In: Myth, Image, Metaphor in the Cultures of Asia and Africa, Proceedings of the 6th AllPolish Conference of Oriental Studies. Edited by Marek M. Dziekan, Sylwia Filipowska and
Ewa Siemieniec-Gołaś. Warsaw: Elipsa, 2121. 301–373.
RAJESH KUMAR SINGH
Ex-Senior Fellow, Ministry of Culture, Govt. of India
ORCID: 0000-0003-4309-4943
CIRCA 465 CE
AND THE STUPA-SHRINE OF THE AJANTA CAVE 11
A
: This essay presents a snapshot of the year 465 . Place: South Asia
with the Ajanta Cave 11 in focus. A history is narrated here to describe how the
stupa-shrine (naos) of the Ajanta Cave 11 was conceived, and added to what was
earlier a dormitory; how the Hunnic disturbance in Eurasia created migrations of
people including the one when the Bactrian and Gandhāran people migrated towards
Middle India and the Deccan on one hand and the Tarim / Taklamakan region on
the other. The war refugees, I argue, introduced the concept of multi-functional or
composite architecture in India. It was to bring under the same roof the three major
functions of monastic architecture: dwelling, congregation, and worship. It was an
experiment that resulted in the creation of Cave 11’s naos. It was a quadrant and
flat-roofed naos. The idea was at once accepted across all the contemporary rockcut sites. The article shows how the displaced people the “foreigners” from Eastern
Mediterranean, ancient Turkey, Iran, Bactria and Gandhāra, swiftly changed the
Buddhism of Middle India and the Deccan from an earlier form of Śrāvakayāna
that insisted on the stupa cult to a different one that insisted on the cultic worship
of the Buddha and Bodhisattva figures. The latter was a confluence of the traditional
Sarvāstivāda and Bodhisattvayāna, which acquired the name of Mūlasarvāstivāda
towards the seventh century CE.
K
: Ajanta Cave 11, Buddhist naos, Buddhist rock-cut architecture, Buddhist
archaeology, Indian Buddhism, cetiyaghara, stupa, Mulasarvastivada, Alchon Huns,
Gandhara
302
Rajesh Kumar Singh
I
*
The Ajanta1 Cave 11 (Plates 16.3–4; 23.1; 28–29)2 is a relatively neglected
edifice in the Ajanta studies as compared to the painted cave temples.3 In 1880,
James Fergusson and James Burgess described the edifice in about one and
a half page.4 Then in 1883, Burgess devoted about half a page more, and also
published the ground plan.5 In both the reports, the authors were at pains to
date the monument because of the coexistence of some very early and some
very late features. About a century later, Walter M. Spink analyzed the edifice
most thoroughly. He revealed tomes of new data and facts.6 Because the cave’s
development was so complicated, the description of the developments is also
bound to be complicated. Therefore, a bird’s eye view of the timeline may be
had by looking at Spink’s time chart of Ajanta, and looking for the entry on
Cave 11 there (Pl. 15). Very roughly, it would convey the following gist:
The Ajanta Cave 11 developed during c. 462–478 . The work was started
vigorously in c. 462 . There was less vigorous excavation during late 463–
467 . Then, during 468 and early 469 the work was vigorous again before being
discontinued from the middle of c. 469 . No excavation happened afterwards.
The inner shrines or naoi7 of Ajanta and related sites were all conceived during
466 . However, the development of the Cave 11’s naos started in mid-468
and discontinued from mid-469 . About a decade later the work in the naos
was resumed during early 478 . After Hari Sena’s death in c. 477
there was
Conventions: (1) All the dates are expressed in circa and Christian Era unless specified otherwise.
(2) The spellings and diacritical marks for historical persons, places, dynasties, tribes, or clans are after
SB (431–437), SBC (431–437), HG (669–700), and Archaeological Survey of India (1964, II: 357–379)
in this order of priority. (3) Where inconsistency is found this study supersedes the author’s earlier works.
1 Ajanta caves: 20°33’ N, 75°42’ E.
2 Pls. 1–13 are after those in S18, S20a, S20b, and S20c (143–153). Notations have been removed
due to want of space. Revised notations would be found in Singh (Forthc., ‚description of plates’).
3 The Ajanta Caves 1, 2, 9, 10, 16, and 17 have received more attention due to the paintings,
although some other caves also have extant paintings. For a comprehensive corpus of the narrative
paintings, vide Schlingloff (1999) and (2013); and of the non-narrative paintings, vide Zin (2003a) and
(2003b).
4 F80 (294–295).
5 B83 (47, pls. XXVIII.2, XXXVII.1).
6 Spink (1968), W07 (141–161, fig. 14), W09 (36–37, pl. 90, fig. 14), W14 (pls. 1–3, 44, 83, 122,
230, 258, 259, 265–266), W17 (144, 244–248, 274–275, 339, 375–377).
7 Other synonyms may be: in English: sanctum sanctorum; in Pali, Prakrit, and Apabhramśa: cetiyaghara; in Sanskrit: caityagṛha, stūpamandiram, or garbhagṛha, etc. I will often use the word naos and
pronaos (shrine-antechamber) ahead for brevity.
*
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
303
“anxious consolidation” during 478
when “major patrons started to leave” the
sites. Thereafter, two years (479–480 ) are allocated for the “Period of Disruption”
when the so-called “intrusive imagery” were painted or carved in blank areas of
the caves that were living or had been consecrated.
It was a path breaking study that was full of new data, evidence, and logically
derived interpretations of whatever data that was available to the scholar.
Thereafter in 2014, another attempt was made by a doctoral student to study the
cave again from a chronological perspective.8
All the above studies suffer from a serious problem. They were all done from
inside the box, so to speak. The researchers did not look outside the box, i.e. outside
the region(s) of the site(s) concerned. They did not look beyond the present or past
borders of the sites, regions, and geographies. They did not evaluate the Hunnic
factor. The whole story changes when the Huns are placed into the picture (Pl. 14).
Actually they must be placed into the picture because it answers all the questions
that were evading answers so far. Every piece of a great jigsaw puzzle (fifthcentury Middle Asia, South Asia, and Central Asia) seems now to fit into place so
remarkably well, especially in the light of certain latest studies.9
Therefore, we are re-studying the monument here to tell a new story of how
the cave, particularly the naos, was developed. The account presented here is
substantially different from the former studies, for it is focussed on the inaugural
circumstances of the edifice as well as the naos (Pl. 28.6). We intend to describe
how the sanctum sanctorum was conceived; how the excavation was initiated;
what was the motive, plan, and inspiration; what problems were encountered; how
the problems were resolved; what changes were made to the original plans; what
failures were encountered; what milestones were achieved; how the developments
catalyzed the architectonics of the times; how various experiments were conducted;
how they influenced the later developments of the Buddhist rock-cut temples;
how it laid the foundations of what is known today as the garbhagṛha10 (sanctum
sanctorum) of the Indian temple architecture; how it became the centre, the pivot,
the axis mundi, the raison d’etre of a sacred space wherein resides the divinity.
Performing a cardinal function inside an architectural space, the sanctum
sanctorum dictates everything else. The story in our Ajanta Cave 11 – and indeed in
Singh (2014, I: ch. XVII, 309–329, 404–406; II: figs. 25, 95–120).
S20a, S20b, S20c, and S21.
10 Sanskrit architectural term for the sanctum sanctorum of the Indian temples. Literally, garbha
means the womb, and ghaṛa means a chamber, abode, house, home, residence, edifice, etc. For a discussion, vide Singh (2014, I: 38–41).
8
9
304
Rajesh Kumar Singh
every other rock-cut temple of the times – was no different. In fact, the story began
from there. What is that story? We are going to present here a new narration with
some new facts, factors, and revelations. In the process, we shall limit ourselves to
the early years of the cave’s development to describe how things were conceived
and inaugurated. Then, we shall focus on the year 465
when the first naos was
conceived and excavation began. So much of knowledge is still waiting to be
revealed about those times. Hence, the curtains will be raised here from at least
some of the secrets of history so that our subject is lit in a new light, and give us
some bits of unprecedented insights into what exactly happened in that year in
South Asia, and why it should matter to the researcher.
C
T
460–
C
’ L
465
:O
E
Spink made a remarkable suggestion that the Ajanta Cave 11 as well as its naos
(Pl. 28.6–7) were among the first such initiatives11 during Period II (c. 460–
468 ) of the Indian rock-cut architecture (Pls. 1.4; 3.4).12 The edifice belonged
to an architectural typology called the Type LM2bii (Pl. 4),13 which was initially
programmed for about fifteen edifices at Ajanta (Pl. 6: figs. 10, 15–19, 21–24,
26–32),14 two at Bagh15 (Pl. 6.11–12), one at Ghaṭotkaca (Pl. 6.25),16 three
at Dharashiva17 (Pls. 6.13; 19.1–3), some at Kondavite18 (Pl. 19.8), one in
S07 (151–152), S14 (54).
S20b (14–20).
13 The classification chart given in the Pl. 4 has been further revised after S18 (fig. 3), S20a (pl. 3),
S20b (pl. 3), S20c (pl. 3). Even the acronyms and nomenclature of architectural terms of the typologies
have been revised. The new nomenclatures are after the first initial(s) of texts or inscriptions in Pali,
Prakrit, or Sanskrit. For example, ‘L’ in the name of a typology stands for leṇa / layaṇa (rock-cut dwelling), ‘M’ for maṭapa / maṇḍapa (hall or pavilion), ‘C’ for cetiyaghara (literally, house of worship), etc.
14 S12b (45–46).
15 Bagh caves: 22°22’ N, 77°78’ E.
16 The Ghaṭotkaca caves are situated near Janjālā village, Soegāon taluka, Aurangabad district,
Maharashtra state, India. The site is about 14 km west of Ajanta as the crow flies, but about 47 km by
car. There are about three caves out of which the largest one is considered the most significant.
17 Osmanabad district, Maharashtra, India. The site has 7 caves.
18 Kondavite or ‘Mahākālī caves’: 19.130390°, 72.873765°. There are 19 caves as per the current
ASI records, but ‘about 20 excavations’ as per Nagaraju (1981, 237), and 16 caves as per F80 (360) and
B83a (71, pls. XLII–XLIII).
11
12
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
305
Lamgaon19 (Pl. 19.7), and a later one (?) in Harishchandragad.20 The plan Type
LM2bii was chosen to perpetuate the type of monastic dormitories that was
prevalent through the preceding Period I, c. 250
– c. 325
(Pls. 1.1; 3.1).21
22
The same type was now revived during c. 460
for drafting the blueprint of
23
the first batch of the caves of Period II. Most of the edifices were inaugurated
during c. 460–462 ,24 some during c. 463–466 ,25 Cave 29 during 469 ,
and the rest26 towards the end of Period III,27 c. 473–477
(Pls. 1.6; 3.6; 15).
At the initial planning stage (460 ), some edifices were planned large,
some of medium size, and others were smaller. However, regardless of the size,
the layout was common.28 The same layout with advanced variations was being
followed in some caves of Bagh (Pls. 6.11–12; 18.1–2), Ghaṭotkaca (Pls. 6.25;
8.5), Dharashiva caves (Pls. 6.13; 19.1–3), Ankai-Tankai (Pl. 19.4–6), Lamgaon
(Pl. 19.7), at least one example at Pālé29 (Pl. 6.14), and one at Kondavite (Pl. 19.8).
Situated in Goa, the site has two rock-cut monuments: one Buddhist and another Śaivite. In my
opinion, the former is older and the latter is later. The former may be called Cave 1 and the later may
be called Cave 2.
20 Ahmednagar district, Maharashtra. The site’s Kedareshwar Cave is our interest here.
21 For an outline of Period I, vide S20b (10–11).
22 The inaugural year has been variously dated in current research: (A) c. 460
: S20c (123), S21
(43–45); (B) c. 462
(Pl. 15): S20a (26), S20b (14–16); and (C) c. 461 : Singh (I: 134–135, 193,
259–260, 393, 396–397).
23 If the Ajanta Cave 7 (Pls. 6.23; 8.3; 16.2) exudes a different layout it was not originally so. It was
also planned as the Type LM2bii, which was later converted into the Type LC2 (Pls. 4; 16.1–2). Spink
(1985, fig. 22), (2017, 187) demonstrated that the cave also had a hall in the original plan, which was later
omitted due to unknown reason. The change of plan is datable to c. 466 after the duo-functional layout
of the Banoṭī cave (20.4430900° N, 075.3124130° E), Phase I (Pls. 7.11; 10.4; 12.5; 30.1). The duofunctional layout must have been after the similar types in Aparānta that have no hall (Pls. 10.2–3; 7–9).
24 Ajanta Caves 4, 5, Lower 6, 7, 8, 11, 15, 16, 17, 18, 19, 20, 25, 26, and 27; cf. c. 462–463
by Spink (Pl. 15).
25 Ajanta caves 1, 2, Upper 6, 21, 23, 24, Ghaṭotkaca, and Banoṭī cave; cf. Spink’s timeline (Pl. 15).
26 Caves 3, 14, 22, and 28 (Pl. 15).
27 For an outline of Period III, vide S20b (24–30).
28 It was something like the traditional dharmaśālās (lodgings / inns) of today that are seen at pilgrimage centres in South Asia. Older dharmaśālās followed a catuśālā (rooms on four sides of an inner
court) kind of design with a frontcourt, porch, main gate, inner courtyard, and cells all around. They
could be multi-storied.
29 Situated near Pālé in Mahāḍ taluka of Raigad district, Maharashtra (Pl. 18.4). Fergusson and
Burgess (1880, 209) called it the ‘Mhār caves near Pala.’ Then, Burgess (1883a, vii, 18, pl. IX) called
it ‘Pala caves near Mahāḍ or Mhār.’ However, for the people of the region the site is called ‘Gandhāra
Pālé.’ As to why the nearby Gandhari River is called as such may be explored in view of a Gandhāran
episode that is revealed in this essay. For the ASI the site is simply ‘Pala’ or ‘Pale’ – without diacritical
marks. I am following the form ‘Pālé’ in this essay – with the diacritical marks. The sites has 28 caves,
vide F80 (209), B83a (18–19), and N81 (250).
19
306
Rajesh Kumar Singh
The reason of reviving an older layout from Period I was the community-style
dwelling needs. There was no need of lavish decorations, carvings, or paintings.
There was no need of any sanctum sanctorum, because the function of worship
was usually reserved for the temples alone. For the purpose of worship there were
two older temples, the Ajanta caves 9 and 10 (Pls. 6.1–2; 24.1–2). Additionally,
there was now planned two new temples, caves 19 and 26 (Pls. 6.7–8; 17;
24.4–5). They were also simultaneously initiated during c. 460–462
(Pl. 15).
The (original) layout of the new cetiyagharas was also based on the older
prototypes of Period I,30 i.e. the Type MC1c (Pls. 4; 6.7–8). There was no
provision or need of monastic cells or wing-like structures in them. If such
features are seen in them today it is because they were afterthoughts.31 The
initial thinking was simple. The layouts had to be guided by the basic needs,
function, and tradition rather than by individual tastes of the respective patrons
(Table 5). Therefore, it was not surprising if the original layout of Cave 11 was
that of a simple dormitory (leṇa-maṭapa) (Pl. 6.10). According to my study the
original plan had a smaller porch with two pillars only, a smaller hall,32 and
six cells in the interior for monastic dwelling.33 The layout was a typical Type
LM2bii something like the Nasik Cave 19 (Pl. 4) or the Ajanta Caves 12 and 13
(Pl. 6.4–5) added with a pillared porch and pillared hall. In a later course, the
planning of cells was re-configured. Four cells were added to the porch area,
which was to compensate for the loss of four cells on the rear right wall. It is
difficult to say whether the four interior pillars were there in the original plan or
whether they were afterthoughts (Pl. 16.4). The overall layout, with the interior
pillars, is so similar to those of the Jain Dharashiva Cave 4 (Pl. 19.3), Kondavite
Cave 1334 (Pl. 19.8), Dhamnar caves35 6 and 10 (Pl. 19.9), Kedareshvara Cave36
at Harishchandragad,37 Lamgaon Cave 2 at Goa38 (Pl. 19.7), and the Ankai-Tankai
S07 (233, 316–317), S09 (9).
For how the ‘wings’ were added to the Ajanta Cave 26, vide S12a (45–47, 52–65), S07 (311,
317–342), S06 (22–96). For how the wings were added to the Ajanta Cave 19, vide S07 (233–248).
32 The hall measures ‘37 feet wide by 28 deep and 10 high:’ F80 (294).
33 Cf. ‘nine cells,’ S07 (146).
34 F80 (361), B83a (71, pl. XLII).
35 Dhamnar caves: 24°12’ N, 75°30 E.
36 F80 (479) and B83b (57) assigned the caves to the 9th–10th centuries.
37 Few km from Nanaghat, 26 km from Junnar. For a brief survey: F80 (477–479), B83b (57,
pls. XLVII–XLVIII).
38 Thakur (2017, fig. 14).
30
31
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
307
Jain Caves: Cave 2 (ground floor), Cave 3, and 4 (Pl. 19.4–6).39 All of them have
four pillars in the centre of the respective halls. None of the pillars is highly
ornate. Except the Kedareshvara Cave (of later period) all the rest indicate an
early stage in the evolution of the plan and pillars. There must have been a traffic
among these sites – from Ajanta to Goa, etc.
When the excavation work started in our Ajanta Cave 11, there kept coming
many problems. The planners had to introduce many changes from time to time.
Some changes were innovative, others forced by compulsions. Certain errors
were also committed. The work was repeatedly halted.40 There was also a hiatus
that lasted in this cave for almost a decade, c. 469–477
(Pl. 15). The work
halted in 469
did not resume until 478 . When it did so it was only for
some reworking in the shrine and for intrusive paintings in the interior and porch.
It will be another task to describe all the problems and developments.41 So, we
must leave those details aside. What matters is that the planners were able to
find a solution to each problem. This was in spite of the fact that the masons
were doing this type of work for the first time in their lives. They had no prior
experience of excavating a mountain cliff, which was because of the Hiatus I
(c. 326–459 ) that had spanned for about 133 years42 (Pls. 1.3; 3.3). Due to
the prolonged lull period, the tradition and knowhow of the rock-cut architecture
was nearing extinction.43
M
R
By early 465
most of the interior had been excavated. Now it was time
to excavate the cells for which the plan had been substantially modified. The
new plan was to carve four cells on the rear wall,44 three on the left wall, and
none on the right wall (Pl. 16.4). It was, of course, asymmetry, which must not
have been there in the original plan, because the architects of those times had
a penchant for symmetry. What was on the left of the Monument’s axis had to
be on the right of the axis – subject to feasibility. So, the question arises: Why
did they deliberately plan asymmetry in the revamped layout? It was because the
There are seven Jain caves of 10th–12th c. at the Ankai-Tankai Hills near Manmad; vide F80
(505–508, pl. XCIV), B83b (58–59, pls. XII, XLIX).
40 Singh (2014, I: 318–329, 404–406), cf. S07 (141–161).
41 S07 (141–160), cf. Singh (2014, 309–329).
42 S20b (12–14).
43 Spink (1981, 111).
44 The left rear cell ‘rL2’ (Pl. 16.4) was possibly an addendum, vide S07 (144).
39
308
Rajesh Kumar Singh
planners faced a rude shock one day to see that there was no scope to excavate
any cell on the right walls; neither in the hall nor in the porch.
It was because of a mistake committed right from the times when the façade
was begun from c. 460–461 . The mistake was due to measuring the space
between our Cave 11 and the flanking caves: No. 10 on the right, and No. 12 on
the left (Pl. 23.1). It seems that before beginning the excavation work they had
only measured the space between caves 10 and 12 as it was visible on the exterior
face of the cliff. They had not estimated the situation inside the interior. They
had not realised that the interior cells were going to require a lot of additional
space, much more than what was visible on the face of the cliff. The dimension
of a cell used to be about 7 × 7 feet, or slightly more, or less. In addition, at
least 1 foot of thickness was required for the cells’ front walls doorways. So, it
adds up to 8 feet on the left and 8 feet on the right of the floor plan. Total 16
feet will be the increased width of the floor plan in the interior. If the visible
width of the façade was 37 feet, the required width inside (including the cells)
will be 37 + (8 × 2 = 16) = 53 feet. Moreover, to keep a safe distance from the
neighbouring caves, some more extra space had to be measured. This simple but
important point the planners had certainly missed to pre-calculate. As a result,
they found after excavating the interior that there was no scope to excavate all
the cells as originally planned. Consequently, forced by the unexpected situation
they had to make some compromising adjustments. They planned some new cells
and excavated them at most unusual locations. The oddly located cells are seen on
either side of the porch. That such a sequence of events and adjustments happened
may also be corroborated by a plethora of in-situ evidence, which unequivocally
suggest that the porch cells were afterthoughts. They were meant to compensate
the loss of cells in the interior right wall, which in turn was due to another factor
relating to the angle of the cave’s axis. Let us understand it a little bit.
As noted earlier, the cave’s original plan was the Type LM2bii (Pl. 4),
essentially a dormitory. The plan had a porch, a hall, and equal number of cells
on the left, rear, and right interior walls (Pl. 16.3). In other words, the right
interior wall was also meant to have cells, which of course required adequate
space over there.
However, there had been committed a second mistake. It was regarding the
cave’s Cartesian Y-axis, which turned out to be approximately 195° S-SW45 while
45
Author’s measurement; cf. ‘193.5°’ by Vahia, et al. (2017, 808).
Circa 465
309
and The Stupa-Shrine of the Ajanta Cave 11
that of the adjacent Cave 10 it was 190° S-SW on azimuth.46 It means that the
Y-axes of the two caves are not parallel;47 they are not perpendicular on the Cartesian
X-axis. The conflicting angles are difficult to notice from outside, or inside, until
and unless you inspect the matter. That is why the planners of our Cave 11 had
not suspected the problem. They realised the problem only after the hall had been
excavated. Fortunately, they noticed in time that the (interior) right wall was about
to slam into the left vault of the Cave 10 (Pl. 16.3). The gap between the two caves
had narrowed down to a few feet only. This meagre space was simply not enough
for excavating the right interior cells of our Cave 11. So, the plan of carving cells on
the right wall was promptly aborted. Perhaps, a warning signal had also come from
the Caves 25 and 26 (Pl. 23.2), which were also simultaneously being excavated
during 464–465
(Pl. 15). In those caves, an accident had already perhaps taken
place. While excavating the Cell D on the left side of the porch of the Cave 25, the
masons had accidently narrowed through the right upper vault of Cave 26. They had
hammered an accidental hole.48 If you stand inside that cell (which is there on the
left side of Cave 25’s porch) you can see the hole; you can peep through it, which
will show you the nave of the adjacent Cave 26. If you stand inside the nave of
Cave 26, the hole can be spotted in the front right corner of the ribbed vault. As
a result of the accident the work was aborted in the concerned cell of the Cave 25.
The whole left area of the porch was then abandoned. This mishap impacted the
entire further development of that cave. Something like that also happened in our
Cave 11. The work had been halted yet again, even if briefly.
E
E
I: F
B
465
M
F
I
G
G
,E .
K
,
While such a disappointing scenario was prevailing in our Cave 11 there had
started another development of another kind that impacted nearly half of India.
Many caravans of people, merchants, and monks had likely begun to migrate from
Bactria and Greater Gandhāra towards Middle India and the Deccan (Pls. 1.4a;
46
47
48
Author’s measurement; cf. ‘199°’ by Vahia, et al. (2017, 808).
S07 (142–144).
S12a (figs. 18–19).
310
Rajesh Kumar Singh
3.4a).49 They were displaced from the north-west of India. There had been a sort
of internecine warfare going on among the Kidarite, Alchon, and Hephthalite
Huns. During c. 451–454 , the Alchons had pushed out the Kidarites from
the Oxus Valley, Gandhāra, Taxila, western Punjab, and Kashmir.50 However,
in c. 464–465
the Hephthalites had expelled the Alchons from the Greater
Gandhāra region.51 This triggered an exodus of people in all possible directions;
the Buddhist establishments were among the worst affected, for they had the
maximum wealth in the temples52 attracting the maximum plundering and looting.
The refugee crisis may be called the Buddhist Exodus I, which may be ascribed
to c. 465–467
(Table 2.2; Pls. 1.4a; 3.4a).53 The displaced people were unable
to flee to the north, south, or west of Bactria and Greater Gandhāra because
those areas were already like cauldrons of fire. Even to the east, the situation
was disturbed up to the Indus River. They had to think of some faraway lands
to seek refuge and safety. A likely destination was the Tarim / Taklamakan area
in the north-east of Bactria and Gandhāra. The peripheral settlements on the
Desert were least attractive for the invaders that had insatiable craving for gold.54
It was not yet disturbed by c. 465–467 . The disturbance happened later on
when some of the oasis towns were attacked by the Hephthalites.55
Another likely destination was India, particularly the inner pockets of India,
i.e. Middle India, the Deccan, and Aparānta. The conclusion of our research is
that a substantial number of refugees had definitely reached Middle India, the
Deccan, and Aparānta during early 465 . At the time, the regions south of
Narmada offered the best hideouts to the terrorized refugees. They did not settle
anywhere close to the north-western frontiers of Skandagupta’s Empire. Needless
to say, whenever war refugees migrate en masse they cross over to a land that
could provide them refuge, food, shelter, and safety. A climate of peace is the
number one precondition. The scenario in India at the time was that the Narmada
valleys and the Vindhyan Ranges had vast stretches of deep jungles and endless
green mountains as gateway to the great Western Ghats. They might not have
S20a (27–28), S20b (17–18), S20c (126–128). Singh (Forthc., ch. II).
S20a (15–20), S20c (125), S21 (4–5, 12).
51 Zeimal (1996, 128), Kurbanov (2010, fig. 87), S20a (27–28), S21 (12 n. 49), SB (xv), SBC
(17–58).
52 SB (63, 91).
53 S20c (125–128).
54 SB (232, 264, 301). For a study of the Huns vis a vis gold, vide SB (p. 19, 29, 31, 36, 38, 57,
58, 134, 135, 174, 175, 176, 177, 270, 366, 373, 376, 396, 400, 401, 403).
55 Called Yeda, or hu, or Hua in the sources. SB (50–55).
49
50
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
311
been the most ideal places for settlement. However, the jungles and mountainous
landscapes were the safest hideouts for the harried monastics who could muster
courage to cope with the wildlife but not with the iconoclast, inhuman, barbarians,
which the Alchons Huns appear to have been at the time. We can say this not
from the angle of the invaders but from the angle of the invaded. The new Gupta
emperor, Skandagupta, r. 455–469 56 (Pl. 14) – who was then heading the
Grand Alliance of the Indian Empires57 – had indeed restored the peace. This
was after the two previous wars when the Alchon Huns were repulsed at great
costs of the lives of many kings.58 The first Hunnic invasion has been ascribed
to c. 447 ,59 and the second to c. 455 60 (Pl. 14).
It was not just the peaceful climate, even the religious climate was
incentivising, because the Vākāṭaka Empire under Hari Ṣeṇa, r. 459(?)–477(?)
had also adopted the multi-religious state policy of the Guptas. It is clear from
the current research that Hari Ṣeṇa had become an emperor par excellence. His
boundaries had vastly expanded to include the following countries: Vidarbha,
Ṛṣika (roughly Khandesh), Avanti, Lāṭa, Aparānta, Trikūṭa, Aśmaka, Mūlaka,
and Kuntala (Pl. 14).61 Some of these countries and many others have been
named in the Ajanta Cave 16 inscription, v. 18 as the lands ‘subjugated’ by the
S20a (21–22), S21 (17 n. 80–81).
It was a strategic and military alliance between the Gupta Empire, Vākāṭaka Empire, their vassals,
and other kings. For the new theory, vide S21 (16–17, 21–22, 37, 46, 51–53). The alliance’s aim was to
defend against the Early Alchon Huns, to create a security architecture against the Hunnic attacks. The
alliance was somewhat like how NATO was formed. But, the Alchon Huns already had a ‘NATO system’
of their own that was much more formidable, experienced, and ferocious.
58 ‘Skandaguptasya śānte varṣé / tranquil reign of Skandagupta:’ vs. 3–4 of Kahāum Stone Pillar
Inscription of Skandagupta dated 460/461 , vide Siddham (IN00034), B81 (No. 29), Sircar (1965,
No. 26), and F88 (No. 15).
59 S21 (12–22).
60 S21 (22–34).
61 M63 (108, 110): ‘स क
ु ु लाव् किलঢ়कोसलिআकूटलाटा [परा ] जािनमान [।] U – U — U – U
ि – U – [18 ||] / sa kuntalāvantikaliṉgakōsalatrikūṭalāṭāndhra[parānta]
शौय१ िवुतानिप िनद८ शगु णाित
jānimān [|] U – U — U – U śauryaviśrutānapi svanirdeśaguṉāti – U – [18 ||].’ Sircar (1965, 453) read
‘शौय१ िवुतान् ’ as – ‘U नैरधृतान् ’. ‘परा / parānta’ is supplied by Gokhale (1991, 9–10), (1992, 277). For
Singh’s study on how Hari Ṣeṇa occupied these lands, vide S21 (chs. VI, XI). The following are the
place names as identified in S65 (453, n. 2) and M63 (107): – Kuntala: Kannada country, district round
Banavāsi. Avanti: Western Mālwā, capital Ujjain. Kalinga: The country between Mahānadī and Godāvarī;
in the narrow sense Purī-Ganjam region of Orissā. Kōsala: South Kosala or modern Chattisgaḍh with
adjoining parts of Orissa, including Rājpur-Sambalpur-Bilāspur region. Trikūṭa: Country to the west of
Nāsik. Lāṭa: Central and southern Gujarāt, between Mahi and Tāpi; Nausāri-Bharuch region. Āndhra:
The Telugu speaking country to the south of Godāvarī. Aparānta: The northern Koṇkan. Cf. mapping of
these regions by Weiner (1977, 11), reproduced in W17 (178), W14 (xv).
56
57
312
Rajesh Kumar Singh
Vākāṭaka Mahārāj Hari Ṣeṇa. The security shield of the Alliance covered a large
part of India: From the Irāvati (Ravi) and Śatadru (Sutlej) Rivers in northern
India to the Krishna River in the southern Deccan, and from Bengal in the east
to Saurāṣṭra in the west. Such a vast stretch of land offered extensive sanctuary
to the refugees, for the rest of Eurasia was highly disturbed including China.
The refugees could not have settled near the Sindhu (Indus), Vitasta (Jhelum),
Candrabhāga (Chenab), Irāvati, or Śatadru Rivers, for those regions had been
invaded in the past. In spite of all measures of border fencing / guarding, the
borderlands of any state are always prone to fresh intrusions / attacks. After
all, it was along the said borderlands and rivers where Skandagupta’s father,
Kumāragupta I,62 r. 41563–44764 (Pl. 14), and then his uncle, Ghaṭotkacagupta,
r. 416–447–455,65 were likely killed by the same enemies—the Alchon Huns led
possibly by Mahāṣāhi Khīṅgīla.66
Although the migrants were indeed refugees they were not perhaps treated as
“aliens” / “foreigners” or burdens. They had not come from an unfamiliar cultural
background. They belonged to the same religious and cultural tapestry that stretched
from Gandhāra to the Deccan, thanks to Vaiṣṇavism, Śaivism, Buddhism, and
Jainism, which had a history and footprints up to Balkh in Afghanistan. Forced
occupation of their lands, looting of their properties, and burning of their temples
would have been a common sight in Bactria, Greater Gandhāra,67 and Sindh.
There are reasons to believe that some of the migrants also started going towards
Kucha, which was a northern oasis town on the Taklamakan Desert. They did
not at first go to Kucha when the Exodus I started in c. 465 . It was in the
subsequent year (466 ) when some refugees appear to have gone towards
Kucha (Table 2.2; Pls. 1.4a; 3.4a). The migrations of people created rapid cultural
exchanges. In Kucha, many transformations began from the archaeological Period
II (c. 466–474 ), which must be attributed to this Exodus I (Pl. 3.4a). There
started in Kucha a similar pattern of architectural adaptations as in India, i.e.
62 The Gupta genealogy has been a work in progress. For a recent version, vide W05 (135–137).
Cf. earlier versions: F88 (17), Majumdar and Altekar (1954, 173–175), B81 (1–89), Chakrabarti (1996,
191), Zeimal (1996, 127), and B97 (170).
63 Based on the Sanchi inscription of Gupta Era 93 (412/413
) and Bilsad inscription of
96
(415 / 416 ), vide F88 (29, 42), S65 (280–285), and cf. W05 (136 n. 16).
64 W05 (135–137).
65 S21 (13 n. 54–57).
66 S21 (12–25).
67 SB (xv–xvi, 50–55).
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
313
typological conversions of the existing caves. A common theme was introduced
at both the sites, i.e. the concept of multi-functional and composite architecture.68
The monks, laity, and merchants who came to India did not presumably come
empty-handed. When you are a refugee fleeing from the homeland, you may leave
everything behind, but not your gold, jewellery, currency, favourite scripture,
God (depicted in some form that is easy to carry), etc. Once in the new lands,
the migrants took interest in the ongoing developments of rock-cut monuments
across various sites. Two years later in c. 468 there came Exodus II (Table 2.6;
Pl. 3.4b), which witnessed a greater displacement of people. They mixed up with
the local people so creatively and constructively that they became instruments
of great change. This fact was metaphorically expressed by the depictions of the
“foreigners” in the arts of Bagh and Ajanta. It included the depiction of their
cultures, costumes, ethnicities, worldviews, scholasticisms, texts, mythologies,
legends, and other cultural elements.69 This is the explanation to the ghost
question as to what the ‘foreigners’ are doing in the arts of Bagh and Ajanta.
68 The matter requires attention from scholars of Kucha studies. Giuseppe Vignato has made a chart
that indicates that the layout of many caves of Kucha were changed during what he calls Periods II, III,
and IV, vide Howard and Vignato (2014, 174–175), Vignato (2006, 52–54). The architectural changes
were begun from Period II: ibid. 174. But there is no date to it, because Viganto is generally hesitant to
assign any dates to his periodisation unless vaguely. He does not tell us when his Period II began and
ended. He does not tell us when his Period III began, but he did tell us once that ‘the year 550 . . could
be considered the dividing point of the site into earlier (first to third) and later (fourth) phases’ (Vignato
2006, 50). Whatever be the end of the Period III, the beginning has been recently assigned to c. 475
(Pl. 3.7): vide S20a (37–38; tables 3–4; pls. I.7, X), S20b (tables 1, 5; pls. I.7, 10), S20c (pls. I.7, 10).
It must be linked to the Exoduses III and IV that emanated from Gandhāra and India towards different regions including Kucha (Table 2.8–9; Pl. 3: 5, 7). As far as the beginning of Kucha’s Period II is
concerned, it was certainly catalysed by Exodus I (Table 2.2, Pl. 3.4a). The migrants from Bactria and
Gandhāra introduced, in India as well as in Kucha, the concept of multi-functional architecture among
other things. This happened during c. 465–466 . This date is crucial for Kucha as well as Middle India
and the Deccan. It brought significant changes in northern Tarim / Taklamakan regions as well as India.
On these grounds, Kizil’s Period II may be ascribed to c. 466–474
(Pl. 3.4a). It was a small time span
of 8–9 years. More displacement of people happened later on. There came a time when large number
of migrants went all around from Bactria, Gandhāra, Middle India, and the Deccan. It happened during
c. 477
that has been termed Exodus IV (Table 2.9, Pl. 3.7).
69 None of the avadānas painted in Ajanta and Bagh could be prior to Exodus I, c. 465–467
(Table 2.2; Pls. 1.4a; 3.4a). A host of other features and ideas also came from Gandhāra. The proposition
also concurs with Spink’s ‘Chronology of the Paintings at Ajanta,’ vide S07 (392–393), S09 (figs. 42–43).
According to the same, the first paintings were done in 468
starting from the Ajanta caves 11, Lower
6, 16, and 20. However, our studies indicate that at least some narrative paintings had already begun
before 468 . They were more ‘classical’ or traditional drawn from the Pali jātakas, for after 467
the
north-western MSV scholasticisms, image cult, Bodhisattvayāna, etc. had taken over the regions south
of the Vindhyan Range. The fact not only applies to Ajanta but also to Bagh and other contemporary
sites (Table 1).
314
Rajesh Kumar Singh
Among the many fresh ideas that the migrants brought, one was that of the
tri-functional architecture. It was common in the north-west of India for many
centuries.70 Tri-functional edifices were three times more economical, useful,
and versatile. They required only one-third of the funding, human resources,
and turnaround time. The greatest advantage was the architectural integrity and
versatility that it offered. It had included all the major functions under the same
roof: worship, congregation, and dwelling. Other associated functions could also
be carried out, such as resting, dining, or meditation.
We are not the first one to notice that it was a new plan at Ajanta, etc.
Sheila L. Weiner had noticed it. She could think of two possible sources of
inspiration. One was Nagarjunakonda.71 The other was Haḍḍā in Afghanistan.72
The first suggestion about Nagarjunakonda is questionable on two counts. First,
Andhra, like the Deccan, had also undergone the same lull period, i.e. Hiatus I,
c. 326 – c. 459 73 (Pls. 1.3; 3.3). It was a kind of Dark Age for the Buddhist
rock-cut monuments of Dakṣiṇāpatha lasting for about 133 years. Therefore, it
is highly unlikely that a ruptured site would catalyse a revived site. The second
improbability comes from architectural typology. Two Ajanta monuments were of
the Type MC1c (Pl. 4): the Caves 19 and 26 (Pls. 6.7–8; 17; 24.4–5). This type
was there in Nagarjunakonda but in a miniaturized format. Moreover, it was never
independent, but was an architectonic prefix to the monasteries having a pillared
hall with cells (Pl. 5.9–10). In that sense, the Nagarjunakonda typology was also
LMC2 + MC1c. It was also tri-functional, composite, and integrated. However,
the greatest distinction was the fact that there were multiple axes. The areas of
worship had own axis and the congregation hall had another axis. In fact, the
two axes were not the same or parallel, but perpendicular to each other. This is
For a better understanding of how various ideas on art and architecture travelled from one culture
and civilisation to another, more studies are necessary on fifth-century Greater Gandhāra, the Indian
subcontinent, West Asia, and Central Asia. We need to build on the studies by Majumdar and Altekar
(1954, 107), Weiner (1977, 42, 48), Divakaran (1989), Callieri (1997–1998, cat. 7.39, pl. 24), Behrendt
(2007, figs. 10, 12, 13, 27), Brancaccio and Behrendt (2006, 85–86, 89, 91, 93, 96, 98), Khan (2008),
Kim and Han (2011), Lo Muzio (2012), (2014), Zin (2013), Compareti (2014), Rienjang and Stewart
(2018, 75, 94, 151, 153, 156–157, 160–162), Brancaccio (2018, 62–65), and Ciordia (2020). The following may also be consulted: Ashraf Khan and Mahmood-ul-Hassan (2004), Lo Muzio (2008, figs. 4–7,
9, 13), Khan (2016), and Gandhāran Studies journal being published from Pakistan.
Floor plans of some of the sites have been collated on Plates 20–22; and a list of some of the other
sites can be found in Table 7.
71 Weiner (1977, 44–46, fig. 4, 109, 111).
72 Weiner (1977, 47–47, fig. 5, 110–113).
73 For more on Hiatus I, vide S20b (12–14).
70
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
315
not what is seen in Ajanta, Bagh, etc. There were at least 15 other monuments in
Ajanta and two at Bagh that had started as Types LM2bii, but were now being
converted into the Type LMC2 + S4.
In short, whereas the typology in Nagarjunakonda was LMC2 + MC1c, the
Ajanta typology of at least 15 caves was Type LMC2 + S4. The mismatch proves
that these are different buildings, based on different plans and principles, and
should not even be compared together. There were two shrines in Nagarjunakonda
whereas in Ajanta, Bagh, etc. there was a single shrine. It was located inside the
hall and on the axis of the cave. It was located at the deepest end of the axis
carved into the rear wall. Another point of contrast is that the stupa and the image
both were enshrined separately in different locations of the edifice; they were not
synthesized together as in the Ajanta caves 19 and 26. The metaphor is indicative
of the cultic practices. Whereas in Nagarjunakonda the duality was retained, in
fact preserved and nurtured, at Ajanta and Bagh the duality was synthesized
or abolished. The foregoing factors are enough to rule out the possibility that
Nagarjunakonda inspired the change in Ajanta.
Now we come to the other suggestion by Weiner that the said architectural
layout of Ajanta, etc. – the Type LMC2 (Pls. 4; 5.10–36) – had some connection
with Haḍḍā in Afghanistan. This suggestion appears to be quite possible although
Weiner or other scholars after here who felt the same were unable to describe the
details and circumstances that made it possible. Our study indicates that the Type
LMC2 + S4 (Pl. 4) of Bagh, Ajanta, etc. was prevalent only in the north-west of
India: in Taxila, Gandhāra, Haḍḍā, Miran, etc. This layout was there, albeit the
format was masonry architecture.
A counter argument may be made that such composite buildings are found
in plenty at many Buddhist archaeological sites along the Indo-Gangetic plains
and also in Nepal. The argument cannot be supported by other factors. There is
no direct evidence to indicate that our planners in Middle India and the Deccan
had a steady contact with the Buddhist sites along the Ganga-Yamuna plains.
Notably, the doctrines of the Mahāyāna and Mūlasarvāstivāda movements were
prevalent in northern India from the first century
itself. However, the same
was not reflected south of Narmada up until c. 468 . We should not commit the
mistake of thinking that there was anything called pan-Indian Buddhism. What
we had were regional schools, sects, and traditions.
The foregoing discussion indicates that the architectural typology that was
prevalent in the north-west of India had to be the one to inspire the typological
change in Ajanta, Bagh, etc. To this possibility we have proposed that the said
316
Rajesh Kumar Singh
change happened due to Exodus I, 465–467
(Table 2.2; Pls. 1.4a; 3.4a).
The realisation that there was Exodus I, II, III, and IV is going to be dependent
on whether the reader is well familiar with many areas studies and disciplines;
whether the reader is abreast with the current researches in the following
domains: (a) The Ajanta studies; (b) The Gupta studies; (c) The Vākāṭaka
studies; (d) The Traikūṭaka studies; (e) The Aulikara and Parivrajaka studies;
(f) The Hunnic studies with focus on Early Alchon Huns; (g) The Kuchean
studies; and (h) The Gandhāran studies. Such a reader will not doubt that there
did take place a spate of typological conversions from dormitories to dormitorycum-temples; and that such conversions began during c. 465 and 466 .74
The movement pervaded across all the sites (Table 1).
P
C
-G
/S
-S
/N
The tri-functional layout demanded the addendum of a naos in the rock-cut
dormitories. So, there began a massive programme of retrofitting a naos inside
every dormitory across some two dozen odd sites that we know of (Table 1;
Pls. 25–27).75 The Ajanta Cave 11 (Pls. 6.10; 16.3; 23.1; 26.1) was among the
first edifices that were converted from the Type LM2bii to the Type LMC2 +
S4 (Pl. 4). Its shrine (Pls. 26.1; 28.6–7) was among the first ones to have been
initiated anywhere in Middle India or the Deccan.76 The shrine’s original blueprint
did not have any image in it. It was simply, and strictly, a stupa-shrine.77 The
same was the case of every other shrine that was simultaneously planned or
initiated during 465–467
(Pls. 6.7–32; 25–27). The stupa is hidden at the back
of the Buddha image (Pl. 28.7).78 The stupa was never completed. The floor is
raised up in the back area (Pls. 28.6; 29.2), which in my opinion was to reserve
a matrix of rock for carving a base for the drum as in the Bagh Caves 2 and 4
(Pls. 18.1–2; 27.5–6). They wanted to carve the base at a later stage, a task that
Cf. ‘466–468’
(S09, fig. 41); ‘466’
(S14, 65), (S09, 26), (S07, 50); ‘467’
(S07, 153);
‘468’
(S14, 54), (S07, 389), (S09, fig. 39). Spink argued that Bagh Cave 2 inspired the stupa of the
Ajanta Cave 11, vide S14 (54 n. 57), S17 (31–36).
75 For Spink’s excellent documentation of many such naoi, vide S14 (65–108). Note, however, that
there is a difference of up to 2 years between his dating and my dating.
76 S14 (54).
77 S09 (pl. 90), S14 (1).
78 The shrine areas of the Ajanta caves have restricted entry. So, for studying the shrine, the reader
is advised to use Google Map’s ‘Street view’ feature that shows things in 3D. Currently, it is the best
available tool for in-situ research.
74
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
317
was never finished. Notice also that some parts of the stupa are missing: the
umbrella (Skr. chatra), the shaft of the umbrella, the harmikā, and vedikā. Why
these are parts missing? Let us probe the matter.
If you imagine Ajanta in c. 465 , there were about 20 dormitories (layaṇas)
that were in different stages of progress.79 The task of adding a stupa-shrine
(cetiyaghara) was difficult, particularly in the larger caves. So, smaller edifices
(Caves 5, 11, and 15) were first chosen for the addition of shrines. Our Cave 11 was
not only small in dimension, it was also situated next to the ancient cetiyaghara
Cave 10. Proximity to Cave 10 was also a deciding factor. Moreover, the cave’s
excavation had also reached an advanced stage; the hall had been largely scooped
out; the walls had been levelled; and the masons were working on the cells. In
fact, some of the cells had been carved out including the rear ones.
It was at that point of time (early 465 ) when there started to happen the
Buddhist Exodus I, c. 465–467 (Table 2.2; Pl. 3.4a). War refugees from Bactria
and Greater Gandhāra had begun to arrive in Middle India and the Deccan.80
When the migrant monastics found that the Śrāvakayānists of Ajanta were
making big rock-cut temples and numerous rock-cut dormitories, they found it
strange. It was strange because separate edifices were being made for worship
and dormitories. Whereas the migrants knew better solution; they had composite
architecture in their lands. It was evaluated that the function of worship could
very well accommodated within the dormitories, and conversely, the function
of dwelling could be added to the worship halls – Caves 19 and 26 (Pls. 6.7–8;
17.1–2). The suggested scheme was a hybrid concept in terms of function and
utility. Although such composite and integrated architecture were definitely made
in the past, but they had perhaps been long forgotten. The experiments were
carried out in some caves during Period I.2 (Pl. 5) but those experiments had
remained on the margins only.
So, when the concept was re-introduced in the fifth century , it appeared
a nice idea. However, the planners had to evaluate the feasibility. Essentially the
task was to add a sanctum sanctorum to the dormitories, which were quadrant and
flat-roofed. The dormitories were duo-functional (congregation and dwelling),
which were now to be converted into tri-functional edifices with the addition of
a shrine. It was a comprehensive architectural solution to provide for dwelling,
Ajanta Caves 1, 2, 4, 5, Lower 6, Upper 6, 7, 8, 11, 15, 16, 17, 20, 21, 23, 24, 25, 26 lower right
wing, 26 lower left wing, and 27 (Pls. 4, 15).
80 S20a (27–28), S20b (17–18), S20c (126–128).
79
318
Rajesh Kumar Singh
congregation, and worship under the same roof. The new layout had to be the
Type LMC2 + S4 (Pl. 4). If the clergy was in charge of the ‘town planning’ and
major decisions, it approved the tri-functional layout. It was then left for planners,
donors, and workers to implement the scheme. Consequently, everyone in Middle
India and the Deccan began to convert the leṇas (dormitories) or leṇa-maṭapas
(dwelling + congregation) into layaṇa-caityagṛhas or layaṇa-maṭapa-caityagṛhas
(dwelling + congregation + worship). The idea was so insistent and powerful, and
the advantages were so attractive, that the planners of Ajanta, Bagh, etc. went
back to the drawing board. They began to re-draft how exactly a stupa-shrine
could be added to the dormitories under excavation.81 It was all about wanting to
have what may logically be called the quadrantile and flat-roofed cetiyaghara.82
The tri-functional layout turned out to be a game changer. It eliminated the
need of excavating multiple buildings for different functions. It would also have
been realised that the apsidal and vaulted temples were far more demanding on
time, budget, skilled workforce, and execution. What was worse, they addressed
only two functions: worship and congregation. Until that time (c. 465 ) the
cetiyagharas of India seldom had any cells. They were also duo-functional
(worship and congregation). It was, therefore, realised at Ajanta that the four
temples were not enough: Caves 10, 9, 26, and 19, in this chronological order
(Pl. 24.1–5). Now they knew that every edifice could be turned into a temple,
call it whatever: Leṇa-cetiyaghara / layaṇa-caityagṛha / Type LMC2 / templecum-convent (Pls. 4; 6.10–36). Thus, the year 465
was a turning point in
83
rock-cut architecture.
There were 17 such edifices in Ajanta: Caves 1, 2, 4, Lower 6, Upper 6, 7, 8, 11, 15, 16, 17,
20, 21, 23, 24, 26 lower right wing, and 26 lower left wing; at least two in Bagh; one in Ghaṭotkaca; at
least one in Pālé, and at least two in Dharashiva.
82 Many edifices at Ajanta, Bagh, and Ghaṭotkaca were originally planned as Type LM2bii
(Pls. 6.10–32), and two edifices were planned as Type MC1c (Pl. 5.7–8). The former type was meant
for dwelling, congregation, resting/dining whereas the latter type was meant for congregation and worship (Pl. 4). The edifices that had begun as Type LM2bii – e.g. Caves 1, 2, 4, 6, 16, 17, etc. – were
converted or attempted to be converted into the Type LMC2 (Pl. 4). Whereas those that had begun as
Type MC1c – Caves 19 and 26 – were converted into LMC1 (Pl. 6.7–8). Such conversions have been
thoroughly deduced and painstakingly analysed by Spink (2009, 25–28), (2017, 31–36).
83 ‘466’
: S09 (25–28), S17 (31–36); cf. ‘465:’ S20c (124–125, 138).
81
Circa 465
319
and The Stupa-Shrine of the Ajanta Cave 11
L
E
465
C
F
-G
/N
Drawing a blueprint was one task; translating it in a monolithic context was
quite another. Materialisation of the new ideas required many adaptations. First,
a new system of decoration was needed. The original plan was merely that of
a dormitory, which did not require much decoration. However, a leṇa-cetiyaghara
(Prakrit)84 was surely going to require one. The newly planned naos had to
be positioned on the axis of the cave, and to be carved in the rear wall. The
problem was that the masons had already excavated the rear cells: rL1, rL2,
rR1, and the central one that was later converted into a naos (Pl. 16.3–4). What
to do with them was a ghost question. They were coming on the way of the
shrine. In fact, they were occupying the space that was required for the shrine.
The only way out was to transform one of the central cells into a shrine. For
this, a cell was chosen that was closest to the axis, i.e. the cells between rL1
and rR1 (Pl. 16.3–4). There was a need to completely transform that cell in
such a way as to create a stupa-shrine. Yes, indeed, what they wanted was
a stupa-shrine; they did not have any image in mind. The image was carved
later on. For augmenting the interior space of the cell they needed to increase
the depth, width, and height. In the process, the stupa needed to be revealed. The
concept of a naos for anthropomorphic Buddha had not yet germinated in Middle
India, the Deccan, and western India. It was not going to be introduced until
c. 468
(Table 2.7, Pl. 3.4b).85 In 465 , our planners did not know that they
would be making further changes. At the time the Buddhists of Middle India, the
Deccan, and western India followed an orthodox Śrāvakayāna school. They were
strictly the adherents of the stupa cult. They were not Mūlasarvāstivādins (yet).
Consider the fact that in the earlier times we do not come across the depiction of
Śākyamuni, Bodhisattva, Avalokiteśvara, Hārītī, Dipankar Buddha, MāṇibhadraPūrṇabhadra, Saṃsāracakra (Wheel of Life) pralambapādāsana mudrā, avadānas
(from MSV), and other legends (from non-canonical Sanskrit literature) in the
rock-cut monuments of Middle India, the Deccan, and western India.86 It appears
Or layaṇa-caityagṛha (Skr.) or temple-cum-monastery (Eng.).
S09 (figs. 41, 45), S14 (459–460, 463–464), S17 (31–34, 414).
86 Among the exceptions are the tiny Buddha and Bodhisattva images carved in Kanheri Cave 3
during the first century . Those too were likely sponsored by donors from the north-west of India,
vide Leese (1979, 93).
84
85
320
Rajesh Kumar Singh
that there was hardly any depiction of anything connected directly to the MSV
scriptures prior to this date. This was true of Middle India, Western India, and
the Deccan. It was also more or less true of the whole of India.
When depictions from the traditional MSV started from c. 468 CE onwards
it was like a tsunami of Buddha images. However, a point must be noted that the
Bodhisattvas were still not introduced until c. 470 CE. It took a while for them
to find the general acceptance, which happened by c. 475 CE. There also came
a time when they became rather compulsory by c. 477 CE. The phenomenon was
witnessed across all the other rock-cut sites. The point that needs to be underlined
is that all these sea changes happened when the first hammer was struck in our
Cave 11 for carve what was to be the first Buddha image of Middle India, western
India, and the Deccan.87
Going back to 465 , nobody had any inkling that they will soon abort the
plan or excavation of the stupas for the sake of newly approved Buddhas and
Bodhisattvas. So, all the plans during c. 465–467 were focussed on excavating
the stupa-shrines. The fact is applicable not only for Ajanta and Bagh but also for
other sites wherever rock-cut caves were being excavated (Table 1). The Ajanta
caves 19 and 26 (Pls. 6.7–8; 17; 24.4–5), which were the only ones that were
originally planned as ceityagharas, were also meant to have just the stupas or
cetiyas in them. There was no provision of any image in them. The two temples
were planned rather like the older cetiyagharas, Caves 9 and 10 (Pls. 6.1–2;
24.1–2), excavated during the pre-Christian Era. The Buddha images seen in these
four temples (Caves 9, 10, 19, and 26) were not even conceived prior to c. 466 .
As to how exactly to begin the excavation work, it appears that the planners and
the masons had no clarity. They did not have a ready formula, or knowledge of an
established procedure, as to how to carve or excavate a cetiya-ghara / stupa-shrine
in a monolithic context. Some errors were also committed in the process. Even the
workmanship shows the lack of expertise. Observe, e.g. how the pillars have been
carved (Pl. 28: 2, 3, 5); how the windows have different designs (Pl. 28.3); how
the porch area has been so unusually treated (Pl. 28.2); how the doorways, jambs
(Pl. 28.3), and cell doorsteps (Pl. 28.5) have been carved. Lack of great workmanship is
also observed in the windows, doors, floors, benches, door fittings, stupa, ambulatory
(Pls. 28.6; 29.1), sculptures, and angle of axis (Pl. 16.3). If you study the cave’s early
development, you will spot enough evidence – too many to be detailed here – to
indicate just how inexperienced the planners and masons were initially.
87
S20a (31, 57) S09 (fig. 45), S14 (77, 81, 82, 90, 97, 460).
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
P
321
P
Even for expert carvers, it would never have been easy to carve out a hemispherical
stupa. Getting a rotund shape, dome, or drum is always so difficult in a monolithic
context. You needed to define by chisel and hammer some of the following parts
of the stupa while excavating downwards from the top: 1. The chatra (umbrella);
2. The inverted stepped pyramid capital; 3. The harmikā / vedikā; 4. The aṇḍa
(dome), 5. The meḍhī (drum); and 6. The base of the stupa with or without
mouldings. However, not all the six parts are found in all the stupas. A survey of
the early rock-cut stupas can show that there was no consistency.88 Our stupa of
Cave 11 has the parts mentioned at No. 2, 4, and 5 above, but not No. 1, 3, and 6.
Most baffling point is that there is no umbrella (Pl. 29). Often, the umbrellas
were carved (in relief) in the ceiling itself. They were also wooden in many
cases, as in Karle Cave 889 (Pl. 30.2). However, what is visible in our Cave 11
tells another story. Not only the carved umbrella is absent in the ceiling, we also
do not see any shaft or socket for it. Even the harmikā and vedikā are absent.
Most importantly, there is not enough space between the ceiling and the dome to
accommodate these elements. The extant space is hardly one foot (Pl. 29: 1, 4).
In that space, they had probably blocked out the inverted-stepped-pyramid-capital
(Pl. 29.5). Remnants of the blocked out matrix of rock for the same can be
sighted between the ceiling and the dome (Pl. 29: 3, 5). However, most of it
(towards the rear side) was cut away when the plans had changed; when the
stupa cult was unexpectedly supplanted by the image cult from c. 468 onwards
(Table 2.7). Then, a Buddha image was carved on the front of our stupa. The
two gandharva figures (Pl. 29.3) were also carved from the same matrix of rock
that was reserved for the inverted-stepped-pyramid-capital. If you inspect the rear
sides of the gandharvas’ legs you will be able to deduce that the legs could be
carved because there was the reserved matrix of rock (Pl. 29: 3, 5).
The form of a stupa is such that if you wish to create it inside a monolithic
chamber you will need to leave out equal spaces in all directions. However, before
you get to the stupa, you will have to first carve the chamber—or, maybe both
together. You cannot excavate the chamber without dealing with the stupa, and vice
versa. The process of monolithic excavation is always from the top downwards,
which is just the opposite of what we do in a building or a construction work.
88
89
Dehejia (1972, 76–77), Nagaraju (1981, 72–77).
N81 (pl. 129).
322
Rajesh Kumar Singh
So, beginning from the ceiling level, one had to excavate towards the floor.
Simultaneously, beginning from the front or from the shrine’s doorway, one had
to excavate towards the rear wall. To do this from inside of what was earlier
a cell of Cave 11, which now needed to be expanded, first you had to determine
the expected depth, width, and height of the chamber (ghara). Second, you had
to somehow mark the centre of the chamber. Then, sketch a circle around the
centre for the umbrella. This would automatically leave out an equal measure of
space on all the sides. The centre so defined was called the yaṣṭi90 or yūpa91 in
the ancient times. It was the axis mundi. In masonry stupas, it was a wooden or
stone pillar or shaft placed in the vertical centre of the stupa around which the
stupa was erected. Here too, a yaṣṭi or yūpa was needed to be carved, or at least
marked, in the ceiling’s centre. The centre of the chamber was also the centre
of the umbrella and the stupa. That is why the umbrella was the first thing to be
carved in the rock-cut stupas. It used to be carved in relief in the ceiling (Pl. 30.3).
However, this was not done in our Cave 11. Did they not know the process? Was
there nobody to supervise the masons? Now the problem was how to excavate
the stupa when the centre was not marked in the ceiling? Surely, it was a blunder,
the first evidence that the carvers were novices. They did not even anticipate that
there was a risk in proceeding further without a handbook, so to speak.92 They
did not even carve the entire ceiling. This can be said if you observe the rear part
of the ceiling, which is not angular but semi-circular (Pls. 28.6; 29.1). It is a new
point of consideration. Why would they want a semi-circular / apsidal back? It
was strange because such an apsidal back is not found in such naoi of the fifth
and sixth century (Pls. 8–13) that are flat-roofed and quadrant, no matter what
the site (Table 1). It is also not there in the older Ajanta Cave 9 (Pls. 6.1; 24.1).
The plan of the four worship halls were based on archetypal prototypes with
a barrel vault over the nave and hemispherical ceilings at the back (Pls. 6:1–2;
7–8; 24). Three of them have apsidal floors: The Ajanta caves 10, 19, and 26
(Pls. 6.1; 7; 8). However, the floor of the Ajanta Cave 9 (Pl. 6.2) is not apsidal
but angular at the back. Against such variations what was the logic or rational
of carving a flat and apsidal ceiling and an apsidal floor? Why this unique
configuration? We have no answer. We cannot explain why the first stupa-shrine
Yaṣṭi (Skr.): staff, stick, flag-staff, stem, pole, pillar, etc. (Monier-Williams 2002, 848).
Yūpa (Skr.): a sacrificial post or stake, column erected in honour of victory, pillar, trophy, etc.
(Monier-Williams 2002, 856).
92 The fact that there was no such ‘handbook’ in the initial years has also been indicated by Spink
(S09, 12), (S17, 3).
90
91
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
323
in about 133 years had to be so adventurous.93 The only conclusion we are able
to deduce under the circumstances is that the workers – and supervisors if there
were any – were novices; they were working without a ‘handbook.’ They did
not have a clear procedure in place. That is why they were doing odd things.
As to why they did not create any yaṣṭi, they must not have spotted the yaṣṭi
in the adjacent Cave 10 (Pl. 24.3). The yaṣṭi is visible today but it was surely
not visible then. It was hidden or concealed under the woodwork and paintings,
which must have completely covered it. All the wooden beams and rafters, and
the layers of the plasters or paintings on them are gone now. So, we are able to
see the yaṣṭi today.
If we inspect the chamber of our Cave 11, more evidence of poor workmanship
is visible. The walls have not been properly levelled. In fact, the left wall is
crooked; it also has an accidental hole through which you can peep into the
adjacent cell. The meḍhī (drum) is not exactly perpendicular; it is tilted at places;
the circumference is greater on the lower side than the upper side (Pl. 29.2). The
shape of the dome also exhibits lack of perfection.
The only items displaying fine workmanship are the sculptures: The Buddha
figure, the gandharva figures, and the kneeling worshipper (Pl. 28.6). Why is that
so? Was there a different hand working on them? If you investigate the matter,
you will come to the same conclusion as Spink and I did that the images were
not planned at all when the stupa was being done. They belong to a different date
in the future.94 They were going to be carved about two years later in c. 468
(Table 2.7). They were not even conceived in c. 465
when the stupa was
being excavated.
A
(
N ?)
So, leaving the sculptures aside, let us now attend to the area behind the stupa,
which looks like an incomplete semi-circular ambulatory (Pls. 26.1; 28.6; 29.1).
The floor over there is raised up. Crude chisel marks are all over: on the floor,
walls, and back of the stupa. There is absolutely no sign of completeness; no race
of plaster or painting is there. The rear space behind the stupa is not angular
but semi-circular, or apsidal, whatever you call it. In this respect, it is like the
rear ambulatory of the adjacent Cave 10 (Pls. 6.1; 16; 24.2). It has already been
suggested that the very conception of our Cave 11 was to serve as a residential
93
94
The gap was due to Hiatus I (Pls. 1.3, 3.3), vide S20b (12–14).
S07 (150–152).
324
Rajesh Kumar Singh
adjunct or companion to the older Cave 10,95 which has an apsidal ambulatory.
This is a notable point, for the other ancient temple, Cave 9, has a quadrant floor
plan; the rear of the stupa is angular; it is not apsidal (Pl. 6.2). We can deduce
that the planners of our Cave 11 were looking at Cave 10 as a model, and not
Cave 9. Now let us come to the question whether what looks like an ambulatory
is indeed an ambulatory meant for circumambulation. The question is raised
from the viewpoint of function, which as we know always dictates the form of
architecture. Form-follows-function is a principle in design discipline, architecture,
mechanics, structuralism, and even life sciences.
If you survey the leṇa-cetiyagharas up to Period IV you will find a startling
fact that a majority of them does not have any ambulatory (Pls. 5–10). The
few monuments that do have ambulatories belong to Period I.2 (Pl. 5) and II.1
(Pl. 6.7–17; 7.1–8). From Period II.2 to IV (Pls. 7.9–11; 8–9) no ambulatories
were generally made. However, the ambulatory returned from the Kuchean
Period III (Pl. 11) and the Indian Periods V(1) and V(2) (Pls. 12–13). It did
so as a quadrantile ambulatory type: in Kucha as well as in India. The apsidal
format was avoided.96 Thus the whole approach to ambulatory in terms of form
and function was changed radically in the said later periods.
Let us come now to the question of use or usage. From the beginning up to
Period IV, some surprising observations could be made. You will find that most
of the so-called ambulatories were neither used nor perhaps meant to be used.
Although some ambulatories might have been used through Period I (Pl. 5.6–19)
it was not the case through Periods II and III (Pls. 6–10). Consider, for example,
that there are altogether 38 flat-roofed and quadrantile naoi across the fifth-century
rock-cut sites (Pls. 6.10–36; 25–27). Out of them only 8 (about 22.2%) have the
so-called ambulatories.97 The remainder 30 naoi (about 77.7%) do not have any
such thing. Even in the caves where it is present there is a different logic to why
it is there. The logic is that those naoi were initiated for the stupa, and not for the
anthropomorphic Buddha. The form of the stupa is such that it naturally generates
or expects a circular path or enclosure around it, which may or may not be used
for circumambulation. However, during Period II, when the stupas, or plans of the
stupas, were replaced by Buddha images (c. 468 onwards), there was no longer
carved any ambulatory afterwards. Moreover, those ambulatories that were under
excavation, the work in them was aborted and never resumed (Pls. 6.18–20; 7;
95
96
97
F80 (295).
An exception was made in the case of Ellora Cave 10 (F80, pl. LXII).
Ajanta caves 11, 16, Lower 6, and 17 (Pl. 6: figs. 10, 15–17; Pl. 7: figs. 1, 6–8).
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
325
9–11). Conclusion? The Buddha, it seems, did not require circumambulation;
it was not a necessary ritual. The fact is applicable to all the 24-plus sites that
were developing during Periods II and III (Table 1; Pls. 8–10). It also applies
to Nagarjunakonda98 (Pl. 5.9–10). Even in such shrines where the work on the
ambulatories had been completed before 468 , i.e. before the arrival of the
Buddha image cult, there is evidence that those ambulatories were never put to
use (Pl. 7).99 In the later course, because either the stupa was completely replaced
or supplanted by the Buddha image, or the Buddha image was carved on the front
faces of such stupas as in Caves 19 and 26 (Pls. 24.4–5), the Buddha figure had
become the central object of veneration, i.e. the main divinity (from c. 468
onwards). The stupa was no longer enjoying the status of the main divinity. With
the stupa gone, the purpose and function of circumambulation was also gone.
Next we observe as if the image did not require any circumambulation. We are
led to deduce by the pattern that even in the apsidal temples – Caves 19 and 26
(Pls. 17; 24.4–5) – no such ritual would have been performed.
Such aspects may be observed even in the earlier times. There were hardly
such ambulatories that had convenient circumambulation paths. Most of them
have constricted spaces; passage through them would not have been easy;
certainly not for groups of people on daily basis (Pl. 5); lighting is so abysmal.
Most ambulatories do not have plaster or paintings. Only rarely the paintings are
found on the rear walls, as in the Ajanta Cave 9. It is clear that the ambulatory
areas were not given a priority by the fifth-century painters. So, the question
surfaces again: If these are the facts then why the so-called ambulatories were
made at all? If you carefully inspect the published plans or physically survey the
in situ evidence it would be revealed that the so-called ambulatories were actually
the results of a process; they were by-products of a procedure, i.e. the procedure
involved in carving a stupa. This will be understood when we inquire why the
‘ambulatory’ of our Cave 11 was never completed (Pls. 28.6; 29.1–2). The fact
is that the planners were least concerned about anything called ambulatory. They
were rather focussed on the stupa. While carving the rear part of the stupa inside
a monolithic chamber, the space behind was automatically created. So, it is simply
a round space without any intended function. Later on, when the stupa cult was
replaced by the Buddha cult, they carved the Buddha image, the gandharva
figures, and the worshipper on the front face of the stupa (Pl. 28.6). However,
98
99
Weiner (1977, 44–46).
S07 (39, 154–155, 157).
326
Rajesh Kumar Singh
they did not even touch any more the incomplete round space at the back. At
this point a counter argument may be made that the work was perhaps halted so
that the ambulatory was never completed.
The argument cannot be valid, for they had plenty of time to complete the
so-called ambulatory had they really wanted. They had plenty of time to complete
the images and paint them too, not only once, but two times. There are two
layers of paint (Pl. 29: 1, 3). If you further inspect the naos as well as the
entire cave, it will be found that the cave temple was also put to prolonged
use and worship.100 The cave was plastered (Pl. 28.5), and many areas were
also painted (Pl. 28) – albeit in different times.101 It is true that the work was
repeatedly halted in the cave for various reasons. But there was always enough
time, especially when the cave was in use for many years, for the ambulatory to
be completed had they really wanted. The reason why it was never completed
was only because the circumambulation was no longer or never a necessary ritual
for the Śrāvakayānists of Middle India and the Deccan that created the rock-cut
caves. Pradakṣiṇā (circumambulation) was not the real motive or function of the
space at the back of the stupas. It is the same situation that will be found in all
the other naoi of the times wherever there seems to be an ambulatory (Pl. 7.1–8).
M
S
-A
A point should now be noted that there is no pronaos or shrine-antechamber in
some of the cave temples including our Cave 11 (Pls. 7.1–4). Why is that so?
It has been recently suggested that the concept of pronaos was brought to India
– or, more precisely, was re-introduced south of Vindhyachal – during the last
quarter of c. 466
(Table 3.6).102 However, the stupa-shrine of our Cave 11
was already excavated in the previous year (late 465 ). At that time no one
had heard about any such thing. No one had known about some of the pronaoi
that existed in the earlier times, e.g. in the Nasik Cave 17,103 Kuḍā Cave 29,
and Kuḍā Cave 24104 (Pl. 5: 1, 14–15). Belonging to the second century , the
S07 (149, 157, 159, 160).
According to Spink, the paintings of Cave 11 belong to four different dates: 468, early 469,
early 478(?), and mid-478–480 , vide S07 (393), S17 (418).
102 S20c (124).
103 S20c (120–122).
104 In his article on shrine-antechambers, Singh (2020c, pl. 4.13–14) reproduced the plans of the
Kuḍā caves 29 and 24 (Pl. 5.14–15) but he made no discussion on them at all. That is a notable error
in his research.
100
101
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
327
excavation of the Nasik cave was aborted even before the shrine could be fully
completed. Because the intended relief stupa in the shrine of that cave was never
actually carved, the edifice would not have been consecrated or worshipped. The
cave must have been abandoned. So, no wonder if nobody knew about it, or about
the experimental shrine-antechamber inside it.105 However, no such justification
can be made with respect to the Kuḍā caves 29 and 24 (Pl. 5.14–15). So, it
does seem like the earliest planners of Period II in Ajanta, Bagh, Dharashiva,
and Pālé were really ignorant about the antechambers of the said Kuḍā caves.
It was only for this reason that the earliest naos of Period II (Pl. 7.1–4) do not
have any pronaos.
As to the antechamber of the contemporary Bagh Cave 2 (Pls. 6.12, 7.2, 15,
18.2, 27.5), it was still not conceived at the time (late 465 ).106 According to our
study that pronaos was carved during the last quarter of c. 466 , i.e. about one
year later (Table 3.6). In late 465 , the masons in the Bagh Cave 2 would still
have been busy in excavating the hall; they were still engaged in scooping out the
rear part of the hall, which is so huge. It was not until the last quarter of 466
that there was introduced in that cave the concept of the shrine-antechamber.107
The harbingers of the novel concept were the migrants that kept pouring into
Middle India and the Deccan from the ravaged lands of Bactria and Greater
Gandhāra. They were war refugees. In their homelands, shrine-antechambers were
common in Buddhist complexes and other temples.108
C
This study expands the horizons of our knowledge about fifth-century South Asia.
Some aspects and factors head not come to our notice as far as the published
scholarship is concerned. By examining a single rock-cut monument of the fifth
century—and within that just the inner shrine of that monument—we are able
to understand and observe a much wider picture as to what was the cultural,
political, military, and religious background. It is a completely new background
that had escaped our attention.
S20c (120–122).
Cf. a different view of Weiner (1977, 42).
107 S20c (122–128).
108 S20c (128 n. 36). One of the earliest pronaoi closer to India is found in the Jaṇḍiāl Temple in
Taxila (Pl. 20.1) dated to the Scytho-Parthian epoch (S. J. Marshall 1936, 105, pl. XVI).
105
106
328
Rajesh Kumar Singh
The study revealed that the inner shrine of the Ajanta Cave 11 was started
in circa 460
like numerous other rock-cut monuments when Vākāṭaka Hari
Ṣeṇa was ruling in northern Deccan and Skandagupta in northern India. After
some years when the cave excavation was still in progress, there started troubles
in Bactria and Greater Gandhāra. The Hephthalites had pushed out the Alchon
/ Alkhan Huns around c. 465 . Due to that disturbance Buddhist migrations
started from Bactria and Greater Gandhāra. Some of the displaced people went
inter alia towards Central Asia and Middle India. These lands were safe, at least
during those years. The mass migration has been called here the Gandhāran
Buddhist Exodus I of 465–477 .
When the monastics came to India they introduced many facets of the Gandhāran
culture and traditions. One of them was the Gandhāran architectural tradition
that was being followed in the monasteries over there. There was a particular
system of architecture, which was all-inclusive in terms of functionalities. The
three major functions of monastic dwelling, congregation, and worship coexisted
under the same roof. This system of architecture did not exist in Middle India
and the Deccan until that time. Because more than 17 rock-cut edifices in Ajanta
alone, and many more at other sites, were being developed, which were not trifunctional, but bi-functional. They were either places of worship and congregation
or of congregation and dwelling. It was after the refugees that came over there that
the idea of the tri-functional architecture, or a composite or hybrid architecture,
was introduced in Middle India (Bagh) and the Deccan (Ajanta, etc.). As a result,
there began many experiments throughout middle India and the Deccan. All the
patrons who were involved in making such architecture started to add a naos or
shrine to the ongoing excavation of the dormitories.
This article shows how this difficult task was achieved; what were the
problems encountered; what the challenges were; what solutions were arrived
at; how this task was practically executed. It also tells us that within a few years
there came a second wave of mass migrations from the same regions, which has
been called here the Gandhāran Buddhist Exodus II of 468 .
During Exodus II more new ideas were introduced. One of them was the cultic
worship of the Buddha image, which came from the Mūlasarvāstivāda School.
As a result, the shrine of our Ajanta cave 11, the work on which had been halted
due to many mistakes and errors, sprang to life again. There was the new idea
that the shrine should now have the Buddha figure instead of the stupa. However,
the problem was how to carve a Buddha figure in a monolithic setup when the
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
329
shrine was already excavated with a half-carved stupa? They did not want the
stupa anymore for sure. But the problem was that they could not have removed
the stupa altogether. Where else would they carve the image? Some rock, in fact,
enough amount of rock was needed. So, it was a choice to be made. The planners
were forced to retain the stupa, the block of the stupa, up to whatever extent that
it was carved. Then, they carved out the Buddha and attendant figures on the
frontal face of the stupa and left the backside untouched. They did it so smartly
that it is extremely hard to deduce the different layers of activities. The same
thing was being done in every other shrine of every other edifice including the
Ajanta caves 19 and 26. This is a new way of looking at things; of looking at
Ajanta; of looking at the architecture of the times; of decoding the cultic practices
from the rocks – the things that the coins, scriptures, and inscriptions did not
tell us about. Last but not the least, it is a new way of reading the fifth-century
South Asia when religious piety and “cauldrons” co-existed; when the Buddhism
of Middle India and the Deccan was changed forever; when the stupa cult had
given way for the image cult.
330
Rajesh Kumar Singh
T
Table 1. Working inventory of the Indian rock-cut sites/monuments
during Periods II–IV
Quantum of caves
S.No.
Site
Region
Religion
P
I P
II–IV
(c. 250 (c. 460–480 )
T
cradled
–325 ) by Hari Ṣeṇa
1.
Ajanta
Ṛṣika
Buddhist
6
24
30
2.
Ankai-Tankai
Ashmaka/
Trikūṭa
Jain, Hindu
0
4a
4
3.
Aurangabad
Mūlaka?
Buddhist
3
3
6
4.
Bagh
Valkhā
(Avanti)
Buddhist
1 (lost?)
7
8
5.
Banoṭī
Ṛṣika
Buddhist
0
1
1
6.
Deogarh rock Middle
shelter, U.P. India
Buddhist
0
3b
3
7.
Dhamnar,
M.P.
Avanti
Buddhist
11
3
14c
8.
Dharashiva
Aśmaka/
Kuntala?
Jain
10
4
14
9.
Elephanta
caves 2–4
Aparānta
Buddhist
→ Śaivite?
0?
3
3
10.
Ellora caves
27, 20A
Mūlaka?
Buddhist→
Śaivite?
0
2
2
11.
Ghaṭotkaca
Ṛṣika
Buddhist
0
3
3
12.
Junnar
Aparānta
Buddhist
100
1
101
13.
Kanheri
Aparānta
Buddhist
87
33
120d
14.
Kerūmātā
Aparānta
Buddhist
0
1
1
15.
Khaṃbhālīdā
Lāṭa / Saurashtra
Buddhist
5
1
6
16.
Khejadiya
Bhoop, M.P.
Avanti
Buddhist
?
?
28
Circa 465
331
and The Stupa-Shrine of the Ajanta Cave 11
Quantum of caves
S.No.
Site
b
c
d
e
f
Religion
P
I P
II–IV
(c. 250 (c. 460–480 )
T
cradled
–325 ) by Hari Ṣeṇa
17.
Kondavite
Aparānta
Buddhist
3
16
19e
18.
Lamgaon
Caves, Goa
Kuntala
Buddhist,
Śaivite
0
2
2
19.
Loṇāḍ
Aparānta
Buddhist
0
1
1
20.
Maṇḍapeśvar Aparānta
Buddhist→
Śaivite?
0
2
2
21.
Nasik
Trikūṭa
Buddhist
20
5
25
22.
Pālé
Trikūṭa
Buddhist
28
1 (adapted)
28f
23.
Pitalkhora
Mūlaka
Buddhist
10
5
15
24.
Shānā Vākiyā Saurashtra
Buddhist
65
?
65
349
125
501
Total
a
Region
There are eight Jain caves, one of indeterminate affiliation, and one Hindu. Prima facie it
appears to the present author that at least some caves should belong to fifth century ce, or they
had a 5th-century phase, while the other caves are of later period.
There is no rock-cut architecture as such. They are rock-shelters with Buddhist images: carved
and painted. There is also a masonry stupa nearby (Joshi 2017).
ASI (2015, 5).
N81 (190), cf. Pandit (2012, 71).
N81 (237).
N81 (250).
Note: Kotra, M.P. had some related developments (Manuel, forthc.). The caves of other periods or
the sites of other regions are outside the scope.
332
Rajesh Kumar Singh
Table 2. Four Buddhist mass migrations / exoduses across South
and Central Asia between c. 465 and 477
First occurrence
of a feature
(a)
Quadrant and
flat-roofed layout
(dwelling +
congregation)
(b)
Apsidal and
vaulted (worship +
congregation)
S.No.
Date
Exodus
Area
1.
460 –
mid-465
–
Middle India and
the Deccan
–
2.
465–467
Exodus I
Bactria & Gandhāra to
many different lands
including Kucha and
Middle India
3.
Late
465
–
Middle India and the
Deccan
Tri-functional layout
(dwelling + congregation
+ worship)
–
Ditto
Inner shrine with stupa
–
Ditto
shrine-antechamber
4.
5.
Late
465–467
Late
466
6.
468
Exodus II
7.
468
–
8.
9.
Bactria & Gandhāra to
many different lands
including Kucha and
Middle India
Middle India and
the Deccan
–
Buddha figure,
parinirvāṇa theme
Exodus III
= Spink’s
469–472
‘recession’
and ‘hiatus’
Middle India to other
lands including Kucha
Bodhisattvas in Cave 17
Exodus IV
Middle and western
India to Central Asia,
China, Tibet, East Asia,
and Sri Lanka (?)
Bodhisattva,
Avalokiteśvara, avadānas
477
Cf. Spink’s chronology (Pl. 15).
Circa 465
333
and The Stupa-Shrine of the Ajanta Cave 11
Table 3. Chronology of select rock-cut naoi and pronaoi
during “Indian Period II”a
a
b
S.No.
Cave
Inception of
façade / porch
Inception of shrine
Inception
of shrineantechamber
1.
Ajanta 11
460
465, quarter 4
–
2.
Bagh 4
460
466, quarter 1
–
3.
Dharashiva
2
463
466, quarter 2–3
–
4.
Pālé 1
3rd–2nd c.
466, quarter 3–4
–
5.
Ghaṭotkaca
cave
465?
466, quarter 3–4 (relief
stupa in the front aisle)b
–
6.
Bagh 2
461
---------
466, quarter 4
7.
Ajanta 16
461
---------
467, quarter 1
8.
Ajanta 6
Lower
461
---------
467, quarter 2
9.
Ghaṭotkaca
cave
465?
---------
467, quarter 3–4
Cf. S14 (xii), (Pl. 15). After S20c (138).
Cf. c. ‘469’ : S05 (359).
334
Rajesh Kumar Singh
Table 4. Some known donors of the cave temples of Ajanta,
Ghaṭotkaca, Bagh, and Pālé
Site/cave
Main patrons as per
inscription(s) / interpretation
References
Ajanta
Cave 4
‘Māthura…’
C06 (284), Sircar (1959–1960, 26)
Ajanta
Cave 10
‘Vāsithīputa,’ ‘Kanhaka,’
and ‘Dharmadeva’
C06 (295–297), B83a (116),
Ghosh (1967, 244),
Dhavalikar (1968, 149)
Merchant ‘Ghanāmadaḍa’
C06 (309), B83a (116)
Western Vākāṭaka’s Prime
Minister ‘Varāhadeva’
King ‘Dharādhipa’
and his younger brother
‘Ravisāmba’
who died untimely
v. 30: C06 (312), M63 (109, 111),
B83a (126, 127)
Ajanta
Cave 12
Ajanta
Cave 16
Ajanta
Caves
17–20
‘Upendragupta’
Ajanta Cave
26-Complex
Ghaṭotkaca
cave
Pālé
Cave 1
‘Monk Buddhabhadra’
Western Vākāṭaka’s Prime
Minister ‘Varāhadeva’
Phase I (Period I): unknown.
Phase II (Period II):
Varāhadeva?
v. 8–9: M63 (125, 128), S97 (47),
cf. C06 (320)
Spink (1975, 143 n. 2), W07 (47),
W09 (45–48), W17 (416–417)
v. 13: C06 (334),
Yazdani (IV (1955): 116, 118)
v. 18: C06 (339), M63 (117, 119)
S21 (30 n. 124)
Circa 465
335
and The Stupa-Shrine of the Ajanta Cave 11
Table 5. Quantum of Indian rock-cut monuments with and without a naos,
up to Period VII
Period
Caves without sanctum sanctorum
Caves with sanctum sanctorum
I
1450
50
II
10
60
III
0
14
IV
0
0
V-A
0
6
V-B
0
20
VI
0
200
VII
0
300
1460
650
TOTAL
The figures are author’s guesstimate in absence of published data.
Table 6. Index to Pls. 5–13 / development of Buddhist naos
from Period I.2 to Period V(2)
Plates 5–10
Naos type
Type S5
Pl. 11
→
Architecture
LMC2 + variants →
type
Period
Date
Region
Indian Period
I.2–IV
Circa
120–480
→
Pl. 12
Pl. 13
Next stage
→
Next stage
→
Next stage
MC4a
→
MC4b
→
MC4c
Kuchean
Periods
II–III
→
Indian
Indian
→
Period V(1)
Period V(2)
→
Circa
533–570
→
Circa
466–550
W. India, Middle
India, and the
→
Deccan
Kucha
→
→ Aurangabad →
Arrow ‘→’ indicates evolution, transformation, or transmission.
Circa
550–600
Aparānta
& Deccan
336
Rajesh Kumar Singh
Table 7. Other sites not covered in Plates 20–22
Name of Site / Monument
Ref./Plan
G
Mostly Pakistan
Paṭvano Gaṭai rock shelter
-
Shah-ji-ki-Dheri
-
Taxila
Bagh Gai
K02 (pl. 28)
Bhamāla
M51 (pl. 114)
Chir Tope A and Chir Tope B, Khādeṛ Mohṛā (Akhaurī)
M51 (pls. 67–69)
Court D at Tareli
K02 (pl. 7)
Giri
M51 (pls. 81–82)
Jamāl Gaṛhī
K02 (pl. 9)
Jaulian
M51 (pl. 101), K02 (pl.
5)
Jinan Wālī-kī-Ḍherī
-
Mekhkha Sanda
K02 (pl. 8)
Mohṛā Morādu
M51 (pl. 93)
Pippala
M51 (pl. 98)
Shotorak
K02 (pl. 20)
Sirkap
K02 (pl. 1)
Takht-i-Bāhi
K02 (pl. 6)
B
Uzbekistan, Termez
Balalyk Tepe
-
Afghanistan
Mes Aynak
-
Tapa Sardār
KH11 (fig. 8)
Circa 465
337
and The Stupa-Shrine of the Ajanta Cave 11
Name of Site / Monument
X
/C
Ref./Plan
T
Dandan Oilik # 12, 13
KH11 (fig. 11)
Ming-oi near Shorchuk
KH11 (fig. 10)
Miran
A
This study was conducted under Senior Fellowship scheme of CCRT, Ministry of Culture,
Govt. of India (2019–2021; File # CCRT-SF-3-357-2020). However, the spade work
was carried out earlier when I was a Consulting Art Historian with Dharohar, SML,
Udaipur (2017–2019). At that time, Prof. Monika Zin (Leipzig University; and Saxon
Academy of Sciences in Leipzig) had suggested me to investigate the inner shrines of
certain Ajanta caves vis a vis the so-called ‘central pillar caves’ of Kucha. She also
suggested me to probe whether the ‘Hephthalites,’ as she put it, had anything to do
with the destruction of the Guptas, Vākāṭakas, and abandonment of the Ajanta caves.
The outcome of my probe is not something, which she is willing to endorse (personal
communication). Nevertheless I would like to thank her for many scholarly exchanges.
I would like to thank Ms. Latha Kurien Rajeev who provoked me repeatedly some years
back: “Who are those foreigners painted on the ceilings of Ajanta?”. Thanks are also due
to Dr. Manuel J. (DSA, Bhophal Circle, Archaeological Survey of India) for stimulating
discussions, and to Dr. Rajendra Yadav (SA, Mumbai Circle, ASI) for providing the site
plans of the Kondavite and Pāḷé caves.
A
B81
B97
B78
B83a
B82b
C06
Dh84
F80
F88
HG
F
Bhandarkar, et al. (1981)
Bakker (1997)
Burgess (1878)
Burgess (1883a)
Burgess (1883b)
Cohen (2006)
Dhavalikar (1984)
Fergusson and Burgess (1880)
Fleet (1888)
Bakker (2019)
R
W
338
K02
KH11
M51
M63
N81
SB
SBC
S12a
S12b
S18
S20a
S20b
S20c
S21
S65
S97
S05
S06
S07
S09
S14
S17
W05
XQSY
Rajesh Kumar Singh
Kuwayama (2002)
Kim and Han (2011)
Marshall (1951, Vol. III)
Mirashi (1963)
Nagaraju (1981)
Sourcebook, Balogh (2020)
Sourcebook Companion, Bakker (2020)
Singh (2012a)
Singh (2012b)
Singh (2018)
Singh (2020a)
Singh (2020b)
Singh (2020c)
Singh (2021)
Sircar (1965)
Shastri (1997)
Spink (2005)
Spink (2006)
Spink (2007)
Spink (2009)
Spink (2014)
Spink (2017)
Willis (2005)
Xinjiang Qiuci Shiku Yanjiusuo (2000)
R
Archaeological Survey of India. Archaeological Remains: Monuments & Museums. II vols.
Archaeological Survey of India, 1964.
Archaeological Survey of India. Dhamnar: Art, Architecture, and an Innovative Conservation
Module. Brochure, ASI, Bhopal Circle, 2015.
Ashraf Khan, M. and Mahmood-ul-Hassan. „Discovery of Mural Paintings from Jinnan Wali
Dehri, Taxila.” Journal of Asian Civilizations, vol. XXVII, 2004, pp. 14–27.
Bakker, Hans T. The Vākāṭakas: An Essay in Hindu Iconology. Egbert Forsten, 1997.
Bakker, Hans T. Holy Ground, Where Art and Text Meet: Studies in the Cultural History of
India. Brill, 2019.
Bakker, Hans T. The Alkhan: A Hunnic People in South Asia. Companion to Hunnic Peoples
in Central and South Asia: Sources for their Origin and History. Barkhuis, 2020.
Balogh, Dániel, editor. Hunnic Peoples in Central and South Asia: Sources for Their Origin
and History. Barkhuis, 2020.
Barthoux, J. Excavations at Hadda. (Site 404), 1927.
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
339
Behrendt, Kurt. The Art of Gandhara in the Metropolitan Museum of Art. New York, The
Metropolitan Museum of Art, 2007.
Bhandarkar, Devadatta Ramakrishna, Bahadurchand Chhabra, and Govind Swamirao Gai.
Corpus Inscriptionum Indicarum. Vol. III (rev.). Inscriptions of the Early Gupta Kings.
Rev. Archaeological Survey of India, 1981.
Brancaccio, Pia. The Buddhist Caves at Aurangabad: Transformations in Art and Religion.
Brill, 2011.
Brancaccio, Pia. 2018. „The Silk Road and the Cotton Road: Buddhist Art and Practice
Between Central Asia and the Western Deccan.” Vol. II. Archeologia Delle Vie Della
Seta: Percorsi, Immagini e Cultura Materiale, edited. by B. Genito and L. Caterina,
Scienza e Lettera Editore, 2018, pp. 57–73.
Brancaccio, Pia and Kurt Behrendt. Gandhāran Buddhism. UBC Press, 2006.
Burgess, James. Report on the Antiquities in the Bidar and Aurangabad Districts in the
Territories of His Highness the Nizam of Haidarabad…. W. H. Allen, 1878.
Burgess, James. Archaeological Survey of Western India. Vol. IV. Report on the Buddhist
Cave Temples and Their Inscriptions…. Trubner & Co., 1883a.
Burgess, James. Archaeological Survey of Western India. Vol. V. Report on the Elura Cave
Temples and the Brahmanical and Jaina Caves in Western India…. Trubner & Co., 1883b.
Callieri, Pierfrancesco. Seals and Sealings from the North-west of the Indian Subcontinent and
Afghanistan (4th century B.C. – 11th Century A.D.): Local, Indian, Sassanian, GraecoPersian, Sogdian, Roman. Naples, Istituto Universitario Orientale, 1997–1998.
Chakrabarti, K. „The Gupta Kingdom.” History of Civilizations of Central Asia. Vol. III.
The Crossroads of Civilizations: A.D. 250 to 750., edited by B. A. Litvinsky, UNESCO
Publishing, 1996, pp. 188–210.
Ciordia, José M. „The Ship in the Cave: The Greek and Nautical Origin of Buddhist
Architecture.” Journal of Asian Architecture and Building Engineering, XIX (1), 2020,
pp. 48–69.
Cohen, Richard S. „Ajanta’s Inscriptions.” Ajanta: History and Development. Vol. II. Arguments
About Ajanta., edited by Walter M. Spink, Brill, 2006, pp. 273–339.
Compareti, Matteo. „Some Examples of Central Asian Decorative Elements in Ajanta and
Bagh Indian Paintings.” The Silk Road, vol. XII, 2014, pp. 39–48.
Dehejia, Vidya. Early Buddhist Rock Temples: A Chronological Study. Thames and Hudson,
1972.
Dhavalikar, M. K. „New Inscriptions from Ajanta.” Ars Orientalis, vol. VII, 1968, pp. 147–153.
Dhavalikar, M. K. Late Hinayana Caves of Western India. Deccan College Post Graduate
and Research Institute, 1984.
Divakaran, Odile. „Avalokiteśvara — From the North-West to the Western Caves.” East and
West, no. 1/4, 1989, pp. 145–178.
Fergusson, James and James Burgess. The Cave Temples of India. W. H. Allen & Waterloo,
1880.
Fleet, John Faithfull. Corpus Inscriptionum Indicarum. Vol. III. Inscriptions of the Early
Gupta Kings and Their Successors. Superintendent of Government Printing, India, 1888.
Ghosh, Amalanand. „No. 43––Two Early Brahmi Records from Ajanta.” Epigraphia Indica,
vol. XXXVII, 1967, pp. 241–244.
340
Rajesh Kumar Singh
Gokhale, Shobhana. Kanheri Inscriptions. Deccan College Post Graduate and Research
Institute, 1991.
Gokhale, Shobhana. „Epigraphical Evidence on Harisheṇa’s Conquest from Kānherī.” The
Age of the Vākāṭakas, edited by Ajay Mitra Shastri, Harmann Publishing House, 1992,
pp. 269–278.
Howard, Angela F. and Giuseppe Vignato. Archaeological and Visual Sources of Meditation
in the Ancient Monasteries of Kucha. Brill, 2014.
Joshi, M. C. „Rock Shelters at Deogarh, District Lalitpur, UP: Certain Idioms of HinduBuddhist Legacy.” Kalash, vol. XIX–XXII, 2017, pp. 20–27.
Kak, Subhash. „Early Indian Architecture, Temples, and Art.” Migration & Diffusion, VI
(23), 2005, pp. 6–27.
Khan, M. A. and M. Hassan. „A New Discovery in the Taxila Valley: Archaeological Excavations
at the Buddhist Monastery of Jinan wali Dheri.” Gandhara: The Buddhist Heritage of
Pakistan. Legends, Monasteries, and Paradise, edited by Dorothee von Drachenfels and
Christian Luczanits, Kunst- und Ausstellungshalle der BRD, 2008, pp. 302–307.
Khan, Nasim K. „Fresco Paintings from Yakatoot (Peshawar) Gandhāra.” Annual Report
of the International Research Institute for Advanced Buddhology at Soka University,
vol. XIX, 2016, pp. 47–56.
Kim, Young Jae, and Dong Soo Han. „Evolution, Transformation, and Representation in
Buddhist Architecture: The Square Shrines of Buddhist Monasteries in Central Asia after
the Fourth Century.” Architectural Research, XIII (4), 2011, pp. 31–42.
Kurbanov, Aydogdy. „The Hephthalites: Archaeological and Historical Analysis”. PhD Thesis,
Berlin: Free University Berlin, 2010.
Kuwayama, S. Across the Hindukush of the First Millennium: A Collection of the Papers.
Kyoto University, 2002.
Leese, Marilyn. „The Earliest Buddha Icon in Kanheri Cave 3.” Artibus Asiae, vol. XLI (1),
1979, pp. 83–93.
Lo Muzio, Ciro. „Remarks on the Paintings from the Buddhist Monastery of Fayaz Tepe
(Southern Uzbekistan).” Bulletin of the Asia Institute, vol. XXII, 2008, pp. 189–206.
Lo Muzio, Ciro. „Notes on Gandharan Painting.” Glimpses of Indian History and Art:
Reflections on the Past, Perspectives for the Future, edited by Tiziana Lorenzetti and
Fabio Scialpi, Sapienza Universita Editrice, 2012, pp. 319–335.
Lo Muzio, Ciro. „The legacy of Gandhāra in Central Asian painting.” In the Shadow of the
Golden Age: Art and Identity in Asia from Gandhara to the Modern Age, edited by Julia
A. Hegewald, EB Verlag, 2014, pp. 115–138.
Majumdar, Ramesh Chandra and Anant Sadashiv Altekar. The Vākāṭaka-Gupta Age (Circa
200–550 A. D.). 2nd. Motilal Banarsi Dass 1954.
Marshall, John. The Bagh Caves in the Gwalior State. India Society, 1927.
Marshall, John. A Guide to Taxila. 3rd. Manager of Publications, 1936.
Marshall, John. Taxila: An Illustrated Account of Archaeological Excavations... Between 1913
and 1934. III vols. Cambridge University Press, 1951.
Mirashi, V. V. Corpus Inscriptionum Indicarum. Vol. V. Inscriptions of the Vākāṭakas.
Government Epigraphist for India, 1963.
Monier-Williams, Monier. A Sanskrit-English Dictionary. 2nd. Motilal Banarasidass Publishers,
2002.
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
341
Nagaraju, S. Buddhist Architecture of Western India, c. 250 B.C.–c. A.D. 300. Agam Kala
Prakashan, 1981.
Pandit, Suraj. The Age of the Traikutakas: Coins, Inscriptions, and Art. Agam Kala Prakashan
2012.
Ramachandran, T. N. Memoirs of the Archaeological Survey of India. No. 71. 1938.
Nāgārjunakōṇḍā. Archaeological Survey of India, 1953.
Rienjang, Wannaporn and Peter Stewart. Problems of Chronology in Gandhāran Art.
Archaeopress Archaeology, 2018.
Schlingloff, Dieter. Guide to the Ajanta Paintings. Vol. I: Narrative Wall Paintings. Munshiram
Manoharlal Publishers, 1999.
Schlingloff, Dieter. Ajanta: Handbook of the Paintings. Vol. I. Narrative Wall-Paintings.
Translated by Miriam Higgins and Peter Haesner. III vols. IGNCA, 2013.
Shastri, Ajay Mitra. Vākāṭakas: Sources and History. Aryan Books International, 1997.
Singh, Rajesh Kumar. „The Early Development of the Cave 26-Complex at Ajanta.” South
Asian Studies, vol. XXVIII (1), 2012a, pp. 37–68.
Singh, Rajesh Kumar. An Introduction to the Ajanta Caves: With Examples of Six Caves.
Harisena Press, 2012b.
Singh, Rajesh Kumar. „Ajanta’s Antiquity: Sources and Problems. Vol. I–II.” PhD Thesis,
Dept. of Art History & Aesthetics, Faculty of Fine Arts, M. S. University of Baroda,
Vadodara 2014.
Singh, Rajesh Kumar. „Rock-cut architecture of Western India, 1. Momentum I: origins of
‘leṇa cetiyagharas,’ ca. 120–300 CE.” History Today, the Journal of the Indian History
& Culture Society, vol. XIX, 2018, pp. 216–231.
Singh, Rajesh Kumar. Khīṅgīla vs. Buddhist Caves: A Synchronised Chronology of the Early
Alchon Hūṇs, Early Guptas, Vākāṭakas, Traikūṭakas, and Buddhist Caves (ca. 451–480 CE)
= Ajanta Mahāpiṭaka I.2. Hari Sena Press, 2020a.
Singh, Rajesh Kumar. Periodisation of Rock-cut Monuments of India. = Ajanta Mahāpiṭaka
I.3. Hari Sena Press, 2020b.
Singh, Rajesh Kumar. „The Earliest Two and a Half Shrine-antechambers of India.” Rocznik
Orientalistyczny / The Yearbook of Oriental Studies, vol. LXXIII (1), 2020c, pp. 119–153.
Singh, Rajesh Kumar. Early Hunnic Invasions and Abandonment of Buddhist Temples in
India. = Ajanta Mahāpiṭaka I.4. Munshiram Manoharlal Publishers, 2021.
Singh, Rajesh Kumar. Select Stupa-shrines from Ajanta, Bagh, Dharashiva, and Pala. Forthc.
Sircar, Dinesh Chandra. „Inscription in Cave IV at Ajanta.” Epigraphia Indica, vol. XXXIII,
1959–1960, pp. 259–262.
Sircar, Dinesh Chandra. Select Inscriptions Bearing on Indian History and Civilization. Vol. 1.
From the Sixth Century B.C. to the Sixth Century A.D. 2nd. University of Calcutta, 1965.
Soundararajan, K. V. Nagarjunakonda (1954-60). Vol. 2. The Historical Period. = Memoirs
of the Archaeological Survey of India. Archaeological Survey of India, 2006.
Spink, Walter M. „Ajanta’s Chronology: The Problem of Cave Eleven.” Ars Orientalis, 1968,
pp. 155–168.
Spink, Walter M. „Ajantā’s Chronology: The Crucial Cave.” Ars Orientalis, 1975 (2017),
vol. X, 1975, pp. 143–169.
342
Rajesh Kumar Singh
Spink, Walter M. „Ajanta’s Chronology: Politics and Patronage.” Kalādarśana: American
Studies in the Art of India, edited by J. G. Williams, IBH Publishing, AIIS 1981 (2017),
pp. 109–126.
Spink, Walter M. „Ajanta’s Chronology: Cave 7’s Twice-Born Buddha.” Studies in Buddhist
Art of South Asia, edited by A K Narain, Kanak Publications 1985 (2017), pp. 103–116.
Spink, Walter M. Ajanta: History and Development. Vol. I. The End of the Golden Age.
Brill, 2005.
Spink, Walter M. Ajanta: History and Development. Vol. II. Arguments About Ajanta. Brill,
2006.
Spink, Walter M. Ajanta: History and Development. Vol. V. Cave by Cave. Brill, 2007.
Spink, Walter M. Ajanta: History and Development. Vol. IV. Painting, Sculpture, Architecture,
Year by Year. Brill, 2009.
Spink, Walter M. Ajanta: History and Development. Vol. VI. Defining Features. Brill, 2014.
Spink, Walter M. Ajanta. History and Development. Vol. VII. Bagh, Dandin, Cells and Cell
Doorways. Brill, 2017.
Tarzi, Zemaryalai. „Préservation des Sites de Hadda: Monastères Bouddhiques de Tape Shotor
et de Tape Tope Kalan.” Mari. Annales de Recherche Interdisciplinaires, vol. VIII, 1997,
pp. 207–222.
Thakur, Mrinmayee. „Rock-Cut Architecture: A Case of Caves in Goa.” Goa College of
Architecture Research Journal, vol. I, 2017, pp. 1–10.
Vahia, M. N., A. P. Jamkhedkar and Parag Majajan. „Astronomical Orientation of Caves
in Ajanta and Ellora.” Heritage: Journal of Multidisciplinary Studies in Archaeology,
vol. V, 2017, pp. 806–812.
Vignato, Giuseppe. „Archaeological survey of Kizil: it’s groups of caves, districts, chronology
and Buddhist schools.” East and West, vol. LVI (4), 2006, pp. 359–416.
Weiner, Sheila L. Ajanta: Its Place in Buddhist Art. Uniersity of California Press, 1977.
Willis, Michael. „Later Gupta History: Inscriptions, Coins, and Historical Ideology.” Journal
of the Royal Asiatic Society of Great Britain & Ireland, vol. XV (2), 2005, pp. 131–150.
Xinjiang Qiuci Shiku Yanjiusuo. Kezier Shiku Neirong Zonglu / Comprehensive Record of
Contents of Qizil Grottoes. Xinjiang Qiuci Shiku Yanjiusuo, 2000.
Yazdani, Gulam. Ajanta: Monochrome Reproductions of the Ajanta Frescoes Based on
Photography... IV vols. Oxford University Press, 1931–1952.
Zeimal, E. V. „The Kidarite Kingdom in Central Asia.” History of Civilizations of Central
Asia. Vol. 3. The Crossroads of Civilizations: A.D. 250 to 750., edited by B. A. Litvinsky,
UNESCO Publishing, 1996, pp. 123–137.
Zin, Monika. Guide to the Ajanta Paintings. Vol. II. Devotional and Ornamental Paintings.
Munshiram Manoharlal Publishers Pvt. Limited, 2003a.
Zin, Monika. Handbuch der Malereien / Handbook of the Paintings. Vol. II. Devotionale und
Ornamentale Malereien. II vols. Harrassowitz, 2003b.
Zin, Monika. „Buddhist Narrative Depictions in Āndhra, Gandhāra and Kucha – Similarities
and Differences that Favour a Theory About a Lost ‘Gandhāran School of Paintings’.”
Buddhism and Art in Gandhāra and Kucha, edited by Miyaji Akira, Ryukoku University,
2013, pp. 35–66.
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
PLATES
343
344
Rajesh Kumar Singh
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
345
346
Rajesh Kumar Singh
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
347
348
Rajesh Kumar Singh
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
349
350
Rajesh Kumar Singh
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
351
352
Rajesh Kumar Singh
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
353
354
Rajesh Kumar Singh
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
355
356
Rajesh Kumar Singh
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
357
358
Rajesh Kumar Singh
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
359
360
Rajesh Kumar Singh
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
361
362
Rajesh Kumar Singh
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
363
364
Rajesh Kumar Singh
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
365
366
Rajesh Kumar Singh
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
367
368
Rajesh Kumar Singh
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
369
370
Rajesh Kumar Singh
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
371
372
Rajesh Kumar Singh
Circa 465
and The Stupa-Shrine of the Ajanta Cave 11
373
MIT, OBRAZ, METAFORA
W KULTURACH AZJI I AFRYKI
MYTH, IMAGE, METAPHOR
IN THE CULTURES
OF ASIA AND AFRICA
Polskie Towarzystwo Orientalistyczne
Wydział Orientalistyczny Uniwersytetu Warszawskiego
Fundacja Karaimskie Dziedzictwo
MIT, OBRAZ, METAFORA
W KULTURACH AZJI I AFRYKI
MATERIAŁY
VI OGÓLNOPOLSKIEJ KONFERENCJI
ORIENTALISTYCZNEJ
WARSZAWA, 19–20 KWIETNIA 2021 R.
Redakcja naukowa
Marek M. Dziekan, Sylwia Filipowska, Ewa Siemieniec-Gołaś
Warszawa 2021
Polish Society for Oriental Studies
Faculty of Oriental Studies
Foundation Karaim Heritage
MYTH, IMAGE, METAPHOR
IN THE CULTURES
OF ASIA AND AFRICA
PROCEEDINGS OF THE 6TH ALL-POLISH CONFERENCE
OF ORIENTAL STUDIES
WARSAW, APRIL 19–20, 2021
Edited by
Marek M. Dziekan, Sylwia Filipowska, Ewa Siemieniec-Gołaś
Warsaw 2021
Recenzenci / Reviewers
Agata Bareja-Starzyńska, Renata Czekalska, Marzenna Czerniak-Drożdżowicz,
Sebastian Gadomski, Kamila Junik, Joanna Jurewicz, Lidia Kasarełło,
Magdalena Kubarek, Agnieszka Kuczkiewicz-Fraś, Katarzyna Marciniak,
Halina Marlewicz, Marek Mejor, Barbara Podolak, Marzanna Pomorska,
Małgorzata Religa, Kamila Stanek, Danuta Stasik, Jerzy Tulisow,
Grażyna Zając, Estera Żeromska
Publikacja zrealizowana w ramach programu „Doskonała Nauka” MEiN
UMOWA Nr DNK/SP/465019/2020
Publication financed by the Polish Ministry of Education and Science
Opracowanie redakcyjne i korekta
Hanna Januszewska
Projekt okładki
Zofia Lasocka
© Copyright by Authors and Dom Wydawniczy Elipsa
Warszawa 2021
ISBN 978-83-8017-417-7
Realizacja wydawnicza:
Dom Wydawniczy ELIPSA
ul. Inflancka 15/198, 00-189 Warszawa
tel. 22 635 03 01
e-mail: elipsa@elipsa.pl, www.elipsa.pl
SPIS TREŚCI
Wstęp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
13
1. M ,
,
Kamila Hildebrand
Wiersz Aš-Šayib (Stary człowiek) Muḥammada Ibn Rāšida Āl Maktūma
w świetle teorii metafory pojęciowej wraz z ukazaniem podobieństw
leksykalno-semantycznych języka nabaṭī i współczesnej poezji ZEA . . . . . . .
19
Karolina Wanda Olszowska
Wpływy chrześcijańskie w twórczości Halide Edip Adıvar . . . . . . . . . . . . . . .
37
Kamila Stanek
Językowy obraz bohatera i bohaterstwa w przysłowiach tureckich . . . . . . . .
51
Teresa Miążek
Między mitem, legendą a rzeczywistością w twórczości Agieja
– pomosty w strukturze narracji . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
73
Joanna Gruszewska
Metafory nibbany (nibbāna) i sansary (saṃsāra) w Pieśni Sumedhy
(Therīgāthā 450–524) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
99
Pi-Chun Chou
Między egzotyzacją a udomowieniem – wyzwania dla polskiego tłumacza
klasycznej chińskiej powieści na przykładzie postaci kobiecych
w Śnie Czerwonego Pawilonu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
Ewa Rynarzewska
Mit we współczesnej literaturze koreańskiej: Memil kkot p’il muryŏp
(Gdy dojrzewa gryka) Yi Hyo-sŏka (1907–1942) . . . . . . . . . . . . . . . . . . . . . . 135
Iwona Kordzińska-Nawrocka
Obraz kobiecej zazdrości w japońskiej tradycji literackiej . . . . . . . . . . . . . . . 151
6
Spis treści
2. M ,
,
Nihad Fottouh
Reminiscences of the myth of Isis in representations of the Egyptian mother:
Egyptian President El-Sisi’s supporting remarks . . . . . . . . . . . . . . . . . . . . . . . 175
Melaike Huseyin
Reality behind the Myth and Image of Turkish Women Depicted
by the European Orientalists (19th to 20th century) . . . . . . . . . . . . . . . . . . . 211
Shahla Kazimova
Aszykowie – strażnicy duszy narodu. W poszukiwaniu tożsamości aszyków
w kontekście kultury azerbejdżańskiej . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
Michael Knüppel
Aspects of the motive of cannibalism in Northern Eurasia, and their
reflections in Uralic, Altaic and “Palaeosiberian” languages . . . . . . . . . . . . 243
Blanka Katarzyna Dżugaj
Mit na ekranie – nowe oblicza motywów mitologicznych i epickich
w kinie indyjskim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
Karolina Kłoszewska
Motywy mitologiczne w tamilskich pieśniach religijnych
na przykładzie kultów bóstw żeńskich . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
Rayesh Kumar Singh
Circa 465
and the Stupa-Shrine of the Ajanta Cave 11 . . . . . . . . . . . . . . 301
Małgorzata Sobczyk
Buddyjska symbolika eschatologiczna w inscenizacji orszaku Amitābhy . . . . 375
Jakub Zamorski
„Zen” wraca do Chin – narodowość i nowoczesność
we współczesnym chińskim obrazie buddyzmu chan . . . . . . . . . . . . . . . . . . . . 393
CONTENTS
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
13
1. M
,I
,M
L
Kamila Hildebrand
The Poem Aš-Šayib (The Old Man) by Muḥammad Ibn Rāšid Āl Maktūm
in the Light of the Theory of Conceptual Metaphor along with Showing
the Lexical and Semantic Similarities of the Nabaṭī Language
and Contemporary Poetry of the UAE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
19
Karolina Wanda Olszowska
Christian Influences in the Works of Halide Edip Adıvar . . . . . . . . . . . . . . .
37
Kamila Stanek
The Linguistic Image of a Hero and a Heroism in Turkish Proverbs . . . . . .
51
Teresa Miążek
Between Myth, Legend and Reality in Agyey’s Works – Bridges
in the Structure of the Narrative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
73
Joanna Gruszewska
The Metaphor of Nibbana (Nibbāna) and Sansara (Saṃsāra)
in Sumedha Song (Therīgāthā 450–524) . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
99
Pi-Chun-Chou
Between Foreignization and Domestication – Challenges for Polish
Translator of a Classical Chinese Novel on the Example of Female
Characters in The Dream of the Red Chamber . . . . . . . . . . . . . . . . . . . . . . . 117
Ewa Rynarzewska
Myth in Contemporary Korean Literature: Memil kkot p’il muryŏp
(When the Buckweat Blooms) by Yi Hyo-sŏk (1907–1942) . . . . . . . . . . . . . . . 135
Iwona Kordzińska-Nawrocka
The Image of Female Jealousy in the Japanese Literary Tradition . . . . . . . . 151
8
Contents
2. M
,I
,M
C
R
Nihad Fottouh
Reminiscences of the Myth of Isis in RepresentationS of the Egyptian
Mother: Egyptian President El-Sisi’s Supporting Remarks . . . . . . . . . . . . . . . 175
Melaike Huseyin
Reality behind the Myth and Image of Turkish Women Depicted
by the European Orientalists (19th to 20th Century) . . . . . . . . . . . . . . . . . . . 211
Shahla Kazimova
Ashiks – Guardians of the Nation’s soul. In Search of Ashik’s Identity
in the Context of Azerbaijani Culture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
Michael Knüppel
Aspects of the Motive of Cannibalism in Northern Eurasia, and Their
Reflections in Uralic, Altaic and “Palaeosiberian” Languages . . . . . . . . . . . 243
Blanka Katarzyna Dżugaj
Myth on the Screen – new Faces of Mythological and Epic Themes
in Indian Cinema . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
Karolina Kłoszewska
Mythological Themes in Tamil Religious Songs on the Example of Worship
of Female Deities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
Rayesh Kumar Singh
Circa 465
and the Stupa-Shrine of the Ajanta Cave 11 . . . . . . . . . . . . . . 301
Małgorzata Sobczyk
Buddhist Eschatological Symbolism in the Staging of Amitābha’s
Procession . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375
Jakub Zamorski
“Zen” Returns to China – Nation and Modernity in the Contemporary
Chinese Image of Chan Buddhism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393
PREFACE
T
he book that we present to the Reader is the fruit of the 6th All-Polish
Conference of Oriental Studies, “Myth, Image, Metaphor in Cultures of Asia
and Africa”, which took place on-line in Warsaw on 19–20 April, 2021. The
conference was organised by the Polish Oriental Society in cooperation with the
Faculty of Oriental Studies of the University of Warsaw and the Karaim Heritage
Foundation. The conference was subsidized by the Ministry of Education and
Science within the framework of the “Excellent Science” programme.
The conference was another part in a series, but for the first time it was an
international one, which is partly reflected in the present volume with several
papers by foreign guests. This time, therefore, the conference became a forum
for the exchange of ideas between not only representatives of Polish Orientalist
centres but also universities from Germany, Italy, Russia, Ukraine, China and
Egypt, among the others. As usual, Orientalist conferences are also a meeting
place for representatives of different generations of researchers. This year, due
to the pandemic, this important element of our annual meetings has become
impossible, so this book seems all the more important. This on-line meeting will
thus take on a more traditional form – a meeting based on traditional reading,
which has retained its permanent – in our opinion – significance all this time.
The theme of the conference directed the contributions mainly towards the
humanities, although it is also in the area of social sciences that the issues of
myth and image are beginning to attract more and more attention of researchers.
In this way, this volume reflects a more “traditional” understanding of Oriental
studies, focused primarily on literary and cultural issues. However, there are also
texts of an inter- or transdisciplinary character which take into account issues of
art, religion and film.
The research presented in the volume comprises several fields of Oriental
studies: Arabic studies, Turkology, Altaic studies, Indology (with its internal
14
Preface
differentiation: Sanskritology, Tamil studies, and Hindi culture), Korean studies,
Sinology and Japanese studies.
The conference-related keywords: myth, image and metaphor, as well as
symbol, which does not resonate directly in the conference title, but is closely
related to the other items, are extremely capacious and they are analysed by
researchers dealing with various aspects of all cultures of the world.
The methodology of the humanities which has been developing for centuries
in relation to these issues, starting from the most ancient one, which can be
regarded as a simple description, through structuralism, postcolonial studies to
the so-called “new humanities” or intersectional studies, includes references
to the notions mentioned in the conference theme because they constitute the
elements shaping the human understanding of the world almost since the dawn of
human thought. In each of us, consciously or unconsciously, myths, images and
metaphors are embedded in us and they change at the pace at which the world
and we ourselves change. They shape our vision of the world – our vision of
ourselves and of the “Other”.
It is not without reason that I emphasise the issue of the “Other”, who can also
be called “Alien”. The study of these “other” and “alien” cultures is, after all, the
main task of Oriental studies, no matter how one can understand it. This is what
traditional philology and broader humanistic research conducted by Orientalists
will be concerned with. The study of foreign cultures builds a certain image
which is a sort of a mirror, sometimes a fun-house mirror, in which we can see
ourselves and perceive in it the solution to our own problems.
The book Myth, Image, Metaphor in the Cultures of Asia and Africa is divided
into two parts, Myth, Image, Metaphor in Literature and Myth, Image, Metaphor
in Culture and Religion, which feature seventeen texts. The titles of both parts,
reflecting their substantive content, rather unintentionally coincide with two
of the three leading disciplines among Orientalists working within the field of
humanities, according to the classification introduced by the “Constitution for
Science”.
The authors of the individual studies obviously focus on the specific issues
that currently constitute the subject of their research. However, the are bound by
the conference keywords, which have proven to be very appealing and pertinent.
What is important, the presented analyses clearly indicate that there is no single
understanding of either myth or metaphor both from the methodological point of
view and in the context of particular field of Oriental studies.
Preface
15
Therefore, once again, the opinion known among Orientalists is confirmed
– one Orient does not exist.
The same remark may be applied to myth – it may simply be an element of
some traditional mythological system, it may be a factor that strengthens identity
or directly creates it. An image, in turn, may be directly comprehensible but it
may also possess the power of myth or symbol and owing to it, it is beyond the
reach of the recipients who are not prepared to understand it.
Myth can be adopted or transformed, it can also be created, almost here
and now, for the purposes of the moment, for example in a political activity. It
can be dressed in the garb of modernity, as in a contemporary film. An image
can sometimes be viewed and enjoyed for its beauty but sometimes, however,
it needs to be transformed, as in artistic translation. But an image can also be
linguistic, constituting a non-transferable element of the linguistic image of the
world developed over centuries, sometimes millennia, by various cultures and
civilisations. Myths, interweaving with symbols and images, may assume the form
of rituals, of which – on the other hand – they may be an emanation. They can
also be reflected, for example, in cartography. Both myth and images, metaphors
and symbols may undergo constant semantic transformations conditioned by the
changes in the consciousness of societies and their knowledge about themselves
and others. Suppressed and exterminated, sometimes by (e.g. political) force, they
may turn out to be stronger than any power and return against it in an externally
changed form, but at the same time retaining their proper, original, basic sense.
The themes to which we draw attention in the above paragraph are not
accidental – these are themes that appear in the individual texts contained in this
book. In my opinion, they indicate the richness of the content and encourage one
to read, not selectively but comprehensively. Although the topics presented by
the authors sometimes seem to be disassociated from one another, the leading
theme, however, seems to influence the content to such an extent that we obtain
a kind of mosaic which we would like to supplement with subsequent fragments.
Undoubtedly, it will never be a complete mosaic. Such a task is impossible.
However, a fascinating picture may emerge, which each of the readers will want
to complete in his own further research.
This is precisely one of the main aims of this book.
Marek M. Dziekan
Ewa Siemieniec-Gołaś