2013 年度

Issues in the History of Indian Buddhism

龍谷大学アジア仏教文化研究センター
Issues in the History of Indian Buddhism

Giovanni Verardi



 Giovanni Verardi

Issues in the History of Indian Buddhism

In December 2011, I was invited by the Rynjkoku University in Kyoto to present Hard-
ships and Downfall of Buddhism in India, a book of mine that had been published some
months before (Verardi 2011b). In the course of the lecture, I first illustrated the contents
of the book, and then I focused on a few issues that seemed to me particularly worthy of
attention. The assumptions of the discussion are the following:
a) In ancient India there was not a single model of society but, rather, two oppos-
ing models of social and economic relations that coexisted for a long time in conflict. In-
clusive paradigms should be rejected ௅ in particular, the idea, created by western scholars
and Indian historians who felt the need to provide their country with a national ideology,
of a unified India where Brahmanism and Buddhism are almost interchangeable.
b) One of these models, Buddhism, kept its initial antinomial character to the
end.1 Only if we understand Buddhism as a system opposing the nómos established by
the Brahmanical tradition, we can overcome the difficulty of connecting the VajrayƗna to
the earlier traditions. The VajrayƗna recovered, in a changed situation, the most radical
antinomian positions of the beginning.
c) The history of ancient India involved a high rate of violence, much of which
caused by the imposition of the var۬a state. Violence found a tenuous disguise in myth
and iconographies. We still lack a methodology allowing us to use all the un-Rankean
sources, though the latter tell us a lot about violence, keeping memory of otherwise for-
gotten events. Orthodox Brahmans wrote history by resorting to metaphors and allegories,

1
This is the reason why I prefer to limit the discussion to Buddhism, initially less radical than Jainism,
without addressing the complex discourse of the segmented “Ğrama۬a model”.

1
refusing to adopt the logic of chronologies and positive data, whose documentary power
they knew but which they judged counterproductive to their purposes.
Buddhism is a religious system, interpretable as an antinomial model, that in the
historical process was obliged, by ideology and violence, to dramatically and continu-
ously reposition itself. This resulted in being suppressed or being cornered into subaltern
positions. For example, a part of the Ğrama۬as started rather early to organize themselves
according to a community model paralleling the Brahmanical priesthood.
Early Buddhism had interpreted the needs of the merchant class and of the urban
manufacturing classes that had come to the fore in the third century BC when India came
into contact and became part of the Hellenistic world.2 At first, the Buddhists found sup-
port in the anti-establishment, “enlightened” despotism of an AĞoka, and then boomed
between the late second century BC and the early third century AD. Later on, when the
var۬a state became a fait accompli, Buddhism continued to give voice to the social and
ethnic sectors opposing the agricultural, self-referential model of the tƯrthikas. The idea
of state and society that the Buddhists had in mind was compatible with the varied
peoples inhabiting the subcontinent, whereas the Brahmanical model implied their forced
incorporation into the boundaries of an agrarian society organized according to the rules
of var۬ƗĞramadharma. The latter was opposed not only by the trading classes, to which
operational freedom was necessary, and by the natives who saw themselves downgraded
to the lower peasantry ranks, but also by the Brahmans who had joined Buddhism. There
was a dramatic split in the brƗhma۬avar۬a that was never healed: that a part of the
Brahmans opposed the fundamentals of their own var۬a could not be accepted.

1. The Guptas, and the Buddhists in Checkmate

The first question we address is the break represented in Indian history by the Gupta state
and that sort of re-nationalization of the Indian world which meant the expulsion of the
Buddhists from political power and economic initiative. The ArthaĞƗstra, now better at-

2
This may seem a position derived from outdated paradigms that read the history of India through Greco-
Roman (more Greek than Roman) glasses. Not so; the fact remains that the opening of ancient India to the
outside, with the predictable internal reactions, coincided with, and largely depended on the breakout to-
wards the east first by Alexander and his successors and then by the Roman republic and empire. At a
macro-historical level, Toynbee’s view, which includes India in the Hellenized world, is founded, except
for matters of detail (see e.g. Toynbee 1967: esp. 114, 178௅79, 180௅81 and passim).

2
tributed to the fourth century AD (cf. Willis 2009: 62) helps clarifying the policy fol-
lowed by the dynasty. The government of the rightful, namely orthodox king is based on
da۬‫ڲ‬anƯti (policy exercised through coercion) and rƗjadharma (the king’s duty to pre-
serve the four var۬as). It was in the Gupta period and its aftermath that the “Kali Age lit-
erature” was produced, whose targets are the merchants and artisans, responsible for so-
cial disorder. The vaiĞyas refused to offer sacrifices, depriving the Brahmans of their fees,
and whole regions were under the control of the pƗsa۬‫ڲ‬as, the heretics. In connection
with this, many Brahmans had apostatized.3 Some scholars have acknowledged the upris-
ing of the subject castes and the intense hostility between Brahmans and Ğnjdras in this
period4 but have not recognized that these two facts are one and the same thing with the
Buddhist hegemony over society and the reaction against the religion of Dharma, respec-
tively. Between the late third and the fifth century, Buddhism was not the only dissident
voice, but was held responsible for having exercised hegemony over change and was
identified with it.
The Guptas were aĞvamedhin kings, but around AD 400 a crucial readjustment
took place. At Udayagiri (Vidisa), the ritual capital of the dynasty, we observe the emer-
gence of BhƗgavatism not only as the state religion but as a theistic system capable of
uniting forces with the other major system, Sivaism (as we call it now), while keeping a
strategic tie with the Veda. The audacity of the operation is measured by the fact that
even lower cults based on magic and black magic, represented by the Atharvaveda, were
included into the synthesis: the purohita of the Guptas was a KƗpƗlika (Willis 2009: 169
ff.). The great Brahmanical myths were created or reshaped in this period as a function of
the colonization of lands and the strengthening of the new powers that found representa-
tion in the theistic systems. Hence the myth of Tripura, the three towns of the asuras
(identified with the Buddhists) destroyed by ĝiva; hence King Bali, of Asuric descent,
obliged to give all his lands to Viৢ৆u-VƗmana, who fools him presenting himself, in the
iconographies, as the Buddha (fig. 1); etc.5

3
Cf. e.g. VƗyu PurƗ۬a (I.58.38, 49; vol. 1, pp. 412௅13), where we read that in the Kali age “[t]he
BrƗhma৆as do not perform Yajñas” and “perform obeisance to Snjdras” (for the whole passage on the Kali
Yuga see I.58. 31௅70).
4
Notably R.C. Sharma in his study on the Kali Age literature (Sharma 1982).
5
Some of these myths are analyzed in this perspective in Verardi (2011b:162ff., 265ff. and passim). Re-
garding the story of Viৢ৆u Trivikrama, there is an earlier representation of VƗmana as Buddha in Cave II at

3
 The international crisis of the third century had greatly favoured the agrarian pol-
icy of the Brahmans. We observe a rapid decline in Indian urbanization (Sharma 1987)
and the replacement of manufacturing and trading towns with tƯrthas. Rural India was
largely colonized by the PƗĞupatas, who acted as priests or pujƗris, as shown by a very
large set of archaeological finds from all over northern India. They unmistakably depict
either Sivaite ascetics or ĝiva himself, as is shown by their jƗ‫ܒ‬as and, in many a case, by
the third eye (fig. 2).6 There were groups of PƗĞupatas who identified with ĝiva’s bull
(van Troy 1990: 7), and clay images of bulls are particularly numerous among the many
thousands of small cultic objects that we find throughout the Ganges Valley in the early
centuries AD (Verardi 2011b: 145௅46).
A few Buddhist sources provide us with some information on the state of the reli-
gion in the fourth century. In Asaৄga’s Bodhisattvabhnjmi, cruel kings and ministers are
mentioned along with the accusation of the property of the saۨgha or a stnjpa being taken
away;7 in the fourth-fifth century Sanskrit redaction of the Lotus Snjtra, monks are re-
ported to be expelled from vihƗras, to be imprisoned and punished.8 We also have the
story of the Siূhala monks MahƗnƗman and Upasena who reported that, at the time of
Samudragupta, in such a large country as JambudvƯpa there was no place for them to live
(cf. Law 1967: 29, 97b, 2; p. 15). According to the later testimony of the
MañjuĞrƯmnjlakalpa, Samudragupta, sacrificing animals and surrounded by bad counci-
lors, greatly committed sin, his kingdom being inundated by “carping logicians, vile
Brahmins” (cf. Jayaswal 1937: 48).
The Guptas took upon themselves the task to remedy the confusion of castes (i.e.
the dynamics of a not exclusively agrarian economy), repeatedly condemned in Brah-
manical texts, that had occurred in the previous centuries. The need to compromise under
the increasingly strong Brahmanical pressure meant for the Buddhists the acceptance of
caste rules, despite the fact that they rejected them in principle and, when possible, in the

Badami, where the Dwarf is shown with elongated ears and u‫۬܈‬Ư‫܈‬Ɨ (cf. ibidem: 164, fig. 2). This and the
example in fig. 1 disclose the extent to which iconographies can contextualize myths.
6
A terracotta head which can be identified with that of ĝiva was found in the excavations at Sringa-
verapura (Allahabad District). B.B. Lal has dated it to the first century AD and has described it in the detail
(Lal 1993: 111௅15).
7
Cf. McFarlane (1995: 194), quoting from Unrai Wogihara’s edition of the text (Tǀkyǀ 1930௅36, 165௅7).
8
Saddharmapu۬‫ڲ‬arƯka Snjtra (12.3௅21; pp. 259௅61); Nattier (2008: esp. 20 ff.; 2003: esp. 48 ff.) has shown
that the Sanskrit text we have is a post-third century redaction. It can arguably be dated to Gupta times.

4
reality.9 Faxian has a story set in PƗ৬aliputra regarding the Buddhist master Raivata, a
Brahman by caste, who “led a pure and solitary life”. The king of the country revered
him as his guru, but whenever he went to visit him, he did not dare to sit beside him. “If
the king, from a feeling of love and veneration, grasped his hand … the Brahman would
immediately wash it” (Giles 1923: 46).10 Raivata considered the contact with who was
arguably a petty king of lower birth polluting, but the people of his country revered the
guru because he “diffuse[d] widely the Faith in Buddha, so that the heretics were unable
to persecute the priesthood” (ibidem). They considered Raivata one of them on the basis
of his social conduct, not of his ideas (a minor, resorbable factor), leaving the monastic
community in peace. A perfect trap, for the Buddhists.
Society could not be controlled without the performance of rituals (the rites of
passage, in particular), and a part of the Buddhists realized that they could not leave them
in the hands of the orthodox. Rituals could not be celebrated in disagreement with the
rules established in the dharmaĞƗstras, and only Brahmans were entitled to perform them.
This is the reason why the early bodhisattvas are said to be born in upper castes and be,
quite often, Brahmans. It remains to be determined with accuracy what were the milieus
in which this important shift (which does not necessarily mean surrender to the estab-
lishment) took place: as already mentioned, the fracture within the first var۬a was very
serious and fraught with dramatic consequences.
In the context of the rapid dissolution of the Buddhist hegemony between the late
second and the third century, we also observe an emerging polarity between the ara۬ya
and grƗma monks. The ara۬ya bodhisattvas accused the grƗma monks to be involved in
agriculture and trade and settle in villages. The Questions of RƗ‫܈‬trapƗla depicts a situa-
tion where not only grƗma bodhisattvas but also “corrupt” monks had established them-
selves on an independent or semi-independent basis. They would not pay homage to the
teachers and saints (cf. RƗ‫܈‬trapƗlaparip‫܀‬cchƗ Snjtra: 99; p. 130), and regarding their
functions as performers of rites, they performed services for people in the household
“[t]aking up the banner of the Buddha” (ibidem: 179; p. 138). The “corrupt monks” kept
cows, horses, asses, livestock, male and female slaves, and did not distinguish between
their own property and what belonged to the stnjpa and the saۨgha. Moreover, “they

9
The reader is referred to Eltschinger (2000) for a close examination of the issue.
10
For the travelogue of Faxian, see now Max Deeg’s edition, translation and comment of the text (for the
Raivata episode, Deeg 2005: 548).

5
would have wives, sons, and daughters just like a householder” (ibidem: 180௅81, 183; p.
138). Thus family-based institutions run by Buddhist priests seem to have been already in
existence at a rather early age. Many monasteries had been forcibly abandoned, and the
guiding capacity of the leading monasteries was severely reduced. There were no coun-
cils entitled to impose, at least to an extent, a set of doctrines and social behaviours in
line with the country’s re-ruralisation. It is not a question of HinayƗna/MahƗyƗna dichot-
omy. Xuanzang provides us with surprising testimony on the state of the clergy. In Sind,
he noted a group of adherents to the Little Vehicle who shaved their heads and wore the
ka‫܈‬Ɨya robes of the bhik‫܈‬us, but were engaged in the ordinary affairs of lay life. In the
past, they had followed the monastic rules, but the “changed times” had brought them to
live as worldly people, “with sons and grandsons” (Beal 1884, II: 273௅74). Xuanzang
was a staunch supporter of the MahƗyƗna, which was a better strategy (yƗna) to re-assert
Buddhist hegemony after that the strategy of earlier Buddhism (HinayƗna, as it was la-
beled) in countering the comeback of the tƯrthikas had failed. To his eyes, the MahƗyƗna
also meant reasserting the rules of the Vinaya after a long period of decline and compro-
mise.11
It was only from the mid-fifth century that the Buddhists started regaining part of
their lost power. At Mathura and Sarnath new images of the Buddha started being again
produced (see e.g. Gottfried Williams 1982: figs. 60 ff., 85 ff.), after that image making
had come to a halt or had been severely reduced in the fourth century; the saۨgha reor-
ganized itself at Ajanta in the second half of the fifth century,12 etc. The Gupta dynasty
was near its end, and new perspectives opened for the Buddhists in northern India even
though the Hnj৆as, entering in the subcontinent, espoused the religious policy of the Gup-

11
This, I now believe, is the correct view. Previously it had seemed to me, influenced as I was by the dec-
ades-long debate about the origins of the MahƗyƗna, that within the latter two trends could be distin-
guished: the subaltern and compromising one and the rigorist one (à la Xuanzang). Now I prefer to think
that the MahƗyƗna reacted as a whole to what were considered the inadequacies and degenerations of the
ancient yƗna, unable to meet the challenges of the time and often ready to compromise.
12
Walter Spink has devoted a number of works on the Buddhist revival at Ajanta in the second half of the
fifth century (cf. especially Spink 2005௅7), maintaining that it was due to the patronage of the VƗkƗ৬aka
king Hariৢe৆a. But Hariৢe৆a, like the other rulers of the dynasty, was an orthodox, aĞvamedhin king who, if
that was the case, might have acted by proxy through his wives to protect Buddhism, according to a well-
documented procedure. Spink’s construction of Ajanta as a royal VƗkƗ৬aka site has been questioned, for
different reasons, by Bakker (1997: 37 ff.).

6
tas.13 The true heirs of the latter were, however, in the Deccan: the southern kingdoms es-
tablished themselves along the lines laid down in the fourth century, and it was from
them that a slow but steady comeback towards north would start. The role played by
South India in the history of the country is generally underestimated, mainly because of
the difficulty for scholars to access the written sources, numerous though they are.

2. Politics and Doctrinal Debates

Xuanzang must have praised the MahƗyƗna also because it had developed a dialectical
system able to withstand the offensive of the tƯrthikas. In the Xiyuji there are frequent
references to the public debates held between Buddhists and tƯrthikas, such as those held
at ĝrughna, PrayƗga, PƗ৬aliputra, in Kar৆asuvar৆a, etc. (Beal 1884, I: 186௅87; II: 230௅31,
96௅97, 202௅4): the saۨghas would spring up again (with the building of new monaster-
ies, the resumption of economic activities of the laity, etc.) as a result of the victory of
the Buddhist vƗdins. Yijing reports that Buddhist controversialists, after studying at
NƗlandƗ or ValabhƯ could “discuss possible and impossible doctrines”, acquiring confi-
dence in themselves. Thus they could “proceed to the king’s court to lay down before it
the sharp weapon (of their abilities)” and show their political talent, “seeking to be ap-
pointed in the practical government” (Takakusu 1896: 177). An authority such as NƗgƗr-
juna already had the role to propitiate kings and persuade people to accept Buddhism and
discard the Vedic faith (Scherrer-Schaub 2007: 762௅63). The stakes of doctrinal debates
was political power since at least Gupta times. In the Benares subjected to Gupta rule, the
complete reorganization of the educational and academic life became possible when the
BhƗgavatas won in a crucial debate over the VijñƗnavƗda logicians. The YogƗcƗrins had
to abandon their establishment in Bhadaini, where their monastery was arguably disman-
tled and a ku۬‫ڲ‬a, the LolƗrka Ku৆ঌa, which is still there, was realized on the site
(Agrawala 1983: 5௅6).

13
Suffice here to mention Dhanyaviৢ৆u’s Boar inscription at the BhƗgavata site of Eran, which opens men-
tioning the great fame and lustre of ToramƗ৆a who rules the earth, and the Gwalior inscription of Mihira-
kula, where the king declares himself to be a worshipper of PaĞupati (Sircar 1965: no. 55, pp. 420௅22; no.
57, pp. 424௅26).

7
 After the Guptas, inter-religious controversies14 acquired an even greater impor-
tance. The development of Indian logic may be construed as a function of the political
need of finding support at court, on pain of loss of patronage, expulsion from a given ter-
ritory, and the risk, which became more and more concrete, of a death sentence, already
attested to in the Xiyuji (Beal 1984, II, 97௅99, 262௅64). ƖnvƯk‫܈‬ikƯ, the science of inferen-
tial reasoning, central to doctrinal controversies, was not connected to philosophy, but
pertained to the domain of nƯtiĞƗstra and rƗjadharma, and was thus strictly related to
kingship (Hakker 1958: 64௅67). Debates were regarded as judicial procedures, as appears
from the strict rules that had to be followed and by the fact that they occurred only in the
courts of kings and judges appointed by the king. The legal dimension of Indian debates
eludes us, but we can say that the judiciary passed under the control of the theistic
movements, in parallel to what was happening in the Mediterranean, where, after Theo-
dosius, the disputations with those who now were considered heretics were no longer
theological debates but legal trials (Humfress 2007: 248 ff.).15 It further appears that in
India debates increasingly acquired the nature of ordeals.
When the BhƗgavatas replaced the Sivaites in Kanchipuram, we would see the
Buddhists impaled after being sentenced to death by the Pallava king Nandivarman II (c.
AD 730௅95), a resolute persecutor of Ğrama۬as. The temple of Vaiku৆৬ha PerumƗশ with
the attached gallery decorated with panels illustrating the deeds of the sovereign was
built in the second half of the eighth century in accord with the prescriptions of the
PƗñcarƗtra Ɩgama (Hudson 2002: 134). The king is shown seated in a pavilion where he
has delivered judgement, arguably after a doctrinal debate (fig. 3). One of the early three
ƖlvƗrs is depicted in the adjacent panel, and we know that the ƖlvƗrs, particularly
Tirumaৄkai and To৆৬ara৬ippo৬i, went very far in their anxiety to condition the policy of
the ruler and get rid of the Ğrama۬as. In one of his hymns, To৆৬ara৬ippo৬i, a contempo-
rary of king Nandivarman, upholds that it is his duty to chop off the heads of the
Ğrama۬as. All this was functional to the distribution of lands after the clearing of all per-

14
It goes without saying that I find unacceptable to speak of “inter-sectarian” debates and struggles when
Buddhism or a form of Buddhism are included in the pile: with “sects” we mean a part of a single, momen-
tarily fragmented but re-unifiable system.
15
Emperor Theodosius declared Christianity state religion in AD 380; hence, also a different attitude to-
wards positions that had coexisted, though in conflict, in the early Church but were now considered hereti-
cal.

8
sons whose observances were not in accordance to dharma (the dharma of the ortho-
dox).16
A number of sources, mostly un-Rankean, attest that the losing party in a debate
was obliged to leave a given territory. This happened to the Buddhists with increased
frequency after the death of Harৢavardhana (AD 647), when Buddhism risked collapse.
In the hagiographic material, both Brahmanical and Jain, as well as in TƗranƗtha, we also
have several examples of the necessity of capturing the skill of argumentation of the op-
ponents by breaking into a Buddhist monastery or, conversely, into a Brahmanical ma‫ܒ‬ha
in disguise. 17 The stories mentioning mutilation or the death penalty inflicted to the
Buddhists have been dismissed as apocryphal or as belonging to a much later time (the
thirteenth century, mainly), but if we consider the evidence from the “data base” (ar-
chaeology, epigraphy, iconography), we realize that violent means to silence one own
opponents have a long story behind them. A few examples will suffice to make the point
clear.
In Kerala, king CƝramƗৈ PerumƗশ, who initially favoured Buddhism, was per-
suaded by the Brahmans to organize a public debate with the aim of overcoming the
Buddhists: the tongues of the losers should be cut out. The dispute ended favourably to
the Brahmans, and the tongues of the Buddhists were cut out and they were banished the
country. The PerumƗশs were a local lineage of kings who were given kৢatriya status by
the Brahmans and became supporters of orthodoxy. The king of the story is probably the
eighth-century CƝramƗৈ PerumƗশ of Makǀtai (modern Kodungallur) in the CƝra country
(Zvelebil 1975: 142௅43), and the debate, with the resulting distribution of lands to the
Brahmans, took place at Trikkariyur (Verardi 2011b: 25௅26, 219).18 Control over society
was exercised through a network of temples, the rulers being often figureheads with little

16
The reader is referred to Minakshi (1938: 171௅72) for the identification of the impaled as Buddhists, the
role of the ƖlvƗrs and the political-religious significance of the king’s actions.
17
See e.g. TƗranƗtha’s account of DharmakƯrti disguising himself as a servant in order “to learn the secret
teachings of the tƯrthika systems” (TƗranƗtha: 89A; p. 230) or the story of the Jain guru Haribhadra, two
disciples of whom went to study logic at BodhgayƗ in disguise and were killed when they were discovered.
On the different versions of the latter story, see Granoff (1989; I do not think that these stories are just
“legends”), and see Verardi (2011b: 210 ff.) for an attempt at contextualizing historically the various
strategies for success.
18
The story is reported in the KƝralǀlpatti and in a number of local purƗ۬as and other medieval texts (see
e.g. Taylor 1838: 183). See now Veluthat (2006) on the use of such sources as the KƝralǀlpatti for histori-
cal purposes.

9
real power outside of what was ceded to them by the temple priests. A bold and visible
Brahman oligarchy was often thinly disguised as a monarchy (Narayanan 2002: 116). In
the case of king CƝramƗৈ, none of our sources is “positive”, and iconographic evidence is
also lacking, and yet it is possible to reconstruct the picture.
The same can be said ௅ we are in Tamil Nadu ௅ of certain stories pertaining to the
earlier (Sivaite) wave of theistic reaction against the Ğrama۬as. The great saint Campan-
tar [Sambandar], one of the early NƗyanƗr who lived in the second half of the seventh
century and a Brahman by caste, caused the head of the vƗdin Buddhanandi to roll on the
ground by the potency of a song during a debate at Butamankalam in the middle Kaveri
plain.19 Despite the death of their elder, the Buddhists insisted on another debate to be
held, centred on mok‫܈‬a. It ended with the victory of Campantar and with the conversion
of the Buddhists to the Sivaite faith. Going through the story, we understand that the
Sivaite crowd that welcomed Campantar on his arrival in town created the conditions to
cause the death of the Buddhist elder. Intimidations and preventive acts were meant to
avoid major clashes. The story is told at length in the Periya Pura۬am,20 through which,
as has been observed, we are inevitably led to read all the earlier evidence (Gros 1984): it
is difficult, however, to imagine a community of Buddhists being cleared away from the
main settlement of a fertile area such as Butamankalam in twelfth-century
Cǀ঺ama৆৬alam: the story cannot but refer to an earlier period. There is ampler evidence
on the anti-Ğrama۬a operations carried out by the Sivaites in the late seventh௅early
eighth century (see below). They came as a consequence of the speeding-up of the Brah-
manical settlement policy, thanks to which the most fertile, generally riverine regions,
were firstly colonized, and then, step by step, the less productive and even the marginal
areas. Debates, legitimation of power and occupation of lands are aspects of a coordi-
nated process. We would better understand the unwinding of Indian medieval history by
putting aside the idea of an India forced in the cage of a “feudal” phase. Indian Marxist
historians have done a great service to the development of the discipline, but I do not
think that the category “feudalism”, which is the object of reconsideration even in Europe
(Reynolds 2004), is of great help for us to understand medieval India, of which it is

19
It is identified with present Budalur near Tirukkattupalli in Thanjavur district, and, tentatively, with the
native place of the fifth-century thera Buddhadatta (Dikshitar 1931: 689, n. 1).
20
Periya PurƗ۬am (33.904௅25; vol. 2, pp. 176௅181). See Verardi (2011b: 214 ff.) for the historical fram-
ing of this and similar facts. In the same locality, Campantar vanquished another Buddhist vƗdin, ĝƗriputra,
and defeated once again the Buddhists at Sattamankai (Ramachandran 1954: 7, 10).

10
probably not a structural component. The tƯrthikas did hold power and lands, very often
usurping the function of the kৢatriyas. For all the possible parallels in the ownership and
management of lands and estates, even Karl Marx was looking for an “Asiatic” mode of
production. Inexistent as the latter is, this suggests that it is more appropriate to look for
the operative mechanisms of a system within it.

3. Level of Confrontation

There are many examples of the escalating level of violence against Buddhist institutions
in the eighth century and later. Besides the intervention of kings, militias were formed in
ĝivaite ma‫ܒ‬has and in BhƗgavata circles as well to knock down the power of the
Ğrama۬as (Verardi 2011b: 231ff.). There are no appreciable differences among the vari-
ous regions of India but for the tools used and, quite often, for the different nature of the
documentation at our disposal. The dynamics observable in the Valley of Kathmandu, in
Orissa, at Ellora, and, again, in Kanchipuram, may help clarifying the point.
A hymn engraved on the Garuঌa pillar at AndigrƗma in ViĞalnagara (early Kath-
mandu) praises the ‫܈܀‬i VyƗsa for having cured the evils described in the Kali Age litera-
ture: “men had taken to atheism”, and “the false logicians (i.e. the Buddhist vƗdins) were
suppressing the truth”. These “disciples of the Sugata” are called “crooked distorters of
this world” (Regmi 1983: no. XXVII; see here fig. 4). The excavations carried out in the
adjacent area have shown that a stnjpa was dismantled after AD 749, that the area was
thoroughly cleared of all Buddhist memory, and that a Viৢ৆u temple was erected on the
Buddhist monuments razed to the ground (figs. 5, 6). The Garuঌa pillar was erected on a
base made with the bricks robbed from the stnjpa (Verardi 1992, I: 88௅90; see here fig. 7).
AndigrƗma became Harigaon, the village of the temple of Hari whose revenues were as-
signed to the maintenance of the Viৢ৆u temple.21 BhƗgavatism established itself in the
Valley as the hegemonic power, but the role played by the Sivaites in curbing down the
Ğrama۬as was not less effective. The evidence is provided by the transformation into
li۪gas of votive stnjpas, usually known as “Licchavi stnjpas” from the name of the royal
family that ruled over Nepal for a long time. They are provided with niches, but these are
empty because their images were chiselled away, and they have been deprived of their

21
The place name AndigrƗma is found in an inscription dated AD 749 from the excavated site (Verardi
1992, I: 143௅44).

11
finials (Wiesner 1980): an example comes from Chabahil, the CƗrumatƯ VihƗra said to
have been erected by the daughter of AĞoka (figs. 8௅9).
The local vaۨĞavƗlƯs keep memory of the conflict that cornered the Buddhists in
a subaltern position, the responsibility of past events being attributed, as commonly done
in these texts, to ĝaৄkarƗcƗrya. Although the saint probably never set foot in Nepal, we
cannot discard the information that these texts provide, because it integrates the evidence
given above. Some followers of Buddhism fled, we read, and some others were put to
death. Bhik‫܈‬u۬Ưs were forced to marry, monks were equated to g‫܀‬hastas, sacred texts
were destroyed. Yet, in some places, ĝaৄkara was obliged to leave the BauddhƗmargƯs as
temple priests (Wright 1877: 118௅23). We have, in essence, an account of the formation
and characteristics of NewƗr Buddhism, forcibly subaltern and yet strongly identitarian.
In Orissa, we have the chance to see the KƗpƗlikas in action. Reviled in Brah-
manical texts for the dirty jobs they executed, they were functionally connected with the
clean castes through mechanisms where the Brahmans, normally only associated with the
dak‫܈‬i۬abhƗva or performance of the five mahƗyajñas, could take part to heterodox forms
of worship, though with very strict limitations (van Kooij 1972: 29). The KƗpƗlikas, be-
sides exercising magic and black magic, had the task of getting rid of “learned Brahmans”
(Lorenzen 1972: 74௅77). The latter could hardly be the vaidikas or the tƯrthikas imbued
with ritualistic doctrine: the only learned Brahmans whose suppression makes sense were
those who had apostatized, especially the Buddhist controversialists who kept the ortho-
dox in check. The bhrnj۬ahan was the gravest of sins, and the mahƗvrata of the KƗ-
pƗlikas was exactly the penance for removing it (ibidem: 77௅81). The split personality of
these ascetics reflects the equally split personality of ĝiva, who, according to a little un-
derstood myth reported in the Matsya PurƗ۬a (184.81௅94; vol. 2, pp. 168௅69), cuts
BrahmƗ’s fifth head, resplendent for the asceticism. In the ĝiva PurƗ۬a, it is KƗla-
bhairava, one of the vƗma aspects of ĝiva and reference God of the KƗpƗlikas together
with CƗmu৆ঌƗ, the “left” aspect of the Goddess, who commits the murder
(RudrasaۨhitƗ: II.xxxiv.52; p. 434). ĝiva is condemned to wander from tƯrtha to tƯrtha
until he is delivered from his sin in Benares.22 It was ĝiva’s deed that the KƗpƗlikas re-
enacted, thus moving from one place to the other and ready to kill and be pardoned again.

22
Lorenzen (1972: 77௅81) examines the passage of the Matsya PurƗ۬a and provides references to other
relevant material. The VƗra৆ƗsƯ MƗhƗtmya of the Matsya PurƗ۬a has long been thought to be the earliest in
the literature, but see now Bakker and Isaacson (2004).

12
 The KƗpƗlikas are represented in a frieze in the VaitƗl Deul at Bhubaneshwar, na-
ked and with ka‫ܒ‬hva۪gas, i.e. sticks surmounted by the skull of the victims (fig. 10). In
this temple, they are associated with other groups of Sivaites, a central position being
given here to LakulƯĞa, who deeply restructured the movement of the PƗĞupatas in the
second or third century AD.23 The presiding deity in the garbhag‫܀‬ha of the temple is
CƗmu৆ঌƗ, flanked on both sides by the mƗt‫܀‬kƗs. Among the terrific forms of ĝiva, we
note a form of Bhairava sitting in a fighting posture, holding a sacrificial knife: a severed
head, with the feature of the Buddha, lies in front of the God, and two chopped heads are
depicted on a tripod on the pedestal (fig. 11).24 Outside the temple, there is a much worn,
reworked Buddhist sculpture serving as base of a ynjpa, the sacrificial post (fig. 12), testi-
fying to the sacrifices – including human sacrifices – offered up to the Goddess.25 It was
made reusing on purpose an architectural feature from some destroyed Buddhist edifice
(cf. Verardi 2011b: 235 and esp. note 222 on p. 261).
In the eighth century, the brahmanization of the Deccan was a done deal, and it
was time to break through to the north: Orissa found itself caught between the orthodox
push from the South and the new Buddhist power of the PƗlas. This situation put the stall
in Indian history for centuries, and did nothing but worsen. In both geostrategic and sym-
bolic terms, the objective was Magadha, the cradle of Buddhism, adjacent to both Bengal
and Orissa. Looking at the images of CƗmu৆ঌƗ scattered all over Orissa, we often see her
triumphantly seated either on a naked, dead man ௅ an asura often identifiable with a
Ğrama۬a,26 or on a tribal rƗjƗ threatened by the settlement policy of the tƯrthikas and
supporter of the Ğrama۬as. A stele from Kalamisri (fig. 13) makes us understand who
suppressed the heretics and their allies: a KƗpƗlika is still hitting, while the prey animals
begin to feed on the corpse ௅ primarily the owl, the vƗhana of the Goddess.

23
Opinions on LakulƯĞa, his religious role, the epoch when he lived, and even his historical reality, are ex-
tremely varied. I limit myself to recall, with Debala Mitra (1984: 104), the MathurƗ Gupta inscription of
AD 380, from which it may be inferred that a militant form of ĝivaism was already in existence.
24
The best description of this temple and the relevant iconographic material is still that provided by Pani-
grahi (1961: esp. 32 ff., 77௅81, 233௅34).
25
For the textual evidence on the balidƗna, see KƗlikƗ PurƗ۬a, 55; vol. 2, pp. 385 ff. This purƗ۬a is dated
to the mid-ninth century. See also the apocryphal story in the biography of Xuanzang where the pilgrim is
said to have run the risk of being sacrificed to the Goddess (Beal 1911: 86௅87).
26
Some of the dead, besides being naked in that they do not wear the cloth of the Veda, show elongated
ears and hair shaved or nearly shaved.

13
 In the majority of cases iconographies are not so explicit, and even less explicit is,
in general, the epigraphic record, not to speak of the many elusive myths, but integrating
the various sources to create a credible scenario is possible. In the epigraphic record and
literary works, the opposition lion/elephant is mentioned very often: elephants are de-
scribed as the enemies won in battle by the conquering lions. We have whole temples de-
fended by rampart lions, and other built on crushed elephants. In the KailƗĞanƗtha temple
at Kanchipuram, where rampart lions are ubiquitous, King RƗjasiূha Pallava (c. AD
700௅730) states that Puruৢottama (Viৢ৆u) was born to rescue from the ocean of sin the
sinking people swallowed by the monster known as the Kali Age, and defines himself
that pious king who has made the universe obedient to his orders: he has proved a royal
lion (RƗjasiূha) to the troops of the elephants of his foes (cf. Hultzsch 1890: no. 24, v. 8,
11; p. 14). The reference to the Kali Age brings us to wonder about the meaning of a re-
lief with ĝiva as a teacher seating under a tree with a couple of deer (fig. 14), a scene
which is not simply modelled on that of the Buddha’s first sermon but is intended to re-
place it: the elephant head below is a cut-off head that we also see under the other temple
panels that dipict ĝiva fighting the asuras. The meaning of lions and elephants, semanti-
cally ambiguous, becomes clear: their polarity appears, in context, as the opposition be-
tween the victorious orthodox kings (the lions) and the defeated Buddhists (the ele-
phants), the Buddha being the Elephant by definition.27
The best-known symbol of elephant hunting and killing is the
GajƗsuraূhƗramnjrti of ĝiva. The myth, set in Benares, is narrated in the Knjrma PurƗ۬a
in relation to the K৚ttivƗseĞvara Liৄga, the li۪ga of the Lord who is clad in the garment of
the elephant’s skin: the God has killed the daitya who had taken the form of an elephant,
and has made its hide his robe (Knjrma PurƗ۬a: 1.32.16௅18; vol.1, p. 253). Here the myth
allegorizes the clash between the Sivaite devotees and the Ğrama۬as, whose elder, the
Elephant, was killed as an intimidation act aimed at dispersing his followers. The Bud-
dhists-elephants are the major obstruction to the complete affirmation of ĝiva, and they
must be removed, or obliged to submit.28 The poems of Campantar explain this clearly:
“Those Buddhists and mad Jains may slander speak. / Such speech befits the wanderers
from the way. / But he who came to earth and begged for alms, / He is the thief who stole

27
On the tropos Buddhists/elephants, see discussion in Verardi (2011b: 221ff.). See also id. (1999/2000)
for the identity Buddha-elephant.
28
As emphasized by the same text a little below (l. 21), the Kali Yuga “is terrible and full of evil”.

14
my heart away. / The raging elephant charged down at him; / O marvel! He but took and
wore his hide” (Kingsbury and Phillips 1921: no. 18; p. 27).
The exploits of ĝiva and Viৢ৆u against the asuras and their mythicization in the
devƗsura war barely conceal the identity Devas/Brahmans and Asuras/Heretics
(pƗsa۬‫ڲ‬as and nƗstikas). In a number of texts, the heretics are described as nagdas, na-
ked,29 because, as said above, they are deprived of the cloth of the Veda. This is the rea-
son why we can identify the asuras killed by KƗlabhairava with the help of Ga৆eĞa and
the Goddesses in the yajñaĞƗlƗ (fig. 15) of the mammoth KailƗĞanƗtha temple at Ellora in
upper Deccan. The Goddesses take directly part in the suppression of the asuras fighting
against them and drinking their blood, as is explained in the DevƯ MƗhƗtmya (8.7௅62; pp.
63௅67). What is then the yajñaĞƗlƗ? It may be interpreted either as the actual place where
special yƗjñas, centred on the ritual killing of religious enemies were performed or as the
stone rendition of the temporary structure erected outside the temple where the heretics
were or had been executed.

4. The VajrayƗna Revolt

The GuhyasamƗja Tantra delivers a radical attack to social institutions and socially ac-
cepted behaviours maintaining that no one can succeed in obtaining perfection through
difficult, painful processes and that one can succeed easily through the satisfaction of all
desires. MahƗvairocana declares that the outcaste, workers and similar people find per-
fection in this yƗna, which is likewise for those who do not respect the life of the others,
who delight in telling falsehood, who steal other people’s property, who continuously
take recourse to sensual pleasures, etc. (GuhyasamƗja Tantra: V. 3௅7; cf. Gnoli 1983, pp.
631௅32).30 One would say, at first glance, that this text actively favours the establishment
of a collapsed, broken-up society: the addressees of the initiates are not only the outcaste,
but the murderers, thieves, liars, immoral and incestuous persons, the defamers. However,
many of these epithets were just those given by the orthodox to their opponents in order

29
See e.g. VƗyu PurƗ۬a (II.16.31; vol. 2, p. 613). The text makes the example, among others, of Brahmans
who shave their hair “for nothing”.
30
For a complete translation of this tantra, the reader is referred to Gäng (1988).

15
to defame them, and we probably have to see in the text a deliberate provocation.31 We
should examine Indian VajrayƗna in its initial, generating context, limiting the level of
symbolic explanation to a minimum, and further considering that we mostly deal with
metaphors, not with symbols in the proper sense.
The authors of the GuhyasamƗja Tantra and other texts addressed social sectors
only partly superimposable to those on which the Buddhists had staked to establish their
hegemony in the past. The “bourgeoisie” of trade and industry had disappeared a long
time but for Bengal, and the attempt of the MahƗyƗna to counteract the agricultural poli-
cy of the tƯrthikas on their own ground had failed. The lands owned by the monasteries
were deeply infiltrated by theistic and caste practices, even where, as in Bengal, the
monastic institutions could not easily be blackmailed with the threat of destruction. It is
not possible, however, to explain the rise and success of the VajrayƗna abstracting from
the rise and success of PƗla power. In the second half of the eighth century, throughout
the ninth century and even later, the PƗlas succeeded in controlling a large part of the
subcontinent and provided a safety net to the advocates of the new yƗna. DharmapƗla
(AD 775௅800) conquered geostrategic Kanauj, and extended his control on the PanjƗb
and GandhƗra (Kielhorn 1896௅97: v. 12) and on parts of the Deccan, and his successor
DevapƗla (c. 800௅840) was equally successful: with his successors, too, it seemed that
northern India, and even parts of the South, could be restored to the religion of Dharma.
The distinctive traits of the VajrayƗna are the legitimation of violence in order to
defend the religion and the acquisition at the theoretic and not only factual level of sexual
behaviours in open opposition to the caste-and-family social model of the tƯrthikas. Re-
garding violence, the Buddhists had learnt the lesson imparted to them by the violent
eight-century events. A rethinking of the whole matter had started even before, arguably
in connection with the violence experienced in Gupta times,32 but now the GuhyasamƗja
Tantra explicitly invited concentration on the three-pronged vajra “that paralyzes all the

31
I no longer think that they were simply paradoxical (Verardi 2011b: 304), but that the followers of the
VajrayƗna made their own the insults of the tƯrthikas, transforming them into a weapon (as if to say: we are
murderers? Well, yes, that’s right: here we are…).
32
According to Asaৄga’s Bodhisattvabhnjmi, the Bodhisattva is ready to be reborn in hell to kill those po-
tentially responsible for the slaying of innocent beings. He would fight to get back the property of the
saۨgha or a stnjpa robbed by “thieves and bandits” (cf. MacFarlane 1995: 194, quoting from Unrai Wogi-
hara’s edition of the text, Tokyo 1930௅36, 165௅7). With the latter, the armed groups organized by the
tƯrthikas in order to intimidate the monks are probably alluded to (see Verardi 2011b: 231 ff.).

16
non-Buddhist teachers” projecting it on the head of the enemy, which will not prevail
against the buddhasainya, the army of the Buddha (cf. Davidson 2002: 193௅94). In the
SarvatathƗgatatattva Saۨgraha, VajrapƗ৆i warns the TathƗgatas against the existence of
criminals such as MaheĞvara and other Brahmanical gods, who are summoned by a man-
tra. ĝiva is eventually annihilated by VajrapƗ৆i, by whom the other gods take also refuge.
He abandons his form of MahƗdeva and is reborn, entering the ma۬‫ڲ‬ala with another
name (cf. id. 1991: 200௅2). The ma۬‫ڲ‬ala is thus the conceptualization of a territorial
space where Brahmanical power must be terminated with violent means if necessary to
allow the Buddhists to recreate the Kingdom of Dharma. The “great strategy” of yore, the
MahƗyƗna, was ineffective and obsolete; a strategy of violence, exemplified by the vajra,
was now needed.
The social sectors involved in the VajrayƗna revolt included, besides the outcaste
and lower-caste people exemplified by the social origin of the eighty-four siddhas,33 the
“tribals” who opposed the reclamation of land and their forced annexation to the caste
society. The tribal chiefs of the Vindhyas are the men with long hair, often styled in
elaborate manner, lying dead under the Goddess or with their heads chopped off (fig. 16)
or, in certain cases, surrendering and imploring her clemency. We know them also from
the literature (the ĝabaras are topical) and, if the explanatory model presented here makes
sense, from the very tradition on the origins of the VajrayƗna as well. A tribal chief
probably was king Indrabhnjti of UঌঌiyƗna, where VajrayƗna Buddhism originated (Tucci
1949: 212௅13).34 He may be a mythical king, but myth can be explored and contextual-
ized: it is reasonable to believe that he was one of those chiefs who took advantage of the
situation arisen with the military campaigns of DharmapƗla and the temporary disruption
of the Brahmanical social order imposed in the Northwest by the ৡƗhƯs (Verardi 2011a:
159ff., 168 ff.).
Buddhist images documenting violence as a political means started being pro-
duced as early as the ninth-tenth century in Bihar, where we see AparƗjitƗ triumphing on

33
The reader may go through the CaturaĞƯtisiddha Prav‫܀‬tti, known to us through a Tibetan translation. Cf.
Robinson (1979); discussion in Verardi (2011b: 346௅47).
34
On the VajrayƗna and North-western India, cf. also Davidson (2002: passim). The line of interpretation
according to which the origin of the VajrayƗna should be looked for in eastern India, particularly in Orissa
(identified with UঌঌiyƗ৆a by some authors; see e.g. Sahu 1958: 152 ff.), is due to the overflowing amount
of textual and iconographic documentation that comes from that area ௅ a documentation, however, in
which converge very different materials indiscriminately labelled as “Tantric”.

17
a subjugated Ga৆eĞa (fig. 17). At NƗlandƗ, Trailokyavijaya stands victoriously on ĝiva
and PƗrvatƯ annihilated (fig. 18). On behalf of these iconographies, which would multiply
over time and grow in ferocity, it appears that it is difficult to distinguish between the
siddha movement and institutional Buddhism (the monasteries of NƗlandƗ, for instance,
where similar images were produced; cf. Misra 1998, III: 143, 149 ff.). We actually do
not know how long it took to the shocked Bodhisattvas, namely the great MahƗyƗna
monks, attending the saۨgƯti in the GuhyasamƗja Tantra (V. 8; cf. Gnoli 1983, p. 632) to
accept the the Vajra strategy. But texts, too, could hardly be produced without the active
collaboration of learned monks capable of rethinking the fundamentals of Buddhism.
In the (partly) ritualized sexual practices introduced into the system, scholars have
seen since long a giving in to pressure from the bottom of society. However, rather than a
giving in, we should speak of a conscious, even if risky strategy aimed at becoming fully
credible with the social sectors deemed essential to the revival of Buddhism, at mixing
with them.35 The taking on in person of sexual behaviours undermining social order is
just a reversal of strict monasticism, always criticized as anti-social by the tƯrthikas. We
are in front of a renewed, violent opposition to the caste-and-family model, as intolerably
disruptive as strict monasticism.36 The one and the other are also sharp alternatives to the
compromising householder monks.
The history of the lowermost castes and the untouchables has attracted little atten-
tion in India, where research is inhibited by the preconception according to which the
Brahmans would have created a social system where the conflict rate was low (though at
the price of reinforcing the caste system and creating the barrier of untouchability). This
view is a legacy of nineteenth-century historiography, which created an exotic India
compensating Europe and its violent history. The VajrayƗna revolt ravaged northern In-
dia and was a serious challenge to the establishment. We observe a progressive radicali-

35
This is also true of certain fringes of Sivaism (here we have mentioned the KƗpalikƗs) as has often been
pointed out by scholars, and in Bengal it would be also true of Visnuism, which, using every means, would
inherit the Buddhist masses left without points of reference. But we cannot be satisfied with analyses that
stop at the phenomenological level. As I am trying to explain in these pages, the aims of the Buddhists and
of the theistic movements were very different. The purpose of the latter was to restore a rigid order in soci-
ety.
36
Sometimes we forget the meaning of even the best known events in the Buddha’s life: that SiddhƗrtha, in
contrast with the rules that a twice-born should have followed, left his young wife was an extremely pro-
vocative act.

18
zation of the uprising, and the open-air temples of the yoginƯs may be taken as indicators
of the increasingly difficult social, ethnic and inter-religious situation. These temples are
situated along the critical borderline constituted by the Vindhyas, from eastern Rajasthan
to Bhubaneshwar, and arise in places that are secluded but not distant from the local cen-
tres of power.37 The yoginƯs are a spatial projection of the Goddess in her vƗma aspect,
and in fact, at the centre of the yoginƯ enclosures (fig. 19) there usually is an aedicule
with the images of CƗmu৆ঌƗ and ĝiva in one of his terrific forms. The enclosures are, on
reflection, a functional transformation of the garbhag‫܀‬ha of the VaitƗl Deul, where, as
we have seen, two rows of mƗt‫܀‬kƗs flank CƗmu৆ঌƗ on both sides, and of the yajñaĞƗla of
Ellora, less structured though the latter is. It is unlikely that the people suppressed in
these enclosures were “normal” enemies killed in war: we are out of the logic of the bat-
tlefield, the natural place for the fallen to die; it is more probable that the victims were
killed in odium fidei. They were the protagonists of an insidious guerrilla war whose
stake was BodhgayƗ. That in a number of cases religious enemies were meant, appears in
a tenth-century stele from Bihar where CƗmu৆ঌƗ, after beheading a group of enemies
(mostly men of religion), takes her seat under the pipal tree on a naked asura who has the
shape of the Buddha, whom she has dislodged from his throne. A few other chopped-off
heads, two of which showing u‫۬܈‬Ư‫܈‬Ɨs, are in the sacrificial basin in the lower part of the
stele, together with amputated legs and arms (figs. 20௅21).
The Sivaites were not alone in the attempt at eradicating Buddhism. In the ninth
century, the Viৢ৆uites gained control of GayƗ, where they appointed the GayƗvƗlas, a
lower class of Brahmans performers of ĞrƗddhas. ĝrƗddha rituals are associated with the
instalment of Viৢ৆u-GadƗdhara in the town’s main temple:38 with his “sharp axe”, the
God had hewn “the tree of great sin” (VƗmana PurƗ۬a: 50.15, pp. 424௅25). The tree is
naturally that of the Awakening. In fact, the Visnuites appropriated BodhgayƗ, as appears
from the archaeological documentation (F. Barba in Verardi 2011b: 409௅11) and as is at-
tested to in the Kalki PurƗ۬a, a late composite work preserving an extremely valuable

37
For a description of the yoginƯ temples, I refer the reader to Dehejia (1986) and, in relation to Orissa, to
Donaldson (2002), where he will find a lot of interesting material. These authors would probably not share
my view on the matter, which I have discussed in greater detail elsewhere (Verardi (2011b: 288௅89,
294௅95).
38
Viৢ৆u GadƗdhara, a violent manifestation of the God, is first mentioned in an inscription datable to AD
1058 (Sircar 1964௅65: v. 9). With reference to BodhgayƗ, the frightful nature of Viৢ৆u is made explicit
also in the GayƗsura myth, first analyzed by Rajendralala Mitra (1878: 9 ff.).

19
material. It contains a detailed description of the battle of KƯka৬a (BodhgayƗ/Magadha)
waged by Kalki, the last avatƗra of Viৢ৆u, where the Buddhists are said to have drawn
up in battle order accompanied by “millions of outcaste” (Kalki PurƗ۬a: II.vii. 38; p. 80).

5. The Suppression of Buddhism in Magadha and Eastern India

The “Turuৢkas” started advancing into northern India when the conflict between the
tƯrthikas and the VajrayƗnists was reaching its apex. Indian powers fought against the
Turuৢkas for long, while the Buddhists tried, in places, to benefit from the situation
weakening the tƯrthikas: besides stirring up the lower classes, they often sided with the
Muslims, as had happened in eighth-century Sind, where the good relationships between
Muslims and Buddhists are well documented. 39 When the tƯrthikas realized that they
would never defeat the Turuৢkas in battle, they strove to separate the Muslims from the
Buddhists striking the latter very hard. Finally, the tƯrthikas compromised with the invad-
ers, particularly under the Delhi Sultanate (1206௅1290). The compromise allowed them
to re-establish social order in certain territories, to retain an operational political space
and reorganize themselves.
The actors on the Indian political scene were very numerous, but here we focus
on the GahƗঌavƗlas, who ruled MadhyadeĞa from Kanauj and Benares and were, in
places, supportive of the Buddhists, and on the Senas. The latter were an orthodox dy-
nasty original from South India that put an end to PƗla rule in Bengal and inflicted a mor-
tal blow to the Buddhist monasteries of eastern India. In AD 1128௅29, the GahƗঌavƗla
king Govindacandra donated six villages to the community of ĞƗkyabhik‫܈‬us of the Jeta-
vana at Sravasti (Sahni 1911௅12), a policy followed by his local vassals, too (Kielhorn
1888: 1. 15௅16; cf. Niyogi 1959: 258). Two of his queens were Buddhist ௅ an early
model of differentiated social intervention that was again put to test. Who were the
ĞƗkyabhik‫܈‬us? Our closest chronological and spatial reference are the ĞƗkyabhik‫܈‬us, or
householder monks, of NewƗr Buddhism ௅ a crucial, though little-used “fossil guide” to
Indian medieval Buddhism. 40 It is reasonable to think that the GahƗঌavƗla donations

39
Maclean (1989: esp. 51௅52, 58, but passim). For Sind as a place where the fundamentals were fixed for
the later interaction between Buddhists and Muslims, cf. Verardi (2011b: 350ff.).
40
John Locke, to whom we owe the most valuable contribution to NewƗr Buddhism, maintains that the lat-
ter was probably structured on the presence of householder monks, the ĞƗkyabhik‫܈‬us or ĞƗkyavaۨĞas, since
its inception (Locke 1985: 4, 484).

20
went to the advantage of the more integrated, non-violent sectors of Buddhism. Queen
KumƗradevƯ, although not endowed with all the royal prerogatives (Niyogi 1959: 199),
donated a vihƗra to the sthavƯra of the Buddhist community of Sarnath to honour the
Dharmacakra Jina, whose image she also restored (Konow 1907௅8: v. 20௅23).41
Govindacandra was exposed, from the west, to the attacks by the Muslim rulers of
Panjab, to whom he paid tribute (D.C. Ganguly in R.C. Majumdar 1957: 51), and, from
the east, to those of the hardliner Senas. His successor Jayacandra (c. AD 1170௅93)
found himself in an impasse that led him to radical choices. He owned a white elephant
(Niyogi 1959: 198) ௅ an animal, and a colour, laden with precise symbolism in Bud-
dhism.42 We have a clue regarding his policy from the fact that his dik‫܈‬aguru was ĝrƯmi-
tra, a siddha, whom he praises because at the MahƗbodhi he had recovered numerous lost
scriptures (N. Sanyal 1929: esp. 11.7, 11; Niyogi 1959: 198, 210, 260), arguably those
pertaining to the third setting in motion of the Wheel of the Law. Kanauj was lost, and
the dynasty controlled only parts of its former territories from Benares. It is not difficult
to imagine the opposition that, exactly in Benares, the policy of the king had to meet. At
Sarnath, a very large Brahmanical temple was built in the northern area, where the so-
called “Monastery I” is located. The area appears as a walled compound accessible
through huge gateways, erected on a dismantled Buddhist monastery the inner court of
which was made to coincide with the courtyard of the new building (F. Barba in Verardi
2011b: 427 ff.). “Monastery I” started being called “Dharmacakra JinavihƗra” (Sahni
1923: 28) with reference to the donation of KumƗradevƯ, but the relevant inscription is
not associated with “Monastery I” (Mani 2005௅6). At least fourteen Brahmanical sculp-
tures and architectural fragments were found in the area (cf. F. Barba in Verardi 2011b).
D.R. Sahni, who carried out excavations at Sarnath after John Marshall, understood that
these fragments must have belonged to a Brahmanical temple (Sahni 1914: 243௅44) but
was unable to locate it because he failed to grasp the stratigraphic conundrum. Around
AD 1200, the Lord of Isipatana may be said to have been ĝiva SƗraৄganƗtha (hence SƗr-
nƗth), usurping the park and the deer from ĝƗkyamuni.

41
KumƗradevƯ refers explicitly to the Jina, and does not seem to have paid homage to a form of ĝiva such
as Aূ৚teĞvara that assimilates the Buddha and bestows “the reward of liberation upon women” (Sanderson
2004: 254).
42
The white colour is the sum of all other colours, thus being “central”. This is why the Buddha appears to
Queen MƗyƗ as a white elephant, and why the colour of Vairocana (a hypostasis of the Buddha), whose
position is central in relation to the other Buddhas, is white.

21
 Research work on the Sena dynasty is still limited, and scholars continue to ad-
here to the misleading construct of a unified PƗla-Sena period. The concern of the Senas,
while the Turuৢkas were advancing towards the middle and eastern Ganges valley, was to
establish var۬ƗĞramadharma in the territories that came under their rule. King VallƗ-
lasena (c. 1158௅1179) started abusing the Buddhist merchants, who were numerous in
Bengal. Vallabha Ɩঌhya, the richest merchant and banker of the region and leader of the
SonƗr-VaniƗs, had always financed the king: when the latter showed the intention to
march against Magadha, he refused to do so, and the VaniƗs were forcibly expelled from
Bengal. Those who remained were downgraded: the Brahmans were prohibited from
teaching them and officiating for them (H. P. ShƗstri 1911: 21௅22). The Senas started a
radical administrative-territorial overhaul of the former PƗla territories. Judging from the
vijayak‫܈‬etras established in the various grƗmas, villages were conceived as administra-
tive and political units of the new power and milestones in the creation of a caste agrarian
society. The largest concentration of these new centres was in the navigable sector of
Vaৄga and near the estuarine mouth of the Bay of Bengal (R. Sanyal 2008௅9:101). This
may be construed as a determined attempt at curbing down trade ௅ or at least the trade
that was in Buddhist hands.
The Senas are responsible for the attacks on Buddhist monasteries. We have an
echo of the attack against Somapura/Paharpur from a Nalanda inscription, according to
which “the ascetic Karu৆aĞrƯmitra … when his house was burning, (being) set on fire by
the approaching army of VaৄgƗla, attached (himself) to the pair of lotus feet of the Bud-
dha, (and) went to heaven” (N.G. Majumdar 1931௅32: v. 2௅3). We also know of the at-
tacks against Vikramasila/Antichak, whose end is commonly but groundlessly attributed
to BakhtiyƗr-i KhaljƯ, the military adventurer who first opened to the Muslims the gates
of eastern India in the early thirteenth century and who never passed through the region
of Vikramasila to reach Bengal (Chaudhary 1978: 217௅18). A Brahmanical temple was
built on the Buddhist ruins of the monastery before the arrival of the Muslims, to be
abandoned later on. Other attempts to destroy the monastery had already taken place. An
inscription on a pillar stump mentions a local chief, Sahura of CampƗ, who had dispelled
a planned attack by “the rulers of Baৄga” (Verma 2001, II: 303): evidently, the Senas.43

43
The archaeologists responsible for the excavations at Antichak show the traditional reticence and misun-
derstanding about the dynamics of destruction of the Buddhist monasteries of eastern India. B.S. Verma
(2011: 17) writes that “[t]he introduction of tantra and the worship of large number of gods and goddesses

22
 At OdantapurƯ/Bihar Sharif the monastery, which had already withstood repeated
onslaughts of VallƗlasena, was adjacent to the PƗla camp (Ali 1406 H/1985, I A: 50). The
Muslim attack is narrated in a famous passage of the ܑabaqƗt-i NƗ‫܈‬irƯ (I: 551௅52). The
Muslims were apparently unaware of the real nature of the site, and were upset when
they discovered the truth. There were no soldiers in the fortress, as became clear after the
killing of the monks, and we must think that the Turuৢkas, who based their field opera-
tions on the information provided by local informers ௅ inevitably, tƯrthikas ௅ were led to
a place that was not defended at all. The Ratnodadhi library at NƗlandƗ, we learn from
the Tibetan sources, was put on fire by the tƯrthikas,44 and DharmasvƗmin, who resided
there in 1234௅35, reports a story whose sense is that the Muslims had been instructed to
launch an attack on the monastery, but gave the game away so that the monks could save
themselves (Roerich 1959: 33b௅34a; pp. 31௅32, 93௅94).
The rivalries between Buddhists played a role in accelerating the collapse of the
religion. The monks of Sri Lanka, who had a large part in the management of BodhgayƗ
since Gupta times, were opposed to MahƗyƗnists and VajrayƗnists. DharmasvƗmin was
rebuked for carrying with him a manuscript of the A‫܈‬tasƗhaĞrikƗ PrajñƗpƗramitƗ, which
he had to hide, the censor being the ĞrƗvaka keeper of VajrƗsana proper (ibidem:
19b௅20a; pp. 18௅19, 73௅74). TƗranƗtha reports an episode that he attributes to the time of
DharmapƗla, but which is probably later on behalf of the fact that it is about an image of
Heruka and the “many treatises on Tantra”. The ĞrƗvakas of the Island and their allies
burnt the books, and smashed the image into pieces (TƗranƗtha: 109A, p. 279). Inside the
three gates giving access to the MahƗbodhi Temple and the VajrƗsana, only the sacristans
could sleep: the point is that there were “three hundred sacristans native of Ceylon, who
belong[ed] to the ĝrƗvaka school; other (schools) ha[d] no such right” (Roerich 1959:
19b; pp. 18, 73). Why Indian Buddhist monks and priests were excluded from control of
the VajrƗsana? The only possible explanation is that the foreign monks helped in clean-
sing the site from the supporters and sympathizers of the rebels. The ĞrƗvakas had no fol-
lowing in India, they were not a threat, and could be sent back home at any time.
The reversal of alliances, which caused the Buddhists to fight on two fronts,
against the tƯrthikas and against the Muslims, and eventually be suppressed, is exempli-

in this new form of Buddhism narrowed down the gulf between Buddhism and Hinduism and it did not
take long time for the Buddhists to transfer their allegiance wholely to the Hindu gods and goddesses […]”.
44
dPag bsam ljon bzang I.92 and Index, i (“Kaku৬a Sidha”). Cf. Vidyabhusana (1921: 516).

23
fied by the creation of the state of MithilƗ, which included northern Bihar and a part of
Magadha, with BodhgayƗ. DharmasvƗmin reports that the few monks residing there in
1235 camouflaged the entrance to the MahƗbodhi temple with an image of ĝiva: thereby
the Turuৢkas would not attack it (ibidem: 64). This means that the Muslims spared the re-
ligious buildings of the tƯrthikas, being induced to attack those of the Buddhists. The
kingdom of MithilƗ under the Kar৆Ɨ৬a kings and their successors kept its independence
thanks to the alliance that king Narasiূha made with the SultƗn of Delhi, to whom he
paid tribute: he helped him to capture the whole of South Bihar in the early 13th century.
MithilƗ’s “national” poet VidyƗpati (c. 1352௅1448) would account for the alliance be-
tween king Sakrasiূha (1296௅1316) and ‘AlƗ al-DƯn KhaljƯ. He considered IbrƗhƯm
ShƗh, the SharqƯ ruler of Jaunpur (1402௅1440), second only to God, and his capital a
second AmarƗvatƯ, and persuaded him to organize a military expedition to put an end to
the still chaotic conditions prevailing in the country (Chaudhary 1970: 71). The majority
of historians are reticent on what took place in the region, and we must be content with
either short, apodictic statements or cursory admissions.45
Regarding the Buddhist communities that survived in Bengal and Orissa until the
sixteenth century, their history remains largely to be written.

45
B.J. Hasrat, a distant heir of those who, in Nepal, suffered the backlash of the events, has observed, for
instance, that “… in the plains of India … the prosecution of the Buddhists was furious ௅ root and branch
eradication in fact” (Hasrat 1970: 39, note 1).

24
 References

Agrawala, Vasudeva S. (19832), VƗmana PurƗ۬a: A Study. Varanasi: Banaras Hindu Uni-
versity (1st ed. 1964).
Ali, Muhammad Mohar (1406 H/1985), History of the Muslims in Bengal, vol. IA: Mus-
lim Rule in Bengal (600࣓1170/1203࣓1757); vol. IB: Survey of Administration, Society
and Culture. Riyadh: Imam Muhammad Ibn Sa’njd Islamic University.
Bakker, Hans (1997), The VƗkƗ‫ܒ‬akas: An Essay in Hindu Iconology. Groningen: Egbert
Forsten.
Bakker, Hans, and Harunaga Isaacson (2004), “A Sketch of the Religious History of
VƗrƗ৆asƯ up to the Islamic Conquest and the New Beginning”, in The SkandapurƗ۬a,
Vol. IIA, AdhyƗyas 26௅31.14. The VƗrƗ۬asƯ Cycle. Groningen: Forsten.
Beal, Samuel. 1884. Si-Yu-ki. Buddhist Records of the Western World, Translated from
the Chinese of Hiuen Tsiang (AD 629) by Samuel Beal, 2 vols. London: Trübner & Co.
௅௅௅௅௅ (1911), The Life of Hiuen-Tsiang by the Shaman Hwui Li, with an Introduction:
Containing an Account of the Works of I-Tsing. London: Kegan Paul, Trench, Trübner
& Co.
Chaudhary, Radhakrishna (1978), “Decline of the University of VikramaĞƯla”. Journal of
Indian History 56, 213௅35. Trivandrum.
௅௅௅௅௅ (1970), History of Muslim Rule in Tirhut (1206࣓1765 AD). The Chowkhamba Sanskrit
Studies, 72. Varanasi.
Davidson, Ronald M. (1991), “Reflections on the MaheĞvara Subjugation Myth: Indic
Materials, Sa.skya-pa Apologetics, and the Birth of Heruka”. The Journal of the Inter-
national Association of Buddhist Studies, 14, 197௅235. Madison: International Asso-
ciation of Buddhist Studies.
௅௅௅௅௅ (2002), Indian Esoteric Buddhism: A Social History of the Tantric Movement.
New York: Columbia University Press.
Deeg, Max (2005) Das Gaoseng-Faxian-Zhuan als religionsgeschichtliche Quelle. Der
älteste Bericht eines chinesischen buddhistischen Pilgermönchs über seine Reise nach
Indien mit Übersetzung des Textes. Wiesbaden: Harrassowitz Verlag.
Dehejia, Vidya (1986), YoginƯ Cult and Temples: A Tantric Tradition. New Delhi: Na-
tional Museum.

25
Deva, Krishna (1984) Images of Nepal. New Delhi: Archaeological Survey of India.
DevƯ MƗhƗtmya = Thomas B. Coburn (1991), Encountering the Goddess: A Translation
of the DevƯ MƗhƗtmya and a Study of Its Interpretation. Albany: State University of
New York.
Dikshitar, V.R. Ramachandra (1931), “Buddhism in Tamil Literature”, in Buddhistic
Studies, edited by B.C. Law, 673௅98. Calcutta: Thacker, Spink & Co.
Donaldson, Thomas E. (2002), Tantra and ĝƗkta Art of Orissa, 3 vols. New Delhi: D.K.
Printworld.
Eltschinger, Vincent (2000), « Caste » et philosophie bouddhique. Continuité de
quelques arguments bouddhiques contre le traitement réaliste des dénominations
sociales. Wiener Studien zur Tibetologie und Buddhismuskunde, 47. Universität
Wien.
Gäng, Peter, trans. (1988), Das Tantra der verborgenen Vereinigung: Guhyasamaja-
Tantra. Aus dem Saskrit übersetz und herausgegeben von ࣓࣓࣓࣓࣓. Munich: Eugen Die-
derichs Verlag.
Giles, H[erbert] A[llen] (1923), The Travels of Fa-hsien (399࣓414 AD), or Record of the
Buddhistic Kingdoms. Re-translated by ௅௅௅௅௅. London: Routledge & Kegan Paul.
Gnoli, Raniero (1983), Testi buddhisti in sanscrito. Torino: Unione Tipografico-Editrice
Torinese.
Gottfried Williams, Joanna (1982), The Art of Gupta India: Empire and Province. Prince-
ton: Princeton University Press.
Granoff, Phyllis (1989), “Jain Lives of Haribhadra: An Inquiry into the Sources and Logic
of the Legends”. Journal of Indian Philsophy, 17, 105௅28. Dordrecht.
Gros, François (1984), “Introduction: Pour lire les TƝvƗram / Towards Reading the TƝvƗ-
ram”, in TƝvƗram. Hymnes ĝivaïtes du pays tamoul, vol. I, ÑƗnacampantar, edited by
T.V. Gopal Iyer and F. Gros, v-xxxvi/xxxvii-lviii. Publications de l’Institut Français
d’Indologie, 68/1. Pondichéry.
Hakker, Paul (1958), “ƖnvƯkৢikƯ ”. Wiener Zeitschrift für die Kunde Süd- und Ostasiens
und Archiv fur Indische Philosophie 2, 54௅83. Universität Wien: Indologisches Insti-
tut.
Hasrat, Bikrama Jit (1970), History of Nepal as Told by Its Own and Contemporary
Chroniclers. Edited with a Prolegomena by ௅௅௅௅௅. With a Foreword by Donovan Wil-
liams. Hoshiarpur: V.V. Research Institute Press.

26
Hudson, Dennis (2002), “Early Evidence of the PƗñcarƗtra”, in The Roots of Tantra, ed-
ited by K.A. Harper and R. L. Brown, 133௅67. Albany: State University of New York.
Hultzsch, Eugen (1890), South Indian Inscriptions, Tamil and Sanskrit, from Stone and
Copper-Plate Edicts at Mamallapuram, Kanchipuram, in the North Arcot District, and
Other Parts of the Madras Presidency, Chiefly Collected in 1886࣓87. Madras: Ar-
chaeological Survey of Southern India.
Humfress, Caroline (2007), Orthodoxy and the Courts in Late Antiquity. Oxford: Oxford
University Press.
Jayaswal, K[ashi] P[rasad] (1937), An Imperial History of India in a Sanskrit Text [c.
700 BC–c. 770 AD], with a Special Commentary on Later Gupta Period. With the
Sanskrit Text Revised by RƗhula SƗnk‫܀‬ityƗyana. Patna: Motilal Banarsidass.
KalikƗ PurƗ۬a = Shastri, B[iswa] N[arayan] (1992), The KƗlikƗpurƗ۬a (Text, Introduc-
tion and English Translation with Shloka Index), edited by S. Pratap. Delhi: Nag
Publisher.
Kalki PurƗ۬a = Le Kalki-PurƗ۬a. Première traduction du sanskrit en langue occidentale
de Murari Bhatt et Jean Rémy, suivi d’une étude d’André Préau. Préface de Jean
Varenne. Bibliothèque de l’Unicorne. Milano 1982 : Archè. [Translation of the
Sanskrit text edited by ĝrƯ JƯvƗnanda VidyƗsƗgara BhattƗcharya, Calcutta 1860].
Kielhorn, F[ranz] (1888), “A Buddhist Stone Inscription from Sravasti of [Vikrama]-
Samvat 1276”. The Indian Antiquary: A Journal of Oriental Research, 17, 61௅64.
Bombay.
௅௅௅௅௅ (1896௅97), “Khalimpur Plate of Dharmapaladeva”. Epigraphia Indica, 4, 243௅54.
Calcutta.
Kingsbury, F[rancis], and G[odfrey] E. Phillips (1921), Hymns of the Tamil ĝaivite Saints.
Calcutta: Association Press / London: Oxford University Press.
Konow, Sten (1907௅8), “Sarnath Inscription of Kumaradevi”. Epigraphia Indica, 9,
319௅28. Calcutta.
van Kooij, Karel R. (1972), “Introduction” to: Worship of the Goddess According to the
KƗlikƗpurƗ۬a, Part I: A Translation with an Introduction and Notes on Chapters 54࣓69,
1௅37. Orientalia Rheno-Traiectina, 14. Leiden: Brill.
Knjrma PurƗ۬a = The Knjrma PurƗ۬a. Translated and Annotated by G[anesh] V[asudeo]
Tagare, 2 vols. Delhi 1981: Motilal Banarsidass.
Lal, B[raj] B[asi] (1993), Excavations at ĝ‫܀‬i۪gaverapura (1977࣓86), Volume I. Memoirs
of the Archaeological Survey of India, 88. New Delhi: Archaeological Survey of India.

27
Law, B[imala] C[hurn] (1967), The Mission of Wang Hiuen-ts’e in India (Les Missions de
Wang-Hiuen-Ts’e dans l’Inde). Written in French by M. Sylvain Levi, Translated from the
Original French by S.P. Chatterjee; Edited with Notes by ௅௅௅௅௅. Calcutta: Indian Geo-
graphical Society.
Locke, John K. (1985), Buddhist Monasteries of Nepal: A Survey of the BƗhƗs and BahƯs of
the Kathmandu Valley. Kathmandu: Sahayogi Press.
Lorenzen, David N. (1972), The KƗpƗlikas and KƗlƗmukhas: Two Lost ĝaivite Sects.
Berkeley௅Los Angeles: University of California Press.
Maclean, Derryl N. (1989), Religion and Society in Arab Sind. Monographs and Theoreti-
cal Studies in Sociology and Anthropology in Honour of Nels Anderson, 25. Leiden:
Brill.
Majumdar, N.G. (1931௅32), “NƗlandƗ Inscription of VipulaĞrƯmitra”. Epigraphia Indica,
21, 97௅101. Calcutta.
Majumdar, R.C., ed. (1957), The Struggle for Empire. The History and Culture of the In-
dian People, 5. Bombay: Bharatiya Vidya Bhavan.
Mani, B.R. (2005-6), “The Enigmatic Monastery of KumƗradevƯ at Sarnath: New Identi-
fication”. PrƗgdhƗrƗ. Journal of the U.P. State Archaeological Department, 16,
157௅64. Lucknow.
Matsya PurƗ۬a = The Matsya Puranam. Translated by a Taluqdar of Oudh. New Delhi
19802: Oriental Books Reprint Corporation. (1st ed., Allahabad 1916).
McFarlane, Stewart (1995), “Fighting Bodhisattvas and Inner Warriors: Buddhism and the
Martial Traditions of China and Japan”, in The Buddhist Forum, Volume III, 1991࣓1993.
Papers in Honour and Appreciation of Professor David Seyforet Ruegg’s Contribution
to Indological, Buddhist and Tibetan Studies, edited by T. Skorupski and U. Pagel,
185௅210. New Delhi: Heritage Publishers.
Minakshi, C[adambi] (1938), Administration and Social Life under the Pallavas. Madras
University Historical Series, 13. University of Madras.
Misra, B[haskara] N[atha] (1998), NƗlandƗ, 3 vols. Delhi: B.R. Publishing Corporation.
Mitra, Debala (1984), “LakulƯĞa and Early ĝaiva Temples in Orissa”, in Discourses on
ĝiva: Proceedings of a Symposium on the Nature of Religious Imagery, edited by
M.W. Meister, 103௅18. Bombay: Vakils, Feffer & Simons.
Mitra, Rajendralala (1878), Buddha GayƗ, the Hermitage of ĝƗkya Muni. Calcutta: Ben-
gal Secretarial Press.

28
Narayanan, M.G.S. (2002), “The State in the Era of the CƝramƗn PerumƗশs of Kerala”, in
State and Society in Pre-Modern South India, edited by R. Champakalakshmi, K. Ve-
luthat and T.R. Venugopalan, 111௅19. Thrissur: Cosmo Books.
Nattier, Jan (2008), A Guide to the Earliest Chinese Buddhist Translations: Texts from
the Eastern Han and Three Kingdoms Periods. Tokyo: Soka University, International
Research Institute for Advanced Buddhology.
௅௅௅௅௅ (2003), A Few Good Men: The Bodhisattva Path According to The Inquiry of
Ugra (Ugraparip৚cchƗ). Study and Translation by ࣓࣓࣓࣓࣓. Ann Arbor: University of
Michigan, Institute for the Study of Buddhist Traditions / Honolulu: University of
Hawa’i Press.
Niyogi, Roma (1959), The History of the GƗha‫ڲ‬avƗla Dynasty. Calcutta: Oriental Book
Agency.
dPag bsam ljon bzang = Pag sam jon zang. History of the Rise, Progress and Downfall
of Buddhism in India, by Sumpa Khan-po Yece Pal jor [Sum–pa mKhan-po ye-shes
dpal‘byor ye shes dpal’byor], edited […] by Sarat Chandra Das, 2 vols. Calcutta 1908:
The Presidency Jail Press.
Panigrahi, Krishna Chandra (1961), Archaeological Remains at Bhubaneswar. Calcutta:
Orient Longmans.
Periya PurƗ۬am = St. Sekkizhaar’s Periya Puranam. Translated by T.N. Ramachandran,
2 vols. Thanjavur 1990௅95: Tamil University.
Ramachandran, T.N. (1954), The NƗgapa‫ܒܒ‬i۬am and Other Buddhist Bronzes in the Ma-
dras Museum = Bullettin of the Madras Government Museum, n.s. 7/1. Madras.
RƗ‫܈‬trapƗlaparip‫܀‬cchƗ Snjtra = Boucher, Daniel (2006), Bodhisattvas of the Forest and
the Formation of the Mahayana: A Study and Translation of the RƗৢ৬rapƗla-
parip৚cchƗ-snjtra. Honolulu: University of Hawa’i Press.
Rea, Alexander (1909), Pallava Architecture. Archaeological Survey of India, New Im-
perial Series, 34. Madras: The Superintendent, Government Press.
Regmi, Dilli Rahman (1983), Inscriptions of Ancient Nepal, 3 vols. New Delhi: Abhinav
Publications.
Reynolds, Susan (2004), Feudi e vassalli. Una nuova interpretazione delle fonti
medievali. Roma: Jouvence. (Revised Italian ed. of Fiefs and Vassals, Oxford 1994.)
Robinson, James B. (1979), Buddha’s Lions. The Lives of the Eighty-four Siddhas. Catu-
raĞƯti-siddha-prav৚tti by Abhayadatta. Translated into Tibetan as Grub thob brgyad cu

29
 rtsa bzhi’i lo rgyus by sMon-grub Shes-rab. Translated into English by ௅௅௅௅௅. Berke-
ley: Dharma Publishing.
Roerich, George, trans. (1959), Biography of DharmasvƗmin (Chag lo tsa-ba Chos-rje-
dpal), a Tibetan Monk Pilgrim. Original Tibetan Text Deciphered and Translated by
௅௅௅௅௅. With a Historical and Critical Introduction by A.S. Altekar. Patna: K.P.
Jayaswal Research Institute.
RudrasaۨhitƗ = The ĝiva PurƗ۬a, edited by by J. L. Shastri, vol. 1. Delhi 1970: Motilal
Banarsidass.
Saddharmapu۬‫ڲ‬arƯka Snjtra = The Saddharma-PundarƯka or the Lotus of the True Law.
Translated by H[endrik] Kern. Sacred Books of the East, 21. Oxford 1884: The Clar-
endon Press.
Sahni Daya Ram (1911௅12), “Saheth-Maheth Plate of Govindachandra; [Vikrama]
Samvat 1186”. Epigraphia Indica, 11, 20௅26. Calcutta.
௅௅௅௅௅ (1914), Catalogue of the Museum of Archaeology at SƗrnƗth, with an Introduction
by J. Ph. Vogel. Calcutta: Superindendent Government Printing.
௅௅௅௅௅ (19233), Guide to the Buddhist Ruins of Sarnath, with a Plan of Excavations and Five
Photographic Plates. Simla: Superintendent, Government Central Press.
Sahu, N.K. (1958), Buddhism in Orissa. [Bhubaneshwar]: Utkal University / Cuttack:
Orissa Mission Press.
Sanderson, Alexis (2004), “Religion and the State: ĝaiva Officiants in the Territory of
the King’s Brahmanical Chaplain”. Indo-Iranian Journal, 47, 229௅300.
Dordrecht௅Boston: Reidel Publication Co.
Sanyal, Niradbandhu (1929), “A Buddhist Inscription from Bodh-Gaya of the Reign of
Jayaccandradeva”, V.S.124x. Indian Historical Quarterly, 5, 14௅30. Calcutta.
Sanyal, Rajat (2008௅9), “Geo-Polity in Early Medieval Bengal under the Sena Rule: Re-
Reading Epigraphic Sources”. Journal of Ancient Indian History, 25, 94௅113. Univer-
sity of Calcutta.
Saraswati, S[arasi] K[umar] (1977), TantrayƗna Art: An Album. Calcutta: The Asiatic
Society.
Scherrer-Schaub, Cristina (2007), “Immortality Extolled with Reason: Philosophy and Poli-
tics in NƗgƗrjuna”, in PramƗ৆akirtƯত. Papers Dedicated to Ernst Steinkellner on the
Occasion of His 70th Birthday, edited by B. Kellner et al., 2 vols, 757௅93. Wiener
Studien zur Tibetologie und Buddhismuskunde, 70. Universität Wien.

30
Sharma, Ram Sharan (1982), “The Kali Age: A Period of Social Crisis”, in India: His-
tory and Thought. Essays in Honour of A.L. Bhasham, edited by S.N. Mukherjee,
186௅203. Calcutta: Subarnarekha.
௅௅௅௅௅ (1987), Urban Decay in India (c. 300–c. 1000). New Delhi: Munshiram Mano-
harlal.
ShƗstri, H[ara] P[rasƗd] (1911), “Introduction to Modern Buddhism in Bengal”, in Na-
gendra NƗth Vasu, The Modern Buddhism and its Followers in Orissa, 1௅28. Calcutta:
Hare Press.
Sircar, Dinesh Candra (1964௅65), “Inscriptions of Two Brahmana Rulers of Gaya”. Epi-
graphia Indica, 36, 81௅94. Delhi.
௅௅௅௅௅ (19652), Select Inscriptions Bearing on Indian History and Civilization, vol. 1:
From the Sixth Century B.C. to the Sixth Century A.D. University of Calcutta.
Sivaramamurti, Calambur (1980), India (Storia della scultura nel mondo, vol. 5). [Mila-
no]: Arnoldo Mondadori Editore.
Spink, Walter (2005௅07), Ajanta: History and Development, 5 vols. Leiden௅Boston:
Brill.
ܑabaqƗt-i NƗ‫܈‬irƯ = H[enry] G[eorge] Raverty (1881), ܑabaۘƗt-i-NƗ‫܈‬irƯ: A General His-
tory of the Muhammadan Dynasties of Asia Including Hindustan from AH 194 (810
AD) to AH 658 (1260 AD) and the Irruption of the Infidel Mughals into Islam by
MaulƗnƗ, MinhƗj-ud-DƯn, Abnj-‘Umar-i-‘U‫܈‬mƗn. Translated from Original Persian
Manuscripts by ௅௅௅௅௅, 2 vols. Calcutta: The Asiatic Society.
Takakusu, J[unjirǀ] (1896), A Record of the Buddhist Religion as Practised in India and
the Malay Archipelago (AD 671࣓695) by I-Tsing: Translated by ௅௅௅௅௅, with a Letter
of F. Max Müller. Oxford: The Clarendon Press.
TƗranƗtha = TƗranƗtha’s History of Buddhism in India. Translated from Tibetan by Lama
Chimpa Alaka Chattopadhyaya, edited by Debiprasad Chattopadhyaya. Simla 1970:
Indian Institute of Advanced Study.
Taylor, William (1838), “Examination and Analysis of the Mackenzie Manuscripts De-
posited in the Madras College Library”. Journal of the Asiatic Society of Bengal, 7,
105௅31, 173௅92. Calcutta: The Asiatic Society.
Toynbee, Arnold J. (19675), Il mondo ellenico. Torino: Einaudi. (Original ed.: Hellenism.
The History of a Civilization. London 1959).
Tucci, Giuseppe (1949), Tibetan Painted Scrolls, 2 vols. and portfolio. Roma: Libreria
dello Stato.

31
VƗmana PurƗ۬a = The VƗmana PurƗ۬a with English Translation. Edited by Anand Swa-
rup Gupta; Translated by Satyamsu Mohan Mukhopadhyaya et al. Varanasi 1968: All
India Kashiraj Trust.
van Troy, J. (1990), “The Radical Social Protest of the PƗĞupatas”. Indica, 27, 1௅10.
Bombay.
VƗyu PurƗ۬a = The VƗyu PurƗ۬a. Translated and Annotated by G[anesh] V[asudeo] Ta-
gare, 2 vols. Delhi 1975: Motilal Banarsidass.
Veluthat, Kesavan (2006), “History in the Construction of a Regional History”, in Medie-
val India, Problems and Possibilities: Essays in Honour of A.R. Kulkarni, edited by R.
Seshan, 79௅100. Jaipur: Rawat Publications.
Verardi, Giovanni (1992), Excavations at Harigaon, Kathmandu: Final Report. (IsMEO
Reports and Memoirs 25), 2 vols. Rome: Istituto Italiano per il Medio ed Estremo
Oriente.
௅௅௅௅௅ (1999/2000), “The Buddha௅Elephant”. Silk Road and Archaeology, 6 (= Papers
in Honour of Francine Tissot), edited by E. Errington and O. Bopearachchi, 69௅74.
Kamakura.
௅௅௅௅௅ (2007), Excavations at Gotihawa and Pipri, Kapilbastu District, Nepal. Roma:
Istituto Italiano per l’Africa e l’Oriente.
௅௅௅௅௅ (2011a), “Buddhism in North-western India and Eastern Afghanistan, Sixth to
Ninth Century AD”. Zinbun, 43, 147௅83. Kyoto: Kyoto University, Jinbun Kagaku
Kenkyusho.
௅௅௅௅௅ (2011b), Hardships and Downfall of Buddhism in India. Appendices by Federica
Barba. New Delhi: Manohar.
Verma, B.S. (2001), “Excavations at Antichak: The Ruins of the Ancient VikramĞƯla Univer-
sity”, in ĝrƯ Subrahma۬ya Sm‫܀‬tƯ. Essays on Indian Pre-history, Proto-history, Archae-
ology, Art, Architecture, Epigraphy, Numismatics, Crafts, Iconography and Conservation
(Dr. Raviprolu Subrahmanyam Commemoration Volumes), edited by I.K. Sarma and B.
Vidyadhara Rao, 2 vols., 299௅308. New Delhi: Sundeep Prakashan.
௅௅௅௅௅௅ (2011), Antichak Excavations ࣓ 2 (1971࣓1981). Memoirs of the Archaeological Sur-
vey of India, 102. New Delhi: Archaeological Survey of India.
Vidyabhusana, Satis Candra (1921), A History of Indian Logic: Ancient, Medieval and
Modern Schools. University of Calcutta.
Wiesner, Ulrich (1980), “Nepalese Votive Stnjpas of the Licchavi Period: The Empty
Niche”, in The Stnjpa: Its Religious, Historical and Architectural Significance, edited

32
 by A.L. Dallapiccola in collaboration with S. Zingel-Avé Lallemant, 166௅74. Beiträge
zur Südasienforschung, Südasien Institut, Universität Heidelberg, 55. Wiesbaden:
Franz Steiner Verlag.
Willis, Michael (2009), The Archaeology of Ritual: Temples and the Establishments of
the Gods. Cambridge: Cambridge University Press.
Wright, Daniel (1877), History of NepƗl. Translated from the ParbatiyƗ by MunshƯ Shew
Shunker Singh and Pandit ShrƯ GunƗnand, with an Introductory Sketch of the Country
and People of NepƗl by the Editor. London: Cambridge University Press.
Zvelebil, K[amil] V[eith] (1975), Tamil Literature. Handbuch der Orientalistik, 2 Abt., 2
Band, 1 Abschnitt. Leiden௅Köln: Brill.

33
Figures

Fig. 1. Namakkal, Tamil Nadu. De-

tail of Trivikrama panel with
VƗmana as the Buddha, showing
elongated ears and shaved head. 8th
century. From Sivaramamurti (1980:
131).

Fig. 2. Pipri, Kapilbastu District (Nepal).

Heads of Sivaite ascetics found in the
excavations in association with other cul-
tic material. 2nd௅3rd century AD. From
Verardi (2011b: 146).

34
 Fig. 3. Kanchipuram. Gal-
lery in the Vaiku৆৬ha
PerumƗশ temple com-
pound. King Nandivarman
II Pallavamalla adminis-
tering justice and impale-
ment of Buddhists. Mid-
8th century. From Verardi
(2011b: 226, fig. 4).

Fig. 4. Harigaon, Kathmandu. Rubbing

of inscription on Garu‫ڲ‬astambha. Sec-
ond half of the 8th century. From
Regmi (1983, III: XXVII A, B).

35
 Figs. 5a, b. Harigaon,
Kathmandu. Lowermost
chambers of stnjpa razed to
ground after AD 749, with
the original filling of sand
(fig. 5a) and after emptying
the ku۬‫ڲ‬as (fig. 5b). Photo
IsIAO, Rome.

Fig. 6. Harigaon, Kathmandu. Section showing floor of Viৢ৆u temple covering the
foundations of Buddhist monuments. Simplified rendering of the stratigraphy of the
excavation (G. Verardi and G. Veneroso).

36
 Fig. 7. Harigaon, Kathmandu.
Trial-trench at the base of
Garu‫ڲ‬astambha, showing
brick foundation. Photo
IsIAO, Rome.

Fig. 8. Chabahil, Kathmandu. Fig. 9. Detail of preceding, showing

Stnjpa with finial removed and niche with chiselled-off image. From
empty niches. 8th century. From Wiesner (1980: fig. XII/4).
Wiesner (1980: fig. XII/3).

37
 Fig. 10. Bhubaneshwar, VaitƗl
Deul. Detail of frieze with KƗ-
pƗlikas. Second half of the 8th
century. From Donaldson (2002,
III: fig. 627).

Fig. 11. Bhubaneshwar, VaitƗl Deul. Fig. 12. Bhubaneshwar, VaitƗl

Bhairava and beheaded asura in the Deul. Reworked Buddhist architec-
shape of the Buddha. Second half of tural piece used as base of ynjpa
the 8th century. From Verardi outside temple door. From
(2011b: 236, fig. 6). Donaldson (2002, III: fig. 1).

38
Fig. 13. Kalamisri, Orissa. CƗmu৆ঌƗ seated on tribal chief killed by
a KƗpƗlika. 9th century. From Donaldson (2002, III: fig. 261).

39
 Fig. 14. Kanchipuram,
Temple of KailaĞanƗ-
tha. ĝiva as teacher un-
der a tree with two deer.
Chopped elephant head
below. Mid-8th century.

Fig. 15. Ellora, Temple

of KailƗĞanƗtha. Detail
of yajñaĞƗla with
Bhairava, Ga৆eĞa, a
Goddess and two dead
asuras. First half of the
8th century. From Ver-
ardi (2011b: 270, fig. 9).

40
 Fig. 16. Hirapur, Bhubaneswar.
Temple of the Sixty-four yoginƯs.
YoginƯ holding skull-cup and knife
and standing on chopped-off head of
tribal. Mid-10th century. From De-
hejia (1986: 102).

Fig. 17. Bihar. AparƗjitƗ subjugates

Ga৆eĞa. 10th century. From Saras-
wati (1977: fig. 184).

41
 Fig. 18. NƗlandƗ. Trailokyavijaya.
10th௅11th century.

Fig. 19. Ranipur-Jharial, Orissa. Temple of the yoginƯs. Mid-10th

century. From Dehejia (1986: fig. on p. 41).

42
 Fig. 20. Bihar. CƗmu৆ঌƗ seated on
dead asura in the shape of the Bud-
dha. 10th century. Courtesy National
Museum, New Delhi.

Fig. 21. Detail of pre-

ceding, with dead a-
sura/Buddha and vic-
tims.

43
 Issues in the History of Indian Buddhism
Giovanni Verardi


৅ ষ   ফা঩

৅ষ঻   ඁદপ৾॔४॔บઇધ৲ଢ଼஢७থॱ‫ش‬

ౣ ૐ   ඁદপ৾॔४॔บઇધ৲ଢ଼஢७থॱ‫ش‬

ક ਚ   ‫  ٿ‬਎੃৘ৣ਎ય೻૖ৢপ૝ূোপੵ঴୞৉भ

ඁદপ৾প૝৾ༀஜ຋ைమ

ਗ਼ ਵ    )$; 

(PDLO   EDUF‫ٯ‬DGU\XNRNXDFMS‫ ك਀৻ق‬85/ KWWSEDUFU\XNRNXDFMS

ഀ ླྀ   ઙૄভ঺ িরউজথॺ


ஜ୿ƸŴ૨ᢿᅹ‫ܖ‬Ⴞᅶᇌ‫ܖٻ‬৆ဦႎᄂᆮؕႴ࢟঺
 ૅੲʙಅžǢǸǢᜂ‫؏ע‬ƴƓƚǔʿ૙Ʒ‫ٶ‬ಮࣱƱ
ƦƷྵˊႎӧᏡࣱƷዮӳႎᄂᆮſᲢ᳸࠰
ࡇᲣƴǑǔᄂᆮ঺ௐƷɟᢿưƋǔŵ