,
lasse sich nur gedrngt unterbringen, 5 4 wem aber nicht nur die lockere Schreib-
weise des Statthalternamens, sondern auch die Nichterwhnung einer lykischen
Legation in der inschriftlich berlieferten L a u f b a h n des L. Neratius Marcellus
a u f s bestimmteste widersprechen. 5 5
Nach alledem spricht also das meiste dafr, da der Neratius der
I n s c h r i f t von Tlos d a s Kognomen Priscus gefhrt hat. E r h a t t e entweder
n u r die tria -nomina, L., bzw. M. (oder hnliches) Neratius Priscus, oder aber
zwei nomina. In diesem Fall knnte sein Vollname beispielweise auch L. (M.)
H i r r i u s Neratius Priscus luten. Aber auch dann, wenn er n u r die im vorigen
vorgeschlagene tria nomina gefhrt htte, knnte er schon deswegen nicht
50
Die nhere B e g r n d u n g s. unten, 459 f f .
51
L. N e r a t i u s P r i s c u s d. .: D 1033, 1034. A E 1927, 117; 1949, 23. AFA C X X
H . - L . N e r a t i u s Priscus d. J . : D 1034, A E 1969/1970, 152 und die S p t e r e n etwa vor-
greifend, Dig. 48.8,6. N e r a t i u s Priscus, d e r J u r i s t : Dig. 1.2,2,53 (u. a.), V. H a d r .
18,1. Die E r g n z u n g des teils getilgten N a m e n s eines . . .]us Priscus \ cos auf der
h e x a g o n a l e n Basis von S a v a r i a ( R I U 36) auf L. N e r a t i u s Priscus [L. B a l l a , Die Inschrift
eines Senators aus S a v a r i a , E S 4 (1967), 61] t r i f f t k a u m das Richtige. D a s nomen dieses
S e n a t o r s war gewi k r z e r , d a f r spricht die Proportioniertheit d e r Inschrift. Dieser
K o n s u l a r war kein N e r a t i u s .
62
Vgl. die oben, A n m . 28 zit. Stellen, f e r n e r Plin., E p . 3.8,1. Dig. 33.7,12,43.
63
Die lteren Zeugnisse f r diese beiden cognomina des M. H i r r i u s F r o n t o Neratius
P a n s a sind in R E X V I 2545 zusammengestellt. Neues Material bringen, Bull. 1958, no 294
(das K o g n o m e n P a n s a ergnzt), M. T o r e l l i , T h e cursus bonorum of M. Hirrius F r o n t o
N e r a t i u s Pansa, J R S 58 (1968), 170 ff. (beide cognomina sind ergnzt, d o c h knnen diese
n i c h t bezweifelt werden).
54
D a s Gleiche gilt a u c h von dem K o g n o m e n im Genitiv, 0QVTCOVO.
58
W a s m a n in d e r Kleine Pauly 4 (1972), 66 liest, da n m l i c h dieser Marcellus
m i t d e m in der I n s c h r i f t v o n Tlos e r w h n t e n N e r a t i u s identisch ist u n d da dieser die
L e g a t i o n in Lykien u n d P a m p h y l i e n ca 92 94 innegehabt h a t . sind g a n z unbegrndete
Behauptungen.

D I E ZWEI L. N E R A T I I PRISCI 421
mit dem einen oder dem anderen Homonymer der saepinatischen Bauinschrift
identisch sein, weil in deren bekannten L a u f b a h n e n eine Legation in L y k i e n
u n d Pamphylien nicht erwhnt wird.
Der lykische Statthalter [. ] Neratius [Priscus] d r f t e also entweder m i t
dem gleichnamigen Jurist, oder mit einem vierten, bisjetzt noch u n b e k a n n t e n
Homonymer identisch sein. Das zweite Glied dieser Alternative kann m a n
aber nur dann aufrecht erhalten, wenn es beweisbar wre, da der J u r i s t auf
keinem Fall mit dem Neratius der Inschrift aus Tlos identisch sein kann. Dies
liege auf der H a n d , wenn die besagte Inschrift entweder in die vor-, oder die
nachdomitianische Zeit datierbar wre. Der Jurist, dessen G e b u r t s d a t u m
den Spteren vorgreifend auf die fnfziger J a h r e verlegt werden kann,5
d r f t e nmlich seine in Einzelheiten zwar noch nicht aufgeklrten, aber wegen
seines Suffektkonsulats k a u m bezweifelbare prtorische L a u f b a h n normaler-
weise noch unter Domitian hinterlegen. Die Inschrift von Tlos lt sich jedoch,
wie wir es gleich sehen werden, ebenfalls in die domitianische Zeit einordnen,
wodurch ihre Beziehung auf den Jurist beinahe gewi u n d die A n n a h m e
eines vierten Homonymers unntig wird.
III
Die Inschrift von Tlos enthlt ein wichtiges Datierungselement in d e r

Bezeichnung des Statthalters als txatorrj. J e n e kaiserzeitliche Inschriften
aus dem lykisch-pamphylischen Bereich, in denen die hiesigen Gemeinden,
oder der lykische Bund, 5 7 oder Privatleute den Statthalter mit diesem, d e m
lateinischen iuridicus entsprechenden Titel 58 als den gerechten Rechtspfleger
beehren, gehren in die flavisch-traianische Zeit. Das frheste diesbezgliche
Zeugnis s t a m m t nmlich aus dem 1. Regierungsjahr Vespasians, 59 das spteste
aber, das in einer akephalen Inschrift von Tlos erhalten geblieben ist, aus der
Zeit zwischen 103 u n d 115. Alle brigen datierbaren Beispiele gehren in die
zwei letzten J a h r z e h n t e des 1. Jh.1 Daneben kommt schon in frhtraianischer
Zeit die Bezeichnung, 6 fjyepwv vor,2 die nachher von der spttraianischer
Zeit ab, ist vorherrschend geworden.3 Nach diesem epigraphischen B e s t a n d
58
b e r diese F r a g e s. eingehender unten, 462.
8
' b e r dies s. J . D e i n i n g e r , Die Provinziallandtage der rmischen Kaiserzeit,
Vestigia 6, Mnchen 1966, 73 ff.
58
Cf. D. M a g i e , De Romanorum iuris publici sacrique vocabulis sollemnibus in
Ghraecum sermonem conversis, Leipzig 1905, 28. R o s e n b e r g , s. v. iuridicus, R E X 1151,
ferner D 8842, A E 1966, 472, etc.
58
I I 131 (Sex. Mrcius Priscus, 69/70).
80
I I 571 (Ignotus). T r a i a n ist schon Dacicus, aber noch nicht Parthicus.
Zur Datierimg s. P . K n e i s s l , Die Siegestitulaturen d e r rmischer Kaiser. H y p o m n e m a t a
25, Gttingen 1969, 198.
81
I I 21 (T. A u r . Quietus, 80/81). D 8818 (P. Baebius Italicus, 85). I I
570 (L. Domitius Apollinaris, ca 9 3 - 9 6 ) . I I 594 (L. Iul. Marinus. ca 9 8 - 9 9 ) .
82
I I 421 (C. Trebonius Proculus M e t t i u s Modestus, ca 9 9 - 1 0 2 ) .
83
S. u. a. die D o k u m e n t e des O p r a m o a s d e n k m a l s in Rhodiapolis, I G R I I I 739,
I I 907, passim.
Acta Antiqua Aeademiae Scientiarum Hungarian 23, 1980

422 T. NAGY
k o n n t e der Neratius der Inschrift von Tlos die Provinz Lycia u n d P a m p h y l i a

frhestens unter Vespasian, sptestens aber unter Traian verwalten. Die im
letzten Jahrzehnte wiederholt sorgfltig zusammengestellte Statthalterliste
dieser Provinz64 ermglicht nun weitere Przisierungen.
Die fast vollstndige Liste u n t e r d e n Flaviern zeigt nmlich n u r eine
einzige m e h r j h r i g e L c k e , die zwischen d e n Legationen des P . Baebius Ita-
l i c u s u n d d e s A . I u l i u s Q u a d r a t u s n a c h 8 5 u n d v o r 90, e t w a f r d i e A m t s j a h r e
8 6 / 8 7 8 8 / 8 9 v o r h a n d e n ist.65 I n d e r R e g i e r u n g s z e i t T r a i a n s z e i g t sich aber
n a c h Q. P o m p e i u s F a l c o u n d v o r T i . Iulius F r u g i , zwischen 105/106 u n d 112/113
eine n o c h grere L c k e , die der A m t s d a u e r z u m i n d e s t zweier Statthalter
66
entspreche. U n s e r Neratius k n n t e also allein nur e n t w e d e r u n t e r D o m i t i a n ,
z w i s c h e n 86/87 u n d 88/89, o d e r u n t e r T r a i a n , zwischen 105/106 u n d 112/113
die Provinz Lycia u n d P a m p h y l i a innehaben. Auch schon deswegen darf m a n
s e i n e I d e n t i t t m i t M . H i r r i u s N e r a t i u s P a n s a ( s u f f . ?73) n i c h t e r n s t l i c h e r w -
gen. E i n e L e g a t i o n in der l e t z t e r w h n t e n Zwischenzeit w r e jedoch a u c h f r
d e n J u r i s t zweifelsohne schon z u s p t . E s b e d e u t e n m l i c h , d a er d a s h c h s t e
A m t f r h e s t e n s u m 109/110, s p t e s t e n s a b e r n u r in d e n l e t z t e n Regierungs-
jahren Traians angetreten htte. Der erstere Zeitansatz fllt aber schon dadurch
64
W . E c k , Senatoren, 112 ff., s. a u c h 244. B . K r e i l e r , . O. (Anm. 49), 103 f.
B. Thomae, LP I I . 2 ( 1 9 7 7 ) , NO 3 0 .
65
W . E c k , . O. 138 f. rechnet m i t dieser L c k e zwischen 86/87 u n d 90/91. A. I u l .
Q u a d r a t u s ist aber n a c h d e n J a h r e n 87 u n d 88 im J . 89 n u r bis z u m 19. Mai in R o m
b e z e u g t ( A F A CXVT sq. H.). Die auf die zweite H l f t e dieses J a h r e s bezgliche Arval-
a k t e n sind zwar nicht e r h a l t e n gebheben, d o c h die N i c h t e r w h n u n g des A. Iul. Quad-
r a t u s in d e n Teilnehmerlisten der A r v a l b r d e r a m 3. u n d 22. J a n u a r , ferner a m 27. 28.
Mai 90 ( A F A C X X I V sq. H.) m a c h t es b e i n a h e gewi, d a Q u a d r a t u s b e f a n d sich in der
e r s t e n H l f t e dieses J a h r e s schon nicht m e h r in R o m . E r h a t die Regierung in L y k i e n
u n d P a m p h y l i e n also wohl schon in der z w e i t e n H l f t e des J . 89 b e r n o m m e n . Seine
hiesige L e g a t i o n d a u e r t e gewi zumindest bis z u m E n d e 92, als er a n der Z u s a m m e n k u n f t
d e r A r v a l b r d e r a m 16. November 92 n o c h n i c h t anwesend w a r . Die L e g a t i o n des
P . B a e b i u s Italicus ist d u r c h I G R I I I 548 f r d a s J a h r 85 einwandfrei bezeugt. Diese
I n s c h r i f t w u r d e gewi vor 14 September 85 (4. trib. pot. u n d 11. K o n s u l a t Domitians)
gesetzt. Die R i c h t i g k e i t des u n t e r e n Zeitansatzes, April 85, h n g t jedoch d a v o n ab,
d b D o m i t i a n seine 9. imperatorische A k k l a m a t i o n tatschlich in diesem M o n a t ange-
n o m m e n h t t e , wie d a f r in BMC I I 314 a r g u m e n t i e r t wird. W i r knnen a b e r W . E c k
d a r i n u n b e d i n g t zustimmen, d a Baebius I t a l i c u s seine lykische Legation f r h e s t e n s n u r
u m die M i t t e 84 a n t r e t e n konnte, d a er i m C h a t t e n k r i e g D o m i t i a n s (a. 83) noch die
legio X I I I I Gemina Martia Victrix k o m m a n d i e r t e (IGR I I I 551 = D 8818). Die Zeit-
d a u e r seiner Legation in Lykien ist jedoch ungewi. Jedenfalls k n n t e e n t w e d e r P .
B a e b i u s I t a h e u s oder sein u n m i t t e l b a r e r Nechfolger n u r zwei J a h r e lang die P r o v i n z
v e r w a l t e n . Die steife A n w e n d u n g d r e i j h r i g e r L e g a t i o n b e w h r t sich nicht i m m e r .
66
Q. P o m p e i u s Falco h a t an der Spitze der legio V Macdonien im 1. D a k e r k r i e g
T r a i a n s t e i l g e n o m m e n u n d militrische A u s z e i c h n u n g e n verdient. D a n a c h u n d vor seinem
S u f f e k t k o n s u l a t im J . 108 h a t er zuerst L y k i e n u n d P a m p h y l i e n , bald die P r o v i n z I u d a e a
v e r w a l t e t (D 1035, 1036). Die f r diese zwei Legationen von W . E c k , Senatoren, 160 f.
vorgeschlagenen D a t e n , 1 0 2 / 1 0 3 - 1 0 4 / 1 0 5 , b z w . 1 0 5 / 1 0 6 - 1 0 6 / 1 0 7 t r e f f e n allem Anschein
n a c h d a s Richtige. Ti. Iul. F r u g i wird in d e n I n s c h r i f t e n des O p r a m o a s d e n k m a l s
w i e d e r h o l t als der Legat der Provinz L y c i a u n d P a m p h y l i a e r w h n t : I I 905, I F , 2;
I I B , 14; I I F , 4; I V E , 10; VB, 16. E . R i t t e r l i n g , Zur Zeitsbestimmung einiger U r k u n -
d e n v o m O p r a m o a s d e n k m a l , R h . M. 73 (1920), 35 ff., u n d n a c h i h m D . M a g i e , R o m .
R u l e , I I . 1599 d a t i e r t e n die S t a t t h a l t e r s c h a f t des Iul. F r u g i auf d a s J . 114. W . E c k ,
. O. 117 u n d n a c h i h m B . T h o m a e , L P I I . 2, n 30, p. 8 auf die A m t s j a h r e 113/115.

DIE ZWEI L. NERATII PRISCI 423
weg, d a aus den J a h r e n 109114 alle a m t s f h r e n d e K o n s u l n b e k a n n t sind,'7

u n d u n t e r diesen der N a m e unseres N e r a t i u s fehlt. Aus d e m zweiten Zeit-
a n s a t z l a s s e n s i c h w e g e n h n l i c h e n G r n d e n e n t w e d e r n u r d a s J . 115, o d e r
d i e J a h r e n a c h 116 i n B e t r a c h t z i e h e n . S e i n B r u d e r , M a r c e l l u s , w a r n u n
s c h o n i m J . 95 S u f f e k t k o n s u l . E i n 19-, o d e r n o c h m e h r j h r i g e s I n t e r v a l l
zwischen den K o n s u l a t e n zweier B r d e r stehe aber i n der F r h p r i n z i p a t s z e i t
ganz vereinzelt da,68 u n d l t sich a u c h b e z g l i c h des J u r i s t e n u n d seines
B r u d e r s k a u m ernstlich erwgen. M a n d a r f also a u s diesem U m s t a n d w o h l
mit R e c h t die Folgern ng ziehen, d a d e r J u r i s t u n t e r T r a i a n nicht eine S t a t t -
h a l t e r s c h a f t in L y k i e n b e k l e i d e n k o n n t e .
67
Der bei A. DEGRASSI, FC 34 noch fehlende Suffektkonsul des J . 114 ist inzwi-
schen d u r c h das Militrdiplom von Pisarevo in der Person des L. Caesennius Sospes
b e k a n n t geworden. Cf. M. M. RAFSAET, R M D 14.
68
I n den julisch-claudischen Zeiten kommen hchstens 1012jhrige Intervalle
vor. Beispiele d a f r :
Zehnjhriges Intervall: C. Sulpicius Galba (ord. 22), L. Livius Ocella Ser. Sulpicius
Galba (ord. 33). A. DEGRASSI, FC 9 f. P I R S 721 u n d 723. Suet., Galba, 3,4.
Elfjhriges Intervall: P. Cornelius Scipio (ord. 56), P . Cornelius Scipio Asiaticus
(sujf. 68). A. DEGRASSI, FC 15 und 18. P I R 2 1439, 1440.
Zwlfjhriges Intervall: L. Seius Tubero (sujj. 18), L. Aelius Seianus (ord. 31).
A. DEGRASSI, FC 8 und 10. Tubero und Seianus waren aber Halbbrder. Zu den bestrit-
tenen Familienverhltnissen dieser Mnner s. jetzt D. HENNIG, L. Ael., Seianus, Vestigia
21, Mnchen 1973, 15 f.
U n t e r den Iuliern und Claudiern wie es die Nachfolgenden exemplifizieren mgen,
berwiegen aber die gleichzeitigen, oder die zeitlich dicht aufeinanderfolgenden Konsu-
laten der B r d e r :
Cossus Lentulus (ord. 25), Cn. Lentulus Gaetulicus (ord. 26). A. DEGRASSI, FC 9.
P I R 2 1381, 1390.
L. Cassius Longinus (ord. 30), C. Cassius Longinus, der bekannte J u r i s t ( s t i f f . 30).
A . DEGRASSI, F C 10. P I R 2 5 0 3 u n d 501.
F a u s t u s Cornelius Sulla (stiff. 31), L. Cornelius Sulla Felix (ord. 33). A. DEGRASSI,
FC 1. c. P I R 2 1459 und 1465.
A. Vitellius (sujj. 32), L. Vitellius (ord. 34). A. DEGRASSI, FC 1. C. Vgl. Suet,, Vit. 2.
T. Statilius T a u r u s (ord. 44), T. Statilius Taurus Corvinus (ord. 45). A. DEGRASSI,
FC 12. P I R S 618 und 595.
A. Vitellius (ord. 48), L. Vitellius ( s u f f . 48). A. DEGRASSI, FC 14.
C. Vellerns Paterculus (sujf. 60 ?), L. Vellerns Paterculus (sujf. 61?). A. DEGRASSI,
FC 16 f. Vgl. R. HANSLIK, S. V. Vellerns, R E V I I I A (1955), 660.
I n der Flavierzeit bedeuteten vier J a h r e schon ein greres Intervall:
L. Volusius Saturninus (ord. 87), Q. Volusius Saturninus (ord. 92). A E 1949, 23.
Beide waren die Shne des Q. Volusius Saturninus (ord. 56). S. die Stemma: P I R V 603.
Meist haben jedoch auch in der Flavierzeit die Brder gleichzeitig oder aber im
raschen Nacheinander das hchste A m t bekleidet. Von den Mitgliedern des kaiserbchen
Hauses abgesehen (T. Caesar Vespasianus ord. 70, Caesar Domitianus s u f f . 71) die fratres
Curvii (Mart,, 5, 28; vgl. Plin., Ep. 8.18,4), Cn. Domitips . . . Lucanus und Cn. Domi-
tius . . . Tullus, haben die Fasces um 77/78 gleichzeitig oder eher dicht nacheinander
gefhrt. D 990, 991. I R T 527, 528. W . ECK, Senatoren, 91 f.
Von dem Bruderpaar, Sex. Vettulenus Cerialis und C. Vettulenus Civica Cerialis
[cf. H . NESSEDHAUF, CIL X V I p. 20, n. 5. W . ECK, S. v. Vettulenus, R E Suppl. X I V
(1974), 842 f., no 1; 846, no 3] drfte der Erstere nach den datierten Posten seiner p r t o -
rischen und konsularen Laufbahn den Konsulat im J . 72 oder 73 bekleiden. E r w a r
n m b c h legtus legionis V Maeedonicae zumindest zwischen 68 70; Jos., B J 3,310;
6,237. Nach der Einnahme Jerusalems (im September 70) wurde er von Titus dekoriert
(D 988) und mit der legio X Fretensis als S t a t t h a l t e r (/fetrrp) in Iudaea zurckgelassen,
wo er frhestens um die Mitte 71 von Lucilius Bassus abgelst wurde: Jos., B J 7,17 + 163.
CIL X V I 1 4 - 1 6 . - Am 28. April 75 ist Sex. Vettulenus Cerialis als S t a t t h a l t e r der

424 T. NAGY
Wre es also mglich den Neratius der Inschrift von Tlos eben deswegen
f r einen vierten H o m o n y m e n , vielleicht den Vater des L. Neratius C. f.
Proculus ( s u f f . . 145) 69 zu halten? Die A n n a h m e eines vierten Neratius ist
jedoch solange nicht ntig, bis sich bezglich des Juristen nicht auer allem
Zweifel herausstelle, d a er eine Legatur in Lykien u n d Pamphylien nicht
einmal zwischen 86/87 u n d 88/89 bekleiden konnte.
Der im allgemeinen angenommene Lebensalter des Juristen scheint in
der T a t auf dem ersten Blick gegen seine Legation in Lykien u n d Pamphylien
auch in dem angezeigten Zeitabschnitt sprechen. Wenn nmlich wir jener
herrschenden Ansicht zustimmen, da sein Bruder, L. Neratius Marcellus,
in der zweiten H l f t e der fnfzigeren J a h r e n , u m 56/57, 70 oder genau im J . 59
geboren ist, 71 u n d Priscus der jngere B r u d e r gewesen war, 72 in diesem F a l l
k n n t e er normalerweise in seinem ca 26 27. Lebensjahr die Legation in
einer kaiserlichen prtorischen Provinz k a u m antreten. 7 3 Das jngere Alter
des Neratius Priscus grndet sich aber einzig auf jene mehr als bestreitbare
A n n a h m e , da er mit dem fast gleichnamigen u n d frher mit dem lteren,
neulich aber mit d e m jngeren L. Neratius Priscus gleichgestellten Suffekt-
k o n s u l des J . 97 identisch ist und folglich wohl mit einigen J a h r e n jnger
gewesen war, als sein Bruder, der schon im J . 95 zum hchsten Amt gelangt ist.
D a sich aber die I d e n t i t t des Juristen weder mit dem lteren, noch mit dem
jngeren L. N e r a t i u s Priscus aufrecht erhalten lt, wird auch seine vermeint-
liche Identitt mit dem fast gleichnamigen Suffektkonsul des J . 97 sehr frag-
lich. E s ist nmlich beinahe gewi, da jener Neratius, der die fasces in diesem
J a h r gefhrt h a t , wurde in den nachfolgenden J a h r e n von Traian nach-
einander in zwei wichtigen kaiserlichen Provinzen mit Legation betraut.
Dieser Senator, wie d a r a u f im Vorigen schon hingewiesen wurde, konnte jedoch
vor 106 kaum als consiliarius Traiani am obersten Gerichtshof ttig sein.
D e r Ratgeber Traians bei den Kognitionsverfahren war aber gewi der J u r i s t
Neratius Priscus. Folglich ist es ebenso gewi, da der damals in den Provinzen
ttige Homonymer nicht der Jurist sein konnte. D a m i t flle aber der einzige
vermeintliche Beleg weg, der bisjetzt f r das jngere Lebensalter des Juristen
P r o v i n z Moesia bezeugt (M. M. R a p s a e x , . O. 2), somit k o n n t e er seine hiesige Legation

schon in der zweiten H l f t e des J . 74 a n t r e t e n . E i n neues Frg. der F O (AE 1968, 6)
bezeugt, da in der e r s t e n H l f t e dieses J a h r e s Cerialis nicht d a s A m t eines Suffekt-
konsuls besetzt h a t . So bleiben f r seinen K o n s u l a t n u r die im obigen angedeuteten
J a h r e n 72, oder 73 brig. Sein Bruder erhielt d a s hchste A m t m i t Bercksichtigung
seines Prokonsulats in Asia (im J . 87/88, oder 88/89), sptestens im J . 75. A. S t e i n
(Die Legaten von Moesien, Budapest 1940, 34) n i m m t das J . 76 an, whrend W . E c k ,
( R E Suppl. X I V . 1. c.) h l t auch ein f r h e r e s D a t u m (?74) f r mglieh.
69
D 1076. Z u r D a t i e r u n g G. A l f l d y , K o n s u l a t u n d Senatorenstand u n t e r d e n
Antoninen, B o n n 1977, 140.
"> E. G r o a g , R E X V I 2593.
71
R . Syme, H e r m e s , 1957, 483, 491 (u. a.).
72
R . Syme, . O. 492. G. C a m o d e c a , . O. 30 f.
73
Vgl. bes. J . M o r r i s , Leges Annales u n d e r t h e Principat, L F 87 (1964), 316 ff.

D I E ZWEI L. NERATII PR I SCI 425
ins Feld gefhrt werden konnte. Nach dessen Eliminierung stehe nichts dem
entgegengesetzten Ansicht im Wege, da nmlich der Jurist, als der ltere
Bruder betrachtet werden kann, der wohl mit einigen Jahren frher, als
Marcellus, um die Mitte der Fnfzigerjhre geboren ist. Jene mgliche Ein-
wand, da in diesem Fall es schwer verstndlich wre, wie Neratius Priscus
als ein fast schon Achzigjhriger u m die J . 133/134 juristisch noch t t i g sein
knnte, 7 4 b t mit Hinweis auf die Geistesfrische einiger literarisch oder im
ffentlichen Leben noch ttigen Alton 75 ihre Beweiskraft stark ein.
Der mit dem spteren Kaiser, M. Ulp. Traianus (geb. 53)7e etwa fast
gleichaltrige Neratius Priscus knnte normalerweise schon um die Mitte der
Achzigerjahre die P r t u r bekleiden u n d danach unvermittelt die lykische
Legation bernehmen. Eine solche, zweifelsohne ungewhnliche mterlauf-
bahn 7 7 stehe doch in der Flavierzeit keinesfalls ganz einzigartig da. Ein Ignotus,
der als Statthalter in der Provinz Lycia u n d P a m p h y l i a unter Titus bezeugt
ist 78 und der wegen seiner Amtszeit mit dem in dieser Provinz zwischen 1. Juli
80 u n d 30. J u n i 81 belegten Statthalter, T. Aur. Quietus 7 9 (//. 82) wohl
74
Dig. 44.2,11 prol. b e r d a s hier e r w h n t e Senatus Consultum s. v. M e i n h a r t ,
Die D a t i e r u n g des SC Tertullianum, ZSS 83 (1966), 110 ff. b e r Neratius, 126 f.
75
M. Terentius V a r r o h a t sich genau in seinem achzigsten Lebensjahr a n d a s
Schreiben seines landwirtschaftlichen Fachwerkes d r a n g e m a c h t (Varro, r . r . 1,1).
T. Livius, geb. 59 v. Z., gest. 17 u. Z., h a t in seinen drei letzten Lebensjahren, also schon
als ein Siebziger, die letzten 22 B c h e r seines groangelegten historischen Werkes nieder-
geschrieben, bzw. -diktiert. P e r . C X X I . K l o t z , s. v. Livius n 9, R E X I I I 816 ff., bes.
818. Q. H a t e r i u s blieb promptus et popularis orator usque ad XC prope aetatis annum
Hieron., Chron. 172 g ( H e l m ) . Vgl. auch Tac., ann. 4,61: Q. Haterius, famlia senatoria,
eloquentiae, quoad vixit, celebratae. T . Vestricius Spurinna, in seinem 77. L e b e n s j a h r
schrieb noch outraque lingua lyrica doctissime. Plin., E p . 3,1,7. Vgl. H . B a r d o n , L a
l i t t r a t u r e latine inconnue, I I . P a r i s 1956, 232 , usw., u s w .
76
R . H a n s l i k , . O. (Anm. 32), 1035 f., wo ist noch nachzutragen, E p i t . 13,14.
Die abweichende D a t i e r u n g von J . M o r r i s , . O. (Anm. 73), 328: c. 56/7, ist nich ber-
zeugend.
77
Name Prtorische mter vor der lykischen Legatur Belege
. Caristanius leg. pr. pr. Ponti et Bithyniae \ leg. leg. IX D 9485

Fronto Hisp.
P . Baebius leg. leg. X I I I I G. M. V. D 8818
Italicus
A. Iulius leg. pr. pr.1 Ponti et Bithyniae \ leg. pr. pr. D 8819, 8819a
Quadratus Aeiae bis2 | leg. Aug. Cappadociae \ procos. AE 1966, 463
Cretae et Cyrenae
L. Iulius leg. pr. pr. prov. Cypri \ leg. pr. pr. prov. D 1026
Marinus Ponti et Bithyniae \ curator viae Tibur-
tinae 1 leg. Aug. leg. XI Cl. p. f .
1 2
sc. proconsulis i.e. sub. duobus proconsulibus. P I R 2 J 507.
78
I G R I I I 724, noch im Leben des Kaisers Titus in MYRA gesetzt.
78
I G R I I I 690 = D 8796. Vgl. P I R 2 A 1592.

426 T. NAGY
identisch sein kann, 8 0 wurde mit dieser Legation ebenfalls unmittelbar nach
seiner P r t u r b e a u f t r a g t . Auch sonst, wie es von einem guten Kenner der
frhkaiserzeitlichen senatorischen Laufbahnschemen schon etwa vor 40 J a h r e
t r e f f e n d bemerkt wurde, alle Vorschriften ber die Amterbesetzung in kaiser-
lichen R o m sind ja n u r regulae, Regeln, die durch Ausnahmen besttigt wer-
den. 81 Auch die L a u f b a h n des Juristen mag f r die Richtigkeit dieser Erken-
n u n g sprechen.
Der lykischen L e g a t u r unseres Neratius unter Domitian stehen andere
K a n d i d a t e n gar nicht im Wege. Die Statthalterschaft eines Unbekannten
w u r d e zwar alternative in die Regierungszeit dieses Kaisers, oder in die jene
T r a i a n s versetzt. 82 Von diesen Datierungen ist aber wohl n u r die letztere
v e r t r e t b a r . Dieser Ignotus wird nmlich in seiner akephalen Cursusinschrift
aus Tlos, als ixaiorrj betitelt, 8 3 und auerdem noch bei der Aufzhlung
der von ihm bekleideten m t e r , als zweimaliger Legionskommandant erwhnt.
Diese zwei Datierungselemente insgesamt verweisen die lykisch-pamphylische
Legation dieses U n b e k a n n t e n ziemlich gewi in die Regierungszeit Traians. 84
I n diesem Unbekannten knnen wir also allem Anschein nach den einen der
S t a t t h a l t e r aus dem Zeitraum zwischen 105/106 und 112/113 erkennen. Frag-
lich sind dagegen die Amtszeiten der zwei nchstfolgenden Ignoti, von denen
der E r s t e nur mit gewisser Wahrscheinlichkeit in die Regierungszeit Traians
verweisen mag. 85 Bezglich des Zweiten 86 liegen aber keine Anhaltspunkte
einer nheren Datierung vor. E s ist jedoch auch wohl mglich, da diese beiden
U n b e k a n n t e n mit dem einem oder anderem der schon namentlich bekannten
S t a t t h a l t e r identisch sind. Sie knnen jedenfalls nicht gegen die Statthalter-
s c h a f t des Juristen in Lykien zwischen 86/87 u n d 88/89 als Gegenargumente
gelten.
Die obigen erlauben die Folgerung zu ziehen, da hnlich wie Pannnia,
auch die Provinz Lycia u n d P a m p h y l i a zweimal von einem Neratius verwaltet
80
B. K r e i l e r , . O. (Anm. 49), 110. Skeptisch B. T h o m a e , L P I I . 2, no 30, p. 13.
D a jedoch der cursus bonorum, des C. Caristanius F r o n t o (D 9485) bezeugt, da dieser
S e n a t o r den T. Aur. Quietus noch im Leben des Kaisers Titus, also wohl u m die Mitte
des J . 81 von dieser L e g a t i o n abgelst h a t (s. auch W . E c k , Senatoren, 130), ist eine an-
d e r e Identifizierung k a u m mglich.
81
St. B r a s s l o f f , Zu I n s c h r i f t e n aus der rmischen Kaiserzeit, S e r t a Hoffilleriana,
Z a g r a b i a e 1940, 239.
82
W . E c k , . O . 2 2 8 .
83
I G R I I I 5 5 8 - I I , 569.
84
Der Titel txaiorrj f r den S t a t t h a l t e r ist im lykisch pamphylischen Bereich
n a c h d e r Regierungszeit T r a i a n s nicht bezeugt. S. oben, Anm. 60. u n d 61. Andernteils,
ein zweimaliges L e g i o n s k o m m a n d o in der p r t o r i s c h e n L a u f b a h n der Senatoren k a n n
m a n , v o n dem Vierkaiser j h r abgesehen, nicht v o r der Regierungszeit T r a i a n belegen.
Die f r h e s t e n Beispiele f r zwei Legionskommandos bieten die K a r r i e r e des T. Iul.
M a x i m u s . . . Manlianus (D 1016), ferner eines Ignotus (D 1021) u n d des A. Larcius Priscus
(D 1055) aus traianischer Zeit. Alle andere Beispiele gehren schon in die spteren Zeiten.
S. die diesbezgliche Liste bei G. A l f l d y , Legionslegaten, 78 ff.
85
I I 573, m i t E r g n z u n g auf diesen Kaisername.
86
I I 574 (kurz u n d fragmentarisch).

wurde. Die pannonische Legation haben die Mitglieder der plebejischen Linie
der senatorischen Familie der Neratier aus Saepinum zweimal innegehabt,
whrend die lykische die zur anderen Linie dieser Familie angehrigen Sena-
toren, die vom Kaiser Vespasian u n t e r die Patrizier erhoben wurden. 87
IV
Eine Legation in der kaiserlichen, prtorischen Provinz Lycia u n d

P a m p h y l i a b e d e u t e t e unter den Flaviern regelmig gleich den Abschlu d e r
prtorischen L a u f b a h n des betreffenden Senators. 8 8 D e r von dieser L e g a t i o n
abgetretene S t a t t h a l t e r knnte sich schon als der A n w r t e r eines S u f f e k t -
konsulats b e t r a c h t e n . Dieses hchste A m t h a t bekanntlich auch der J u r i s t
bekleidet. Somit e r h e b t sich die Frage, w a n n er seinen K o n s u l a t a n t r e t e n
konnte. U n t e r den Flaviern lt sich kein bestimmtes Intervall zwischen d e r
lykischen Legation u n d dem K o n s u l a t beobachten. D e r letztere folgte dieser
L e g a t u r zeitlich entweder u n m i t t e l b a r oder aber nach einem 2 4-jhrigen
Intervall. 8 9 Wenn also Neratius Priscus, der J u r i s t , Lykien verwaltet htte, d a n n
8
' N a c h der Unbewiesenheit jeder A r t einer Adoption bezglich des J u r i s t e n k a n n
aucli die Zugehrigkeit dieses Senators z u m P a t r i z i a t k a u m fraglich sein.
88
Bis jetzt is keine Ausnahme v o r h a n d e n . L. Iulius Marinus Caecilius S i m p l e x
w a r schon u n t e r T r a i a n vor seinem K o n s u l a t (a. 101) noch m i t d e m Prokonsulat d e r
Provinz Achaia b e t r a u t . W . E c k , . O. 148 f., m i t A n m . 152.
89
Von den S t a t t h a l t e r n Lykiens u n t e r Vespasian sind eigentlich nur die L e g a t i o n
das Sex. MarciusPriscus [68/69 und 69/70],sowie der Konsulat des M. Petronius U m b r i n u s
(sujj. 81) d a t i e r t . F u r die erste s. B. THOMAE, L P I I . 2, no 30, p. 3, mit allen Belegen,
f r den zweiten A. D e g r a s s i , FC 24. W e n n n u n Mrcius Priscus die Fasces im J . 79
g e f h r t (vgl. C I L X V I 25) und P e t r o n i u s U m b r i n u s die Provinz Lycia uns P a m p h y l i a
in den A m t s j a h r e n 76/77 und 77/78 (W. E c k , Senatoren, 123 f.) verwaltet h t t e n , d a n n
erstrecken sich die Intervalle in diesen beiden Fllen auf 8, bzw. 3 (2) J a h r e .
E t w a s besser sind wir bor die I n t e r v a l l e zwischen der lykischen Legatur u n d d e m
K o n s u l a t aus den Regierungsjahren des Titus u n d Domitians u n t e r r i c h t e t . Unsere dies-
bezglichen Kenntnisse lassen sich in der nachfolgenden tabellarischen bersicht z u s a m -
menfassen:
Name Belegtes Amtsjahr Letztes Amtsjahr Konsulat Intervall
T. A u r . Quietus 80/81 1 80/81 2 82 3 kein

. Caristanius F r o n t o 81/82* ca 83/84 5
906 7 (5)7
8 8
P . Baebius Italicus 85 85/86? 90 4 (?3)
A. Iul. Q u a d r a t e s 89/90" 92/93?" 94 3?(kein?)
L. Domitius
Apollinaris ca 95/96 11 97 12 -(kein?)
1: s. oben, S. 463, m i t A n m . 79. - 2: s. oben, S. 464, m i t A n m . 80. - 3: CIL X V I 28.

4: s. oben, S. 464, m i t A n m . 80. - 5: W . E c k , . O. 133. - 6: A E 1949, 23. - 7: Die
erste N u m m e r gibt d a s Intervall von d e m bezeugten J a h r der lykischen Legation an,
die zweite in K l a m m e r n jenes von d e m o f t n u r angenommenen letzten A m t s j a h r .
8: s. oben, S. 460, m i t A n m . 65. 9: s. die vorige Anm. 10: A. D e g r a s s i , FC 28.
- 11: W . E c k , . O . , 146. - 12: A E 1975, 132.

428 T. NAGY
darf m a n nach den Obigen mit gewissem R e c h t erwarten, d a auch er den

K o n s u l a t entweder im auf seiner Legation unmittelbar folgenden J a h r oder
mit etlichen J a h r e n spter angetreten h t t e .
Alle a m t s f h r e n d e Konsuln sind n u n aus dem J . 90 bekannt. 9 0 E i n
N e r a t i u s Priscus ist u n t e r ihnen nicht befindlich. Dagegen kommen seine zwei
Vorgnger in der lykischen Legatur, C. Caristanius F r o n t o u n d P . Baebius
Italicus in der Konsulliste dieses J a h r e s vor. Ein willkommenes Wink, d a
unser Neratius Priscus die fasces etwa s p t e r fhren mte. Die Konsuln a u s
den J a h r e n 91, 92 u n d 94 96 sind aber gleichfalls alle bekannt. 9 1 Ein Neratius
Priscus wird in diesen nicht erwhnt. W o h l begegnet uns u n t e r die Suffekt-
konsuln des J . 94 A. Iulius Quadratus, der unmittelbare Nachfolger unseres
N e r a t i u s in der lykischen Legation. N a c h diesem zweiten W i n k kann k a u m
ein Zweifel darber obwalten, da unser Neratius seinen Konsulat nur im
J . 93, d. h. nach einem dreijhrigen I n t e r v a l l nach seiner lykischen Legation,
bekleiden konnte.
Die Namen der Konsuln dieses J a h r e s sind bekanntlich nur teils ber-
liefert. Die F(asti) P(otentini) erhalten von den Suffektkonsuln dieses J a h r e s
allein die Namen des Avidius Qfuietus] u n d eines Corn[elius . . . aus den zwei-
ten u n d dritten Nundinien des J a h r e s . Aber unabhngig davon, ob dieser
Corn [elius . . . mit dem C. Cornelius R a r u s Sextius Na[so ?], dem Proconsul
Africae in einem nher unbekannten J a h r unter Traian identisch wre oder
nicht, 9 2 bleibe ein Platz f r den Konsulat des Neratius Priscus in diesem J a h r
noch in beiden Nundinien.
In den vorangehenden Jahren 92 u n d 91, wie auch in den nachfolgenden
J a h r e n 94 und 95 waren jhrlich je zwei Suffektkonsulpaaren im Amte. Auer-
d e m in den J a h r e n 92 u n d 95, als Domitian seinen 16., bzw. 17. Konsulat beklei-
dete, t r a t am Idus I a n u a r . and die Stelle des Kaisers noch ein Konsul. I n den
J a h r e n 91 und 94 amtierten also jhrlich 6, whrend in den J a h r e n 92 u n d 95
7 K o n s u l n , davon jhrlich 4, bzw. 5 Suffektkonsuln. Nach diesen gesicherten
A n g a b e n der Konsullisten erschien es als das Wahrscheinlichste, da auch im
J . 93, als Domitian einen eponymen K o n s u l a t nicht bernommen hat, ebenfalls
n u r 6 Konsuln, unter diesen 4 consules suffecti im Amt gewesen waren.9211
N u n h a t aber F . Zevi auf dem Mnchener Epigraphischen Kongresse
ein neues Fragment der F O eingezeigt, wo die letzten Buchstaben von 5 Suf-
fektkonsulpaaren erhalten geblieben sind u n d hat dieses Bruchstck dem J . 93
90
A E 1949, 23. A. D e g r a s s i , F C 27.
91
A. D e g r a s s i , F C 27 f. Vgl. L. V i d m a n , F O 16.
92
Die I d e n t i t t , b e h a u p t e t von R . S y m e , J R S 43 (1963), 153 h l t auch W . E c k ,
. O., 56, A n m . 6 als ziemlich gesichert. S k e p t i s c h ist dagegen B e n g t T h o m a s s o n , s. v .
Cornelius, R E Suppl. I X (1962), 22 f., n o 297a u n d Opusc. R o m . 7 (1969), 169. A. D e g -
r a s s i , F C 28 u n d H.-G. P f l a u m , Scripta m i n o r a , I . Paris 1978, 31 d e n k e n bezglich des
K o n s u l a t s des C. Cornel. R a r u s alternative a n d a s J . 93, oder 97. Gegen d a s letztere
D a t u m h a t aber zuletzt F . Z e v i , . O. ( A n m . 17), 137 Stellung g e n o m m e n .
2a y g l w E c K ) a 5 6

zugewiesen. 93 Wenn sich diese D a t i e r u n g bewhre, d a n n wren noch zumindest

8 Suffektkonsuln aus dem J . 93 u n b e k a n n t . F r diese 8, wenn nicht n u r 2
Pltze, hat nun die Forschung nach der Verffentlichung des F r a g m e n t s
X l l l d . der F O und des besagten F r a g m e n t s des F P die Namen der nachfol-
genden 11 Senatoren vorgeschlagen : 94
1. L. Dasumius (Hadrianus?), 9 5 2. A. Egrilius Plarianus Pater, 9 8 3. L.
Iulius Ursus Servianus, 97 4. Publicius (oder Publilius) Tullus, 98 5. Sallustius
Lucullus, 9 9 6. Silius Italicus, 1 0 0 7. Cn. Suellius Flaccus, 1 0 1 8. ein nher nicht
identifizierter Ummidius, 1 0 2 9. Vibius Varus, 103 10. L. Iulius Mar[inus ?], 104
11. T. Vestricius Spurinna. 1 0 5
Die berbewerbung f r die noch unbesetzten 2 oder 8 Konsulstellen im
J . 93 ist unbestritten. Auf die mit dieser in Zusammenhang stehende Frage, da
von den angefhrten 11 K a n d i d a t e n welche Senatoren sich bei einer nheren
P r f u n g bewhren, versuchen die Nachfolgenden eine A n t w o r t zu geben.
1. U n t e r dem Namen des ersten Kompetenten verstecken sich wohl
eigentlich zwei Persnlichkeiten : ein L. Dasumius . . ., proconsul Asiae u n t e r
Traian in einem unbekannten J a h r , 1 0 8 u n d ein [. . .] Hadrianus, der in einem
der J a h r e 104 113 ebenfalls diese Provinz als Prokonsul verwaltet hat. 1 0 7
Die I d e n t i t t dieser zwei, nicht mit ihren Vollnamen berlieferten Prokonsuln,
lt sich zurzeit nicht beweisen. 108 [ . . . ] Hadrianus d r f t e allenfalls mit R c k -
93
F. ZEVI, Nuovi f r a m m e n t i dei F a s t i Ostienses, A k t e n des VI. I n t e r n a t . K o n -
gresses f r Griechische und Lateinische E p i g r a p h i k Mnchen 1972 (1973), 438.
94
I n einem f r h e r e n Stand d e r F o r s c h u n g h a t m a n d u r c h die falsche D a t i e r u n g
d e r Militrdiplomen von Salona u n d Negovanovci (CIL X V I 38 u n d 39) die K o n s u l n
des J . 94 f r jene des vorangehenden J a h r e s gehalten. S. e t w a : W . LIEBENAM, Fasti
consulares imperii Romani, Bonn 1909, 17 u n d die Artikel ber die betreffenden S e n a t o r e n
in P I R u n d R E .
96
A . DEGRASSI, F C 2 8 : 9 3 / 1 0 0 c . R . SYME, T a c i t u s I I . O x f o r d 1 9 5 8 , 6 6 4 f . u n d
W . ECK, . O. 166 sind f r das J . 93.
96
A . DEGRASSI, F C . I . E . : 9 3 9 7 ? V g l . W . E C K , . O . 5 6 , A n m . 6 .
97
A. DEGRASSI, FC 27: p r i m a dell'87 (o nel 93).
93
A . DEGRASSI, F C 2 8 : 9 3 / 1 0 0 C.
99
A . DEGRASSI, F C 2 2 4 : p r i m a d e l l ' 8 6 (O n e l 9 3 ) .
100
A . DEGRASSI, F C 2 8 u n d 2 2 7 : 9 3 ?
101
W . ECK, Befrderungskriterien der senatorischen L a u f b a h n , A N R W I I . 1,
B e r l i n - N e w York 1974, 200, Anm. 201: mglicherweise 93.
102
R . SYME, J R S 43 (1953), 160. IDEM, The Ummidii, Histria, 17 (1968), 72 ff.
bes. 82 f.
103
R . SYME, J R S 43 (1953), I.e. (perhaps). S. auch L. VIDMAN, F O 47 f.
104
A. DEGRASSI, FC 26: p r i m a dell'87 (o nel 93). Das letztere D a t u m b e f h r w o r t e n
H . NESSELHAUF, G n o m o n , 2 6 ( 1 9 5 4 ) , 2 6 7 , R . SYME, T a c i t u s , I I 638.
105
H . NESSELHAUF, . O. 270: in einem der J a h r e 84 86 oder 93.
406
CIG 2876, P I R 2 D 14.
107
D a C. Aquilius Proculus f r d a s Prokonsulatsj h r 103/104 bezeugt ist (W. ECK,
u . O . 161, Anm. 214), erstreckt sich der D a t i e r u n g s w e r t der zwischen 103 und 11 3 g e p r g t e n
Mnzen aus T h y a t i r a selbstverstndlich n u r auf die im obigen T e x t angegebene Zeitspanne.
108
Nach dem Vorgang E . GROAG'S, P I R 2 D Add. p. X I a r g u m e n t i e r t e f r die
I d e n t i t t wiederholt R . SYME, J R S 43 (1953), 156, Tacitus, I I 636 (u. a.), der den Pro-
konsulat des L. Dasumius [ H a d r i a n u s ? ] auf das A m t s j a h r 106/107, w h r e n d dessen
K o n s u l a t auf das J . 93 verlegt h a t . V o r b e h a l t e sind dagegen in P I R 2 H 5. Ganz zurck-
h a l t e n d i s t B . THOMAE, L P I I . 2 ( 1 9 7 8 ) , 2 9 .

430 T. NAGY
sieht a u f die datierten Mnzen von T h y a t i r a einer-, u n d auf das allgemeinen

1213-jhrige I n t e r v a l l zwischen K o n s u l a t u n d P r o k o n s u l a t andrerseits das
e r s t e r e A m t frhestens i m J . 93, sptestens aber in einem der J a h r e 98 100
bekleiden. 1 0 9 F r L. D a s u m i u s . . . kann m a n jedoch noch die obige, zumindest
f a k u l t a t i v e Mglichkeit wegen seiner vllig u n b e k a n n t e n Amtszeit in Asia
n i c h t in Rechnung n e h m e n . 1 1 0
2. F r den wahrscheinlichen K o n s u l a t des A. Egrilius Plarianus P a t e r
im J . 93 berief sich A. Degrassi auf eine I n s c h r i f t von Ostia. 1 1 1 Aus der ange-
f h r t e n Inschrift geht jedoch nur soviel hervor, da Egrilius P a t e r zur Zeit,
als dieses beschriebenes D e n k m a l aufgerichtet wurde, schon Konsul gewesen
war. Nheres ber d a s J a h r seines K o n s u l a t s lt sich aber aus dem besagten
T e x t nicht ermitteln. Gegen einen mglichen K o n s u l a t dieses Senators im
J . 93 spricht a u e r d e m noch, da sein Sohn erst im J . 128 z u m K o n s u l a t
g e l a n g t ist. 112 W e n n also Egrilius P a t e r die Fasces schon im J . 93 g e f h r t
h t t e , daraus ergebe sich ein 34-jhriges I n t e r v a l l zwischen seinem und seines
Sohnes K o n s u l a t s d a t e n , was aber in den J a h r z e h n t e n u m die Wende des
1. z u m 2. J h t . ganz vereinzelt da stehe. 1 1 3 Dies in B e t r a c h t ziehend emp-
f i e h l t es sich den K o n s u l a t des Egrilius P a t e r mit einigen J a h r e n nach 97 zu
verlegen. J a , die neuesten Forschungsergebnisse scheinen d a f r sprechen, da
der a m 1. Okt. 128 belegte Suffektkonsul, A. Egrilius Plarianus, ist nicht m i t
d e m jngeren, sondern m i t dem lteren A. Egrilius Plarianus, also mit dem
P a t e r identisch sei. 1 1 3 a
3. L. Iulius U r s u s Servianus, der Schwager des spteren Kaisers H a d r i a n ,
ist n a c h der einleuchtenden Beweisfhrung von R . Syme 1 1 4 identisch mit dem
109
Das J . 97 fllo weg, wenn sich die v o n F . Zevi, . O. (Anm. 17), 137 f r dieses
J a h r vorgeschlagene K o n s u l l i s t e bewhre.
no Wenn j e m a n d d o c h auf der I d e n t i t t bestehe und den K o n s u l a t dieses Sena-
t o r s auf das J . 93 verlege, der v e r f a h r e richtiger, diesen Senator n i c h t als L. Dasumius
[ H a d r i a n u s ? ] , sondern i m Gegenteil als [L. D a s u m i u s ? ] H a d r i a n u s benennen. Eine mg-
liche Bekleidung des K o n s u l a t s in diesem J a h r folgt nmlich a u s den besagten Mnz-
l e g e n d e n nur fr [ . . . ] H a d r i a n u s . Die S t a t t h a l t e r s c h a f t des L. Dasumius . . . in der
P r o v i n z Asia ist dagegen n i c h t nher d a t i e r b a r .
111
CIL X I V 399. E . G r o a g , P I R 2 E 48 h a t gegen H . D e s s a u ad. tit. festgestellt,
d a die richtige A r t i k u l i e r i m g des Textes in den Z. 2 3 ist wie f o l g t : A. Egrili Plariani
Patris, p(atroni) c(oloniae), co(n)sulis. F r diesen Vollnamen bietet n u n ein willkommenes
n e u e s Beleg, AE 1955, 173.
112
A. D e g r a s s i , F C 37. Aber nicht dieser, sondern sein Onkel, adoptiert von einem
M. Acilius Priscus, hie M. Acilius Priscus Egrilius Plarianus. S. die S t e m m a der Familie
bei H . B l o c h , A Ostie, N d S 1953, 264.
113
S. die A u s f h r u n g e n von W . E c k , Senatoren, 67 f., A n m . 73, der von d e m
durchschnittlichen 2530-jhrigen Intervall n u r 4 Ausnahmeflle registriert, aber n u r
f r e i n krzeres I n t e r v a l l . E i n anderes Bild zeigt sich freilich in der frhantoninischen Zeit,
als ein Intervall von 34 o d e r noch mehreren J a h r e n ziemlich h u f i g v o r k a m . S. die dies-
bezgliche tabellarische Zusammenstellung bei G. A l f l d y , K o n s u l a t , 323 f.
,13a
f . Zevi, N u o v i documenti epigrafici sugli Egrili Ostiensi, M E F R 82 (1970),
279ff., bes. 300f. Vgl. M. C o r b i e r , L'aerarium Saturai et Vaerarium militare, P a r i s -
R o m a 1974,169 f. z u g e s t i m m t vom H.-G. P f l a u m , Les fastes de la province de Narbon-
naise, P a r i s 1978, 20.
114
R . Syme, J R S 43 (1953), 156. I d e m , Tacitus, I I . 636. S. auch, P I R 2 J 569
u n d 631.

D I E ZWEI L. N E R A T I I PRISCI 431
Ser. lui. Servianus ( s u f f . 90). Die Namensverschiedenheit erklrt in diesem

Falle gengend eine nach 90 und vor 102 erfolgte Adoption des Ser. lui.
Servianus durch den L. Iul. Ursus ( s u f f . 84).
4 5. Als Suffektkonsuln im J . 93 knnen dagegen weder Puhlilius (oder
Publicius) Tullus, noch Silius Italicus in Evidenz halten. Der erste von diesen
verdankte ihre Existenz einer verlesenen Mnzlegende, die richtig den N a m e
des L. Baebius Tullus ( s u f f . 95) u n d Prokonsul in Asia ca 100/110 erhalten
hat. 1 1 5 ber den zweiten K a n d i d a t , den angeblichen Sohn des b e k a n n t e n
Dichters, Ti. Catius Asconius Silius Italicus, h a t schon E . Groag das Wesent-
liche mitgeteilt. 116 Der Erstgeborene des Dichters, dessen Konsulat Martialis
in einem Epigramm beglckwnscht, 1 1 7 hie L. Silius Decianus und h a t die
fasces im J . 94 gefhrt. 1 1 8
6. Der nchste Kompetent, Sallustius Lucullus, hat zwischen 93 96,
bzw. 92/93 die Provinz Britannia verwaltet. 1 1 9 Wohl sein zweiter Nachfolger
in dieser Legatur, P . Metilius Nepos ( s u f f . 91), wurde im J . 98 durch T. Avidius
Quietus ( s u f f . 93) von dieser Post abgelst. 120 Die Konsulatsdaten der letzt-
genannten zwei Senatoren zeigen schon an u n d f r sich gengend, da der
Konsulat des Sallustius Lucullus d r f t e vor 91, ja die erhaltene Konsulliste
der F P bercksichtigend, vor 87 liegen. 121
7. Eine nhere Betrachtung erwnscht der neulich auf das J . 93 vor-
geschlagene Suffektkonsulat des Cn. Suellius Flaccus. 122 Dieser Vorschlag
grndet sich auf die Legatur des Flaccus in Numidien, wo es laut einer Inschrift
aus der Nhe von Syrtis, 123 als leg. Aug. propr. (sie !) f r das A m t s j a h r 86/87
in der Tat bezeugt ist. 124 Wie es schon lngst e r k a n n t wurde, die mit dem
Legionskommando verknpfte Legation in der Provinz Numidia e r f f n e t e
dem hiesigen Befehlshaber meist den Weg zum Konsulat. Von den zwei
bekannten Vorgngern des Flaccus aus der Regierungszeit Domitians, L. Tet-
tius Iulianus bekleidete schon im auf seiner numidischen Legation folgenden
J a h r , C. Octavius Tidius Tossianus L. Iavolenus Priscus aber nach einjhrigem
115
P I R 2 29.
116
E . G r o a g , s. v . Silius, NO 14, R E I I I A (1927), 78.
117
Mart., 8.66. Z u r D a t i e r u n g : R . H a n s l i k , Die n e u e n F a s t e n f r a g m e n t e von Ostia,
W S 63 (1948), 117 ff., bes. 123.
118
A E 1949, 23. Vgl. R . Syme, J R S 43 (1953), 157. Flschlieh (D) gibt d a s P r -
n o m e n A. D e g r a s s i , F C 227 an.
119
A. E . B i r l e y , T h e R o m a n Governors in B r i t a i n , E S 4, 1967, 68. W . E c k ,
. O. 143.
120
CIL X V I 43.
121
W e n n sich jedoch seine I d e n t i t t m i t d e m P . Sallustius Blaesus b e w h r e
[cf. R . Syme, J R S 43 (1953), 158], d a n n h t t e er die Fasces schon im J a h r 89 g e f h r t .
A. D e g r a s s i , FC 27.
122
S. oben, A n m . 101.
123
P . R o m a n e l l i , Ti e iserizioni Tripolitane di interesse storico, E p i g r a p h i c a ,
1 (1939), 111 = A E 1940, 70 = I R T 854.
y g i G T h o m a s s o n , Die S t a t t h a l t e r der rmischen Provinzen N o r d a f r i k a s ,
I I . L u n d 1960, 158 f., u n d Opusc. R o m . 7 (1969), 180.

432 T. NAGY
Intervall das hchste Amt. 1 2 5 Wenn auch Ausnahmsflle prinzipiell nicht

geleugnet werden knnen, 126 ist es doch im gegebenen Falle ziemlich unwahr-
scheinlich, da Suellius Flaccus seinen Konsulat erst nach einem fnfjhrigen
Intervall nur im J . 93 angetreten htte. Das berechtigte Bedenken verstrkt
noch mehr der Umstand, da die L a u f b a h n des Cn. Suellius Flaccus nach seiner
numidischen Legatur ganz unbekannt ist, ja wissen wir nicht einmal, ob er
das hchste Amt berhaupt wirklich bekleidet htte. Ein anderer Vorschlag
also, d a nmlich Flaccus noch als Anwrter auf den Konsulat in Ungnade
gefallen, oder gestorben ist, scheint eine grere Wahrscheinlichkeit zu haben. 127
8 9. Vibius Varus und ein gewisser Ummidius scheiden nach der Neuda-
tierung des Fragments X X X I . der FO auf das J . 115 aus der Reihe der com-
petitor es ebenfalls aus. 128
10. Als einen ziemlich gut beglaubten Anwrter auf einen Suffektkon-
sulat des J . 93 darf man dagegen den L. Iul. Mar[inus?] ansehen, der in der
zweiten Hlfte der achzigeren Jahre, sptestens im J . 89/90 die Provinz Pontus
und Bithynia verwaltet hat 1 2 9 und im J a n u a r 97 schon als konsularer Statt-
halter in der Provinz Moesia inferior bezeugt ist. 130 Nach diesen Daten mte
Iul. Mar[inus?] die fasces unbedingt im J . 93 fhren, da sein Name fehlt in
den vollstndigen Konsullisten der Jahre 87 92 und 94 96. 131
11. Ebenso spricht das meiste dafr, da T. Vestricius Spurinna seinen
ersten Suffektkonsulat im J . 93 bekleidet hat. E r war bekanntlich cos. s u f f .
I I im J . 98.132 Aus der Liste der Suffektkonsuln dieses Jahres folgt nun unge-
zwungen, da Spurinna das hchste Amt zum ersten Mal unbedingt nach 84
125
Zu den Legation dieser Senatoren in Numidien s. B. E. THOMASSON, Statthalter,
I I 157 f., zu ihren Konsulaten s. A. DEGRASSI, FC 25 f.
128
S. die im obigen, S. 464 Bemerkten.
127
hnlich beurteilte diese Frage schon E . BIRLEY, J R S 52 (1962), 224.
128
F . ZEVI, . O . ( A n m . 93), 4 3 8 .
129
Erwhnt seinen hiesigen Prokonsulat die inschriftlich berlieferte L a u f b a h n
seines Sohnes (D 1026), der u n t e r seinem Vater legtus pro praetore war, gewi vor 91.
E . GROAG, s. v. Iulius, no 341, R E X 670. P I R 2 J 408. G. ALFLDY, Legionslegaten, 20.
W . E C K , . O . 140. B . K R E I L E R , . O . ( A n m . 4 9 ) , 143.
130
CIL XVI 41. Der in seiner Zeit noch berechtigte Zweifel, ob Mar[inus?] einzig
a u f g r u n d des Fundortes des fragmentierten Militrdiploms, gef. in der rmischen Festung
v o n Salsovia, das nicht einmal den Namen der Provinz enthielt, den S t a t t h a l t e r n Unter-
moesiens zugerechnet werden knne [s. A. STEIN, Die Legaten von Moesien, 58, B.
THOMAE, L P (Moesia, Dacia, Thracia), Gothoburgi 1977, 13], ist nach dem Bekannt-
w e r d e n des Militrdiploms von Viminacium [SI. DUSANIC M. R. VASIC, An Upper
Moesian Diploma of A . D . 96, Chiron, 7 (1977), 291 f.] gegenstandlos geworden. Cn.
P i n a r i u s Aemilius Cicatricula Pompeius Longinus ist mit diesem am 12. Juli 96 aus-
gestellten Diplom wohl auch f r das A m t s j a h r 96/97 on Moesia superior als S t a t t h a l t e r
b e z e u g t , und Mar[inus?], der laut dem Diplom von Salsovia seinen Post schon vor
J a n u a r 97 besetzen d r f t e , k o n n t e folglich nur die Provinz Moesia inferior verwalten.
131
S. a u c h H . NESSELHAUF, . O. ( A n m . 104), 267 u n d W . ECK, . O . 73, A n m . 109.
D e r letzteer Forscher r echnet auch mit dei' Mglichkeit, da Mar[inus?] schon vor 87
consul gewesen sein knnte. E s ist aber schon deswegen ganz unwahrscheinlich, weil sein
Vorgnger in dem P r o k o n s u l a t von Pontus u n d Bithynia, Ti. Iul. Celsus Polemaeanus
d e n Suffektkonsulat erst im J a h r 92 bekleidet h a t . A. DEGRASSI, FC 28.
132
A . DEGRASSI, F C 29.

antreten mte. Die erste Stelle u n t e r den Suffektkonsuln dieses Jahres n i m m t

nmlich der Neupatrizier Cn. Domitius . . . Tullus ein, 133 der seinen ersten
Suffektkonsulat im J . 77, oder 78 bekleidete. 134 Nach diesem Patrizier, der
am Idus Ianuar, an die Stelle des Kaisers Nerva getreten ist, folgen in der Liste
drei Plebejer, alle als cos. s u f f . I I . Zuerst, Sex. Iulius Frontinus, cos. s u f f .
erstmals im J . 73. 135 Dann, L. Iulius Ursus, cos. s u f f . das erste Mal im J . 84, 134
u n d als der dritte, T. Vestricius Spurinna, der nach dieser Nacheinanderfolge
gewi als der rangjngere unter diese Plebejern kenntlich gemacht wird.
Seinen ersten Suffektkonsulat h a t also Spurinna beinahe gewi nicht vor 84,
etwa noch unter Vespasian, 137 sondern nach 84 bekleidet. Das J a h r 85 k o m m t
aber aufgrund der neueren Forschungsergebnissen 1 3 8 nicht in Frage. W e n n
Spurinna die Fasces im J . 86 gefhrt u n d danach seine von Plinius bezeugte
Legation in Untergermanien bernommen htte, 1 3 9 d a n n wrde er zurzeit
des Saturninus-Aufstandes der S t a t t h a l t e r dieser Provinz gewesen. Der S t a t t -
halter in dieser Provinz war aber damals bekanntlich A. Bucius L a p p i u s
Maximus, 1 4 0 der Suffektkonsul im letzten nundinum des J . 87. 141 Spurinna
d r f t e also normalerweise spter, als Lappius Maximus, aber unbedingt vor
98 die Fasces fhren, da er schon zum zweiten Mal das hchste Amt erlangt
h a t . Das J . 97 fllt jedoch weg 142 u n d ebenso wenig kommen selbstverstnd-
lich auch die J a h r e 94 96 und 87 92 in Betracht. N a c h alledem bleibe also
nur das J . 93 f r seinen ersten Suffektkonsulat brig. Spurinna war damals
zwar schon fast ein Siebziger. 143 Dieser hohe Lebensalter schliee aber den
obigen Zeitansatz keinesfalls aus. Domitian hat im J . 96 den neunzigjhrigen
C. Manlius Valens zum eponymen Konsulat zugelassen 144 und Kaiser N e r v a
133
E r k a n n t von R . Syme, J R S 43 (1953), 156. Vgl. auch H . N e s s e l h a u f , . O. 270,
L. V i d m a n n , F O 51 f.
134
S. oben, A n m . 68.
134
PIR J 322, W . E c k , . O., 73? A. D e g r a s s i , FC 21: T74. Wenn Q. Petillius
Cerialis seinen zweiten Suffektkonsulat n i c h t in absentia bekleidet h t t e , d a n n m t e
F r o n t i n u s , der u n m i t t e l b a r e Nachfolger des Q. Cerialis in der Provinz B r i t a n n i a (Tac.,
Agr. 17), die Fasces sptestens im J . 73 f h r e n und so wre der R e s t . . .]ON[. . . eines
K o n s u l n a m e n s in d e n Fasti feriarum Latinarum (I. I . 1 . 1,2) richtiger nicht auf d e n
N a m e n des Sex. Iulius Fr]on[tinus, s o n d e r n wohl auf jenen des M. Hirrius F r ] o n [ t o zu
ergnzen. Cf. auch J . M o r r i s , . O. 328, 331.
ne PIR2 j 630, m i t vollstndigem A p p a r a t .
137
So W . E c k , . O. 58, Anm. 22. Sonst, Spurinna war im J . 69 kein Legionslegat,
sondern nach Tac., hist. 2.18,1 der Befehlshaber einer gemischten othonianischen T r u p p e .
S. auch, R . S a x e r , Untersuchungen zu d e n Vexillationen des rmischen Kaiserheeres
von Augustus bis Diokletian, E S 1, K l n Graz 1967, 17.
738
S. M o d u g n o - S . P a n c i e r a - F . Z e v i , Riv. storica dell'Antiehit, 3 (1973),
87 ff., bes. 106 = A E 1975, 131.
139
Plin., E p . 2.7,12. Vgl. auch E p . 3.1,12: (Spurinna) provincias rexit, wo der
P l u r a l eine hnliche b e r t r e i b u n g sein k a n n , wie z. B. in Plin., Paneg. 14,3, bei der
E r w h n u n g T r a i a n ' s K o m m a n d o in H i s p a n i a ber legiones.
no P I R I R 8 4 ) m i t a l l e n gelegen.
141
A E 1949, 23.
141
Vgl. F . Z e v i , . O. (Anm. 17).
143
Geb. u m das J a h r 24, Plin., E p . 3.1,10.
144
Cassius Dio, 67.14,5 ( B o i s s e v a i n , I I I . 182).

484 T. NAGY
im folgenden J a h r den zumindest in seinem 83. Lebensjahr stehenden L.Ver-

ginius R u f u s zum Kollegen im Konsulat gewhlt. 1 4 5
Von den im Obigen kurz besprochenen elf K a n d i d a t e n f r je einen Suffekt-
k o n s u l a t des J . 93 bewhren sich also eigentlich nur zwei Senatoren : T. Vestri-
cius Spurinna u n d L . Iulius Mar[inus?]. Alle beiden knnen neben T. Avidius
Quietus und dem Cornfelius . . . mit gewissem Recht in Evidenz halten. Als
der F n f t e wre zu diesen [ . ] Neratius [Priscus] anzureihen, der als S t a t t -
h a l t e r Lykiens zwischen 86/87 und 88/89 das hchste Amt, dessen Bekleidung
von einer Digestenstelle verbrgt wird, n u r in diesem J a h r antreten konnte.
E i n e n mglichen E i n w a n d gegen seinen K o n s u l a t in diesem J a h r darf man
a b e r an dieser Stelle nicht gnzlich auer A c h t lassen.
U n t e r Domitian galt es nmlich gewissermaen als Regel, da ein P a t r i -
zier sein Amt als Suffektkonsul u n m i t t e l b a r nach den consulee ordinarii
angetreten hat. 1 4 6 Die Fasti Potentini vermerken aber u n t e r den Suffekt-
konsuln des J . 93 als den Ersten den Plebejer (T.) Avidius Qfuietus], woraus
n a c h den Obigen logisch folge, da auch die brigen Suffektkonsuln dieses
J a h r e s zur plebejischen Nobilitt angehrt htten. Der J u r i s t als Patrizier
k o n n t e also bei strickter Anwendung der obigen Regel schwerlich in der Kon-
sulliste dieses J a h r e s einen Platz finden. E s ist aber nicht zu schwer, diesen
auf dem ersten Blick berechtigten E i n w a n d eben aufgrund der F P zu ber-
winden.
Der Text dieser F a s t i ist bekanntlich sehr fehlerhaft. Nicht nur die Namen
der amtierten K o n s u l n sind oft verschrieben, 1 4 7 sondern, was bis auf den
heutigen Tag weniger beachtet wurde, auch die Konsulnamen in den einzelnen
Nundinien wurden etliche Male miteinander verwechselt. Zwei Beispiele
gengen. Die F P registrieren beim J . 88 im d r i t t e n nundinum das Konsulpaar,
Sex. Iulius Sparsus, M. Otacilius Catulus. I n einem offiziellen Dokumente,
d e m Militrdiplome von Muhowo, 148 k o m m e n diese Konsulnamen dagegen
in umgekehrter Reihenfolge vor. Das Gleiche gilt f r das J . 90, wo im vor-
letzten nundinum liest man in den F P : Cn. Pompeius Longin(us), L. Pullaienus
Pollio. Die offizielle, umgekehrte Reihenfolge bewahrte das Militrdiplom von
Mainz. 149 Nach diesen Beispielen, die wohl noch vermehrt werden knnen,
scheint es berechtigt zu sein, auch im zweiten nundinum des J . 93 mit einer
hnlicher Verwechslung zu rechnen.
Zuletzt, die Datierung des im Obigen schon erwhnten neuen Fragments
der F O mit f n f Suffektkonsulnpaaren auf das J . 93 150 beseitige auch das
letzte Hindernis aus dem Wege des Konsulats unseres Neratius zu diesem
145
Plin., E p . 2.1,4.
146
H . N e s s e l h a u t , . O. (Anm. 104), 269. W . E c k , . O. 64.
147
Bemerkt schon v o n N. A l f i e r i , . O. (Anm. 16) in der Verffentlichung der F P .
148
CIL X V I 35.
149
CIL X V I 36.
159
F . ZEVI, . O. (Anm. 93), 438.

D I E ZWEI L. NERATII PRISCI 435
Zeitpunkt. Von den mit diesem Fastenfragment wohl bezeugten 10 Suffekt-

konsuln sind zwei, nmlich (T.) Avidius Quietus u n d ein Corn.[elius . . . durch
die F P belegt. Drei weitere Senatoren, T. Vestricius Spurinna, L. Iulius
Marfinus ?] und [. ] Neratius [Priscus] lassen sich ferner m i t der grten Wahr-
scheinlichkeit, fast mit Gewiheit als die Konsuln dieses Jahres in Evidenz
halten. Als der Sechste, trete zu diesen 5 Senatoren mglicherweise noch
[L. Dasumius?] H a d r i a n u s hinzu. Vier Suffektkonsuln bleiben also noch
unbekannt.
Aufgrund der Gesagten lt sich die Lebensgeschichte u n d die ffentliche
L a u f b a h n des Juristen bis zum Konsulat, wie es folgt, vorzulegen : geb. u m
die Mitte der fnfziger J a h r e , P r t o r u m 85, kaiserlicher L e g a t in der Provinz
Lycia u n d P a m p h y l i a zwischen 86/87 und 88/89, Suffektkonsul im J . 93,
etwa in seinem 38., oder 39. Lebensjahr. Dieses Alter k a n n gleich auch f r
den Patriziat des Juristen sprechen. Von den brigen flavischen Neupatriziern,
deren Lebensdaten genauer bekannt sind, h a t nmlich der ltere Zeitgenosse
des Juristen, Cn. Iulius Agricola (geb. am 13. J u n i 40, s u f f . 77) 151 in seinem 37.,
oder 38., whrend der mit dem Juristen fast gleichaltrige M. Ulpius Traianus
(geb. am 18. September 53, ord. 91) 152 in seinem begonnenen 38. Lebensjahr
den Konsulat angetreten.
Unserem Neratius Priscus wurden konsulare A m t e r , soweit wir wissen,
weder unter Domitian, noch u n t e r den spteren Kaisern anvertraut ; wohl
h a t er aber als der R e p r s e n t a n t der prokulianischen Rechtsschule eine
Ttigkeit im Gerichtshof Traians entfaltet. 1 5 3 Unter H a d r i a n wurde unser J u r i s t
in den von diesem Kaiser umgestalteten Gerichtshof 1 5 4 zwar noch fters
einberufen, 155 das besondere Vertrauen Hadrians in der Rechtspflege h a t
jedoch neben dem sabinian ischen Rechtsschule angehrigen P . Slvius Iulia-
nus 15e schon ein jngerer Vertreter der Prokulianer, der zur Generation H a d -
rians angehrige P . Iuventius Celsus ( s u f f . 117 ?)157 genossen. 158 Der Lebenslauf
des Neratius Priscus lt sich bis zu den Jahren 133/134 verfolgen. 159
161
P I R 2 J 126.
142
R . H a n s l i k , . O. (Anm. 32), 1036 f. Anders J . M o r r i s , . O. (Anm. 73), 328.
143
Dig. 37.12,5 + 1.2,2,53. W . K u n k e l , . O., 144 f. b e r die Rechtsschulen
s. j e t z t auch die Artikel von D. L i e b s , . o . (Anm. 37), 197 ff.
144
J . C r o o k , Consilium prineipis, Cambridge 1955, 56 ff. W . K u n k e l , . O. 291 f f .
144
V. H a d r . 18,1: Cum iudicaret, in consilio habuit amicos suos aut comits
solum sed iuris consultes et praecipiue Iuventium Celsum, Salvium Iulianum, Neratium
Priscum aliosque, . . .
148
Sein cursus bonorum : D 8973. S. dazu T h . Mommsen, G. Sehr. 2, 1 ff. Das Leben
u n d W e r k des Salvius Iulianus w u r d e zuletzt behandelt v o n E . B u n d , A N R W I I . 15
(1976), 408 ff. ber d a s strittige G e b u r t s d a t u m s. auch G. A l e l d y , K o n s u l a t 44, Anm. 30.
157
E r ist a m 19. J u l i 114 noch als prtorischer S t a t t h a l t e r in der Provinz T h r a c i a
bezeugt (M. M. R o x a n , R M D 14). Seine hiesige Legation d a u e r t e wohl bis 115/116.
Vgl. W . E c k , Senatoren, 176 f. Anders P I R 2 J 882.
158
Dies zeigt ziemlich klar die Nacheinanderfolge der N a m e n der Juristen in d e m
oben, Anm. 155. a n g e f h r t e n Text. Die zwei ersten Stellen n e h m e n hier die V e r t r e t e r
der prokulianischen, bzw. sabinianischen Rechtsschulrichtungen ein, P . Iuventius Celsus,
respektive P . Salvius Iulianus. N e r a t i u s Priscus wird erst n a c h ihnen als der d r i t t e

436 t. nagy: die zwei l. neratii prisci
E i n e wichtige Station dieses Lebenslaufs war der schwer bestreitbare

K o n s u l a t des J u r i s t e n im J . 93., der a u c h f r die Beurteilung der Konsulate-
d a t e n der anderen zwei L. Neratii Prisci vom Belang ist. D i e A n n a h m e des im
v o r i g e n vorgeschlagenen Zeitansatzes bezghch des K o n s u l a t s des Juristen
k a n n nmlich erst jeden Zweifel d a r b e r aufheben, da von seinen plebejischen
N a m e n s v e t t e r n d e r ltere L. N e r a t i u s Priscus das h c h s t e A m t im J . 87,
w h r e n d dessen gleichnamiger Sohn jenen im J . 97 a n g e t r e t e n haben. Die
B e h a n d l u n g der n a c h ihren K o n s u l a t e n bekleideten m t e r dieser zwei Neratier
in P a n n o n i n , sei mir aber gestattet, einem nachfolgenden A u f s a t z aufbewahren.
Budapest.
e r w h n t . Dieser hadrianzeitlichen B e w e r t u n g des Celsus u n d des Priscus folgt offen-

sichtlich auch P o m p o n i u s in Dig. 1.2.2, 53: Celso patri Celsus filius et Priscus Neratius.
D e r jngere Celsus w a r praetor im J . 107, insofern der Proze de repetundis gegen Varenus
R u f u s im Senat (Plin., E p . 6,5) in den e r s t e n Monaten des J . 107 begonnen h a t , wie es
R . H a n s l i k , s. v. Varenus, no 7, R E V I I I A (1955), 376 sehr wahrscheinlich gemacht
h a t . A. N . S h e r w i n - W h i t e , . O. (Anm. 36) 399 f. hlt dagegen a u c h das Ende 106 f r
wahrscheinlich. Jedenfalls, wenn P . I u v e n t i u s Celsus die P r t u r minima aetate, d. h. in
seinem 30. Lebensjahr angetreten h t t e , d a n n flle sein G e b u r t s d a t u m ins J . 76 (es ist
a u c h d a s Geburtsjahr Hadrians), oder ins J . 77. Sein zweiter K o n s u l a t i m J . 129 (A. D e g -
r a s s i , F C 37) zeigt f e r n e r gengend, d a er d a s Wohlwollen H a d r i a n s im groen Ma
genossen h a t .
159
S. oben, S. 463 u n d Anm. 74.

*
CS. TTTSSY
THE VARIANTS OF THE FRAME STORY OF

THE UKASAPTATI
I n an earlier p a p e r 1 1 arrived a t the conclusion t h a t t h e appearance of t h e

so-called cunning stories among t h e stories in the ukasaptati 2 can be r e g a r d e d
as t h e birth of the genre of short story within S a n s k r i t literature. 3
However, t h e principal f r a m e story of the S u k a s a p t a t i 4 i.e. not t h a t of
t h e single stories, b u t t h a t of t h e whole work is a fairy-tale story, where,
4
C s . TTTSSY: A n t . Tan. 15 (1968) 221 foil, a n d Acta A n t . Hung. 17 (1969)
433 foil.
2
The t e x t editions of the Sanskrit variants of the Sukasaptati are : R . SCHMIDT :
Der Textus ornatior der ukasaptati. Kritisch hgg. von . ABayA I. Cl. 21 (1901) 2.
A b t . Mnchen 1898. 317416; R . SCHMIDT: Die ukasaptati. Textus simplicior.
AKM 10 (1897) No. 1. Leipzig 1893. X + 2 1 3 ; and notes to this t e x t : R. SCHMIDT: ZDMG
48 (1894) 580628; R . SCHMIDT: Der Textus simplicior der Sukasaptati in der Recen-
sion der Handschrift A. ZDMG 59 (1900) 517547 and ZDMG 55 (1901) 1 44 (Textus
elegantior) ; u k a s a p t a t i . Dilli 1959. 152 (Textus simplicior) ; RAMKANTA TRIPTH :
Sukasaptati. E d . with comm. by .Varanasi 1966. 288 (Textus simplicior). The trans-
lations of the Sanskrit variants of t h e Sukasaptati are : R. SCHMIDT : Die ukasap-
tati. (Textus ornatior.) Aus dem Sanskrit bersetzt von . S t u t t g a r t 1899. I V + 1 4 9 ;
R . SCHMIDT : Die ukasaptati. (Textus simplicior.) Aus dem Sanskrit bersetzt von
. Kiel 1894. 111+102 ; R . SCHMIDT: ukasaptati. D a s indische Papageienbuch. I n :
Meisterwerke Orientalischer Literaturen. I I I . Bd. Mnchen 1913. I X I V + 1 120 (Tex-
t u s simplicior) (Einleitung, bersetzung) ; . .
. . . . . .
1960. 173 (Textussimplicior) ; Cs. TTTSSY: Sukaszaptati. Budapest 1962. 249
(Textus ornatior) (Translation, postscript, notes) ; W. MORGENROTH : Das Papageien-
buch. Berlin 1968, 324 (Textus ornatior u n d 16 Erzhlungen aus dem Textus simplicior)
(bersetzung, Nachwort, Anmerkungen).
3
These cunning stories centre around a witty a n d unexpected turn taking place
during a rescue from great straits or a solution of an intricate problem. Their essence
is this unexpected t u r n , a n d it is not t h e victoiy of t r u t h t h a t is important here (as,
for instance in t h e tale of the clever rabbit taken over from t h e P a n c a t a n t r a (Tantr-
khyyika a and I. 6, Pane. t. s. a n d t. o. I . 8., u k . t. . 40, t. s. 31, t. e. 33.), b u t t h e
disentanglement from t h e difficulty, which is always successful by any means. This
ever victorious character of the cunning story reflects an optimism founded on prosperity.
I n contrast to t h e transcendental solutions of the folktale realism prevails here.
4
As for m e I distinguish not only two Sanskrit variants of the Sukasaptati [i.e.
textus ornatior, (abbrev. : t. o.) and textus simplicior (abbrev. : t...)], but three ones, t h a t
is to say I regard t h e textus elegantior (abbrev. : t. .), published by Schmidt based on t h e
manuscript A as a separate variant [cf. Cs. TTTSSY: A n t . T a n . 14 (1967) 249 a n d
Acta A n t . H u n g . 16 (1968) 447]. The literature, however, evidently on tho basis of t h e
slightly misleading title of the edition by Schmidt of the t e x t of manuscript A usually
mentions two variants of text, see M. WINTERNITZ : Geschichte der indischen L i t t e r a t u r .
I I I . Loipzig 1920. 343 a n d id.: A History of Indian Literature. Transi, by SUBHADRA
JHA. Vol. I I I . P a r t I . DelhiVaranasiPatria 19 7 72. 4 1 6 ; A . B . K E I T H : A H i s t o r y

438 CS. TTTSSY
a m o n g others, a p a r r o t occurs.5 Miraculous elements a n d magic p l a y a m a j o r

role in fairy-tale. These fairy-tales t u r n into adventurous a n d interesting
s t o r i e s , h a v i n g t h e e l e m e n t s of t h e w o r l d o f i r r a t i o n a l i t y . N e v e r t h e l e s s t h e
a i m o f t h e p r e s e n t p a p e r is n o t t o d e a l w i t h t h e c h a r a c t e r i s t i c s of t h e f a i r y -
tales, n o r t o discuss all p a r r o t stories,6 b u t m e r e l y to e x a m i n e t h e p a r r o t
s t o r i e s occurring in t h e S u k a s a p t a t i v a r i a n t s a n d its later versions, a n d serv-
ing as their frame story.
T h e f r a m e s t o r y of t h e textus ornatior v a r i a n t of t h e S u k a s a p t a t i 7 is t h e
following :
A s o n , M a d a n a s e n a is b o r n t o H a r a d a t t a , t h e r i c h m e r c h a n t , a s a r e w a r d
f o r h i s v e n e r a t i o n of S i v a . H a r a d a t t a o b t a i n s t h e P r a b h v a t ! of r a d i a n t b e a u t y
a s w i f e f o r h i s s o n , a n d a s a r e s u l t of h e r b e a u t y M a d a n a s e n a n e g l e c t s e v e n h i s
d u t i e s a s a m e r c h a n t . H a r a d a t t a , his f a t h e r g r o w s d e s p e r a t e seeing t h a t his son
n e g l e c t s h i s d u t y u n t i l T r i v i k r a m a , o n e of h i s f r i e n d s g i v e s h i m t h e a d v i c e t o t r y
a n d b r i n g a b o u t a c h a n g e i n h i s s o n ' s a t t i t u d e w i t h t h e h e l p of t h e p a r r o t G u n a -
s g a r a a n d of h i s w i f e , t h e s'ri&-bird M a l a y a v a t , w h o w e r e c h a n g e d i n t o b i r d s
b e c a u s e of t h e c u r s e of I n d r a .
T r i v i k r a m a g o e s o n t o e x p o u n d t h e s t o r y of t h i s t r a n s f o r m a t i o n t o H a r a -
d a t t a . T h e p a r r o t u s e d t o b e V i s v v a s u , t h e G a n d h a r v a k i n g a n d t h e n a m e of
of Sanskrit Literature. London 1961. 290 ; . : Letterature antiche dell'India.

Milano 1969. 202 ; . .
. . . . . . 1960. 7; W . MORGEN-
ROTH : op. cit. 282 ; b u t see also R . SCHMIDT : ZDMG 54 (1900) 515. F r o m t h e viewpoint
of t h e present work, however, it is of no consequence whether textus simplicior and textus
elegantior are regarded as two separate variants or not, because t h e introductory a n d the
concluding parts of t h e f r a m e story are missing in t h e textus elegantior, comprising only
stories 2 53.
5
T h e single stories (so t h e story of Blasarasvat or also t h a t of another parrot,
D h r t a c a k o r a , who occurs only in the t. o. variant) are each told by t h e parrot of t h e
s e v e n t y tales to his mistress, P r a b h v a t i , in order to prevent her f r o m immorality by
n a r r a t i n g a most often immoral story. In relation to the frame of these single stories
t h i s p a r r o t chooses t h e m o s t ingenious device in the textus ornatior variant, see Cs.
TTTSSY : A n t . Tan. 10 (1963) 155 foil, and A c t a Orient. Hung. 18 (1965) 227 foil.
6
See about these other Indian, Persian, Turkish, etc. Sukasaptati-versions named
u k a b h a t t a r , u k a b a h o t t a r ! a n d Tota-itihsa, T o t a - k a h n and Tti-nme: R. SCHMIDT:
Vier Erzhlungen aus der ukasaptati. Kiel 1890. 6 foil., id.: ZDMG 45 (1891) 629 foil,
a n d ZDMG 46 (1892) 664 foil., id.: ber den W e r t h des Sanskritstudiums. S t u t t g a r t
1898. 4, W . PERTSCH: ZDMG 21 (1867) 506 foil, a n d ZDMG 22 (1868) 568, J . HERTEE:
b e r die S u v b a h u t t a r k a t h . I n : Festschrift Windisch. Leipzig 1914.138 foil., J . HERTEL:
D a s P a n c a t a n t r a . Seine Geschichte und seine Verbreitung. LeipzigBerlin 1914. passim,
M. WINTERNITZ: Geschichte der indischen L i t t e r a t u r . I I I . 346 foil., id.: A History of
I n d i a n Literature. Vol. I I I . P a r t I. 419 foil., A. B. KEITH, op. cit. 262 foil, a n d 359 a n d
J . RYPKA: Iranische Literaturgeschichte. Leipzig 1959. 235. See a b o u t the Persian
collection of tales Sindbd-nme (see J . RYPKA: toc. cit.) going back to a n Indian work,
a n d its Arabian, Syrian, Greek, Hebrew, Old Spanish, Middle Latin versions: M. WINTER-
NITZ: Geschichte der indischen Litteratur. I I I . 348, id.: A History of I n d i a n Literature.
Vol. I I I . P a r t I. 421 foil., A. B. KEITH: op. c. 360 foil, and esp. B. E . PERRY: The
Origin of the Book of Sindbad. Berlin 1960.
7
R . Schmidt accounts t h e slightly in mdias res start of t h e f r a m e story for t h e
loss of t h e beginning of t h e t e x t , see R . SCHMIDT : Der Textus ornatior der ukasaptati.
E i n B e i t r a g zur Mrehenkunde. Stuttgart 1896. 1.

T H E VARIANTS OF T H E F R A M E STORY OF T H E SUKASAPTATI 439
his wife was Mlvat. Once they happened to fail to entertain Indra, and when
the god, angry as he was, appeared t h e y turned into birds in alarm, in order to
elude Indra's wrath. Indra, upon this, cursed them t h a t they should retain their
bird shape, and he became conciliated only t h a t much t h a t he promised them
t h a t if they did a service to Madanasena, son of H a r a d a t t a , and if then he re-
leased them in his joy, then they would obtain admission into the world of t h e
immortals again.
After being carried to the house of Madanasena, by means of the story
about Devaarman 8 and Pundarka, t h e parrot actually brings Madanasena t o
reason, and he even sets out in order to increase his riohes. PrabhvatI, who
protests at first despairingly against t h e departure of her husband, is trusted b y
Madanasena to the two birds. Vinayakandarpa, the son of the king, still takes
a fancy to the merchant's young wife spending her days in sorrows and his
female messengers persuade the lady not to spend her life drearily, while her
husband obviously enjoys the pleasures of love with other women. Under t h e
influence of all these, P r a b h v a t I bejewels herself and prepares to set out.
At this time the hen-bird openly opposes her. For this her mistress nearly wrings
her neck, b u t she succeeds in flying a w a y . T h e cock-parrot, however, is shrewder,
wiser and more cautious. Upon the question of his mistress he shows a different
attitude. He seemingly approves of t h e departure of his mistress. He only cau-
tions her to be as wise and clever as t h e hero, or the heroine of a story. Then,
after relating the story he makes his mistress think so long on the solution of
the complicated and difficult situation t h a t at last d a y begins to dawn. Thus,
while P r a b h v a t I usually listened to the story of an unfaithful and false woman,
she herself still does not strike into t h e p a t h of conjugal infidelity, since by t h e
time when the parrot told her the solution at daybreak which she was unable
to find out during t h e whole night, t h e night was over and a t the same time t h e
opportunity to see her lover was also missed.
This introductory frame story is very long. B u t the concluding p a r t is all
the shorter, and it does not correspond to the introductory f r a m e story. This
will be discussed and expounded a t t h e end of this paper.
The introductory f r a m e story 9 of the textus simplicior variant of the u k a -
saptati tallies in its contents with t h a t of the textus ornatior, b u t the story of t h e
curse is not mentioned here, and the P u n d a r k a story is also omitted from t h e
side of Devasarman's story. Nor is t h e a t t i t u d e of the parrot so witty. In fact,
8
See Mahbhrata I I I . 2 0 7 - 2 1 6 .
9
After the two-line poem following Ganea's greeting, a n d after the short intro-
ductory frame story, which states t h a t this is only an abridged version of the story of
the p a r r o t , the first tale begins. Here in t h e introduction t h e n a m e of the merchant is
H a r i d a t t a , t h a t of his wife is Srngrasundar, t h a t of the son is Madanavinoda, later
Madana, that of the son's wife is P r a b h v a t I , who is the daughter of a reSthin n a m e d
Somadatta.

440 CS. TTTSSY
here t h e parrot tells t h e cunning solution only on the condition t h a t P r a b h v a t

promises to stay a t h o m e t h a t night and to go to sleep. 10
T h e concluding p a r t of the textus simplicior completely differs from t h a t
of t h e textus ornatior. I n t h e textus simplicior the concluding p a r t coming a f t e r
t h e 69th story, as a m a t t e r of fact, t h e 70th story itself, is namely exactly cor-
responding to the introductory part. T h e husband, Madana, comes home. His
wife is as kind to him as she had been before t h e time of his departure. However,
t h e parrot starts to allude to the behaviour of t h e wife. U p o n this the wife
begins to praise t h e wisdom of the parrot, the parrot hearing the praise is
a s h a m e d of his hints, a n d P r a b h v a t gradually still starts to confess t h a t dur-
ing t h e absence of her husband she wanted to go to a strange man, and she was
withheld from this b y t h e parrot by means of his cunning talk. T h e man is abash-
ed a t t h e confession a n d he turns to t h e parrot for advice. Then the bird tells
t h e story of M a d a n a m a n j a r i , the G a n d h a r v a Madana's 1 1 daughter, who simi-
larly did not in f a c t c o m m i t conjugal infidelity. 1 2 After listening to this story,
M a d a n a , P r a b h v a t ' s husband forgives, and a t the festival arranged by Hari-
d a t t a , the parrot, t h e hen-bird and Trivikrama being delivered from the curse 13
go t o heaven.
I n the textus elegantior of the S u k a s a p t a t i since in t h e manuscript b o t h
t h e beginning and t h e end are missing this f r a m e story, of course, does not
a p p e a r a t all.
T h e Marthi version of the Sukasaptati, 1 4 as in general, tallies fairly well
w i t h t h e textus simplicior also with regard to the f r a m e story. 1 5
W i t h N a h s a b i a n d Qdir there is another parrot story. Diy'u'd-din Nah-
sabi, who was a contemporary of Hfiz and Sa'di and very likely was living
in t h e court of M u h a m m a d II, finished his work 10 entitled Tt-nme prob-
10
P r a b h v a t , porsuadod by her w a n t o n friends, w a n t s to go to G u n a c a n d r a
according to the textus simplicior. H e is n o t a son of a king in this version. The wife of
t h e p a r r o t acts clumsily a n d flies away in t h i s version, too, in order t o save her life.
11
Of course, t h i s M a d a n a is n o t identical with P r a b h v a t ' s h u s b a n d .
12
A v i d y d h a r a e n j o y e d M a d a n a m a n j a r i , as a consequence of t h e curso of t h e
b r h m a r i a N r a d a . K a n a k a p r a b h a , her h u s b a n d , who w a n t e d to kill his wife therefore,
learned from Candik, t h e goddess in her t e m p l e t h a t M a d a n a m a n j a r i was actually
i n n o c e n t , because t h e v i d y d h a r a had assumed t h e figure of K a n a k a p r a b h a , and there-
fore t h e goddess advises t h e h u s b a n d to forgive.
13
This curse is m e n t i o n e d only here in this version (and in t h e introductory p a r t
it is n o t ) , even here it is n o t given in detail.
14
K . S c h m i d t : S u k a b h a t t a r i . Die Marthi-bersotzung der S u k a s a p t a t i . Marthi
u n d D e u t s c h . 10 (1897) No. 4. Leipzig 1897.
15
Accordingly, w e need not go into details with regard to t h e Marthi version
because of the correspondence between this t e x t a n d the textus simplicior. Tales 71 a n d
72 in t h e Marthi version contain tale 70 of t h e textus simplicior. The tale 71 tells t h e
s t o r y a b o u t the G a n d h a r v a a n d his wife, a n d t h e tale 72 relates t h e story of t h e former
c u r s i n g of the two birds, t h e i r present deliverance f r o m t h e curse a n d their ascension t o
t h e h e a v e n on a celestial carriage, and t h e happiness of Madanasena.
16
See J . R y p k a : Iranische Literaturgeschichte. Leipzig 1959. 235 ; W . P e r t s c h :
Z D M G 21 (1867) 506 ; J . H e r t e l : Das P a n c a t a n t r a . Seino Geschichte und seine Ver-
b r e i t u n g . LeipzigBerlin 1914. 234.

T H E V A R I A N T S O F T H E F R A M E STORY O F T H E S U K A S A P T A T I 441
a b l y in 1330. H e mentions in his work t h a t t h e p a r r o t book had been t r a n s -

lated into Persian a l r e a d y before him by an a u t h o r w h o m he does not n a m e .
Nahsabi's work is very valuable, because besides t h e ukasaptati version he
used, he also m a d e use of other I n d i a n works. M u h a m m a d Qdiri's work,
w r i t t e n similarly in Persian, is an elaboration of t h i s successful Persian work
of NahSabI either, with Pertsch a n d Schmidt, 1 7 f r o m t h e 17th, (or, according
t o Hertel, 1 8 only f r o m the late 18th early 19th c e n t u r y ) .
Qdiri's f i r s t tale 1 9 contains even two parrot stories. The first p a r t of t h e
tale, corresponding t o t h e p a r r o t stories above, relates t h a t once the m e r c h a n t
Meimun set o u t for a long business t o u r . The parrot, w h o together with a n o t h e r
b i r d , is p u t in charge of the house, entertains with his tale the woman, w h o is
s a d because her h u s b a n d left. One d a y , however, a prince f r o m a foreign c o u n t r y
sets eyes on his mistress, and b y t h e aid of female messengers he asks her t o f i x
a n a p p o i n t m e n t . The woman p u t s on her finest clothes, a n d when the other b i r d
disapproves of w h a t she is going to do, she dashes her t o t h e ground, so t h a t she
dies. The p a r r o t is shrewder and he promises to conceal w h a t she does f r o m h e r
husband, just as t h e p a r r o t of bey F e r u h did not tell him anything a b o u t his
mistress. Then, a t his mistress' request, he tells t h e s t o r y of the other p a r r o t ,
which is suitable b o t h to win t h e confidence of his mistress, and, the night being
spent with listening to the story, t o hinder his mistress f r o m going to her lover
t h a t night. This story (the second p a r t of Qdiri's f i r s t tale), 2 0 which being
a n o t h e r version is interesting t o us, goes like this :
A merchant, n a m e d bey F e r u h , on his d e p a r t u r e p u t his wife and all his
goods and chattels in his p a r r o t ' s charge. He was a w a y for a long time, a n d on
t h e other hand t h e w o m a n soon got to know a y o u n g Moghul, and fell in love
with him. She received him every night in her h o m e a n d he stayed there till
morning. All this was going on before the very eyes of t h e parrot, b u t he pre-
t e n d e d to see or hear nothing a t all. After arriving h o m e , t h e merchant t h o r -
oughly interrogated his parrot, who told everything else, b u t did not tell a
single word a b o u t t h e infidelity of the wife. The m e r c h a n t , however, h e a r d
a b o u t the m a t t e r f r o m others, therefore he r e p r i m a n d e d t h e wife and punished
her. The wife, of course, suspected t h e parrot of giving her infidelity a w a y .
Therefore once a t night she tore his feathers, threw t h e p a r r o t out the door, a n d
" W . P e r t s c h : ZDMG 21 (1867) 506; R . S c h m i d t : V o r w o r t zu Tutinamoh. I n :

Meisterwerke Orientalischer L i t e r a t u r e n . I I I . Bd. Mnchen 1913. X V and J . R y p k a : loc. cit.
18
See H e r t e l : D a s P a n c a t a n t r a . 244.
19
. J . L . I k e n : Qudir's T u t i n a m e h . bersetzt v o n . I n : Meisterwerke
Orientahscher L i t e r a t u r e n . I I I . Bd. 121 236. Mnchen 1913. ( I t is the reprint of his
translation edited in 1822.) 123 foil.
20
In t h e second part of t h e first t a l e Meimun's p a r r o t tells the wholo story of t h e
p a r r o t of boy F e r u h . T h e story in N a h Sab and in Qdir is t h e same, so I t h o u g h t it
sufficient to expound t h a t of Qdir. Nahabi, at the end differs only so far as in his
version t h e p a r r o t rises f r o m t h e d e a d because of t h e w o m a n ' s prayers, a n d f r o m
t h i s fact t h e h u s b a n d is driven to t h e conclusion t h a t his wife is iimocent.

442 CS. TTTSSY
t h i n k i n g t h a t she did really destroy t h e bird, she alarmed all t h e servants a n d

m a i d s , saying t h a t t h e p a r r o t was t a k e n a w a y b y the cat. T h e bird, however,
took shelter in the n e a r - b y cemetery in a grave p i t . The m e r c h a n t , when t h e
n e x t d a y he did not f i n d his favourite b i r d , expelled his wife, too. T h e wife went
t o a neighbouring cemetery, because she knew t h a t nobody would give shelter
t o h e r in his house. A t n i g h t the p a r r o t c a m e o u t of the grave a n d told the wom-
an t o c r o p her hair a n d t o spend forty d a y s in t h e temple w i t h o u t a n y food, and
t h e n all her sins will b e forgiven and he himself will reconcile her w i t h her hus-
b a n d . T h e woman t h o u g h t t h a t a pious soul was speaking f r o m t h e grave and
she p e r f o r m e d the instructions. H e r e a f t e r t h e parrot appeared t o t h e woman,
r e p r o a c h e d her, b u t t o l d her t h a t he would reconcile her w i t h t h e r husband,
w h a t h e really did. R e t u r n i n g home to t h e merchant, he told, as h e was answer-
ing his questions t h a t h e was his old p a r r o t , who was taken b y t h e c a t from the
cage a n d closed in his belly. To t h e question of his master how h e could rise
f r o m t h e dead, he answered t h a t it was t h e Almighty who g a v e h i m life again,
seeing t h e woman's f o u r t y d a y long, d e e p sorrow in the cemetery a n d he gave
him t h e c o m m a n d m e n t t o reconcile his m a s t e r with his wife. T h e merchant,
a t t a c h i n g credence t o t h i s story, went for his wife himself, and t h i n k i n g t h a t she
w a s t o t a l l y innocent folded her in his a r m s , brought her back h o m e and t h e y
were living happily w i t h each other h e r e a f t e r .
A Turkish t r a n s l a t i o n or elaboration prepared about one h u n d r e d years
l a t e r is also based on N a h s a b i ' s work. According to Pertsch its a u t h o r is efendi
' A b d u l l a h . 2 1 Nor does this 22 essentially differ from the story of NahsabI and
Qdir. 2 3
21
W . P e r t s c h : Z D M G 21 (1867) 507 foil.
22
This is the f i r s t s t o r y of the f o u r t h n i g h t in t h e Turkish version. See G. R o s e n :
T u t i - N a m e h . Das P a p a g a i e n b u c h . I I I . N a c h der trkischen B e a r b e i t u n g . . . ber-
s e t z t v o n . Leipzig 1858. 30 foil, a n d in : Bibliothek der R o m a n e . X V I I . Bd. Leipzig
sine anno. 24 foil.
23
I n the end t h i s version is, of course, s t a n d i n g nearer to N a h a b i t h a n to Qdiii,
t h e o n l y difference f r o m t h e former being t h a t here we can see a Moslim a d a p t a t i o n of
t h e N a h a b version. I n t h e Turkish version n a m e l y when the woman is going to carry out
t h e i n s t r u c t i o n s of t h e p a r r o t , t h a t she should crop her hair t h e p a r r o t comes forward
a n d r e p r i m a n d s t h e w o m a n for her failuie t o distinguish her friends f r o m t h e enemies,
a n d s h e did wrong to h i m , who did not d e n o u n c e her to her husband, on t h e contrary he
e v e n t r i e d to reassure h i m . H e pointed o u t t h a t t h e woman with t h e foolish act t h a t she
c a s t o u t him brought t r o u b l e to herself as well, a n d t h a t was the cause of her expulsion.
I t is o n l y her stupidity goes the p a r r o t on rebuking her t h a t can be an excuse,
a s n o w also did the w o m a n believe, when he was speaking from b e h i n d t h e idol of Indian
religion, t h a t an idol, m a d e of stone a n d wood, can speak. She should be converted to t h e
true religion, he urges her, a n d then h e would arrange it for her t h a t she would enjoy
h e r h u s b a n d ' s favours again. The w o m a n is really converted to I s l a m a n d the parrot
goes h o m e . The m e r c h a n t , seeing him becomes very glad and asks t h e p a r r o t about his
vicissitudes. The p a r r o t tells him t h a t h e really was dead, but God was gracious to
h i m a n d revived h i m . T h e merchant is astonished at t h a t , b u t t h e p a r r o t explains t h a t
t h e i r I n d i a n deities a r e b u t stone and wood, t h e r e is no soul in t h e m . T h e n t h e merchant,
t o o , is converted to I s l a m . H e goes on to ask t h e parrot why God raised him from the
d e a d . T h e parrot says t h a t t h e wife of t h e m e r c h a n t being converted to t h e Moslim faith
p r a y e d so devoutly to God to help her w i t h t h e proving of her innocence t h a t he was

T H E VARIANTS OF T H E F R A M E STORY OF T H E SUKASAPTATI 443
In the Persian (Nahsab, Qdir) and Turkish versions, which go back to

t h e ukasaptati, a second parrot story can be found. And there is another one
in t h e textus ornatior variant of the ukasaptati. We do not know the first p a r t
of this story, owing to the lacuna f r o m the 65th to the beginning of the 68th tale
in t h e textus ornatior, h u t in tales 68 70 we can read the other parrot story,
which can be regarded as the t h i r d v a r i a n t of the parrot stories.
In tale 68, where the lacuna ends, we can read t h e story known from the
f r a m e tale of Book I I I of the P a n c a t a n t r a , the moral of which is t h a t one cannot
h a v e confidence in the benevolence of t h e old enemy. T h e tale is told by t h e
p a r r o t Dhrtacakora to his master S o m a d a t t a to illustrate this old t r u t h .
In tale 69 our parrot narrates to P r a b h v a t the cunning escape of the
p a r r o t named Dhrtacakora :
Dhrtacakora's story i.e. t h e relating of tale 68 did not strike root in
S o m a d a t t a ' s heart. H e took t h e parrot to the harlot Kmasen, who told t h e
event to her procuress Devasen. U p o n this the latter relaxed and the following
morning she went to the king, a n d before this she gave instructions in the pres-
ence of Kmasen t h a t the parrot should be killed and cooked. Hereafter t h e
maid began to pluck the bird. Then t h e cunning parrot recommended a cosmetic
to t h e maid, and while she went to fetch some other things supposedly necessary
to prepare the cosmetic, the parrot himself escaped through the drain, mean-
while shouting : the cat has t a k e n it, the cat has t a k e n it.24 Hearing this t h e
servant-maid returned, and since she did not find t h e parrot, in alarm she pre-
pared a partridge for the procuress.
I n tale 70 Prabhvat listens to the cunning revenge of D h r t a c a k o r a
f r o m our parrot :
I n fact, this D h r t a c a k o r a went to the temple of iva, before which
K m a s e n danced on every lunar d a y , and there he hid in the cavity of a bilva
tree. Once, when Kmasen's dance came to an end, he began to speak from
there as if he were the god iva, a n d promised to the hetaera t h a t on the next
lunar day a heavenly carriage would come to take her to Mount Kailsa. She
should distribute their property he instructed K m a s e n and both she and
t h e procuress should crop their hair and so they should come there. K m a s e n
gladly performed the order, which she believed to have come from the god, and
in t h e glaring sun they, both of them with bald head, were waiting for the heav-
enly carriage, b u t in vain. And then the c r a f t y and revenging parrot ridicules
t h e hetaera in a sloka :25
given a new life by God, and he was commissioned to attest the innocence and faithfulness
of t h e woman. The husband goes to the temple for his wife himself, and his wife admires
t h e shrewdness and wisdom of the p a r r o t .
24
anaifd anai$n mrjra iti.
25
afhe pratiafham kuryd dare ca tathdaram \
tvay me luUcith pak$h maya te munditam irah \ \

444 CS. TTTSSY
One should r e p a y c r a f t with c r a f t and honour with honour as well !

Y o u plucked o u t m y pinions, I clipped your head.
T h i s third version of the parrot story can be found in t h e Panckhyna-

vrttika. 2 6 This work is a version of the P a n c a t a n t r a written in Old Gujart
prose, in which there a r e a number of stories which do not occur in the Panca-
t a n t r a . I t s author is u n k n o w n , b u t he definitely was a J a i n a monk not versed in
Sanskrit. The only m a n u s c r i p t of his work is t h a t of a certain J i n a v i j a y a which
was w r i t t e n in 1673/74 A.D. Thus t h e P a n c k h y n a v r t t i k a was, naturally
p r e p a r e d a t a time preceding this date. 27 This above-mentioned story about t h e
p a r r o t a n d the hetaera can be summarized as follows r28
A merchant setting out for long commercial tour p u t s his shop in the
charge of his son, M a d a n a and charges his parrot to guard t h e house. Madana,
however, is seduced by a hetaera, who also got much money o u t of him. The par-
rot, therefore, gets a n g r y with her and warns the young merchant against t h e
h e t a e r a . The courtesan learns this from the servant-maid and decides to revenge
her on t h e bird. She asks Madana to bring once his bird along with him. When
he fulfils this wish, t h e hetaera plucks t h e pinions of the bird out, puts it on a
dish a n d goes to fetch a knife. The parrot, however, crawls into the drain, so the
h e t a e r a does not find him. She thinks t h a t the bird was t a k e n away by a cat a n d
she is glad of this. The p a r r o t , having his pinions grown again, leaves his hiding-
place a n d flies into t h e court of a sanctuary. When the hetaera repeatedly prays
before t h e statue of t h e goddess standing there t h a t she should l i f t her with her
b o d y t o heaven, the p a r r o t , in the name of the goddess, enters into conversation
with her and sets her as a condition of her being lifted to heaven to liberate her-
self f r o m the deception of the samsara, to donate her property to the brhma-
nas, t o get blessed by t h e m , to p u t on a white garment, to have her head clipped,
and a f t e r having performed all this, she should come back to t h e court of t h e
s a n c t u a r y and should w a i t until she, the goddess, would t a k e her on a celestial
carriage to heaven. T h e hetaera gladly performs all this, a n d then amidst the
beating of drums she goes to the sanctuary. There she h u m b l y prays that the
goddess should send her a carriage, which would take her to heaven. At this
m o m e n t the parrot ridicules her relating mockingly in a strophe t h a t he had her
hair cropped in r e t u r n t o plucking his feathers out. 29 Then he tells the people
gathered in the court of t h e temple, how the hetaera treated him. Upon this t h e
i n d i g n a n t crowd drags her along in the dust.
26
P a n c k h y n a v r t t i k a N o . 43.
27
H e r t e l : D a s P a n c a t a n t r a . 122.
28
See H e r t e l : D a s P a n c a t a n t r a . 151 foil.
29
This strophe of t h o P a n c k h y n a v r t t i k a is near t h e sloka of t h e textus ornatior
of t h e u k a s a p t a t i , cf. H e r t e l : Das P a n c a t a n t r a . 238 n o t e 1. S e e t h e n o t correct San-
skrit t e x t of this sloka: H e r t e l : P a n c a t a n t r a . 151 n o t e 9.

T H E VARIANTS O F T H E F R A M E STORY O F T H E S U K A S A P T A T I 445
In the collection of tales entitled K a t h r a t n a k r a , too, written by Hema-

vijaya, a jaina monk in 1600/1, 30 a story 3 1 can be found, which is similar to t h e
preceding ones and obviously goes back to some a d a p t a t i o n of the ukasaptati. 3 2
Another version of this story again can be read in t h e manuscript of a
version of t h e textus simplicior of t h e P a n c a t a n t r a , in t h e so called B a n a r a s
manuscript ; 33 where two stories otherwise not occuring in the P a n c a t a n t r a can
be found. According to Hertel, the manuscript itself is a very early one, it is
several hundred years old. 34 Of the two interpolated stories it is the second
which is interesting for us presently and of which Hertel published t h e Sanskrit
text, too. 35 This version differs from the previous ones inasmuch as the parrot of
a merchant prevents a lustful hetaera by prohibitive words from enjoying the
pleasure of love with other men, too. Therefore the hetaera one day, when the
merchant went to visit one of his relatives, plucked t h e pinions of t h e bird out,
and since she thought t h a t t h e bird died, threw it out to t h e street. The parrot,
however, with much effort reached t h e drain t h a t belonged to the temple of
Jlesvara. 3 8 After a time the hetaera also goes to t h a t temple. To her prayer to
the god the parrot replies. H e promises her the heavenly blessedness she asked
for if she dispenses her goods to the beggars, puts on a white garment, has her
hair cropped and on the seventh d a y comes back. Then a heavenly carriage will
come for her he promises to the courtesan to t a k e her to heaven. T h e he-
30
H e r t e l : Das P a n c a t a n t r a . 249.
31
See H e r t e l : Das P a n c a t a n t r a . 237.
32
1 summarize this story as follows :
A m e r c h a n t has a wise p a r r o t , which tries in vain t o dissuade h i m f r o m his love
for a b e a u t i f u l h e t a e r a . The m e r c h a n t tells the woman the p a r r o t ' s warnings, who there-
upon asks h i m for t h e bird, allegedly t o play with him. No sooner has she got hold of
him t h a n she s t a r t s to pluck his f e a t h e r s o u t a n d goes into t h e house to fetch a k n i f e t o
kill him. Meanwhile t h e p a r r o t hides a n d as t h e hetaera when she r e t u r n s does n o t f i n d
him, she is led to t h i n k t h a t some a n i m a l m u s t have eaten h i m u p . On t h e o t h e r h a n d ,
she tells t h e m e r c h a n t t h a t his p a r r o t is flown. The bird, a s his feathers grow again,
hides behind t h e s t a t u e of Candik, t h e h e t a e r a ' s family goddess, a n d in t h e n a m e of
t h e goddess he promises t h e h e t a e r a to fulfil one wish of hers. She doos not suspect t h a t
it is not t h e goddess who is speaking to her a n d asks for everlasting y o u t h , b e a u t y a n d
a lot of m o n e y for herself, a n d besides, t h e power to be able t o compel every m a n t o
obedienco. T h e n t h e parrot gives her the instructions to d i s t r i b u t e all her goods, t o clip
her head, to m o u n t an ass a n d t o ride a r o u n d t h e town to t h e beating of d r u m s , a n d
a t last t o come to t h e temple again w r a p p e d u p in rags. W h e n tho woman has dono all
this, t h e p a r r o t coming forward publicly shouts a variant of t h e alredy known loka a t her.
kfte pratilcrfam kuryn nyyamrgo 'yam Iritah \
bhavatyotpfitau pakau may mundyitam irah 11
One should do a recompens in ease of a deed :
this is proclaimed as the usage of t h e righteousness.
The pinions of m y two wings were plucked by you, I h a d your h e a d clipped.
A f t e r t h i s t h e p a r r o t relates w h a t h a s happened and t h e h e t a e r a a m o n g a shower
of abuses from t h e crowd has t o m o v e to a n o t h e r town.
32
H e r t e l : Das P a n c a t a n t r a . 92 foil.
34
H e r t e l : Das P a n c a t a n t r a . 92 n o t e 3.
35
H e r t e l : Das P a n c a t a n t r a . 95 foil.
38
H e r e Jlevara is a n a m e of Siva, according to Hertel. Gf H e r t e l : Das P a n -
c a t a n t r a . 97 n o t e 2.

446 cs. t t t s s y
t a e r a does everything, a n d on the seventh d a y she returns accompanied by an

immense crowd. Then t h e parrot shouts the well-known sloka a t her, 37 and it is
himself who flies u p in t h e sky.
Hertel, too, mentions t h e version 38 of t h e story to be found in an Old
M a r t h i metric elaboration of t h e P a n c a t a n t r a , the work of N i r m a l a Pthaka 3 9
entitled P a n c o p k h y n a (V. 11). 40
Vidagdhacdmani, t h e parrot of t h e king draws upon himself the hatred
of t h e hetaera, K m a s e n , as he in a certain case advises t h e king to decide
against her, 41 and a t t h e same time he tells a story (V. 12) which together with
t h e other events about t h e a r b i t r a m e n t m a y be omitted. T h e hetaera meditates
revenge. One day she amuses the king so much with her song a n d dance t h a t
t h e king tells her to ask for anything she wants to have. The courtesan then asks
t h e king for the bird, t h e n plucks his pinions o u t and orders a servant-maid to
b u t c h e r him and make a dish of him for her. While the girl goes to fetch a knife,
t h e bird creeps into a drain. The girl, who is unable to f i n d him, buys a pigeon
a n d makes a delicious dish of the pigeon instead of him. W h e n his pinions have
grown again, the parrot flies to the sanctuary of Bhavn, before which K m a -
sen usually dances. H e r e hiding behind the s t a t u e of the goddes he promises
t h e hetaera to t a k e her to heaven if she dispenses her goods, crops her hair a n d
p a i n t s her body black a n d white and dances naked before t h e s t a t u e of the god-
dess. T h e hetaera carries o u t t h e commands in the presence of a huge crowd,
b u t t h e n the parrot coming forwards shouts a little corrupt v a r i a n t of the usual
sloka a t her. 42 After this t h e parrot flies up, and the hetaera goes home ashamed.
And only now I come to discuss t h e concluding f r a m e story of the textus
ornatior of the S u k a s a p t a t i , which as had been mentioned already is very
short and does not correspond to the introductory f r a m e story. This concluding
f r a m e story means a f o u r t h t y p e of variant. According to this on the following
3
' krteh pratikrtam kuryd dhimsite pratihimsitam \
tvay lumcpitau paksau maya mumdpam irah \ \
One should do a recompense of a deed
/ a n d ) a recompense of t h e h a r m in case of a harm!
Y o u had t h e pinions of m y two wings plucked out, I h a d y o u r h e a d clipped.
38
According to H e r t e l t h e version of N i r m a l a P t h a k a can b e t i a c e d back t o
t h e Sanskrit B a n a r a s m a n u s c r i p t a n d it is n o t t h e other way r o u n d , because this story
w i t h N i r m a l a P t h a k a H e r t e l argues is only a f r a m e story of a n other tale. Cf.
H e r t e l : D a s P a n c a t a n t r a . 98. This a r g u m e n t of H e r t e l is not convincing for me, because
t h e story of t h e p a r r o t is j u s t a f r a m e story in t h e above cases a s well.
39
H e r t e l : D a s P a n c a t a n t r a . 23, 262.
40
See H e r t e l : D a s P a n c a t a n t r a . 285 foil.
41
When namely someone tells a friend t h a t he enjoyed K m a s e n , the hetaera
in his dream, K m a s e n learning this d e m a n d s t h e usual fee for this. As she does n o t
receive t h e sum she lodges a complaint against t h e dreamer before t h e king. The k i n g
a t last so to be able to judge t h e case correctly, consults V i d a g d h a c d m a n i , who arbitrates
t h e i m a g e of the fee, reflected in a mirror, for t h e e n j o y m e n t in a d r e a m .
42
athasya pratiafham kuryta daram pranimdaram \
tvay numcate may te mumdite irah | |

T H E VARIANTS OF T H E F R A M E STORY OF T H E S U K A S A P T A T I 447
day, on the one a f t e r the seventieth tale, Vinayakandarpa goes himself to P r a -

bhvat. P r a b h v a t receives him with an amorous glance. Meanwhile, t h e cat 4 3
observes t h e parrot in the cage, and begins to mew. Then t h e parrot says : Why
do you mew, cat ? Neither an enemy nor a thief, this is king N a n d a ' s son, who
does not touch the wife of someone else.44 Hearing these words, Vinayakandar-
pa goes away without a word. 45
In connection to this I must mention also N i r m a l a ravaka's 46 story 4 7
written in b r a j bhkh, which according to Hertel can be traced back to a uka-
s a p t a t i version. According to this king N a n d a at night goes into the house of
Mlat. The parrot becomes indignant a t t h e entrance of t h e king. The king gets
angry, he mentions the cat, and when he sits down on the bed of the yound lady,
the latter prostrating herself before him speaks to him : What has led you here,
my father ? W h y are you not sleeping a t home ? Upon this the king is overcome
with shame.
The main motifs of the four versions of the parrot stories mentioned a b o v e
are as follows :
Type 1 : The parrot guards t h e w i f e of a merchant on a trip, he h i n -
d e r s her from adultery by his t a l e s (introduction of textus ornatior, textus
simplicior and Marth) then he reconciles her with her husband (end of textus
simplicior and Marth).
Type 2 : The w i f e of the m e r c h a n t who is on trip actually c o m m i t s
infidelity w i t h o u t any a t t e m p t s of the parrot to hinder her f r o m it.
The husband learns the infidelity from others, b u t the woman suspects t h e par-
rot of denounce. When the woman in r e v e n g e for this wants to do a w a y
with the parrot and plucks his f e a t h e r s off, he manages to escape. T h e
woman lays t h e disappearance of t h e p a r r o t a t the a t ' s door, b u t her hus-
band expels her for this. The parrot in r e v e n g e cunningly hides in a cemetery
(or temple) and pretends to be a deceased soul (or even a god). He orders t h e
woman to c r o p her hair, b u t at last he brings about a reconciliation between
her and her husband (Nahsabi : 1/2, Qdir : 1/2, Turkish version : 4/1).
43
jighmsu- i.e. desirous of killing.
44
More exactly : the uterine b r o t h e r of wife of another man. This whole s l o k a
is as follows :
kim tvarri rodiii marjra na crir taskarah \
Nandabhpakumro 'yam paranrlsahodarah \\
45
This concluding p a r t of t h e f r a m e s t o r y in the textus ornatior is very s h o r t .
The 300 line long i n t r o d u c t o r y p a r t m a y sorve as a foil to these seven lines. And t h o u g h
t h e solution is definitive : V i n a y a k a n d a r p a leaves P r a b h v a t himself as a result of t h e
humiliating poem of t h e p a r r o t , and so t h e r e is n o point any m o r e to tell other stories
in order to keep his mistress in her conjugal fidelity, this solution does not mention w h a t
h a p p e n s to M a d a n a s e n a a n d his wife w h e n h e r e t u r n s home. T h e fact t h a t t h e b i r d
managed to keep P r a b h v a t in her fidelity, m a y mean t h e d a w n of happiness for t h e
cursed birds, a n d t h a t Madanasena and P r a b h v a t will live h a p p i l y hereafter, b u t n o t
a single word m e n t i o n s this, w h a t the reader m a y rightly miss.
46
H e r t e l : D a s P a n c a t a n t r a . 224 foil.
" See H e r t e l : Das P a n c a t a n t r a . 228 foil.

448 CS. TTTSSY
Type 3 : T h e p a r r o t wants to dissuade a m e r c h a n t from loving a

h e t a e r a. The h e t a e r a wants to t a k e r e v e n g e on t h e parrot. She plucks
his p i n i o n s out, b u t he can escape through the d r a i n . Mostly it arises t h a t
i t m u s t have been a t t h a t has taken him away. The parrot goes to a temple
a n d gives the woman instructions to c r o p her hair, promising her an ascen-
sion to h e a v e n . T h e parrot in the end discovers himself a n d in an every-
where occuring, v e r y similar s i o k a declares publicly before the crowd
t h a t he took r e v e n g e on the hetaera for her wicked maltreatment.
Type 4 : The prince or king g i v e s u p the seduction of the wife of a
m e r c h a n t (or a minister) under the influence of the w o r d s of the p a r r o t
(or t h e woman). T h e a t is mentioned also here.
Hertel believes t h a t the variant of t h e Banaras manuscript is the basis of
t h e above parrot stories, from which t h e others could be easily derive, and t h a t
in t h e original version of the ukasaptati t h e parrot still wanted to prevent a
m e r c h a n t from intercourses with a hetaera, with his seventy tales. The one
called b y me type 2 would be an intermediary form, and the f r a m e of the textus
simplicior, on the other hand is the third phase in the transformation of t h e
f r a m e story. 48
Besides, on t h e basis of two lines of t h e commentary written to the Yoga-
sstra 4 9 a f t e r 1160 A.D. by Hemacandra, born on December 1st, 1088 A.D.,
H e r t e l believes 50 t h a t H e m a c a n d r a knew such a ukasaptati version, in which
t h e p a r r o t is really caught by the cat. These two lines are as follows : Although
you are versed in moral science, you h a v e been caught by this (hetaera), like
t h e parrot narrating seventy stories b y t h e cat. 51 Actually Hertel holds it un-
likely 52 t h a t there would be a lapse of m e m o r y to the d e t r i m e n t of Hemacandra.
As far as t h e remark of H e m a c a n d r a is concerned, this really is so sur-
prising t h a t it is quite understandable t h a t even the possibility of a lapse of
m e m o r y occured to Hertel, which he, however, with due reverence marked,
improbable. We m a y regard this remark of Hemacandra as surprising because
n o t only none of t h e e x t a n t ukasaptati v a r i a n t s contain such a motif b u t be-
sides there is no such a d a p t a t i o n from a later time which would indicate such
a t u r n . The proximity of the parrot and t h e cat (in the f o u r t h type of f r a m e
story), and the f a c t , t h a t just at t h e end of t h e textus ornatior, t h e cat is mewing
48
49
Yogaastram. W i t h t h e C o m m e n t a r y called S v o p a j n a v i v a r a n a by H e m a c a n d r a .
E d . b y r i V i j a y a D h a r m a S r i . Vol. I . F a s c . H E . Calcutta 1910. 444. 41.
60
51
kathsaptatisarnams mrjryyeva uko 'nay |
nltijno 'pi grhlto 'si jagdetyabhayam ca sah \ |
And he ( C a n d a p r a d y o t a ) told A b h a y a : Although y o u are versed in m o r a l
science, y o u h a v e been caught
by this ( c o u r t e s a n ) , like t h e p a r r o t n a r r a t i n g seventy tales b y the cat.
62
H e r t e l : D a s P a f i c a t a n t r a . 236.

T H E V A R I A N T S OF T H E F R A M E STORY OF T H E S U K A S A P T A T I 449
'desirous to kill' 53 the p a r r o t (what is actually beyond hope, because he is pro-

tected, being shut u p in the cage) can, a t best, be the ground for a v a r i a n t
(mentioned by H e m a c a n d r a ) which spoils everything even more, where t h e
protagonist (the parrot) a f t e r struggling for something shrewdly right t h r o u g h
seventy tales, perishes so t h a t the o u t w a r d forces obviously carry out his inten-
tions for all that. 5 4 If, however, there was a v a r i a n t where according to H e m a -
c a n d r a the cat caught the parrot, this alone would be no evidence to p r o v e
t h a t this v a r i a n t must h a v e been the original, and it could be even i f i t exist-
ed just a spoiled, clumsy, later one, b u t in consequence of H e m a c a n d r a ' s
d a t e w r i t t e n before 1180 which happened to be suitable for H e m a c a n d r a ' s
sentence.
However, neither Hertel's opinion a b o u t the chronological evolution of
the f r a m e stories nor his assemption t h a t it is the v a r i a n t of t h e B a n a r a s m a n u -
script which m a y have given rise to all the other forms, can be accepted.
Hertel is a t variance even with himself when he asserts on the one h a n d
t h a t t h e v a r i a n t of t h e B a n a r a s m a n u s c r i p t is the oldest one, a n d then he pre-
sumes on the other h a n d t h a t the chronologically following version is t h a t which
we have introduced as t y p e 3 (where it is still a m e r c h a n t whom the p a r r o t
wants to bar from a hetaera). The development assumed b y Hertel is hardly
conceivable and it is b y no means logical, as t h e following s u m m a r y of his con-
ception can clearly manifest i t : 1. T h e p a r r o t wants t o ensure the fidelity
of t h e hetaera to the m e r c h a n t . 2. The p a r r o t wants to b a r t h e merchant f r o m
t h e hetaera by means of telling seventy tales, though it is a p i t y t h a t here it is
nowhere mentioned t h a t t h e p a r r o t would tell seventy stories, a t best, one or
two. 3. The parrot tries neither with tales nor even w i t h a single word of prohi-
bition to restrain the wife of t h e m e r c h a n t f r o m her lover. 4. T h e parrot telling
seventy tales cunningly a n d shrewdly restrains the wife of t h e merchant f r o m
the infidelity.
Moreover, it is absolutely clear t h a t t h e v a r i a n t of t h e Benares m a n u -
script only represents a different version of our third t y p e . This fact convin-
cingly shows t h a t this m a n u s c r i p t is t h e result of c o n t a m i n a t i o n . The l a t t e r
conclusion will also be corroborated by the observation t h a t some motifs of all
t h e other v a r i a n t s occur in it. The f a c t t h a t this is a very old m a n u s c r i p t proves,
a t best, t h a t a t the t i m e when it was written, all the p a r r o t stories included in it
(i.e. types labelled No. 1 3 respectively) were known t o its a u t h o r . I n this,
namely, the basic idea i.e. t h a t the p a r r o t attaches so m u c h importance to keep
t h e h e t a e r a f a i t h f u l to his master, is clumsy and illogical. Besides, actually
in this version the p a r r o t does not tell a tale, he simply forbids infidelity.
53
See above note 43.
54
I n t h e variants of the third typo if not the word 'caught' t h e expression 'has
taken a w a y ' appears about a (there) non-existent cat, but only as a clever lie of the parrot
or t h e servant-maid.

450 CS. TTTSSY
As to the chronological order of these variations of parrot-stories I a m of

t h e following opinion :55
The shrewd parrot, which speaks a n d takes revenge (type 3), the p a r r o t
escaping cunningly, forbidding nothing, a n d who in the end noble-mindedly for-
gives (type 2), the parrot, who with his seventy tales a n d his shrewdness keeps
t h e merchant's wife f a i t h f u l to her husband (type 1) and t h e parrot who drives
t h e seducer away himself with single humiliating sloka (type 4)5e or the p a r r o t
of tragical fate 57 who is mentioned by H e m a c a n d r a could exist side by side, 58
a n d for us now it is all t h e same which one was earlier. On t h e other h a n d (in
contradiction to the chronology of t h e sequence of frames according to Hertel)
I m a i n t a i n t h a t the f i r s t and basic t y p e of the parrot telling seventy tales was
t y p e 1. The sheer fact, t h a t the different and dissimilar v a r i a n t s of the uka-
s a p t a t i keep without exception t y p e 1 as the frame (except the end of the t.o.),
a n d the different later variants, too, use it for the f r a m e (even if they mention
other variants, too), could be a decisive argument for this assumption.
However, we can still support it b y the following considerations :
The variant (type 2) where the p a r r o t does not even t r y to prevent the
infidelity, nor does he tell tales cannot be the frame story of t h e story of the
p a r r o t telling seventy tales. Besides, I am of the opinion t h a t t y p e 2 is nothing
else t h a n the contamination of types 1 a n d 3. (The motif of the merchant who
l e f t a n d the wife who all the same became unfaithful 5 9 (as the parrot did not tell
tales and did not forbid infidelity) m a y be taken from t y p e 1, and the motif of
revenge from type 3.) A situation where the parrot bars t h e merchant from the
hetaera (type 3) may give opportunity to tell some wise tales b u t scarcely all
55
A positive r e m a r k concerning a n earlier work of m i n e a b o u t this subject, see
J . H a r m a t t a : Ant. T a n . 10 (1967) 241.
66
B u t , a n y w a y , was t h e use of t r o u b l i n g himself with s e v e n t y tales before this ?
57
H e is genially shrewd t h r o u g h s e v e n t y tales t h e r e a f t e r however a eat catches him!
58
Though we t h i n k t h e types 2, 4 a n d 5 to be later t h a n t h e t y p e s 1 a n d 3.
59
A similar tale a p p e a r s a m o n g the Sind bad-stories, too, cf. I . H a h n : A n t . T a n .
10 (1963) 249 foil. I n this S i n d b d v a r i a n t (see F. B a e t h g e n : Sindban. Syrisch u n d
D e u t s c h . Leipzig 1879. 15 foil.) a m a n on leaving home e n t r u s t e d a p a r r o t to tell him
w h a t his wife did while he was away. She t h e n committed a d u l t e r y , w h a t t h e p a r r o t
d i d tell his master when he r e t u r n e d . The w o m a n suspected t h e h a n d m a i d first of betray-
ing her, b u t later she f o u n d out t h a t t h e p a r r o t was her t r a i t o r . So she poured water
over t h e cage covered with a cloth, m a d e flashes and noises in t h e night, so as to m a k e
t h e bird beloive t h a t t h e r e was a t h u n d e r - s t o r m . When the following morning his m a s t e r
a s k e d h i m what h a p p e n e d t h a t night, t h e p a r r o t answered t h a t he could n o t see or hear
a n y t h i n g because of t h e lightnings anil the thunder-claps. As in f a c t there was n o s t o r m
his m a s t e r took him for a liar and i n f u r i a t e d as he was, because he t h o u g h t t h a t the
b i r d only slandered his wife, he slayod him.
Thereupon t h e following m a y be r o m a r k e d : Firstly, as H a h n , too, mentions,
t h i s story is n o t a f r a m e in its own place. On the other h a n d there are significant dif-
ferences as opposed to o u r t y p e 2. T h e p a r r o t here, namely, b e t r a y s the woman, the
w o m a n ' s revenge here succeeds with t h e s t o r m , here the p a r r o t is slain and it is t h e
h u s b a n d himself who slays h i m . Consequently in this version t h e p a r r o t is n o t shrewd,
h e does n o t t a k e revenge on t h e woman or tell tales, he purely a n d simply squeals a n d
p l a y s the informer. H o w e v e r it is one of the innumerable p a r r o t stories, standing nearest
t o t y p e 2 a f t e r all, b u t this could n o t be used as a frame.

t h e variants of t h e frame story of t h e sukasaptati 451
the seventy of them. N o t to mention besides t h a t t h e p o i n t of this v a r i a n t is t h e

revenge of t h e clever parrot, and t h e parrot is victorious by means of t h i s
r e v e n g e a n d n o t the shrewd t a l e s . The t y p e 4 is obviously a spoiled
version of t h e basic idea by an copier, where the only f a i r thing is t h a t a k i n g
voluntarily gives u p the indecent love 80 under t h e influence of a single
admonishment of the parrot or t h e woman, then in t u r n the seventy tales of
t h e parrot are superfluous.
So only t h e first type suits the ukasaptati as a f r a m e . The stories of t h e
ukasaptati contain the s u c c e s s of t h e protagonists achieved by shrewd
wisdom, as I have pointed it out a t t h e beginning of this paper. 6 1 So only t h a t
f r a m e can suit the whole work where the parrot by means of his seventy tales
attains s u c c e s s f u l l y his goal, to guard the fidelity of the wife of his
master. So only a h a p p y end, relating absolute success, suits these seventy tales.
The basic requirement, the occasion to tell the seventy tales, can be found only
in the variants of this type, here it is practical and suitable to amuse the consort
mediating on infidelity with seventy tales and to keep her on the right p a t h
while her husband is away.
The frames of the single stories of the ukasaptati tally completely only
with this f r a m e story, too. A t the inner frames 0 2 we never see any instance of
absentmindedness or inconsistency, which might indicate a later revision.
According to this opinion if there did exist different tale-variants a b o u t
the parrot, this (type 1) is the first and basic variant of the story of the p a r r o t
telling seventy tales so t h a t of the ukasaptati. I t is peculiar to the learned
Indian writers t h a t in connection with this they tell other parrot stories in their
works, and t h e inconsistent or clumsy copying or revision can be accounted for
the concluding p a r t of the textus ornatior, the text of which I consider, on t h e
whole, the most original. 63
For, as we h a v e seen, in the textus ornatior of t h e ukasaptati there are
three variants of the parrot-story to be found. One (type 1) at the beginning,
another one (type 3) in tales 6870 and a third one ( t y p e 4) at the very end.
This may embarrass the reader, as we have mentioned. To solve this problem
my idea is the following:
This v a r i a n t of the ukasaptati, which according to me is the best of all
and nearest to the original despite of being carelessly copied several times,
started, as it is natural, with a parrot-story of type 1 a t its beginning. T o t h e
60
I t is possible t h a t the king became so magnanimous with his renouncement a s
a result of a copying m a d e to a royal order.
81
See above note 3 and Cs. T t t s s y : A n t . Tan. 15 (1968) 221 foil, and A c t a
A n t . H u n g . 17 (1969) 433 foil.
82
Cs. T t t s s y : A n t . Tan. 10 (1963) 155 foil, and A c t a Orient. Hung. 18 (1965)
227 foil.
83
Cs. T t t s s y : A n t . Tan. 21 (1974) 59 foil. id. : a n d A c t a A n t . Hung. 21 (1973)
273 foil. ; id.: A n t . T a n . 23 (1976) 223. ; id.: Acta A n t . H u n g . 24 (1976) 285 foil.

452 CS. TTTSSY
e n d (probably f r o m t a l e 65, and obviously n o t only from tale 68) u p t o tale 70

a n o t h e r p a r r o t - s t o r y (perhaps then a l r e a d y known, b u t never used as t h e
f r a m e of such a long a n d large work; belonging to t y p e 3) w a s inserted pre-
s u m a b l y b y the a u t h o r himself. H o w e v e r , this was n o t m e a n t t o form t h e
closing f r a m e of t h e whole work (as, according to me, it h a s never before
f u n c t i o n e d as t h e f r a m e of the whole u k a s a p t a t i ) . A copyist or adapter,
u n d e r t h e influence of t h i s other p a r r o t - s t o r y , left out t h e earlier e x t a n t clos-
ing p a r t of t h e f r a m e of t y p e 1, after t a l e 70, for after the long work of copying
t h e v a s t collection of tales he no longer remembered the beginning of the work
(or, provided there were two copyists, t h e one copying t h e e n d did not even
k n o w t h e beginning of t h e work). This copyist believed too m u c h two closing
parrot-stories a f t e r each another. A following step t o w a r d s a corrupt t e x t
m i g h t h a v e been t h e elimination of t h e g a m b i t s corresponding to tale 70,
giving rise t h e lacuna of t h e beginning of tales 6568, 64 so as to omit this
second one of t h e two i n t r o d u c t o r y parrot-framestories. A following copyist or
a d a p t e r realized t h a t t a l e 70 does n o t correspond to the beginning of t h e work,
a n d so i t can not s t a n d as a closing f r a m e . This copyist tried t o t a k e u p t h e
t h r e a d of t h e n a r r a t i n g parrot, who e n d e a v o u r e d to hinder his l a d y from t h e
i n n o c e n t love and of P r a b h v a t wishing t o go to his lover. So being un-
a c q u a i n t e d with t h e original closing he w r o t e a new p a r r o t - s t o r y (of t y p e 4)
t o r o u n d off t h e work. E v e n t h a t is possible t h a t he knew a v a r i a n t similar t o
t h a t of N i r m a l a r v a k a , which earlier h a d nothing t o d o w i t h the -
saptati. 8 5
W e m u s t suppose t h a t not only so m a n y parrot-stories or v a r i a n t s of
t h e m m i g h t have existed in the course of t i m e , as Hertel or ourselves know of.
On t h e c o n t r a r y it is obvious t h a t t h e r e were several o t h e r parrot-stories,
which, however, h a v e n o t come down t o us. So Hertel's a t t e m p t in his excellent
book t o relate the remaining variants chronologically, and t o d r a w correspond-
ing conclusions, does n o t necessarily yield correct results. I t is possible t h a t
v a r i a n t s no longer e x t a n t underlied some v a r i a n t s t h a t h a v e come down t o us,
a n d so, because of t h e quondam existence of far earlier v a r i a n t s the whole
s i t u a t i o n m a y have differed considerably f r o m the one e x t r a p o l a t e d by Hertel
f r o m an incomplete material, leaving t h i s incompleteness o u t of account.
64
As the P a n c k h y n a v r t t i k a follows t h e textus ornatior very closely [see Cs.
T t t s s y : A n t . Tan. 21 (1974) 63 foil, a n d A c t a A n t . H u n g . 21 (1978) 277 foil.], t h e
l a c u n a in the t. o. can principally be a m e n d e d o n the basis of this v a r i a n t . B u t Nirmala
P t h a k a ' s story m e n t i o n e d as the last one in t h e type 3 m a y p r o b a b l y be interesting
t h a t t h e king mentioned in t h e t. o. (tale 69) p l a y s in this and only in this variant a role.
I t is possible t h a t between this p a r t of the t. o. a n d the variant of N i r m a l a P t h a k a other
similarities a n d relationships m a y be established.
65
T h e n a m e of t h e king in the v a r i a n t of Nirmala Srvaka is N a n d a . I n the loka
sited in n o t e 44 V i n a y a k a n d a r p a is m e n t i o n e d a s the son of king N a n d a , while earlier
this does n o t happen a n y w h e r e else, the n a m e of his father has never been mentioned.
The n a m e of the king ( N a n d a ) common to b o t h of t h e m m a y h a v e appeared in a story,
where b o t h variants originated from.

T H E VARIANTS OF T H E F R A M E STORY OF T H E S U K A S A P T A T I 453
The parrot occurs in countless other cases in the Sanskrit and in the Indian
literature. I t would be interesting to expound and to systematize the occasional
relations to one another and the types of the other parrot stories, b u t our aim
was to examine the parrot story serving as frame story of the Sukasaptati, to
make an a t t e m p t to determine which one of them was the earliest variant of
the frame story.
Budapest.

J. HORVTH
7. oct. 1911 - 3. fv. 1977
L'estime, l'amiti et la fidlit de souvenir nous e n g a g e n t rendre h o m m a g e

M. J . H o r v t h , n o t r e confrre disparu, m e m b r e du Comit d e r d a c t i o n des Acta A n t i q u a .
E n sa personne, le mdivisme hongrois a perdu l'un de ses spcialistes les mieux f o r m s
et ayant les plus larges connaissances, et d o n t le dcs r e p r s e n t e une perte pnible p o u r
ses confrres, ses disciples et pour tous ceux qui travaillent d a n s la mme b r a n c h e d e
la recherche.
La carrire scientifique de J . H o r v t h tait le sort h e u r e u x et tragique d u s a v a n t :
heureux en ce qu'il p o u v a i t consacrer t o u t e sa vie ses recherches scientifiques q u ' i l
aimait, tragique aussi, en ce qu'il ne lui tait jamais donn d e t r o u v e r les cadres organi-
sationnels, parmi lesquels son activit scientifique aurait p u se dvelopper, en fonction
d e ses capacits. Il a repris la tradition des recherches abordes p a r J . Abel et J . H u s z t i ,
chercheurs minents de la littrature latine en Hongrie m d i v a l e , quand, en a d o p t a n t
les mthodes de la philologie du latin classique, il avait c o m m e n c , en 1941, sa carrire
scientifique avec sa thse de doctorat L'vque Calanus et la Vita Attilae. Ses con-
naissances en matire de philologie et de palographie, f u r e n t compltes en 1941 1942,
Munich, chez P . L e h m a n n , qui f u t , l'poque, un spcialiste minent de la philologie
mdivale.
Son trait intitul Les sources d u C o m m e n t a i r e d ' A n d r a s Pannonius sur les
Cantica Canticorum tmoigne d ' u n e comptence et d ' u n e s p i i t m u i , et montre en m m e
t e m p s que son cercle d'intrt embiasse celui de l ' h u m a n i s m e galement. Sa personnalit
de savant est caractrise p a r une profondeur tendue, l ' t u d e dtaille des sourcos, u n e
logique rigoureuse, et u n esprit lucide, quoi s'ajoutent encore u n dvoument asctique
et le fait d'tre i n t r i e u r e m e n t comme uni avec l'objot de sa recherche, traits caractristi-
ques dont un vrai s a v a n t ne peut p a s se passer. C'est ainsi q u ' i l pouvait s'identifier, a u
niveau spirituel, l'objet de ses recherches, avec Andreas P a n n o n i u s , joueur d e l'pee
de Hunyadi, avec le suprieur chartreux, chou la suite d ' u n e vie m o u v e m e n t e ,
de Hongrie en Italie et qui aprs avoir chang l'pe c o n t r e la solitude de l'ordre, la
rita activa contre la vita contemplativa, trouve dans sa solitude des amis prcieux et u n e
socit plus bavarde, d c o u v r a n t pour lui des secrets . . . assis derrire une table m o d e s t e ,
il cause avec les prophtes, les aptres, et les vanglistes . . . il contemple, au lieu d ' u n
luxe artificiel, la b e a u t d e l'immense univers . . . poursuit d e s tudes rgulires s u r des
traits littraires, c'est l, la sagesse contemplative, crit l ' a u t e u r de lui.
Oui, la sagesse contemplative et la dcouverte des secrets, dterminent l'ros d u
s a v a n t , tournant ce temps-l ses regards vers les problmes stylistiques de la l i t t r a t u r e
latine de l'poque arpadienne. Ce n'est qu'aprs de longues a n n e s de recherches q u e
pouvait natre le m a n u s c r i t embrassant les rsultats de ses recherches, poursuivies d a n s
ce domaine et qui a servi, on 1948, de thse de privat-docent universitaire. Mais l'habili-
tation, cause de la suspension de la p r o c d u r e , ne p u t avoir lieu. Ce n'est que plus t a r d ,
en 1952, qu'il devint candidat s lettres.
Cette m m e anne m a r q u e une a u t r e d a t e i m p o r t a n t e , mme dcidante d e sa
vie. Le Ministre de la Cultuie consentit crer la Chaire de Philologie du latin mdival,
a u sein de la F a c u l t des Lettres do l'Universit Etvs L o r n d , et la Facult le p r s e n t a
en mme temps p o u r le titulariser professeur. Ce n'tait u n e f a u t e ni la Facult, ni
l'Universit, ni mme au Ministre de la Culture s'il ne d e v i n t pas professeur et si la
Chairo de Philologie mdivale no f u t pas cre. Mais cet v n e m e n t ne resta p a s sans
le blesser pour t o u t e la vie et sans faire reculer les recherches mdivales dont les cadres
organiss n'taient t o u j o u r s pas assurs.
Acta Antiqua Arademiae Srientiuruin Hungarirae 28, 1980

456 OBITUARY
Mais, par l'trange ironie du sort, en 1964, deux annes plus t a r d , son ouvrage,
i n t i t u l Les problmes stylistiques de notre littrature latine l'poque arpadienne
lui v a l u t le doctorat s lettres, et, en 1955, le 3e degr d u prix K o s s u t h . Cet ouvrage resta
le p l u s important dans t o u t e son uvre, r s u m a n t et projetant ses recherches antrieures
et postrieures. Thoriquement, le plus g r a n d mrite de cet ouvrage tait sa volont
d ' a d o p t e r les mthodes de recherche stylistique, labores il y a longtemps par la philo-
logie latine classique et appliques l'tranger m m e aux recherches mdivales, l'tude
de la littrature latine de l'poque arpadienne. E t qu'il voulait comprendre et interprter
les u v r e s littraires latines de cette poque d u point de vue de l'volution sociale et
historique. En franchissant le seuil d'une nouvelle poque de l'volution sociale et
historique, nos anciennes traditions historiques, orales et crites, subissent, suivant les
t e m p s , de nouvelles apprciations historiques et critiques, dit-il, et il dmontra, l'aide
d ' u n e analyse minutieuse comment la littrature crite combat, d a n s u n e priode ult-
rieure de l'volution, les ouvrages littraires, r e f l t a n t les ides et les b u t s d'une priode
a n t r i e u r e de l'volution fodale. L'adoption de ces aspects et de ces mthodes apporta
ses r s u l t a t s dans q u a t r e d o m a i n e s : 1. le r a p p o r t des deux lgendes de Grard, 2. le
s t y l e des rcits d'Anonyme, 3. le problme d ' a u t e u r de l'Histoire des H u n s , 4. la figure
de l'vque Nicolas, a u t e u r de la Geste de l'poque d'Andr I.
Sa personnalit savante tait circonspecte et scrupuleuse et u n e fois convaincu
de la justesse d'une a f f i r m a t i o n scientifique, il t a i t prt la dfendre. Ainsi, les discus-
sions provoques par son livre, l'ont men se d b a t t r e et poursuivre ses recherches.
I l a labor en dtail sa thorie concernant l'vque Nicolas, tant, d ' a p r s lui, l'auteur
de la premire Geste (1955), la valeur de source des lgendes de Grard (1957), ses ides
c o n c e r n a n t l'Histoire des H u n s et son a u t e u r (1963), ainsi que ses recherches d'Ano-
n y m e , surtout en ce qui concerne sa personnalit et sa culture (1966) et, finalement,
ses v u e s sur les historiens de l'poque des A n j o u (1971). Les problmes d'Anonyme et
ia lgende de Grard taient repris en 1974, d a n s le cadre d'un trait. P e n d a n t ce temps-l,
il travaillait aussi pour claircir d'autres questions importantes, d o n t celle des Hongrois
noirs (1967), de l'histoire du Sermon funraire (1970) et les institutions politiques turques
d e l ' E t a t hongrois mdival (1970).
E n 1962, une poque nouvelle de sa vie commena: il devint professeur d'uni-
versit la Chaire de Philologie latine de la F a c u l t des Lettres de l'Universit Etvs
L o r n d , mais son activit de professeur tait partage entre les latins classique et mdi-
val. D j auparavant, partir des annes cinquante, il d u t assumer de nouvelles charges :
les t r a v a u x du Dictionnaire de la Latinit hongroise qu'il devait rorganiser et recom-
m e n c e r avec des moyens insuffisants et d a n s des conditions difficiles. L e s blessures ancien-
nes e t rcentes ainsi que le r y t h m e t e n d u du travail, portrent a t t e i n t e son systeme
n e i v e u x . E n 1966, il a quitt la direction des t r a v a u x du Dictionnaire de la Latinit
hongroise et, malgr les sollicitations rptes, il ne voulait plus j a m a i s la reprendre,
m m e s'il suivit la m a r c h e de ce projet avec u n vif intrt aussi p l u s t a r d et il assista
a u x t r a v a u x de la Commission qui f u t charge de l'laboration du p l a n dfinitif du Dic-
tionnaire. Psychiquement, il se retira de plus en plus dans sa solitude chartreuse.
E n 1972, une nouvelle re s'ouvre d a n s sa v i e : la charge d e la direction de la
Chaire de Philologie latine lui tait confi. Ce devoir le rendit visiblement la vita activa,
m a i s lui cota cher. Lui, qui tait appel pour reprsenter la philologie mdivale, devait
consacrer toute son nergie pour organiser et g a r a n t i r l'enseignement d u latin classique
et les recherches scientifiques qui y sont relies. Quelle que f t la q u a n t i t de temps que
ce devoir lui demandt, il traduisit et publia, aprs u n travail zl, la Chronique de Buda
(1973), entreprit un nouveau tiavail scientifique, celui de la publication critique des
u v r e s compltes de J a n u s Pannonius. Ce travail a donn pour r s u l t a t un trait sur
les genres de J a n u s P a n n o n i u s (1973), o il essaie de dchiffrer le secret du pote : Le
secret de J a n u s Pannonius consiste en ce que lui, crit-il, enchant p a r l'esprit de l'anti-
u i t exprimant l'humanisme par des formes classiques, s'est laiss totalement pntr
u m o n d e antique f o r m a n t et formulant ses sentiments et son espiit t o u t fait dans
l'esprit de cette vie antique . . . La posie de J a n u s Pannonius reste d ' u n e grande valeur
j u s q u ' ce que le sens de l'antiquit reste vivant. Ici dj, il discerne nettement que
sans le sens de l'antiquit, la posie de J a n u s ne p e u t , elle non plus, prsenter de valeur.
L e chercheur du latin mdival p u t saisir la base qui tait de toute sa vie le sol, et sans
quoi le secret de J a n u s Pannonius non plus, n ' e t pu tre dchiffr : le latin classique.
E n 1972, son organisme f u t atteint p a r une maladie incurable. Il vit les dernires
a n n e s de sa vie pleinement conscient de l'issu probable. Devant la dernire certitude,
le s a v a n t prudent, presque timide, devint hros, il regarda en face d u sort, travaillant
m m e les dernires heures de sa vie.

OBITUARY 457
IL est disparu. Mais, il nous a lgu ses uvres dcouvrant les valeurs de la c u l t u r e
mdivale de Hongrie, ses b u t s que nous devons atteindre, l'ros du savant, son dvoue-
ment ascetiquo la science et sa solitude d u c h a r t r e u x o le s a v a n t peut t o u j o u r s t r o u v e r
une socit qui lui dvoile ces secrets.
J. Harmatta

D. D. MOUKANOS : '0 rno TOV elvai TWV fiadrjpaTtxcbv dvrtxEi/tvcov xar rv
nXxwva xal TOV ' ( = DIE SEINSART MATHEMATISCHER GE-
B I L D E NACH P L A T O N U N D NACH A R I S T O T E L E S ) . Dissertation. Athen
1979. 127 S.
Zuletzt hat sich die gelehrte Welt im Jahre 1978 im R a h m e n von feierlichen
Konferenzen u n d Zusammenknften des Aristoteles gedacht, der gerade vor 2300 J a h r e n
(in 322 v. u. Z.) gestorben war. E s wurden bei diesen Gelegenheiten die groen Verdienste
des Stagiriten, sein ungeheuerer E i n f l u auf das Denken u n d Forschen der seitdem ver-
flossenen Epochen in zahlreichen Vortrgen und Untersuchungen gewrdigt. Natrlich
konnte dabei auch die Frage nicht unbercksichtigt bleiben : Und wie steht es um Aristote-
les und die Mathematik ? Was hat die Mathematik dem Aristoteles zu verdanken ? Es
gengt diese Fragen zu formulieren, u m der Tatsache gewahr zu werden, da vielleicht
auf gar keinem anderen Gebiet so sehr, als eben was die M a t h e m a t i k betrifft, der Unter-
schied der beiden groen Philosophen der Antike, Piaton u n d Aristoteles, in die Augen
springt. Denn mgen sie zwar beide gleichermaen die zeitgenssische Mathematik sehr
gut gekannt, die Methode dieser Wissenschaft als Vorbild hingestellt wie dios beson-
ders Platon zu t u n gepflegt h a t t e und ihre ei luternden Beispiele hufig der Mathe-
matik entnommen haben, dennoch war die Einstellung ihrer beiden dieser Disziplin
gegenber grundverschieden.
Das Hauptverdienst der tchtigen Athoner Dissertation von D. D. Moukanos,
auf die ich hier nachdrcklich a u f m e r k s a m machen mchte, besteht in der grndlichen
Kenntnis der einschlgigen L i t e r a t u r besonders der philosophiegeschichtlichen
und in der besonnenen, zurckhaltenden Behandlung des Themas. Die Methode ist
philosophisch-historisch o r i e n t i e r t ; gelegentlieh nimmt sie a u c h die Hilfe der analyti-
schen Spraehphilosophe in Anspruch.
Es wrde wohl allzu weit fhren, wenn ich diesmal die Ansichten des Verfassers
im einzelnen schildern und besprechen wollte ; anstatt dessen sei hier nur ein kurzer
berblick der Arbeit geboten.
Jenes Bild der antiken Mathematik, das der gesamten Arbeit zu Grunde liegt,
geht auf den verdienstvollen griechischen Forscher unsorer Tage, E . Stamatis zurck.
D a r u m wird sein Name in der Einleitung, die in mehreren Abschnitten den bisherigen
Stand der Foischung skizziert, u n t e r den erston genannt (S. 14, Anm.). Eine unmittel-
bare Anregung bedeuteten f r Moukanos die Ansichten der sog. Tbinger Schule ; er
hebt besonders eine Arbeit von J . H. K r m e r aus d. J . 1959, u n d eine andere von K . Gaisor
a u s d. J . 1968 hervor ; diese h t t e n das Bild eines bermig mathomatisierten Piaton
entworfen, das sich jedoch aus den Dialogen nicht besttigen liee. Solche Gedanken
scheinen zuerst den Widerspruch des Verfassers ausgelst zu haben. Erst nach einer
kurzen Schilderung dieser R i c h t u n g der Forschung blickt er auf solche lteren Werke
zurck, wie diejenigen von E . F r a n k (1923), J . Stenzel (1924) oder O. Becker (1927).
Doch folgt der Rckblick keineswegs einer chronologischen Ordnung ; es werden hie
und da auch Verffentlichungen der allerletzten J a h r e / m e r als ltere Werke berck-
sichtigt. (Im zeitlichen Durcheinander k o m m t wohl auch jeno bedenkliche Tatsache zum
Ausdruck, da unsere modernen historischen Ansichten sich keineswegs auseinander
oder auch nur nacheinander e n t f a l t e t hatten.
Der Kern der Arbeit besteht eigentlich aus zwei Teilen. Der dritte Teil mehr
ein bloer Anhang ist nur eine neugriechische bersetzung einiger Abschnitte

460 reviews
d e r Metaphysik : 1076 a 8 37 ; 1076 a 38 1077 b 16 ; u n d 1077 b 17 1078 b 6.

Der erste Teil ist in f n f A b s c h n i t t e n (S. 27 71) P i a t o n s Ansichten von der
m a t h e m a t i s c h e n E x i s t e n z gewidmet. Vorausgeschickt wird in zwei k u r z e n Kapiteln (S.
2831) die pythagoreische Lehre, f r die die E n t d e c k u n g der I n k o m m e n s u r a b i l i t t
ausschlaggebend gewesen sei. P i a t o n sei jedoch von dieser lteren L e h r e abgewichen.
( E s ist wrtlich von einer nxXtaiz die Rode.) E r s t P i a t o n h t t e n m l i c h die eindeutige
Scheidung des misthetonir u n d des moeton eingefhrt. E s wird hervorgehoben, d a
n a c h P i a t o n s Auffassung die Existenz m a t h e m a t i s c h e r Gebilde d e r Zahlen u n d der
geometrischen Figuren ontologischen Charakters sei. Moukanos widerlegt dabei nicht
blo Ansichten der T b i n g e r Schule, sondern auch eine S t e l l u n g n a h m e von R . M. H a r e
(1967), u n d eine andere v o n H . Cherniss (aus d. J . 1946 bzw. 1966). Die beiden letzten
A b s c h n i t t e des ersten Teils b e h a n d e l n die sog. berschtzung d e r M a t h e m a t i k in der
A l t e n Akademie u n d d a s Mathematisieren der N a t u r im Timaios, sowie die Stellung-
n a h m e des Aristoteles dagegen.
Die abschlieenden E r r t e r u n g e n des ersten Teils f h r e n schon zur Abhandlung
d e r Aristotelischen Theorie ber die Seinsart m a t h e m a t i s c h e r Gebilde hinber. W a s h a t
n u n d e n Aristoteles v e r a n l a t f r a g t Moukanos zu seiner A b n d e r u n g der Platonischen
L e h r e ? Offenbar erwies sich der Stagirit kritisch (um nicht zu sagen : skeptisch)
n i c h t n u r der Lehre von d e n Ideen, sondern auch der ganzen Dialektik oder mindestens
der in platonischem Sinne aufgefaten Dialektik gegenber. D a r u m entmathematisierte
er die Physis, und so e n t s t a n d seine rein logisch orientierte L e h r e v o n der Mathematik,
a n Stelle von Piatons ontologischer Lehre.
E s geht wohl a u c h schon a u s dieser k n a p p e n Z u s a m m e n f a s s u n g d e r Gedanken von
M o u k a n o s hervor, d a seine Dissertation vorwiegend Aristoteles-orientiert ist. Man
w i r d es dem Philosophen n a t r l i c h nicht belnehmen, d a seine S t r k e gerade in jenem
V e r s t n d n i s liegt, das er d e m Aristoteles entgegenzubringen vermag.
E s bleibt dabei eine o f f e n e Frage, ob angesichts eines solchon Aristotelismus
Piaton und die Mathematik n i c h t notwendigerweise zu kurz k o m m e n m s s e n ? Aber diese
F r a g e soll nicht als V o r w u r f ausgelegt werden. Zweifellos werden a u c h Platoniker u n d
solche, die fr die M a t h e m a t i k leidenschaftlicher interessiert sind, als der nchterne
Stagirit, die schne, g r n d l i c h e u n d sorgfltig zusammengestellte A r b e i t von Moukanos
d a n k b a r und mit N u t z e n in die H a n d n e h m e n .
A. S z a b
P. LIVER: D I E N A C H W I R K U N G D E R A N T I K E N S A K R A L S P R A C H E IM
CHRISTLICHEN GEBET DES LATEINISCHEN UND ITALIENISCHEN
M I T T E L A L T E R S . B e r n , F r a n c k e Verlag, 1979. 434 p . R o m a n i c a Helvetica 89.
R . Liver's V o r h a b e n war die Beziehungen der antiken Sakralsprache wie sie

sich a u s den sprlichen a u t h e n t i s c h e n Sprachdenkmlern der Vorzeit, bzw. religis
inspirierten dichterischen T e x t e n des Goldenen Zeitalters rekonstruieren lt zu den
dichterisch gestalteten lateinischen u n d altitalienischen Gebeten des Mittelalters (Dante
einschhelich) aufzuhellen u n d zu errtern.
D a s Werk gliedert sich in zwei H a u p t t e i l e . Der erste ist beschriebenden Charak-
t e r s : es werden darin die stilistischen u n d s y n t a k t i s c h e n F o r m e n des Gebets a u f g e f h r t ,
vorwiegend an mittelalterlichen Beispielen. Der zweite, aber k r z e r e H a u p t t e i l ist der
A u f g a b e gewidmet, die N a c h w i r k u n g der a n t i k e n Sakralsprache im Mittelalter zu m u s t e r n
u n d bewerten.
Die Zweiteilung, wie sie u n s vorliegt, scheint ungnstig zu sein. I m zweiten Teil
sollte nmlich die Verfasserin, u m die a n t i k e Tradition sichtbar zu m a c h e n , alle Angaben
wiederholen, die sie i m ersten schon aufgezhlt h a t t e u n d zwar einem berflssig
komplizierten System folgend. A n s t a t t ihrer Wiederholung aber b e g n g t e sie sieh m i t
Verweisungen, womit d e m Leser eben d a s Bearbeiten des theoretisch-bewertenden Teiles
erschwert wird, auf diese Weise ist er z u m stndigen Vor- und R c k w r t s b l t t e r n gezwun-
gen. So haben wir den E i n d r u c k , d a der Verfasserin, die zum S a m m e l n u n d Gruppieren
des Materials noch L u s t h a t t e , wenig K r a f t z u r Ausarbeitung der H a u p t a u f g a b e geblie-
ben ist. E s wre zweckmiger gewesen, das W e r k zu a u f z u b a u e n , d a die sprachlichen
Erscheinungen, alphabetisch, bzw. t h e m a t i s c h gegliedert, einzeln der z u m Ziel gesetzten
k o m p a r a t i v e n P r f u n g unterzogen worden wren, u n d die ntigen Folgerungen in einem
k n a p p e n Schlukapitel ihren P l a t z g e f u n d e n h t t e n ; auf solche Weise stnde uns jetzt
ein ntzliches Nachschlagewerk zur Verfgung.

reviews 461
Auch im Z u s a m m e n h a n g mit dem scheinbar imponierenden Material erheben sich

zwei E i n w n d e ; d a nmlich im ersten Teil viele Angaben a n g e f h r t werden o h n e
jedoch Vollstndigkeit zu erstreben , deren a n t i k e Parallelen fehlen und die auf diese
Woise zur K o m p a r a t i o n , zur Verwirklichung der H a u p t a u f g a b e des Buches, von vorn-
herein ungeeignet zu sein scheinen. Andererseits ist das Ausschlieen der P r o s a g e b e t e
theoretisch b e a n s t a n d b a r . Denn so wird ein betrchtlicher Teil der liturgischen T e x t e
der offiziellen Gebote der katholischen Kirche der P r f u n g vllig entzogen ; diese
Tatsache verringert d e n Wert des Werkes u n d n i m m t u n s die Mglichkeit, seine E r g e b -
nisse unmittelbar zu benutzen.
Trotz unserer kritischen Bemerkungen geben wir den Schlustzen der Verfasse-
rin vllig recht : Schon eine Zusammenstellung gebotssprachlichen Materials i m hier
abgesteckten R a h m e n zeigt, wie gefhrlich es ist, die Formulierungen einzelner A u t o r e n
als originale Leistungen zu erklren oder sie auf den Einflu eines bestimmten einzigen
Autors z u r c k f h r e n zu wollen, wie das gerade f r D a n t e immer wieder versucht w i r d .
K a u m in einem a n d e r n sprachlichen Bereich (ausgenommen vielleicht den der R e c h t s -
sprache) ist die Macht der Tradition d e r a r t augenfllig wie in d e m der Sakralsprache
I. B o b o n k a i

INDEX
P. Daffin: On Kaspapyros and the So-Called 'Shore of the Scythians' 1

H. W. Haussig : Die ltesten Nachrichten der griechischen u n d lateinischen Quellen
ber die Routen der Seidenstrae nach Zentral- und Ostasien 9
B. A. LitvinskiyI. It. Pichikyan : Monuments of Art from the Sanctuary of Oxus
(Northern Bactria) 25
I. Borzsk: Zentralasiatische Elemente in dem Alexanderroman 85
. Staviskiy : K a r a Tepe in Old Termez. A Buddhist Religious Centre of the Kushan
Period on the Bank of the Oxus 95
J. Wolski : Les sources de l'poque hellnistique et partho do l'histoire d'Iran. Diffi-
cults de leur interprtation et problmes de leur valuation 137
E. A. Davidovich : The First Hoard of Tetradrachmas of the K u s n a 'Heraios' . . . . 147
I . Gershevch : The Colophon of the NOKONZOK Inscription . . . '. 179
J. Harmatta: Kidara and the Kidarite H u n s in Kamr 185
Ph. Gignoux : Titres et fonctions religieuses sasanides d'aprs les sources syriaques
hagiographiques 191
A. D.H. Bivar : Questions of Interpretation in the Liscriptions of the Sasanian Seals 205
K.Czegldy: Zur Geschichte der Hephthaliten 213
M. Grignaschi : La chute do l'Empire Hephthalite dans les sources byzantines et
perses et le problme des Avar 219
I. Ecsedy : Western Turks in Northern China in the Middle of the 7th Century . . . 249
Bo Utas : The Pahlavi Treatise Avdh sahikh Sakistn or "Wonders and Mag-
nificence of Sistan" 259
M. Maroth : Die Xalag in den arabischen Quellen 269
I . Gershevch : The Bactrian Fragment in Manichean Script 273
I . Gershevch : Beauty as the Living Soul in Iranian Manicheism 281
TP. Sundermann : Probleme der Interpretation manichisch-soghdischer Briefe . . . . 289
M. Maroth : Ptolemaic Elements and Geographical Actuality in al-Jfuwrizrm's
Description of Central Asia 317
I. Frhlich: Historiographie et aggada dans le Liber Antiquitatum Biblicarum du
Pseudo-Philon 353
T. Nagy: Die zwei Neratii Prisci und Neratius Priscus der Jurist 411
Cs. Tttssy : The Variants of the F r a m e Story of the ukasaptati 437
t J . Horvth (7. oct. 1911 3. fev. 1977). (J. Harmatta) 455
D. D. Moukanos : 'O zgono zov elvau xwv ya&ryiazixyv vxixeiyvwv xaz zov HXxwva
xai zov 'Agictzozkr] (Ree. . Szab) 459
R . Liver : Die Nachwirkung der antiken Sakralsprache im christlichen Gebet des
lateinischen und italienischen Mittelalters. (Ree. I. Boronkai) 460
PRINTED IN HUNGARY
Akadmiai Kiad s N y o m d a , Budapest

Fig. 1. Takht-i Sangin. Ground plan of the temple of the Oxus. 3rd century B. C.
Fig. 2. Sheath of akinakes, ivory, 5th c e n t u r y B. C.
Fig. 4a. Altar of Atrosokes with dedication to Oxus, limestone, bronze, 2nd century B . C .
Fuj. lb. Altar of Atrosokos with dedication to Oxus, limestone, bronze, 2nd
century B. C.
Fig. 5. Silenus-Marsyas, playing on a two-branched flute. Bronze
Fig. 6. Head of a Hellenistic leader. Unbaked clay. 3rd century B. C.
Fig. 7. Head of a local aristocrat. Alabaster
Fig. S. Torso in belted chiton. Unbaked clay
Fig. 0. Alexander the Great in lion helmet. Ivory. 3rd century B. C.
Fig. 10. Heracles striking down Silenus. Ivory. E n d of 4th century . C.
Fig. 11. Bouterolle of sheath with representation of iiippocampus. Ivory. 2nd century B.C.
Fig. 12. Bactrians on hunting. Wall of a casket. Ivory
Fig. 10. The fragment of the wall painting composition of the donators
Fig. 11. The dontor. Detail of the wall-painting

Fig. 12a. The fragment of the wall-painting with the Buddha and monks
Fig. 13a. The first stone capitel f r o m K a r a tepe
Fig. 13b The first stone capitel f r o m K a r a tope

Fig. I /. The second stone cupitel from K a r a tej
Fig. 15. Stone and baked clay discs
Fig. 16. The lamps
Fig. 17. The fragment of t h e vessel with the handle in the shape a monkey
Fig. 18. The vessels from the burials
Fig. 10a. Post-Kushan coin, averse
Fig. 19b. Post-Kushan coin, averse

Fj. 19c. Post-Kushan coins, reverse
Fig. 20a. An Early H e p h t h a l i t h e coin, averse
Fig. 201). An Early Hephthalite coin, reverse

b
Fig. 21. Graffiti: a in K u s h a n script; b - bilingual (in K u s h a n and Middle Persian
scripts)
Fig. 22. The upper cell in the complex. Reconstruction by W. Donchik
Fig. 23. The northern corridor of the F - I I cave-temple, a view from entrance
Reconstruction by W. Donchik
Fig. 23. The northern corridor of the P - I I cave-temple, b - view within.
Reconstruction by W . Donchik
Fig. 24. The cell of the P - I I cave-temple, a view f r o m the entrance.
Reconstruction by VV. Donchik
Fig. 24. The cell of the P - I I cave-temple, 6 - view from within.
Reconstruction by W . Donchik
F ig. 25. Three western vaulted aiwans on the southern fringe of the complex courtyard
^ Fayaz-Tepe
Fig. 26. Schematic plan of Kushan Termez

Pl. I. First hoard of the tetradraehmas of the Kusana Heraios coin No. 1
PL I I I . First hoard of the tetradrachmas of the Kusna Horaios coin No. 3
PI IV. First hoard of the tetradrachmas of the Kusna Heraios - coin No. 4
Pl. I . First hoard of the tetradrachmas of the Kusna Heraios coin No. 5
Pl. VI. First hoard of the tetradraohmas of the Kusana Heraios coin No. (i
Pl. VIII. First hoard of the tetradrachmas of the Kusna Heraios coin No. 8
l'hue I
171
Plate III
105 5 1 2 0 1 2 5 1 3 0 1 3 5
Map. 1. Central Asia according t o the Rewriting of Ptolemy by al-Ruwarizmi

Map. 2. Central Asia according to t h e Rewriting of Ptolemy by Suhrb
Die Acta Antiqua verffentlichen Abhandlungen aus dem Bereiche der klassischen
Philologie in deutscher, englischer, franzsischer, russischer und lateinischer Sprache.
Die Acta Antiqua erscheinen in H e f t e n wechselnden Urnfanges. Vier H e f t e bilden
einen B a n d .
Die zur Verffentlichung b e s t i m m t e n Manuskripte sind an folgende Adresso zu
senden:
Acta Antiqua, Budapest 502, rostafik 24.
An die gleiche Anschrift ist auch jede f r die Redaktion und dem Verlag bostimmto
Korrespondenz zu richten.
Bestellbar bei d e m Auenhandels-Unternehmen Kultra (1380 B u d a p e s t (12,
P.O.B. 149) oder bei seinen Auslandsvertretungen.
Les Acta Antiqua paraissent en franais, allemand, anglais, russe et latin et pu-
blient des travaux du domaine de la philologie classique.
Les Acta Antiqua sont publis sous forme de fascicules qui seront runis en volumes.
On est pri d'envoyer les manuscrits destins la rdaction l'adresse suivante:
Acta Antiqua, Budapest 502, Postafik 24.
Toute correspondance doit tre envoye , cette mme adresse.

On peut s'abonner l'Entreprise pour le Commerce Extrieur Kultra (1389
Budapest, P.O.B. 149), ou l'tranger chez tous les reprsentants.
Acta Antiqua -
, , , .
Acta Antiqua . -
.
:
Acta Antiqua, Budapest 502, Postafik 24.
-
.
Kultra (1389 Budapest
62, . . . 149), .
Periodicals o f the H u n g a r i a n Academy of Sciences are obtainable
at the f o l l o w i n g addresses:
AUSTRALIA SWETS S U B S C R I P T I O N S E R V I C E
C . B . D . LIBRARY A N D S U B S C R I P T I O N S E R V I C E 347b Heereweg, Lisse
Box 4886, G.P.O., Sydney N.S.W. 2001
INDIA
C O S M O S B O O K S H O P , 145 Ackland Street
A L L I E D P U B L I S H I N G P R I V A T E LTD., 13/14
St. Kilda (Melbourne), Victoria 3182
Asaf Ali R o a d , New Delhi 110001
AUSTRIA 150 B-6 M o u n t Road, Madras 600002
G L O B U S , HchstSdtplatz 3, 1206 Wien XX I N T E R N A T I O N A L B O O K H O U S E PVT. L T D .
M a d a m e C a m a Road, Bombay 400039
BELGIUM THE STATE TRADING CORPORATION OF
OFFICE INTERNATIONAL DE LIBRAIRIF I N D I A L T D . , Books Import Division, Chandralok
30 Avenue Marnix, 1050 Bruxelles 36 J a n p a t h , New Delhi 110001
LIBRAIRIE DU MONDE ENTIER
ITALY
162 rue du Midi, 1000 Bruxelles
I N T E R S C I E N T I A , Via Mazz 28, 10149 Torino
BULGARIA LIBRERIA COMMISSIONARIA SANSONI, Via
H E M U S , Bulvr Ruszki 6, Sofia L a m a r m o r a 45, 50121 Firenze
S A N T O V A N A S I A , Via M. Macchi 58
CANADA 20124 Milano
P A N N N I A BOOKS, P.O. Box 1017
D . E. A., Via Lima 28, 00198 Roma
Postal Station " B " , Toronto, Ontario XI5T 2T8
CHINA JAPAN
C N P I C O R , Periodical D e p a r t m e n t , P.O. Box 50 K I N O K U N I Y A BOOK-STORE CO. LTD.
Peking 17-7 Shinjuku 3 chme, Shinjuku-ku, Tokyo 160-91
M A R U Z E N C O M P A N Y L T D . , Book Department,
CZECHOSLOVAKIA
P.O. Box 5050 Tokyo International, Tokyo 100-31
M A D ' A R S K K U L T U R A , N r o d n i tfida 22
NAUKA LTD. IMPORT DEPARTMENT
115 66 Praha
2-30-19 M i n a m i Ikebukuro, Toshima-ku, Tokyo 171
PNS D O V O Z TISKU, V i n o h r a d s k 46, Praha 2 KOREA
PNS DOVOZ TLaCE, Bratislava 2
C H U L P A N M U L , Phenjan
DENMARK
E J N A R M U N K S G A A R D , Norregade 6 NORWAY
1165 Copenhagen TANUM-TIDSKRIFT-SENTRALEN A.S., Karl
F E D E R A L REPUBLIC O F G E R M A N Y Johansgatan 41-43, 1000 Oslo
K U N S T U N D WISSEN E R I C H BIEBER POLAND
Postfach 46, 7000 Stuttgart 1 W E G I E R S K I I N S T Y T U T K U L T U R Y , Marszal-
kowska 80, 00-517 Warszawa
FINLAND C K P I W, ul. Towarowa 28, 00-958 Warszawa
A TE E M1 NE N K I R J A K A U P P A , P.O. Box 128
SF-00101 Helsinki 10 ROUMANIA
D. E. P., Bucuregtl
FRANCE
D A W S O N - F R A N C E S. A., B. P. 40, 91121 Palaiseau I L E X I M , Calea Grivitei 64-66, BucurejtI
E U R O P R I O D I Q U E S S. A., 31 Avenue de Ver-
SOVIET U N I O N
sailles, 78170 La Celle St. Cloud
O F F I C E INTERNATIONAL DE D O C U M E N T A - S O J U Z P E C H A T - I M P O R T , Moscow
T I O N E T LIBRAIRIE, 48 rue Gay-Lussac and the post offices in each town
75240 Paris Cedex 05 M E Z H D U N A R O D N A Y A K N I G A , Moscow G-200
SPAIN
GERMAN DEMOCRATIC REPUBLIC
H A U S DER U N G A R I S C H E N K U L T U R D I A Z D E S A N T O S , Lagasca 95, Madrid 6
K a r l Liebknecht-Strae 9, DDR-102 Berlin SWEDEN
D E U T S C H E POST Z E I T U N G S V E R T R I E B S A M T
A L M Q V I S T A N D WIKSELL, G a m l a Brogatan 26
Srae der Pariser K o m m n e 3-4, DDR-104 Berlin
101 20 Stockholm
G R E A T BRITAIN GUMPERTS UNIVERSITETSBOKHANDEL AB
BLACKWELL'S PERIODICALS DIVISION Box 346, 401 25 Gteborg 1
H y t h e Bridge Street, Oxford OX1 2ET SWITZERLAND
BUMPUS, HALDANE A N D MAXWELL LTD.
C o w p e r Works, Olney, Bucks MK46 4BN K A R G E R L I B R I AG, Petersgraben 31, 4011 Basel
C O L L E T ' S H O L D I N G S L T D . , Denington Estate
Wellingborough, Northants NN8 2QT USA
W M . D A W S O N A N D S O N S L T D . , Cannon H o u s e EBSCO S U B S C R I P T I O N S E R V I C E S
Folkstone, Kent CT19 SEE P.O. Box 1943, Birmingham, Alabama 35201
II. K . LEWIS A N D C O . , 136 Gower Street F. W. F A X O N C O M P A N Y , I N C .
London WC1E6BS 15 Southwest P a r k , Westwood Mass. 02090
THE MOORE-COTTRELL SUBSCRIPTION
GREECE A G E N C I E S , North Cohocton, N. Y. 14868
KOSTARAKIS BROTHERS INTERNATIONAL R E A D - M O R E PUBLICATIONS, INC.
B O O K S E L L E R S , 2 H i p p o k r a t o u s Street, Athens-143 140 Cedar Street, New York, N. Y. 10006
HOLLAND STECHERT-MACMILLAN, INC.
M E U L E N H O F F - B R U N A B.V., Beulingstraat 2, 7250 Westfield Avenue, Pennsauken N. J. 08110
Amsterdam YUGOSLAVIA
M A R T I N U S N I J H O F F B.V. J U G O S L O V E N S K A K N J I G A , Terazije 27, Beograd
Lange Voorhout 9-11, Den llaag F O R U M , Vojvode MiSida 1, 21000 Novt Sad
H U ISSN 0 0 4 4 - 5 9 7 5 Index: 26.002

Actaantiqua 28 PDF

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

Actaantiqua 28 PDF

Uploaded by

Copyright:

Available Formats

ACTA ANTIQUA

TOMUS XXVIII FASCICULI 1-4

ACTA ANT. HUNG.

Acta Antiqua arc published in yearly v o l u m e s of four issues by

Publishing House of t h e H u n g a r i a n Academy of Sciences

Manuscripts and editorial correspondence should bo addressed to

Orders should be addressed to

K U L T U R A Foreign Trading Company

or t o its representatives abroad

Acta Antiqua is indexed in Current Contents

P. Daffin : On Kaspapyros and the So-Called 'Shore of t h e Seythians' 1

ON KASPAPYROS AND THE SO-CALLED

1 Acta Antiqua Academiae ScierUiarum Hungurirae 28, 1H80

s t r a t e d t h a t it could n o t be so and t h a t in this case exploration and conquest

2 Acta Antiqua Academiae ScierUiarum Hungurirae 28, 1H80

much in locating them on the map. All we can say is t h a t as Hecataeus p u t s

1* Acta Antiqua Acaderr.iac Scientiarvm Hungaricae 28, 1980

Caroe. 2 0 B u t neither A t a k nor Pesvar suits Herodotus' words any better t h a n

4 Acta Antiqua Academiae ScierUiarum Hungurirae 28, 1H80

Acta Antiqua Acadcmiae ScierUiarum Hungaroae 28 , 1980

6 Acta Antiqua Academiae ScierUiarum Hungurirae 28, 1H80

Bruchstck Herzfeld concluded besagt nicht etwa, da die Skythen an

1*ActaAntiqua Acaderr.iac Scientiarvm Hungaricae 28, 1980

Leaving it to professional Greek scholars to decide whether akt m a y be

Acta Antiqua Academiae Scientiarum Hungaricae 28, 1980

DIE LTESTEN NACHRICHTEN DER GRIECHISCHEN

Die in diesem R a h m e n angesprochenen Berichte aus der rmisch-grieehi-

Die Nordroute und ihre Vorgeschichte

Sowohl die Bodenfunde wie die Nachrichten der schriftlichen Quellen

Acta Antiqua Academiae Scientiarum Hungaricae 28,1980

Die Rezeption des griechischen Streitwagens durch China

Acta Antiqua Academiae Scientiarum Hungaricae 28, 1980

Die R o u t e n der Seidenetrae

R e z e p t i o n in die zweite H l f t e des zweiten J a h r t a u s e n d s gesetzt werden, also

Die Rezeption des griechischen Lwenkopfes als T'ao-t'ieh durch China

Acta Antiqua Academiae Scientiarum Hungaricae 28, 1980

Der Ural als Teilhaber am griechischen Chinahandel

Wagen und Lwenkopf kamen aus der mykenischen K u l t u r . D a m i t wird

2* Acta Antiqua Academiae ScierUiarum Hungaricae 28, 1980

ner, 1 5 von wo sie n a c h Osten abbog. 1 6 D e r Handelsverkehr n a c h China k n n t e

Acta Antiqua Academiae Scientiarum Hungaricae 28, 1980

s t a m m t zum grten Teil erst aus dem 5 . - 4 . J a h r h u n d e r t v. Chr. 20 Die E r h a l -

Herodots Bericht ber die Stationen der Nordroute der Seidenstrae

Da es auch in spterer Zeit hnliche Handelsbeziehungen wie in mykeni-

2* Acta Antiqua Academiae ScierUiarum Hungaricae 28, 1980

b a r a u c h die K a r a w a n e n f h r e r wechseln mute. 2 3 Das ergab sich aus den groen

Acta Antiqua Academiae Scientiarum Hungaricae 28, 1980

Die Strae f h r t e dann von den Argyppaioi weiter nach Osten. 3 0 I m

2* Acta Antiqua Academiae ScierUiarum Hungaricae 28, 1980

der d u r c h Herodot b e k a n n t gewordenen Nachrichten ber die Issedonen bezieht

Der Versuch einer Rekonstruktion der Sdroute der Seidenstrae

E i n e mit Herodots Bericht ber die Nordroute der Seidenstrae vergleich-

Acta Antiqua Academiae Scientiarum Hungaricae 28, 1980

J a h r h u n d e r t v. Chr. 38 Was von ihr vorhanden ist, beschreibt einen Abschnitt,

2* Acta Antiqua Academiae ScierUiarum Hungaricae 28, 1980

Das Chinabild in den Nachrichten von der Sdroute der Seidenstrae

I m Gegensatz zu Herodot erwhnt Strabo 5 3 bei der Beschreibung der

Acta Antiqua Academiae Scientiarum Hungaricae 28, 1980

Die Mittlerrolle der Serer im Handel mit China

Acta Antiqua Academiae Srientiarum Hungaricae 28, 1980

Der kombinierte See- und Landweg der Seidenstrae

Dieser dritte W e g der Seidenstrae setzt sich, was den Seeweg b e t r i f f t ,

Acta Antiqua Academiae Scientiarum Hungaricae 28, 1980