The Painting of the “Hunter-King” at Kakrak: Royal Figure or Divine Being?

Full text

Studio Editoriale Gordini
ANNALI DI CA’ FOSCARI
RIVISTA DELLA FACOLTÀ
DI LINGUE E LETTERATURE STRANIERE
DELL’UNIVERSITÀ CA’ FOSCARI
DI VENEZIA
XLVII, 3 2008
Estratto
M C
 e Painting of the «Hunter-King» at Kakrak: Royal Figure
or Divine Being?, pp. -

Matteo Compareti
THE PAINTING OF THE «HUNTER-KING» AT KAKRAK:
ROYAL FIGURE OR DIVINE BEING?
At the Buddhist site of Kakrak, not far from B¡my¡n in Afghani-
stan, some paintings and a colossal statue of Buddha were found
during the fi rst half of the last century.
1 Among the most intere-
sting and beautiful paintings there is the so-called «Hunter-King»
sitting cross-legged on a throne under a square arch supported by
pillars which is reminiscent of Gandh¡ran architectonical frames.
Every space above a capital presents the image of a white stupa
embellished with ribbons and a tree appears on both sides of the
sitting person (fi g. 1). The size of the painting is 0,60 m. × 0,52
m. and the date most recently proposed is the end of 7th-begin-
ning of the 8th century.
2 Its name is due to the crown and jewels
worn by the presumed king and the bow which he is holding
in a strange position with two hands in front of his chest while
two arrows appear sticking in the ground at his side. Finally, an
open-mouthed dog comes out from the curtain which covers the
throne and the heads of two white ducks (his prey?) or snakes
(Nagas?) 3 can be observed on the right upper side of the throne,
just below the fl oating ribbon typical of Sasanian art.
The painting was kept in the Kabul Museum and in an article
that recently appeared on the Internet, Z. Tarzi stated that it
survived the systematic destruction of ancient Afghan monuments
during the dramatic events which are still affl icting that devastated
Central Asian country.
4 The present study is, however, entirely
1 Hackin, Carl, 1933: pl. LXXII, fi g. 86. For the description of the discovery,
see also: Tarzi, 1977: 186-190.
2 Klimburg-Salter, 1993: 356; Klimburg-Salter, 2003: 7. For U. Jäger the
painting at Kakrak is dated to the 7th century: Jäger, 1988: 192. Tarzi cautiously
proposed the 6th-8th century; see: Tarzi, 1989: 659. Rowland preferred 6th-7th
century because of the Hephtalite features of his crown: Rowland, 1974: 106.
3 Jäger, 1988: 195.
4 See: A.W. Feroozi, Z. Tarzi, «The Impact of War upon Afghanistan’s
Annali di Ca’ Foscari, XLVII, 3 (s. or. 39), 2008 131-147

annali di ca' foscari, xlvii, 3 (. . ), 2008
132
based on two photos published by M. Bussagli and B. Rowland
more than thirty years ago
5 and not on the original painting.
So, all the hypotheses proposed here should be considered with
extreme caution.
1. History of the research
The painting is always referred to as the «Hunter-King» (or, in
French, as «personnage princier»): Roman Ghirshman did not
hesitate to use that name and so did, much more recently, F.
Tissot. 6 Mario Bussagli too concluded that he was not a god
since the triple-crescent crown typical of local kings would point
to a bejewelled royal character fond of hunting, as the bow, ar-
rows and dog suggest.
7 B. Rowland did not appear particularly
Cultural Heritage»: http://www.archaological. org /pdfs/papers/AIA_Afghani-
stan_address_lowers.pdf: page 12 (accessed 14 December 2007).
5 Bussagli, 1963: 39; Rowland, 1974: 113. See also: Tarzi, 1977: pl. XC, fi g.
118; Klimburg-Salter, 1989: pl. LXXXVII, fi g. 114.
6 Ghirshman, 1962: fi g. 426; Tissot, 2006: 116.
7 Bussagli, 1963: 39. For Hackin and Carl he was a princely fi gure too: Tarzi,
1977: 188-190. See also: Hallade, 1972: 422; Klimburg-Salter, 1993: 356.
Fig. 1. The so-called «Hunter-King» painting, Kakrak (Afghanistan). After Row-
land, 1974: 112 (detail).

133
the painting of the «hunter king» at kakrak
convinced of that defi nition although he eventually proposed an
identifi cation with a local prince
8 and T. Talbot Rice described
him as a «deifi ed king». 9 An article entirely dedicated to the
analisys of the «Hunter-King» was published by U. Jäger twenty
years ago. Many interesting ideas have been presented by Jäger es-
pecially on the iconographical features of the princely fi gure which
were possibly taken from the world of the nomads (Chionites or
Hephtalites). The triple-crescent crown of the «Hunter-King» is,
in fact, very common among Hephtalite coins which have been
found in Afghanistan.
10 According to Jäger’s interpretation, the
so-called «Hunter-King» was a symbolic representation of a royal
person who converted to Buddhism and abandoned hunting and
violence in general.
11
New hypotheses which openly differ from the traditional
identifi cation have been recently presented in two articles which
appeared on the Internet. Z. Tarzi was of the opinion that it is
an image of «Bodhisattva Siddartha before the illumination or
during an episode of the jataka».
12 A second scholar, P. Bane-
rjee, preferred to recognize in that painting a representation of a
divine being because of the halo behind his head: he proposed
Revanta who is the son of S∞rya and a hunter too.
13 Tarzi’s
idea is not supported by literary sources and it would be good
to know more about the jataka that he is mentioning. Banerjee’s
arguments could hardly be considered correct. They do not take
into consideration many aspects of Iranian Buddhism and seem to
be too concentrated on searching for Indian elements at Kakrak
and B¡my¡n. Those Afghan sites were defi nitely Buddhist ones
and, as it is well-known, Indian divinities like Indra and Brahma
appear very early in scenes of the life of Buddha. Even this does
not justify a reading totally based on Indian religious elements as
they refer to minor divinities (like Revanta) who were not much
represented in contemporary Indian art. It is also worth observing
8 Rowland, 1964-1965: 252; Rowland, 1974: 106, 112-113.
9 Talbot Rice, 1965: 166.
10 On the crown of the «Hunter-King» and its connection with Hephtalite
rule, see: Kageyama, 2008.
11 Jäger, 1988: 194.
12 http://www.archaological.org/pdfs/papers/AIA_Afghanistan_address_lowers.
pdf: page 12 (accessed 14 December 2007).
13 P. Banerjee, «Revanta from Kakrak (Bamiyan) in Afghanistan»: http://
www.ignca.nic.in/pb0017.htm#_edn1 (accessed 14 December 2007). For a similar
approach to the Kakrak painting, see also: Puri, 1987: 301-302. On Revanta see
also: Carter, 1988.

annali di ca' foscari, xlvii, 3 (. . ), 2008
134
that Revanta was usually represented as an archer riding a horse.
Further, the halo of the so-called «Hunter-King» could hardly be
considered a specifi c divine characteristic because, at the near-by
site of B¡my¡n, at least one monk and several fi gures identifi ed as
possible representatives of the royal (Sasanian?) family presenting
gifts to the Buddhas and Bodhisattvas have halos as well.
14
In our opinion, Jäger’s arguments still constitute the best iden-
tifi cation of the «Hunter-King» at Kakrak and the present paper
is mainly based on his ideas.
2. Remarks on some details
More than one detail deserves special attention for identifying that
fi gure. Certainly the crown with crescents and the jewels point
to an important person who could be identifi ed as a sovereign
although nothing allows us to think that local divinities could not
have worn jewels and crowns as well. If the two animals on one
upper right side of the throne should be considered Nagas as
U. Jäger suggested, then they could have been associated with a
royal fi gure like in Kucha paintings or a divine being as it can
be observed very often in Buddhist art.
The context at Kakrak is defi nitely Buddhist and it cannot
be excluded that the model for the «Hunter-King» could have
been an image of Bodhisattva or one of the so-called Buddha
parée, 15 both of whom are characterized by precious garments
and jewels. The position of the head slightly turned down, the
expression of the big almond eyes and the elongated ears call to
mind the face of a Bodhisattva. Statues of Bodhisattvas wearing
crowns can be observed very often from Afghanistan to Xinjiang,
especially in stucco.
16
One of the main elements which determined the choice of the
name of the painting, namely the bow, presents some problems.
In fact, it is reproduced in a realistic way and it is a weapon
used for hunting and the painter did not neglect details such as
the curved shape and the point where the string had to be po-
sitioned. While the arrows sticking in the ground at his side are
14 Tarzi, 1977: pls. B13-B22, B24-B25; Klimburg-Salter, 1989: pl. IV, fi g. 4;
Bivar, 1998.
15 The «Buddha parée» is very common in the region of B¡my¡n: Rowland,
1961.
16 Béguin, 1992: 95-121.

135
the painting of the «hunter king» at kakrak
represented realistically and, close to the feathers, it is possible
to observe the exact point where the string should have been
put. Moreover, the posture of the hands shows great artistic skill
in rendering a sort of reverential attitude to the archer who is
represented offering the bow as a sign of his peaceful intention
towards the Buddha next to him in the painting. The bow cannot
harm anybody in the position in which it is held. What is really
striking in the representation of such a realistic bow is its size:
it seems to be too small to be useful for hunting or fi ghting
and, if compared with the two big arrows, it appears very clearly
that they do not match at all. Why should such a talented artist
make such a mistake? Some scholars think it was not a mistake
and those who recognize it as a royal scene (like U. Jäger) could
deduce that, in this way, the painter was depicting a king who
had been used to kill and now was converted to the Dharma.
However, it could be considered that the bow in that painting
is just a symbolic attribute, a secondary element whose presence
does not affect the power of its owner who should be considered
a divinity as already suggested by Tarzi and Banerjee.
If the painter really intended to reproduce a royal fi gure in
the act of presenting his weapon to the Buddha, why did he
represent the dog as if barking at the Buddha seeming to threa-
ten him with his open mouth and pointed ears? This apparent
incongruity could only be explained by studying other Central
Asian paintings in which animals were depicted in the same way.
Scholars suspect the existence of sketch-books for artists as has
been postulated by Maršak in one of his most recent studies on
Sogdian painting. The striking similarity of battle scenes in the
monumental paintings at Penjikent with some early Sasanian rock
reliefs led the Russian scholar to think that Sogdian painters had
access to sketch-books in which a wide repertoire of useful images
was found.
17 So, if the dog, partially visible behind the curtains
which cover the seat of the «Hunter-King», is represented in
the way that it is, it has no particular signifi cance. The attitude
of the dog could easily has been reproduced by the artist who
copied the model from one of such sketch-books.
The color of the dog seems to be blue like the background
although it is not possible to exclude that it could have been
darker and during the centuries its original pigmentation has
faded away. Unfortunately, the photos published by Bussagli and
17 Marshak, 2002: 67.

annali di ca' foscari, xlvii, 3 (. . ), 2008
136
Rowland are not very clear although on the lower part of the
legs of the dog there are curved lines which possibly belong to
the object behind it. That object seems to be a part of the seat,
probably covered with a yellow material since there are curved
elements which could be folds. The seat presents a very precise
symmetry while the yellow material of the throne appears much
clearer on the right side. It can be discerned better for the ab-
sence of the dog but at that point the painting is quite damaged.
The bad state of preservation could give the illusion of a second
dog which has almost disappeared. However, it does not seem
the case since the traces of a dog would not be in scale with
the dog on the left of the throne. Next to the column capital on
the right there is something painted in brown which is unfortu-
nately impossible to recognize (possibly a portion of the ribbon
attached to the column?). On the left the same space is occupied
by the two arrows. Therefore, the symmetry in the painting is
not strictly respected.
3. Iranian parallels
In the case of a seat or throne supported by two animal-like
beings, there would have been no doubts in identifying the
«Hunter-King» at Kakrak with a Zoroastrian divinity according
to a typical Iranian religious iconography which is well-known in
Sogdian art.
18 Several 8th century paintings from Penjikent, some
ossuaries and terracotta statuettes and literary sources help us to
compile a reliable list of the attributes and animals associated with
Sogdian divinities. In Sanskrit the symbolic animal and vehicle of
an Indian god is called v¡hana and this term is preferred by scho-
lars for Sogdian divinities too. During the 6th century, Sogdians
artists started to abandon their traditional divine iconography in
order to adopt Indian models. This phenomenon was particularly
studied at Penjikent but it can be considered valid also for other
regions of Sogdiana. The reasons for the occurrence of such a
phenomenon are still a matter of debate among scholars.
19 In-
vestigations on this specifi c problem have started too recently to
allow one to compile a detailed Indo-Sogdian pantheon. It should
18 Azarpay, 1975; Škoda, 1980. Sogdian elements have been already noted
at Dokhtar-e Nuširv¡n, not far from B¡my¡n, so, very close to Kakrak: Mode,
1992.
19 Mode, 1991/92; Grenet, 1994; Compareti, forthcoming.

137
the painting of the «hunter king» at kakrak
be also admitted that the religion professed in Sogdiana is only
partially known and most of the Sogdian religious texts that
have been found outside Sogdiana are Buddhist or Manichaean
translations. The few Sogdian divinities who can be now safely
recognized are usually represented sitting on a throne supported
by their allegoric animals partially covered by the usual curtains
or precious material.
We know that the iconography of the throne supported by
allegoric animals was adopted in Kakrak and B¡my¡n as well.
At the site of Dokhtar-e Nuširv¡n (known also as Nig¡r, Af-
ghanistan) there are remains of a painting representing a haloed
divinity under an arch in typical Iranian garments sitting on a
throne supported by horses (?) and donors (smaller in size) on
his right side (fi g. 2). One of the donors is wearing a caftan em-
bellished with pearl roundels, a typical Iranian textile decoration
which became very popular during 6th-7th century.
20 Only few
fragments of the Dokhtar-e Nuširv¡n painting survive and so, in
the last thirty-fi ve years, scholars have
attempted to identify the main fi gure
who is sitting in the central part of
the scene as Verethragna-Bahr¡m, or
Ahura Mazd¡, or Mithra or, most
recently, Zun (possibly, a local form
for Zurv¡n) on iconographical and
historical bases.
21 The study of his
attributes seems to be the key for
a correct identifi cation. I will soon
discuss these attributes and attempt
a different identifi cation. Here it is
worth insisting on the presence in Do-
khtar-e Nuširv¡n of a painting closer
to Sogdian models than Sasanian or
Indian ones.
20 Compareti, 2004: 260-261; Raspopova, 2006: 63, 65.
21 Maeda, 1970; Mode, 1992; Klimburg-Salter, 1993; Grenet, 1995; Sims-
Williams, 1996: 647-648.
Fig. 2. Painting at Dokhtar-e Nuširv¡n (Nig¡r,
Afghanistan).
After: Klimburg-Sal ter, 1993: fig. 5 (the detail
of the «lion head» above the crown was added by
this author).

annali di ca' foscari, xlvii, 3 (. . ), 2008
138
The throne at Kakrak was most likely intended to show only
one dog but it is not possible to exclude the fact that its mo-
del could have been taken from the representation of a local
Zoroastrian god whose seat was supported by two animals as at
Dokhtar-e Nuširv¡n. It is worth observing once more that the
context at Kakrak is defi nitely Buddhist and so many original
elements of the presumed local model could have been lost or
not represented on purpose.
In Sogdian art, the bow and arrow have been recognized as
typical attributes of Tištrya, the Zoroastrian rain god identifi ed
with the star Sirius, who was often represented together with
Nan¡. 22 The divine couple can be observed on a terracotta os-
suary from Kaška Darya (Uzbekistan) on which, unfortunately
there are no symbolic animals but donors. However, the arrow is
clearly discernible in his lower pair of hands
23 (fi g. 3). Another
representation in Sogdian art can be observed in an 8th century
fragmentary painting from Penjikent, sector XXV, room 12 (fi g.
4). The throne of Tištrya in the painting seems to be supported
by winged fantastic creatures which were hypothetically recon-
structed by the excavators on the few details that can be seen.
Also the arrow could not be distinguished properly and its point
22 Grenet, Marshak, 1998: 10-16; Grenet, Marshak in Berdimuradov,
Samibaev, 2001: 58-63. Sirius is part of the constellation of Canis Major which
is said to be one of the dogs of the constellation of the hunter Orion. In
Avestan literature the arrow and, less often, the bow are described as attributes
of Tištrya: Gnoli, 1963: 238-240; Panaino, 1995: 52-54. Tištrya is also considered
to be the general of the eastern part of the sky in Pahlavi texts: Tafazzoli,
2000: 8.
23 The iconography of Tištrya is partially based on Indian models as it is
particularly evident in the Sogdian ossuary from Kaška Darya: Mode, 1991/92:
fi g. 11; Grenet, Marshak, 1998: fi g. 6.
Fig. 3. Terracotta ossuary from Kaška Darya (Uzbekistan). After: Mode, 1991/92:
fig. 11.

139
the painting of the «hunter king» at kakrak
has been added, in fi gure 4, after a reconstruction proposed by
Grenet and Marshak.
24
At Ghulby¡n (Afghanistan), about 200 kilometers north-west
of B¡my¡n, other painted images of Zoroastrian divinities in local
attire have been found and tentatively identifi ed. According to
F. Grenet, the painting at Ghulby¡n should be dated to 4th-5th
century. 25 The sitting god who is larger than the others has been
identifi ed with Tištrya because of the arrow in his right hand and
a pond fi lled with fi shes under his feet, possibly an allusion to the
Vorukaša Sea (fi g. 5). A pair of spread wings with a crescent on
its top appears above his head possibly as part of a crown. This
seems to be a characteristic of local divinities and, in fact, a very
Fig. 4. Painting from Penjikent, sector XXV, room 12. After: Grenet, Marshak,
1998: fig. 12 (detail).
24 Grenet, Marshak, 1998: fi g. 12.
25 Lee, Grenet, 1998: 82.

annali di ca' foscari, xlvii, 3 (. . ), 2008
140
similar element can be noted at
Dokhtar-e Nuširv¡n where, howe-
ver, a pair of ram’s horns and
a lion head (?)
26 appear above
the spread wings (fi g. 2). If the
crescent at Ghulby¡n could be
considered as an astral allusion
to Tištrya, the ram’s horns could
be associated to Khwarnah as it
is suggested in Pahlavi sources.
This divinity has been identifi ed
by some at Ghulby¡n because
of a horned ram head below his
feet. 27 There is no weapon or
characteristic attribute held by Khwarnah at Ghulby¡n although it
could be supposed that a spear or a caduceus was an appropriate
one as it can be observed in Kuš¡n coinage.
28 Only part of the
animals that are supporting the throne of the god at Dokhtar-e
Nuširv¡n are still visible and their heads too have disappeared.
So it is not excluded that they could have been rams and not
horses. At least two isolated fi gures of ibexes and a pair of ram’s
horns appear on the left of the god at Dokhtar-e Nuširv¡n but
it is not possible to hypothesize because of the fragmentary state
of the painting.
29 It is not even sure that the ram should be the
appropriate v¡hana for Khwarnah (called in Sogdian Farn) and, in
fact, Grenet proposed to attribute to him a throne supported by
Fig. 5. Painting from Ghulby¡n (Af gha-
ni stan). After: Lee, Grenet, 1998: fig. 1
(detail).
26 The lion or monstrous head between the ram’s horns and the spread wings
are not always represented in the reconstruction of the painting at Dokhtar-e
Nuširv¡n which were proposed by different scholars in the past. It was one
of Grenet’s main argument in order to associate that god with Mithra whose
halo is embellished with allegories of planets: Grenet, 1995: 116-118. If the heads
of the animals around the halo could be really identifi ed as allegories for the
planets, then they could be attributed to any great god locally venerated, not
only Mithra.
27 Lee, Grenet, 1998, 83.
28 Carter, 1986.
29 Klimburg-Salter, 1993: fi g. 362. Grenet thinks that the fi gures there should
be considered donors: Grenet, 1995: 108.

141
the painting of the «hunter king» at kakrak
camels or winged camels in Sogdian art.
30 However, to Marshak
the camel is the animal of Verethragna-Bahr¡m. 31 At present it
is not possible to be more specifi c. In the Suishu (Chronicle of
the Sui Dynasty, composed in 636) it is clearly stated that only
the king of Bukhara had a throne in the shape of a camel but
no god is mentioned at all.
32
The painting of Tištrya at Ghulby¡n seems to be based on
another tradition, completely different than the fi gure of the god
on the Kaška Darya ossuary and in 8th century Penjikent mural
paintings but which is quite close to the divinities painted at Jar
Tepa (Uzbekistan), another Sogdian site possibly to be dated to
4th-5th century. The sitting couple larger than the other people
close to them represented in the fragmentary paintings at Jar Tepa
have been identifi ed by B. Marshak and F. Grenet as Nan¡ and
Tištrya. Their association with hunting in Avestan literature could
explain their presence at Ghulby¡n and Jar Tepa where archers
shooting at animals have a prominent position.
33 It seems that
the divine couple at Jar Tepa is sitting on a throne supported by
lions, the usual attribute of Nan¡, who should be identifi ed with
the fi gure wearing long garments on the right. The god sitting
on the same throne with a curved object (possibly a bow?) next
to her could then be identifi ed with Tištrya (fi g. 6). Both divine
and animal heads are missing although the lion on the right next
to Nan¡ seems to be higher than the animal on the left next to
Tištrya. If they were two different animals, then the animal next
30 Grenet, 1995: 108. Also Tanabe supports Grenet’s idea: Tanabe, 1982.
31 Marshak, Raspopova, 1990: 137-144.
32 Lin, 2006: 398. See also: Silvi Antonini, 2006.
33 Grenet, Marshak, 1998: 14-15; Grenet, Marshak in: Berdimuradov,
Samibaev, 2001: 58-63. The female fi gure at Ghulbyan, despite her proximity
to Tištrya and her sitting position, has been identifi ed as a donor, not as a
divinity: Lee, Grenet, 1998: 63.
Fig. 6. Painting from Jar Tepa (Uz be-
kistan). After: Berdimuradov, Sami-
baev, 2001: fig. 5 (detail).

annali di ca' foscari, xlvii, 3 (. . ), 2008
142
to Tištrya could be a dog which is one of the most important
animals for Zoroastrians. Unfortunately, the bad condition of the
paintings does not allow one to be precise. Divine couples sitting
on the same throne, supported by different animals (camel and
ram) are known in later Sogdian paintings. They are all found
at Penjikent and should be dated to 8th century.
34
The association of the dog as a possible v¡hana of Tištrya is
highly speculative.
35 The only precise association between Tištrya
and the dog as his allegoric animal has been proposed by F.
Grenet and G.-J. Pinault in a long article on a painted scroll on
paper representing the decans of the Indian and Iranian zodiac
and other divinities.
36 Some words in Tokharian appear as well
in the scroll allowing one to give a precise interpretation of the
subject and localize its origin in the region of Turfan around 8th-
9th century. Most likely, that painted scroll was a kind of «survey
of Western astrology» prepared for local astrologers or members
of the royal family who used the Chinese zodiac of the twelve
animals and were probably curious about other systems.
37 In
that painted scroll, Indian and Iranian astronomical-astrological
allegories appear intermingled in a complicated way which have
parallels in western literature of the Middle Ages mentioned in
the brilliant commentary by Grenet and Pinault. Some of the di-
vinities in the scroll appear together with their symbolic animals.
In this case too the painting is very fragmentary and just few
details of the image identifi ed with Tištrya and the dog have been
preserved. The divinity is represented in Indian attire wearing a
dhoti and with a halo behind his head but he is holding neither
arrow nor bow. The dog is clearly represented behind the divinity:
its elongated snout, long ears and curly tail point to a hound
(fi g. 7). The absence of the bow and arrow could be due to an
artist unfamiliar with the subjects that he was reproducing. In any
34 Belenitskii, Marshak, 1981: fi g. 7-8; Marshak, Raspopova, 1990: fi g. 16-17.
35 In Avestan literature, Den (a personifi cation of the Zoroastrian religion)
is said to be accompanied by two dogs but she is always described as a lady.
The most interesting image of Den in association with the dog was recognized
in a fragmentary 9th-10th century painting on paper from Dunhuang to be pos-
sibly attributed to Sogdian immigrants who settled there: Grenet, Zhang, 1996.
These ideas have been criticized in two recent studies: Silvi Antonini, 2007;
Lo Muzio, 2006 [2009]: 199.
36 The scroll is known as MIK III 520 and it is kept in the Staatliche Mu-
seen, Preussischer Kulturbesitz, Museum für Indische Kunst, Berlin: Grenet,
Pinault, 1997: 1057-1058, fi g. 1.
37 Grenet, Pinault, 1997: 1022-1027.

143
the painting of the «hunter king» at kakrak
case, bow, arrow and dog represented singularly or together are
also the characteristic attributes of the personifi cation of planet
Mercury whose tutelary divinity, in Iranian astrology, is Tir, that
is to say Tištrya.
38
As it has been observed above, those divinities who belonged
to one cultural milieu could have been represented in local dress
in another. In some cases it is very clear that the adoption of a
foreign iconography for a local god is determined by its common
attributes. This is a normal phenomenon particularly true for Ira-
nian divinities (interpretatio iranica) although it is still unknown
why one iconographical system was preferred instead of another. 39
The Parthians and the Bactrians adopted the Greek iconography
for some of their divinities and the Sogdians used that of India.
There is then to consider the receptiveness of the Central Asian
people to the Mesopotamian heritage of their divinities as is
shown in the case of Nan¡ in Kuš¡n and Sogdian art. All this is
particularly true in the sphere of astronomy and astrology where
the base was clearly Hellenistic-Mesopotamian with very important
Fig. 7. Painted scroll MIK III 520, Museum für Indische
Kunst, Berlin. Sketch after: Grenet, Pinault, 1997: fig.
1 (sketch).
38 Gnoli, 1963; Grenet, Pinault, 1997: 1057-1059; Grenet, Marshak, 1998:
12. There is some confusion between Tir and Tištrya in Zoroastrian literature.
The two names can point to the same divinity or two different (and opposed)
ones: Malandra, 1983: 142-143.
39 The problem is discussed extensively in: Compareti, (forthcoming).
Classical authors often called foreign divinities with Greek (or Latin) names
(interpretatio graeca). The choice of the names was determined in many cases
because of characteristics that those divinities had in common. The most famous
example is given by information on Artemis among the Achaemenids, possibly
to mean Anahita, as described by Herodotus: de Jong, 1997: 268-284. According
to the interpretatio graeca, Tištrya was associated to Hermes or Apollo, and to
Nabu according to the interpretatio mesopotamica: Gnoli, 1963: 242; de Jong,
1997: 297-299. According to ancient Mesopotamian tradition, Nan¡ was con-
nected with Tašmetu, Nabu’s wife, so making clear why the Sogdians adopted
the same association: Azarpay, 1975: 25; Potts, 2001: 24. Curiously, in Kuš¡n
coins Tir-Tištrya is represented according to the Greek iconography of Artemis
and, so, with bow and arrows: Gnoli, 1963: 240.

annali di ca' foscari, xlvii, 3 (. . ), 2008
144
Egyptian elements. As several studies seem to confi rm, the Iranians
were extremely interested in astronomical-astrological matters and
the diffusion of «Western» concepts into India, 40 China and other
regions 41 should most likely be attributed to them. Grenet and
Maršak supposed that the way in which Tištrya is holding the
arrow in Sogdian art is the same as that of Mercury in India
(where he was called Budha).
42
4. Conclusion
Several details seem to point to an identifi cation of the so-called
«Hunter-King» at Kakrak with a divinity and not a sovereign al-
though the crown suggested the contrary to Bussagli and others.
Not only the halo but other attributes such as the bow with
arrows and the dog represented supporting the throne, would
confi rm this new hypothesis. However it is diffi cult to decide if
that god should be identifi ed with Tištrya since the dog is not
reproduced symmetrically on both sides of the throne.
As already observed, Kakrak is a Buddhist site. So it is easier
to assume that he is a personifi cation of the planet Mercury. He
is a curious mixture of iconographical elements: the head is that
of a Bodhisattva wearing the triple-crescent crown but the throne
with the dog is possibly taken from a local image of Tištrya. His
sitting position reminds one of an Indian divinity but it was a
typical characteristic of Sogdian religious iconography too.
43
Furthermore, as the paintings at Kakrak are very fragmentary
it could be supposed that other images of the personifi cation of
planets existed next to representations of Buddhas exactly as at
B¡my¡n where a well-known fi gure of the Sun God was painted
above the head of the eastern colossal Buddha. Grenet always
identifi ed him as Mithra44, although it has been recently argued
about an identifi cation of the «Sun God» at B¡my¡n with a
princely fi gure. 45
40 Pingree, 1963; Pingree, 1965; Pingree, 1973; Markel, 1990; Panaino, 1992.
41 Howard, 1983. The astrological text known in Sanskrit as ≥¡rd∞la-
karn¡vad¡na was partially translated into Chinese by the Parthian Buddhist
missionary An Shigao during the 2nd-3rd century AD: Pingree, 1963; Pingree,
1987: 858-859.
42 Grenet, Marshak in Berdimuradov, Samibaev, 2001: 59.
43 In Buddhist art this is a typical attitude of the so-called «pensive Bod-
hisattva»: Juliano, 2006: 32-33.
44 Grenet, 1993; Grenet, 2001.
45 Tanabe, 2005.

145
the painting of the «hunter king» at kakrak
Scholars are still discussing such paintings and it seems that
a solution to their identifi cation does not exist yet. Since all
the hypothetical arguments exposed here should be considered
within the cultural milieu of Iranian Buddhism, then the fi gures
displaying important links with astronomy and astrology that were
represented according to the iconography of local divinities could
have been used in a different context. The reading of those pain-
tings is extremely complicated. For this reason, at the beginning
of the present paper it was said that the study of the Kakrak
paintings should be done very cautiously.
In this way, U. Jäger’s idea about a symbolic representation
of «converted royalty» could still represent a valid alternative
reading although the iconography of the so-called «Hunter-King»
should be regarded as a mixture of elements belonging to a local
divinity (possibly Zoroastrian) with other elements taken from the
nomadic world. The same mixture of elements could be consi-
dered correct in case of an identifi cation of the «Hunter-King»
as a planetary divinity (Mercury) presenting his weapons to the
Buddha as a sign of renunciation.
Bibliography
G. Azarpay, «Iranian Divinities in Sogdian Painting», in Acta Iranica.
Monumentum H.S. Nyberg, I, Leyden, Téhéran, Liège, 1975, 19-29.
G. Béguin, L’Inde et le monde indianisé au Musée National des Arts
Asiatiques-Guimet, Paris, 1992.
A.M. Belenitskii, B.I. Marshak, «The Paintings of Sogdiana», in G.
Azarpay, Sogdian Painting. The Pictorial Epic in Oriental Art, Berke-
ley, Los Angeles and London, 1981, 11-77.
A.E. Berdimuradov, M.K. Samibaev, «Une nouvelle peinture murale
sogdienne dans le temple de Džartepa II. Avec notes additionnelles
par F. Grenet et B. Marshak», in Studia Iranica, 30, 1, 2001, 45-66.
A.D.H. Bivar, «Sasanian Princes at Bamiyan», in V.S. Curtis, R. Hil-
lenbrand, J.M. Rogers (eds.), The Art and Archaeology of Ancient
Persia. New Light on the Parthian and Sasanian Empires, London,
New York, 1998, 103-110.
M. Bussagli, La peinture de l’Asie Centrale, Genève, 1963.
M.L.Carter, «Trifunctional Pharro», in Studia Iranica, 15, 1, 1986, 89-
98.

annali di ca' foscari, xlvii, 3 (. . ), 2008
146
—, «Revanta, an Indian Cavalier God», Annali dell’Istituto Orientale di
Napoli, 48, 2, 1988, 127-137.
M. Compareti, «The Sasanian and the Sogdian “Pearl Roundel” Design:
Remarks on an Iranian Decorative Pattern», in The Study of Art
History, 6, 2004, 259-272.
— «The Indian Iconography of the Sogdian Divinities: the Archaeologi-
cal and the Textual Evidence», in: Proceedings of the 5th International
Congress on the Archaeology of the Ancient Near East, Madrid, April
3rd-8th 2006, forthcoming.
A. de Jong, Traditions of the Magi. Zoroastrianism in Greek and Latin
Literature, Leiden, New York, Köln, 1997.
R. Ghirshman, Art persan. Parthes et Sassanides, Paris, 1962.
G. Gnoli, «La stella Sirio e l’infl uenza dell’astrologia caldea nell’Iran
antico», in Studi e materiali di storia delle religioni, XXXIV, 1, 1963,
237-245.
F. G renet, «B¡miy¡n and the Mihr Yašt», in Bulletin of the Asia In-
stitute, 7, 1993, 87-94.
—, «The Second of Three Encounters between Zoroastrianism and
Hinduism: Plastic Infl uences in Bactria and Sogdiana (2nd-8th century
A.D.)», in Journal of the Asiatic Society of Bombay. James Darmesteter
(1849–1894) Commemoration Volume, 69, 1994, 41-57.
—, «Mithra at les planètes dans l’Hindukush central: essai d’interprétation
de la peinture de Dokhtar-i Nôshirvân», in Au Carrefour des reli-
gions. Mélanges offerts à Philippe Gignoux. Res orientales, vol. VII,
Bures-sur-Ivette, 1995, 105-119.
—,
«Mithra, dieu iranien : nouvelles données», in TOPOI, 11/1, 2001, 35-58.
—, B.I. Marshak, «Le mythe de Nana dans l’art de la Sogdiane», in
Arts Asiatiques, 53, 1998, 5-18.
—,
G.-J. Pinault, «Contacts de traditions astrologiques de l’Inde et de
l’Iran d’après une peinture des collections de Turfan», in Comptes Ren-
dus de l’Académie des Inscriptions et Belles Lettres, 1997, 1003-1063.
—, Zhang Guangda, «The Last Refuge of the Sogdian Religion: Dun-
huang in the Ninth and Tenth Centuries», in Bulletin of the Asia
Institute, 10, 1996, 175-186.
J. Hackin, J. Carl, «Nouvelles recerches archéologiques à Bamiyan»,
Mémoires de la Délégation Archéologique Française en Afghanistan,
3, 2 volls., Paris, 1933.
M. Hallade, «Indo-Iraniche, correnti», in Enciclopedia Universale
dell’Arte, volume VII, Firenze, 1972, 412-428.
A. Howard, «Some Evidence, Mainly Textual, in Chinese Esoteric Bud-
dhism», in Asiatische Studien, XXXVII, 1, 1983, 104-119.
U. Jäger, «Buddhistische Ikonographie und nomadische Herrscher-
rpräsentation. Zum sogenannten “Jäger-König” von Kakrak bei Bami-
yan/Afghanistan», in H.G. Kippenberge et al., (eds.),Visible Religion.
The Image in Writing, vol. VI, Leiden, 1988, 191-201.

147
the painting of the «hunter king» at kakrak
A.L. Juliano, «Converging Traditions in the Imagery of Yu Hong’s
Sarcophagus: Possible Buddhist Sources», in Journal of Inner Asian
Art and Archaeology, I, 2006, 29-49.
E. Kageyama, «The Winged Crown and the Triple-Crescent Crown in
the Sogdian Funerary Monuments from China: Their Relation to the
Hephtalite Occupation of Central Asia», Journal of the Inner Asian
Art and Archaeology, 2, 2008, 9-20.
D. Klimburg-Salter, The Kingdom of B¡my¡n. Buddhist Art and Culture
of the Hindu Kush, Naples-Rome, 1989.
—, «Dokhtar-i-Noshirwan (Nigar) Reconsidered», Eassays in Honor of
Oleg Grabar. Muqarnas, X, 1993, 355-368.
—, «B¡miy¡n: An Obituary and a Glance Towards the Future», Oriental
Art, XLIX, 1, 2003, 2-12.
J. Lee, F. Grenet, «New Light on Sassanian Paintings at Ghulbiyan,
Faryab Province, Afghanistan», South Asian Studies, 14, 1998, 75-85.
Lin Ying, «Sogdians and the Imitations of Byzantine Gold Coins Un-
earthed from the Heartland of China», in M. Compareti, P. Raffetta,
G. Scarcia (eds.), ¶r¡n ud An™r¡n. Studies Presented to B.I. Maršak
on the Occasion of His 70th Birthday, Venezia, 2006, 387-401.
C. Lo Muzio, «Culti brahmanici a Khotan: note sulle pitture del tempio
D13 a Dandan Oiliq», in Rivista degli Studi Orientali, LXXIX, 1-4,
2006 [2009], 185-201.
K. Maeda, «Notes on the Fragment of the Wall Paintings of Doktar-i
Noshirwan», Journal of the Toyota Technical College, 3, 1970, 95-106
(in Japanese with an English summary).
W.W. M alandra, An Introduction to Ancient Iranian Religion. Readings
from the Avesta and the Achaemenid Inscriptions, Minneapolis, 1983.
S. Markel, «The Genesis of the Indian Planetary Divinities», East and
West, 41, 1990, 173-188.
B.I. Marshak, Legends, Tales, and Fables in the Art of Sogdiana, New
York, 2002.
—, V.I. Raspopova, «Wall Paintings from a House with a Granary.
Panjikent, 1st Quarter of the Eighth Century A.D.», Silk Road Art
and Archaeology, 1, 1990, 123-176.
M. Mode, «Sogdian Gods in Exile - Some Iconographic Evidence from
Khotan in the Light of Recently Excavated Material from Sogdiana»,
Silk Road Art and Archaeology, 2, 1991-92, 179-214.
—, «The Great God of Dokhtar-e Noshirw¡n (Nig¡r)», in East and
West, 42, 1992: 473-483.
A. Panaino, «La diffusione dell’astronomia e dell’astrologia mesopo-
tamica in India attraverso la mediazione iranica», in L’astrologia e la
sua infl uenza nella fi losofi a, nella letteratura e nell’arte dall’età classica
al Rinascimento, Milano, 1992, 9-50.
—, Tištrya. Part II. The Iranian Myth of the Star Sirius, Roma, 1995.
D. Pingree, «Astronomy and Astrology in India and Iran», in Isis, 54,

annali di ca' foscari, xlvii, 3 (. . ), 2008
148
1963, 241-243.
—, «Representation of the Planets in Hindu Astrology», in Indo-Iranian
Journal, 8, 1965, 250-267.
—, «The Mesopotamian origin of the Early Indian Mathematical As-
tronomy», in Journal for the History of Astronomy, 4, 1973, 1-12.
—, «Astrology and Astronomy .i. History of Astronomy in Iran», in E.
Yarshater (ed.), Encyclopædia Iranica, vol. II, London-New York,
1987, 858-862.
D.T. Potts, «Nana in Bactria», in Silk Road Art and Archaeology, VII,
2001, 23-35.
B.N. Puri, Buddhism in Central Asia, New Delhi, 1987.
V.I. Raspopova, «Textiles Represented in Sogdian Murals», in R. Schorta
(ed.), Central Asian Textiles and Their Contexts in the Early Middle
Ages, Riggisberger Berichte, 9, 2006, 61-73.
B. Rowland Jr., «The Bejewelled Buddha in Afghanistan», in Artibus
Asiae, XXIV, 1961, 20-24.
—, «Art Along the Silk Road: A Reappraisal of Central Asian Art.
Review of: Painting of Central Asia by Mario Bussagli», in Harvard
Journal of Asiatic Studies, 25, 1964-1965, 248-264.
—, The Art of Central Asia, New York, 1974.
K. Silvi Antonini, «Ob odnom siuåete v åivopisi dvorca Varahši», in C.
Silvi Antonini, D.K. Mirzaahmedov (eds.), Ancient and Mediaeval
Culture of the Bukhara Oasis, Samarkand-Rome, 2006: 50-54.
—, «Breve nota su un manoscritto Pelliot da Dunhuang», in Rivista
degli Studi Orientali, LXXVIII, 1-4, 2007, 495-500.
N. Sims-Williams, «Nouveaux documents sur l’histoire et la langue de
la Bactriane», Comptes Rendus de l’Académie des Inscriptions et Belles
Lettres, 1996, 633-654.
V.G. Škoda, «K voprosu o kul’tovnyh scenah v sogdijskoj åivopisi»,
Soobšπenija Gosudarstvennogo Ermitaåa, XLV, 1980, 60-63.
A. Tafazzoli, Sasanian Society: Warriors, Scribes, Dehq¡ns, New York,
2000.
T. Talbot Rice, Ancient Arts of Central Asia, London, 1965.
K. Tanabe, «Winged Camelus Bactrianus in Sogdian Art», Oriento, XXV,
1, 1982, 50-72 (in Japanese with an English summary).
—, «An identifi cation of the So-Called S∞rya Image at B¡miy¡n», in
Journal of the West Asian Archaeology, 3, 2005, 15-28 (in Japanese
with an English summary).
Z. Tarzi, L’architecture et le décor rupestre des grottes de B¡miy¡n, 2
volls., Paris, 1977.
—, «B¡my¡n .ii. History and Monuments», in E. Yarshater (ed.), En-
cyclopaedia Iranica, vol. III, London, New York, 1989, 658-660.
F. T issot, Catalogue of the National Museum of Afghanistan 1931-1985,
Paris, 2006.

149
the painting of the «hunter king» at kakrak
ABSTRACT
One of the paintings from the 7th-century Buddhist site of Kakrak
(region of B¡my¡n, Afghanistan) portrays a nimbed hunter holding a
bow and sitting on a throne. A dog is placed in the lower part of the
throne and two white animals partially show in the higher part of the
same seat. His attributes and the triple-crescent crown intimated a royal
fi gure fond of hunting: which explains the epithet of «Hunter-King»
used in the past. Among the most recent studies, U. Jäger proposed
to interpret the painting as the symbolic portrait of a nobleman con-
verted to Buddhism, in whose iconography local and nomadic royal
elements are mixed. His interpretation might be correct; but the local
iconographical elements are, in all likelihood, features taken from the
fi gure of Tištrya, the Zoroastrian rain god who was confused with the
planet Mercury. Another possibility is that the so-called «Hunter-King»
is actually a personifi cation of the planet Mercury in the act of submit-
ting to the Buddha close to him in the same painting.
KEYWORDS
Kakrak. Hunter-King. Tištrya. Mercury.