See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/372521007

9 Tibetan Interpretations of the Opening Verses of Vajraghaṇṭa on the Body

Maṇḍala

Chapter · January 2017

DOI: 10.1163/9789004340503_011

CITATIONS READS

0 3

1 author:

Yael Bentor
Hebrew University of Jerusalem
27 PUBLICATIONS 162 CITATIONS

SEE PROFILE

All content following this page was uploaded by Yael Bentor on 05 September 2024.

The user has requested enhancement of the downloaded file.

 iii

Chinese and Tibetan Esoteric

Buddhism

Edited by

Yael Bentor
Meir Shahar

LEIDEN | BOSTON

For use by the Author only | © 2017 Koninklijke Brill NV

 Contents
Contents v

Contents

Acknowledgements ix
List of Figures x

Introduction 1
Yael Bentor and Meir Shahar

Part 1
Chinese Perspectives on the Origins of Esoteric Buddhism

1 Tantric Subjects: Liturgy and Vision in Chinese Esoteric Ritual

Manuals 17
Charles D. Orzech

2 Spells and Magical Practices as Reflected in the Early Chinese

Buddhist Sources (c. 300–600 CE) and Their Implications for the Rise
and Development of Esoteric Buddhism 41
Henrik H. Sørensen

3 The Terms “Esoteric Teaching” (“Esoteric Buddhism”) and “Tantra” in

Chinese Buddhist Sources 72
Lü Jianfu

Part 2
Chan, Chinese Religion, and Esoteric Buddhism

4 Buddhist Veda and the Rise of Chan 85

Robert H. Sharf

5 A Comparative Approach to Śubhakarasiṃha’s (637–735) “Essentials

of Meditation”: Meditation and Precepts in Eighth-Century
China 121
Lin Pei-ying

6 The Tantric Origins of the Horse King: Hayagrīva and the Chinese
Horse Cult 147
Meir Shahar

For use by the Author only | © 2017 Koninklijke Brill NV

vi Contents

Part 3
Scriptures and Practices in Their Tibetan Context

7 Crazy Wisdom in Moderation: Padampa Sangyé’s Use of

Counterintuitive Methods in Dealing with Negative Mental
States 193
Dan Martin

8 Perception, Body and Selfhood: The Transformation of Embodiment

in the Thod rgal Practice of the “Heart Essence” Tradition 215
Eran Laish

9 Tibetan Interpretations of the Opening Verses of Vajraghaṇṭa on the

Body Maṇḍala 230
Yael Bentor

Part 4
Tibetan Buddhism in China

10 Ming Chinese Translations of Tibetan Tantric Buddhist Texts and the

Buddhist Saṃgha of the Western Regions in Beijing 263
Shen Weirong

11 Sino-Tibetan Buddhism: Continuities and Discontinuities: The Case

of Nenghai’s Legacy in the Contemporary Era 300
Ester Bianchi

Part 5
Esoteric Buddhism in Dunhuang

12 On the Significance of the Ārya-tattvasaṃgraha-sādhanopāyikā and

Its Commentary 321
Jacob P. Dalton

13 Avalokiteśvara and the Dunhuang Dhāraṇī Spells of Salvation in

Childbirth 338
Li Ling and Ma De

For use by the Author only | © 2017 Koninklijke Brill NV

Contents vii

Part 6
Esoteric Buddhism in the Tangut Xixia and Yugur Spheres

14 Notes on the Translation and Transmission of the Saṃpuṭa and

Cakrasaṃvara Tantras in the Xixia Period (1038–1227) 355
Hou Haoran

15 Mongol Rulers, Yugur Subjects, and Tibetan Buddhism 377

Yang Fuxue and Zhang Haijuan

Part 7
Esoteric Buddhism in the Dali Kingdom (Yunnan)

16 The Chinese Origins of Dali Esoteric Buddhism 389

Hou Chong

17 Between China and Tibet: Mahākāla Worship and Esoteric Buddhism

in the Dali Kingdom 402
Megan Bryson

Index 429

For use by the Author only | © 2017 Koninklijke Brill NV

230 Bentor

Chapter 9

Tibetan Interpretations of the Opening Verses of

Vajraghaṇṭa on the Body Maṇḍala*
Yael Bentor

Introduction

This chapter explores the body maṇḍala as it is found in a range of polemical

works. I will use the polemical framework itself to better comprehend the crys-
tallization of the tantric traditions in Tibet that took place between the twelfth
and the fifteenth centuries. It was in this period that Buddhist thought and
practice became systematized in Tibet. Exegetical writing activity was then at
its peak, and Tibetan scholarship developed its own styles of Buddhism within
the framework of different schools. My aim is to identify the factors that stimu-
lated the creation of systems of thought in Tantric Buddhism—an area that
has received much less scholarly attention than Buddhist philosophy, although
these two fields are difficult to neatly divide.

Nyāsa and Body Maṇḍala

Sādhanas of the higher Tantras incorporate several methods for achieving the
goal of becoming a Buddha in this life. In this paper I examine one of these
methods, the practice of the body maṇḍala. Complex techniques for medita-
tion on the body maṇḍala were advanced by Tibetan masters. The simplest of
these is referred to as “emplacement” (nyāsa), and is common to both Buddhist
and Hindu tantric practices. By touching various points on the body while
reciting the corresponding mantras of the deities, the yogi renders the body
divine.1 Nonetheless, disagreement concerning the “degree” of divination
achieved by the practices of nyāsa and the body maṇḍala are found in both
Buddhist and Hindu treatises.

* This research was supported by The Israel Science Foundation, grant no. 401/13.
1 As shown by Gavin Flood, scriptures of the Pāñcarātra Āgama such as the Jayākhya-saṃhita,
maintain that through the meditation on nyāsa, the yogis become equal to the gods, endowed
with supernatural powers, fearless and victorious over death (Flood 2006, 113 and his transla-
tion, 188–193).

© koninklijke brill nv, leiden, 2016 | doi 10.1163/9789004340503_011

For use by the Author only | © 2017 Koninklijke Brill NV
Tibetan Interpretations of the Opening Verses of Vajraghaṇṭa 231

All major Buddhist Tantras of the higher class offer instruction on the prac-
tice of setting deities on the body. These include the Guhyasamāja Tantra,2 its
Explanatory Tantra the Vajramālā,3 the Hevajra Tantra,4 the Abhidhānottara
Tantra,5 and the Saṃpuṭa Tantra.6 Yet with regard to meditation on the body
maṇḍala the cycle of Cakrasaṃvara is generally taken as authoritative by
Tibetan masters. In his Sādhana of Cakrasaṃvara, the Bhagavad-abhisamaya,7
Lūyīpā explains that setting the deities on the psycho-physical constituents of
the body serves to purify them.8 While in this case the deities are Buddhist and
their bodily locations have Buddhist soteriological meaning, Lūyīpā’s Sādhana

2 Matsunaga 1978, chapter 8, verse 9, Tōh. 442, vol. 81 [ca], D. folio 102a, p. 203.4. This line indi-
cates only that one must set the Buddhas and bodhisattvas of the five Tathāgata families along
parts of the body.
3 Tōh. 445, chapter 64, verses 4–6, D. vol. 81 [ca], folio 270a, p. 539.2–3. This Tantra, associated
with the Ārya school (Matsunaga 1977, 118–19), uses the more Buddhist terms of the five ag-
gregates for the locations of the five Tathāgatas, and moreover specifically calls this practice
‘body maṇḍala’, ibid., verse 17c, folio 270b, p. 540.3. While chapter 64 describes the bodily loca-
tions of all thirty-two deities of the Ārya system, chapter 68 delineates which part of the body
transforms into which part of the celestial mansion of the maṇḍala. As Matsunaga ibid., be-
lieves chapter 68 is a later addition to the Tantra, perhaps the correlations of the body with
the maṇḍala palace is a later development.
4 Snellgrove 1959, I.ix. Here female deities are set on the aggregates.
5 Abhidhānottara-tantra, Mngon par brjod pa’i rgyud bla ma, Tōh. 369; chapter 9, edited and
translated into English by Kalff 1979. A similar process of Buddhicizing the body maṇḍala can
be seen also in this Yoginī Tantra regarded as a commentarial Tantra of the Cakrasaṃvara
cycle. Here the instruction to set the sixty-two deities of the maṇḍala on the body while recit-
ing the seed syllables of the twenty-four sacred places (pīṭhas), is immediately followed by
another listing, with a more Buddhist ring, of thirty-two deities, such as the five Buddhas set
on the aggregates, the five vajra ladies of the afflictive emotions set on the senses, and three
Tathāgatas set on the body, speech and mind.
6 Tōh. 381, D. vol. 79 [ga], the 3rd rab byed in the 6th brtag pa.
7 Bhagavad-abhisamaya, Dpal bcom ldan ’das mngon par rtogs pa, Tōh. 1427. For a Sanskrit edi-
tion of the Sanskrit, see Sakurai 1998; and for a study, see Gray 2011. Lūyīpā follows the second
and more Buddhist sounding listing in the Abhidhānottara Tantra. However while in the
Tantra there are apparently two different lists of deities that should be set on the body after
the yogi has generated himself as Heruka with his consort Vajravārāhī, Lūyīpā shifts the second
and more Buddhist sounding list of deities and psycho-physical constituents of the body to
the opening of the Sādhana—this practice serves to transform the yogi into a deity at the very
beginning of the practice. This is one of the seeming paradoxes of numerous Sādhanas, where
only as a deity can the yogi engage in the practice of the deity. See Bentor 1996, 2. This is not
different in Hindu practices as well, as Gavin Flood (2012, 1) puts it: “Indeed, the well known
characteristic of the tantric traditions is that to worship a god one must become a god.”
8 Sakurai 1998, 3. Tōh. 1427, D. folio 186b, p. 372.7.

For use by the Author only | © 2017 Koninklijke Brill NV

232 Bentor

practice and the Pāñcarātra Āgama seem to have similar a goal, namely, the
divinization of the body.
Vajraghaṇṭa,9 a focal figure for the purpose of the present paper, was another
important Indian authority on the cycle of Cakrasaṃvara Tantra. His dozen
works included in the Tengyur received much attention from Tibetan authors
of the Kagyu, Sakya and Gelug schools. Several of these compositions contain
instructions on the body maṇḍala of Cakrasaṃvara.10 Like Lūyīpā, Vajraghaṇṭa
opens his Sādhana manual [Tōh. 1432] with an instruction on the deities that
are invited to enter the body. These deities then create what he calls a “body
maṇḍala.” Below we look more closely at Vajraghaṇṭa’s treatise on the initiation
into the maṇḍala of Cakrasaṃvara, the Cakrasaṃvara-ṣeka-prakriyopadeśa.11

The Opening Verses of Vajraghaṇṭa’s Treatise on Initiation (Seka)

Vajraghaṇṭa begins his exposition of the initiation ritual [Tōh. 1431] with an
elucidation of the maṇḍala the disciples will enter. The first of these verses has
generated much discussion among Tibetan scholars, as we shall soon see.
A further difficulty arises from its different readings in Sanskrit and Tibetan
sources. As the Indian Abhayākaragupta cites it, this verse could be describing
mental maṇḍalas.12 However the Tibetan commentators13 I have consulted
concur14 that this stanza speaks of three types of maṇḍalas: the first line refers

9 Vajraghaṇṭa, Rdo rje dril bu or Dril bu pa, composed two Sādhanas of Cakrasaṃvara, as
well as a treatise on its initiation. These are: 1. the Cakrasaṃvara-sādhana, Dpal ’khor lo
sdom pa’i sgrub pa’i thabs, Tōh. 1432, called at times the longer work, and 2. the
*Cakrasaṃvara-kāya-maṇḍala-abhisamaya, ’Khor lo sdom pa’i lus dkyil gyi mngon rtogs,
Tōh. 1434, called the shorter work. 3. Cakrasaṃvara-ṣeka-prakriyopadeśa, Dpal ’khor lo
sdom pa’i dbang gi bya ba mdor bsdus pa, Tōh. 1431, below Seka.
10 In describing the body maṇḍala in more details in his longer Sādhana, Vajraghaṇṭa clearly
follows the Saṃpuṭa Tantra, the 3rd rab byed in the 6th brtag pa.
11 Cakrasaṃvara-ṣeka-prakriyopadeśa, Dpal ’khor lo sdom pa’i dbang gi bya ba mdor bsdus
pa, Tōh. 1431, below Seka.
12 Abhayākaragupta cites this verse in at least two of his works, the Abhayapaddhati, Tōh.
1654, folio 199b, p. 398.2, also in Chog Dorje 2009, 47 & 181–82, and the Āmnāya-mañjarī,
Tōh. 1198, 23rd Snye ma, folio 214a, p. 427.2. I would like to thank Professor Harunaga Isaa-
cson for pointing out the citation in the Abhayākaragupta and for sending me its Sanskrit
version.
13 Bsod nams rtse mo, De’i dbang gi bya ba mdor bsdus, folio 119a, p. 408.2.1–4; Bu ston, Lus
dkyil dbang chog, folio 5a, p. 393.4–6; Ngor chen, Dril bu pa’i lus dkyil gyi bshad pa, folio
371a, p. 400.3.4–6; Tsong kha pa, Sngags rim chen mo, p. 304, Mkhas grub rje, Bskyed rim
dngos grub rgya mtsho, folios 124b–125a, pp. 250.6–251.4.
14 Regardless of their position in the disputes about the meaning of this verse.

For use by the Author only | © 2017 Koninklijke Brill NV

Tibetan Interpretations of the Opening Verses of Vajraghaṇṭa 233

to painted maṇḍalas, the second to colored powder maṇḍalas and the final
two lines to a maṇḍala naturally present in the body.
As my interest is not in some ‘original’ meaning of the author, but in the
ways his text was understood in Tibet, I will translate the first verse in accor-
dance with this shared Tibetan interpretation:15

[The maṇḍala] is explained as prepared in stages,

depicted in paintings or drawn by way of strings and colored powder,
[while] sentient beings are naturally
present non-dual maṇḍalas.

Tibetan scholars disagree about how to read the last two lines.16

I Are Sentient Beings naturally Present Non-dual maṇḍalas?

A Some Early Tibetan Interpretations

The eminent early Sakya scholar, Bsod nams rtse mo (1142–1182),17 opens his
commentary on Vajraghaṇṭa’s Seka by stating that although maṇḍalas are pres-
ent naturally in the body, people are not ordinarily aware of such a presence.
The goal of the initiation is to make the disciple aware of this presence or “to

15 Vajraghaṇṭa’s Seka, Tōh. 1431, D. folio 219b, p. 438.5–6: རི་མོ[ར]་གནས་པའི་ལས་དང་ནི། །ཐིག་ལས་ཚོན་

དགྱེའི་རིམ་པས་བསྟན། །འགྲོ་བ་འདི་ཉིད་རང་བཞིན་གྱིས། །གྲུབ་པའི་དཀྱིལ་འཁོར་གཉིས་མེད་པ། G.N. and P. have
མོར་ for མོ་ in D and C. For the Skt. see Finot 1934, 62: cittastha-karmma-sūtra-rajaḥ-pāta-
kramoditaṃ, gaṃjagaṃ maṇḍalam evedaṃ prakṛtyā siddham advayam. I would like to
thank Professor Harunaga Isaacson for drawing my attention to this incomplete Sanskrit
manuscript of the Seka-kriyā-krama, that contains the first three verses and part of the
fourth. The Abhayapaddhati, ibid., has janan for gaṃjagaṃ in Finot. Both versions open
with cittastha translated into Tibetan in both of Abhayākaragupta’s works mentioned
above as སེམས་གནས་ and therefore could be understood as ‘abide in the mind’. Similarly
also Karmapa Rang byung rdo rje, Dril bu pa’i dbang gi mtshams sbyor, folio 9a, p. 488.2,
has སེམས་ལ་གནས་པ་. However the readings in the Seka itself as found in the Tengyur are རི་
མོ་གནས་པ་ or རི་མོར་གནས་པ་ that translate citrastha and not cittastha.
16 While the Tibetan translation of the Seka is: འགྲོ་བ་འདི་ཉིད་རང་བཞིན་གྱིས། །གྲུབ་པའི་དཀྱིལ་འཁོར་
གཉིས་མེད་པའོ། in Abhayākaragupta’s works these lines are cited as: འགྲོ་འདི་དཀྱིལ་འཁོར་ཁོ་ན་སྟེ།
།རང་བཞིན་གྱིས་གྲུབ་གཉིས་མེད་པ།. Hence other possible translations of these two lines could be:
“this world is but a maṇḍala, naturally present and non-dual,” “these beings are but
maṇḍalas, naturally present and non-dual,” or “beings are non-dual with the naturally
present maṇḍala,” but our main concern is to understand interpretations of these lines by
Tibetans.
17 De’i dbang gi bya ba mdor bsdus, folio 118a1, p. 407.4.1–4.

For use by the Author only | © 2017 Koninklijke Brill NV

234 Bentor

cause to arise that which has not yet arisen in this way in the [meditator’s]
mind.” For Bsod nams rtse mo, when practitioners develop an awareness of the
maṇḍalas naturally present in the body, they will become suitable vessels for
the practice. Bsod nams rtse mo explains:18

The purpose [of the initiation] is to make present in the mind that which
is not present there—that the sixty-two or thirty-seven deities essentially
abide in all sentient beings that dwell in the three realms (Bsod nams rtse
mo, ibid., 118a2–4).

In other words, the maṇḍala of Cakrasaṃvara abides in all sentient beings.

Bsod nams rtse mo is not alone in holding this position. In his commentary on
the same work, Bu ston Rin chen grub (1290–1364) states:

Although the maṇḍala—with its celestial mansion and deities—is pres-

ent on its own in their body, the ritual of initiation into the naturally
present maṇḍala is for rendering those who do not know that it is present
into vessels suitable for the practice (Lus dkyil dbang chog, folio 1b,
p. 386.3–5).19

Vajraghaṇṭa’s Seka continues:20

18 Bsod nams rtse mo comments here on the opening verse of Vajraghaṇṭa’s Seka: “The
yoginī of the stages of the three wheels abides in any sentient being in the three realm.”
Bsod nams rtse mo’s text is somewhat different from the Derge; he has: འཁོར་ལོ་གསུམ་རིམ་རྣལ་
འབྱོར་མ། །གང་གི་སེམས་ཅན་ཁམས་གསུམ་གནས།, while D. folio 219b, p. 438.4 has: འཁོར་ལོ་རིམ་གསུམ་རྣལ་
འབྱོར་ཉིད། །གང་གི་སེམས་ཅན་ཁམས་གསུམ་གནས།. The Skt. in Finot, ibid. is: tri-cakra-krama-yogataḥ,
yena sattvās tri-dhātu-sthā. The phrase ‘the three wheels’ refers to the maṇḍala of
Cakrasaṃvara in which Heruka and Vajravārāhī are surrounded by the four “Essence
Yoginīs,” and then by the three wheels of mind, speech, and body.
19 In commenting on Vajraghaṇṭa’s verse cited above, Bu ston Rin chen grub explains: “These
aggregates of beings who are sentient are the naturally present maṇḍala in which the sup-
port and supported are nondual or saṃsāra and nirvāṇa are nondual, for the samsaric
aggregates are present in the maṇḍala of the Buddhas.” Ibid., folio 5a, p. 393.5–6.
20 Tōh. 1431, folio 219b, p. 438.6: བཅོས་མ་ཉིད་[གཉིས་]ཀྱི་ངོ་བོ་གང་། །དེ་ནི་གདུལ་བྱའི་དབང་ལས་འདོད། །མཁས་
པའི་བསྒྲུབ་བྱ་དེ་མིན་ཏེ། །ཡང་དག་དོན་མཐོང་གྲོལ་ཕྱིར་རོ། While D.N.G. and C. in the Tengyur have ཉིད་,
Bsod nams rtse mo, Bu ston, Tsong kha pa and Mkhas grub rje have the reading གཉིས་, fully
justified by the Skt. dvaya. Finot, ibid., has kṛtrimaṃ dvaya-rūpaṃ yat tad vineya-vaśān
matam, tad na sādhy …, while the Abhayapaddhati, ibid., has kṛtrimadvaya for kṛtrimaṃ
dvaya in Finot, and [sādhy]āṃ mumukṣūṇāṃ muktir bhūtārtha-darśanād for the text
missing in Finot. Thought the term མཁས་པ་ appears neither in the Sanskrit nor the Tibetan
of the Abhayapaddhati, it is found in the Tibetan translation of Vajraghaṇṭa’s Seka and in
its Tibetan commentaries.

For use by the Author only | © 2017 Koninklijke Brill NV

Tibetan Interpretations of the Opening Verses of Vajraghaṇṭa 235

The nature of the two constructed ones

is intended to accord with the disciple.
These [maṇḍalas] should not be produced for the skillful,
since they will be liberated by seeing the true meaning.

The term “constructed” (or “contrived”) in this verse is understood by all

Tibetan commentators I have consulted to refer to painted and colored pow-
der maṇḍalas, in contrast to the unconstructed or naturally present ones we
will encounter immediately. Bsod nams rtse mo21 explains that the skillful, or
the person endowed with the best faculties, will be liberated by seeing the true
meaning of the natural maṇḍala.22 Bsod nams rtse mo then asks: “How is the
natural maṇḍala present?”23 He replies by citing the next verse in the Seka:24

The maṇḍala of the three wheels is present.

The deities of the maṇḍala are present.
The stages of the initiation to these and so on
Bring to be present that which was never present.

Bsod nams rtse mo reads the first two lines to mean that both the celestial
mansion of the maṇḍala and the deities dwelling within it are present natu-
rally in the body.25 He emphasizes once more that the purpose of the initiation
is not to turn sentient beings into maṇḍalas, but to render them aware of its
continuous presence.26

21 De’i dbang gi bya ba mdor bsdus, folio 119a, p. 408.2.5–6.

22 In his commentary on this verse Bu ston, Lus dkyil dbang chog, folio 5b–6a, pp. 394.7–
395.5, explains what the natural maṇḍala is by citing the Saṃpuṭa Tantra on the body
maṇḍala, Tōh. 381, folios 113b–114a, pp. 226.3–227.1.
23 De’i dbang gi bya ba mdor bsdus, folio 119a, p. 408.2.6.
24 Tōh. 1431, folio 219b, p. 438.6–7: འཁོར་ལོ་གསུམ་གྱི་དཀྱིལ་འཁོར་གྲུབ། །དཀྱིལ་འཁོར་ལྷ་རྣམས་གྲུབ་པའོ། །དེ་ཡི་
དབང་བསྐུར་ལ་སོགས་རིམ། །ཅི་ནས་མ་གྲུབ་ཡང་གྲུབ་བྱེད།
25 He then raises a question: “If essentially the maṇḍala is present in the body in this way,
the initiation is unnecessary; and if it is not essentially present, by visualizing that it is
present, the fruit will not arise. Because this is like a beggar boasting to be a king.” To this
he replies: “This is not the case because [the initiation] does not make something—that
previously has not been essentially present—becoming present later on.” De’i dbang gi
bya ba mdor bsdus, folio 119b, p. 408.3.2–4.
26 Bu ston provides a very similar explanation in his commentary on this verse: “Regarding
the initiation into the naturally present maṇḍala and the stages of entering into the
maṇḍala: though the body is a naturally present maṇḍala, it is not clearly present as an
object in the minds of the disciples. It is not sufficient that essentially [the maṇḍala]

For use by the Author only | © 2017 Koninklijke Brill NV

236 Bentor

This topic is further clarified in the commentaries of Bsod nams rtse mo and
Bu ston on the subsequent verse of the Seka:27

It is intended that those who do not know

the nature of that are to become vessels.
The stages of the initiation and so on
actually achieve on the conventional level.

Bsod nams rtse mo explains here that, those who do not know in their mind that
the nature of the maṇḍala is present in them, will come to know this by means
of the initiation, and therefore will become suitable vessels for the practice.28
He explicates further:

As for example, although there is a treasure below the bed of a poor per-
son, when he does not know that it is there, he will not be liberated from
his poverty. Coming to know that it is there, he will be liberated from his
poverty (De’i dbang gi bya ba mdor bsdus, folio 119b, p. 408.3.4–5).29

And then he quotes the Hevajra Tantra:30

Sentient beings are Buddhas,

but they are obscured by superficial taints.
Once these are removed they are Buddhas.

This verse is often cited by proponents of the theory of Tathāgata-garbha in the

sense of a real Buddha essence, though there are numerous variations on this
theme.31

abides [in the body] without being recognized, it is necessary to make this clearly present
also as an object of the mind.” Lus dkyil dbang chog, folio 6b, p. 396.4–5.
27 D. folio 219b, p. 438.7: དེ་ཡི་རང་བཞིན་མི་ཤེས་རྣམས། །སྣོད་དུ་བྱ་བར་བཞེད་པ་སྟེ། །དབང་བསྐུར་སོགས་རིམ་ཀུན་
རྫོབ་ཏུ། །མངོན་སུམ་དུ་ནི་སྒྲུབ་པའོ།.
28 Bu ston follows him: “Regarding those who do not know that the nature of that body is an
essentially present maṇḍala, who enter into the maṇḍala by means of rituals, and recog-
nize that: it is intended that by becoming aware of this, they become vessels for meditat-
ing on the creation and completion stages.” Lus dkyil dbang chog, folio 6b, p. 396.5–6.
29 See also Bu ston, ibid., folio 6b, p. 396.5.
30 II.iv.69: sattvā buddhā eva kiṃ tu āgantuka-malāvṛtāḥ, tasyāpakarṣaṇāt sattvā buddhā
[eva na saṃśayā].
31 Bsod nams rtse mo then explains the two last lines of Vajraghaṇṭa, by saying that since the
maṇḍala is naturally present, ultimately there is no need to make it present. Still by mak-

For use by the Author only | © 2017 Koninklijke Brill NV

Tibetan Interpretations of the Opening Verses of Vajraghaṇṭa 237

The identification of the body with maṇḍala changes the goal of meditating
on the deities within the body, differentiating it from that of Lūyipa’s Sādhana
as well as the Pāñcarātra Āgama. In the present case the practitioner does not
divinize the body, for the body is already divinized, as in all sentient beings
there is Buddha essence. This novel view of the human body as a naturally
present maṇḍala has yet a further implication, one that concerns moving
beyond visualization. Meditating on the body maṇḍala according to this
method, the yogi works with something that is actual, and even constantly
actualized, in one’s very body.
This aspect of actuality is linked to the way tantric visualizations are thought
to achieve their goal. Those who hold that tantric visualization is a type of
mental fabrication still maintain that it serves a soteriological purpose. But
these scholars then must account for how the goal will materialize by visual-
izing something that is not present, and how the mind differs during the
meditation from that of a beggar boasting to be a king. A prevailing explana-
tion suggests that the disciple “takes the fruit on the path”, that is, that he or she
emulates the goal of her practice until it becomes actualized. These two meth-
ods of visualizing the real and unreal are very different approaches to the role
of visualization during tantric meditations.

B Some Late Tibetan Interpretations

During the 15th century, controversy over the body maṇḍala played a role in
the larger conflict between Mkhas grub rje Dge legs dpal bzang po (1385–1438)
and Ngor chen Kun dga’ bzang po (1382–1456).32 Yet the more famous fracas
related to the body maṇḍala of Hevajra took place only after the discord con-
cerning the interpretation of the first six verses in Vajraghaṇṭa’s Seka in the
context of Cakrasaṃvara. The commentary composed by Ngor chen Kun dga’
bzang po on Vajraghaṇṭa’s Seka, titled Dril bu pa’i lus dkyil gyi bshad pa, is a
polemical work which employs the traditional tripartite Sa skya argumentative
method. This technique consists of three steps: dgag, bzhag, spang, to wit:
refuting the position of an antagonist, presenting one’s own position, and dis-
pelling any remaining objections. The first point Ngor chen discusses is the
nature of the maṇḍala naturally present in sentient beings.

1 Refuting the Position of the Antagonist

Following convention in this genre, Ngor chen maintains the anonymity of his
opponents, referring to them as “other lamas”:

ing it present conventionally the initiation removes the superficial taints. Such an expla-
nation echoes the Tathāgata-garbha theory as well.
32 van der Kuijp 1985a, 1985b; Davidson 1991 and Jackson 2007.

For use by the Author only | © 2017 Koninklijke Brill NV

238 Bentor

Other lamas maintain that dharmadhātu, suchness, or “Sugata essence”

abides constantly and stably in all sentient beings, and that this is the
ultimate truth maṇḍala of Cakrasaṃvara. They also explain that syno-
nyms [of “Sugata essence” and so forth] are unconditioned, the profound
truth of cessation, and emptiness endowed with all the supreme as-
pects. We cannot accept this because if the ultimate truth maṇḍala of
Cakrasaṃvara abides in actuality in the continuum of all sentient beings,
this would contradict the lines of Vajraghaṇṭa: “It is intended that those
who do not know the nature of that are to become vessels.” This is because
if sentient beings are the actual maṇḍala of Cakrasaṃvara, this would
contradict the fact that they themselves do not know their own nature,
and they will not be all-knowing Buddhas (Dril bu pa’i lus dkyil gyi bshad
pa, folio 370a, p. 399.3.1–4).

The view of the “other lamas” recalls that of Bsod nams rtse mo and Bu ston,
but Ngor chen expresses it in distinctively tathāgata-garbha terminology. They
seem to take Vajraghaṇṭa’s line “sentient beings are naturally present non-dual
maṇḍalas,” to mean that ultimate truth maṇḍalas of Cakrasaṃvara abide
within all sentient beings as “Sugata essence.” Disagreeing with them, Ngor
chen points out that according to Vajraghaṇṭa, the role of the initiation is to
render the initiates who, as cited above, “do not know the nature of that” into
suitable “vessels” for the practice of Cakrasaṃvara. Had the ultimate truth
maṇḍalas of Cakrasaṃvara been present in the initiates, they would most cer-
tainly be suitable vessels for the practice. Thus the statement “sentient beings
are naturally present non-dual maṇḍalas,” cannot imply ‘Sugata essence’ or the
presence of Cakrasaṃvara maṇḍalas in all sentient beings. As such, it becomes
difficult to justify the claim that while maṇḍalas are present naturally in the
body, awareness of this presence requires initiation.
It is not surprising that Ngor chen rejected a view that echoed that of Dol po
pa (Shes rab rgyal mtshan 1292–1361). Red mda’ ba (1348–1412) before him and
Go rams pa following him did so as well,33 and even Bu ston did not agree with
Dol po pa (Ruegg 1968). Ngor chen was also not the only early-fifteenth-cen-
tury scholar to make this argument against such interpretation of Vajraghaṇṭa’s
stanza. Tsong kha pa Blo bzang grags pa (1357–1419) wrote on the body maṇḍala
of Cakrasaṃvara in three of his works: Sngags rim chen mo (304), a commen-
tary on the Seka of Vajraghaṇṭa, the Dril dbang,34 and a commentary on the

33 Cabezón and Lobsang Dargyay 2007, 71–77, 97–113, 281–306 and passim.
34 Dril dbang = Rnal ’byor dbang phyug dril bu lugs bde mchog lus dkyil gyi dbang chog rin po
che’i bang mdzod. In this commentary, Tsong kha pa lists as his authorities not only the

For use by the Author only | © 2017 Koninklijke Brill NV

Tibetan Interpretations of the Opening Verses of Vajraghaṇṭa 239

Sādhana of Cakrasaṃvara according to Lūyipā, the ’Dod ’jo.35 The earliest of

these works is likely the Sngags rim chen mo, written in 1405.36 There Tsong kha
pa states:

If beings would have existed from the very beginning as the deity
Cakrasaṃvara, it would be unreasonable to maintain that: “it is intended
that those who do not know the nature of that become vessels,” since this
contradicts their not knowing the body maṇḍala of Cakrasaṃvara
(Sngags rim chen mo, 303–304).37

Thus Tsong kha pa, like Ngor chen, rejects the reading of the line at hand that
maṇḍalas are naturally present in all sentient beings. Tsong kha pa and Ngor
chen share a reason for such rejection: the interpretation contradicts another
statement in the same work.
Without revealing the identity of his opponents, Tsong kha pa states:38

Some say that you visualize something which is already present there
since the very beginning.

Uncharacteristically for him, in his own commentary on Vajraghaṇṭa, Mkhas

grub rje identifies his authorial rival as Bsod nams rtse mo:

In his commentary on the Sādhana by Vajraghaṇṭa, the master Bsod

nams rtse mo teaches that the yogis visualize their bodies existing from
the very beginning as maṇḍalas (Bde dril bskyed rim, folios 8a–9b, pp.
780.6–781.1).

Cakrasaṃvara Tantra and Indian works, but also the practice of the Sa skya masters and
other Tibetan scholars, folio 27a, p. 109.4–5.
35 ’Dod ’jo = Bcom ldan ’das dpal ’khor lo bde mchog gi mngon par rtogs pa’i rgya cher bshad pa
’dod pa ’jo ba.
36 Tshe tan zhabs drung 1982, 211.
37 Similarly in his ’Dod ’jo, Tsong kha pa explains that Vajraghaṇṭa’s line: “Sentient beings are
naturally present non-dual maṇḍalas,” “does not mean that the channels, elements and so
forth of the body abide from the very beginning as the maṇḍala of Heruka. If this were so,
it would contradict another line in the text: “It is intended that those who do not know the
nature of that are to become vessels,” since this contradicts their not knowing that very
maṇḍala of Cakrasaṃvara.” The ’Dod ’jo, folio 122a-b, pp. 441.5–442.1.
38 ཁ་ཅིག་གདོད་མ་ནས་དེར་གྲུབ་པ་གསལ་འདེབས་པ་ཡིན་ཟེར། ’Dod ’jo, folio 121b, p. 440.1.

For use by the Author only | © 2017 Koninklijke Brill NV

240 Bentor

Mkhas grub rje’s disclosure does not surprise us, as we have seen that the “other
lamas” criticized by Ngor chen hold a view not only close to that of Dol po pa,
but also to that of Bsod nams rtse mo and Bu ston. Tsong kha pa, too, likely had
Bsod nams rtse mo in mind when he criticized ‘some’ who “visualize some-
thing which is already present there since the very beginning.” In any case,
both Ngor chen and Tsong kha pa critique a view held by one of the forefathers
of the Sa skya school, thus writing against the tradition of their predecessors. I
say their predecessors, since when Tsong kha pa composed his Sngags rim chen
mo in 1405, he most likely regarded himself as a Sa kya pa, just like Ngor chen,
as the Dga’ ldan monastery had not been built, and the identity of the Dga’ ldan
pa was yet to develop.

2 Presenting the Position of the Protagonist

In presenting his own view, Ngor chen explains the lines in question:

[The maṇḍala] is explained as prepared in stages,

depicted in paintings or drawn by way of strings and colored powder,
[while] sentient beings are naturally
present non-dual maṇḍalas.
The nature of the two constructed ones
is intended to accord with the disciple.
These [maṇḍalas] should not be produced for the skillful,
since they will be liberated by seeing the true meaning.

Ngor chen asserts:

The constructed maṇḍalas are the painted and colored powder maṇḍalas.
They are designated as constructed because the artists and ritual mas-
ters respectively need to newly produce them. On the other hand, the
[coarse] body of the yogi is present and along with it also the channels
and elements [of the subtle body] are present, therefore it is unneces-
sary to re-produce them (Dril bu pa’i lus dkyil gyi bshad pa, folio 371a,
p. 400.1.4–5).

Thus Ngor chen’s position is that the yogi’s body, with its channels and ele-
ments that are the bases for generating the maṇḍala, are not constructed, since
they have been present all along. Tsong kha pa as well maintains precisely this
in at least three of his works. In his Sngags rim chen mo, Tsong kha pa explains
our verse:

For use by the Author only | © 2017 Koninklijke Brill NV

Tibetan Interpretations of the Opening Verses of Vajraghaṇṭa 241

Whereas with regard to the colored powder and painted maṇḍalas, it is

necessary to achieve anew the basis of generation by [drawing] lines,
[applying] colors and so forth, with respect to the body maṇḍala, there is
no need to re-produce the channels and elements that are to be gener-
ated as the deity (Sngags rim chen mo, 304).

Similarly in the ’Dod ’jo, his commentary on the Sādhana of Cakrasaṃvara

according to Lūyipā, Tsong kha pa states:

The meaning of constructed maṇḍala is that the maker must produce

anew the bases for generating it—the emblems and so forth, because
they are not present from the time the body was formed. In the case of
the unconstructed maṇḍala, the bases of producing, the channels and
elements,39 are present since the very formation of the body, and there-
fore they are naturally present (’Dod ’jo, folio 137b, p. 472.2–3).

Thus Tsong kha pa also reads the line at hand to mean that the bases out of
which the body maṇḍala are generated are naturally present in the bodies of
sentient beings. In other words the body maṇḍala is produced from the chan-
nels and elements of the body that have been present in the body from the very
moment it was formed. Such bases are not found in the constructed maṇḍalas
drawn in paintings or in colored powder. For Tsong kha pa,40 if the bodies of
sentient beings had been present from the very beginning as celestial man-
sions, no effort would have been required to liberate them. Moreover such a
position would run up against both instructions on the meditation and scrip-
tural authorities.41 This point is found too in Tsong kha pa’s commentary on
Vajraghaṇṭa’s Seka42, in the work of Mkhas grub rje,43 and, as we have seen, in
the thought of Ngor chen.

39 The female deities of Cakrasaṃvara maṇḍala, the Yoginīs, are generated from the chan-
nels, (rtsa), the male deities, the Dpa’ bo-s, from the elements (khams).
40 ’Dod ’jo, folio 122a-b, pp. 440.6–441.1.
41 Tsong kha pa, ibid., states that meditating by visualizing that our bodies have been pres-
ent from the very beginning as celestial mansions runs counter to instructions to medi-
tate on the body becoming a celestial mansion. This will also contradict the scriptural
authorities of the Abhidhānottara-Tantra, *Kambala (Lwa ba pa), Prajñārakṣita, Tathā-
gatavajra, Vajraghaṇṭa and others.
42 Tsong kha pa adds that “the two” referred to in the term ‘non-duality’ in vajraghaṇṭa’s
stanza are the deities and the bases out of which they are generated, or else to the celes-
tial mansion of the maṇḍala and the deities residing in it. Dril dbang, folio 3a, p. 60.3–6.
43 Bskyed rim dngos grub rgya mtsho, folios 124b–125a, pp. 250.6–251.4.

For use by the Author only | © 2017 Koninklijke Brill NV

242 Bentor

Mkhas grub rje assesses the idea that the bodies of sentient beings have
existed from the very beginning as maṇḍalas in light of the Buddhist path and
its soteriological goals.44 For him such view invalidates the four truths of the
nobles that presume that beings are born into saṃsāric suffering due to their
karma and afflictive emotions. As well, it supposes no distinction between
non-awakened and awakened beings, thus rendering pointless the Buddhist
path. At the same time it contradicts all ordinary experience.45

3 Dispelling Objections
In the third step of the debate, attempting to dispel any remaining objections,
Ngor chen reiterates that the term naturally carries no ultimate sense, but
rather a conventional way of being, as these maṇḍalas do not exist in and of
themselves, but are subject to conditionality and change. There is no other
nature to the bodies of sentient beings beside being the basis for generating
the body maṇḍala.46

C To Conclude
Ngor chen and Tsong kha pa present their positions in strikingly similar ways.
This might lead one to wonder if the two thinkers had knowledge of each oth-
er’s work. The absence of a colophon in Ngor chen’s commentary makes the
chronology of work difficult to determine.47 Ngor chen’s repeated citation of
Tsong kha pa’s Sngags rim chen mo and ’Dod ’jo, however, informs us that he
was familiar with the latter’s views on the matter.48

44 Bskyed rim dngos grub rgya mtsho, folios 116b–118a, pp. 234.1–237.5.
45 “When you meditate on yourself as Vajradhara, while your own experience confirms that
your body is in no way adorned with the major and minor marks of the Buddha, and your
mind does not at all directly realize all phenomena, and at the same time you claim to be
a Buddha, your experience of reality is insane.” Ibid., folio 118a, p. 237.3–4.
46 Dril bu pa’i lus dkyil gyi bshad pa, folio 371a-b, p. 400.1.6–2.6. According to Ngor chen: “As
the bases of generation of the painted and powered colored maṇḍalas are constructed,
they are designated as ‘conventional’, while the basis of generation of the body maṇḍala
is unconstructed, and is thus postulated as ‘ultimate’.” Dril bu pa’i lus dkyil gyi bshad pa,
folio 371a, p. 400.1.6.
47 As Heimbel (2014, 555) notes: “More than thirty texts included in Ngor chen’s collected
works don’t refer to Ngor chen as author or have a colophon at all.”
48 For example, in his Dril bu pa’i lus dkyil gyi bshad pa, folios 372a–373a, pp. 400.3.6–401.1.5,
Ngor chen cites and paraphrases a few paragraphs of Tsong kha pa Sngags rim chen mo,
305, line 18 to 306, line 24. Furthermore, Ngor chen disagrees with the same positions that
Tsong kha pa cited and refuted. For example, just like Tsong kha pa who in his Sngags rim
chen mo, 305, lines 12–15, objects to the method of meditation on the body maṇḍala by
visualizing that the body is substituted by the celestial mansion, in his Dril bu pa’i lus dkyil

For use by the Author only | © 2017 Koninklijke Brill NV

Tibetan Interpretations of the Opening Verses of Vajraghaṇṭa 243

The two authors, Ngor chen and Tsong kha pa, concur with regard to the
issue of ‘Sugata essence’. Consequently, neither is able to embrace the straight-
forward explanation offered by both Bsod nams rtse mo and Bu ston of
Vajraghaṇṭa that maṇḍalas are naturally present in sentient beings. Thus the
authors resort to a non-literal interpretation of our line.
We see, then, that while these scholars engaged in vigorous dispute, they
selectively incorporated each other’s views in their own writings. On the spe-
cific point of maṇḍalas naturally present in the body of sentient beings, Ngor
chen held with Tsong kha pa and not with Bsod nams rtse mo, who belonged
to the Sa skya school and in fact was one of its five forefathers. Thus the early
15th century saw much polemic among Tibetan scholars, as well as not a small
degree of constructive dialogue.

II The Roles of the Three Types of maṇḍalas

By and large, then, Tibetan scholars maintain that Vajraghaṇṭa speaks about
three types of maṇḍalas; to wit, two constructed ones (i.e., the painted and
colored powder maṇḍalas) and an unconstructed one (which is naturally pres-
ent in the bodies of all sentient beings). We shall now learn, however, that the
function of these maṇḍalas does not enjoy a similar scholarly consensus.

A Some Earlier Tibetan Interpretations

Bsod nams rtse mo maintains that the painted and colored powder maṇḍalas
are meant for disciples endowed with intermediate and lesser faculties, while
the naturally present maṇḍalas are intended for disciples endowed with the
best faculties. To recall, Vajraghaṇṭa states:

[The maṇḍala] is explained as prepared in stages,

depicted in paintings or drawn by way of strings and colored powder,
[while] sentient beings are naturally
present non-dual maṇḍalas.

Here is Bsod nams rtse mo’s explanation of this verse in terms of the mental
capacity of the disciple:

gyi bshad pa, folio 374a, p. 401.3.4, Ngor chen as well rejected such a method. Furthermore,
in his Dril bu pa’i lus dkyil gyi bshad pa, folios 373b–374a, p. 401.2.6–3.2, Ngor chen cites
Tsong kha pa’s ’Dod ’jo folio 122b, p. 442.5–6, and on folios 378b–379a, pp. 403.4.6–404.1.1,
Ngor chen cites the ’Dod ’jo, folio 125b, p. 448.3 and so forth.

For use by the Author only | © 2017 Koninklijke Brill NV

244 Bentor

1. Since no definite understanding will arise in the minds of persons

endowed with ordinary faculties, in the absence of paintings, etcet-
era, with actual bodies of deities endowed with faces and arms,
Vajraghaṇṭa taught: depicted in paintings, and so forth.
2. Maṇḍalas of the emblems of the deities drawn with lines on the
ground and colored powder and so forth are preferable for develop-
ing faith. Thus Vajraghaṇṭa taught: drawn by way of strings and
colored powder.
3. The maṇḍalas naturally present in the body are intended for those
with sharp faculties. Therefore Vajraghaṇṭa taught: Sentient beings are
naturally present non-dual maṇḍalas (De’i dbang gi bya ba mdor
bsdus, folio 119a, p. 408.2.2–4).

Bu ston provides an almost identical explanation,49 while Rang byung rdo rje,
the third Karmapa, 1284–1339, too holds a similar view:

The body maṇḍala was taught for the sake of disciples endowed with the
best faculties, while the constructed maṇḍala of colored powder was
taught for the sake of the intermediate and the lesser disciples (Lus kyi
dkyil ’khor gyi ’thad pa lung sbyor, folio 24b, p. 519.1–2).

Bsod nams rtse mo, Bu ston and Rang byung rdo rje alike choose to raise this
point at the very beginning of their commentaries on the works of Vajraghaṇṭa.
According to Bsod nams rtse mo:

49 Bu ston explains this in his commentary on the following verse of Vajraghaṇṭa: “The
nature of the two constructed ones is intended to accord with the disciple. These
[maṇḍalas] should not be produced for the skillful, since they will be liberated by seeing
the true meaning.” Bu ston explains: “1. For the sake of disciples endowed with lesser
faculties who do not develop faith unless there are outer deities with faces and arms,
Vajradhara taught [painted maṇḍalas], and 2. for the sake of disciples endowed with
intermediate faculties who have faith even when there is no depiction of faces and arms,
because they recognize syllables, emblems and so forth drawn in colored powder as dei-
ties, Vajradhara taught [colored powder maṇḍalas]. 3. But these two constructed ones are
not for the skillful disciples endowed with the very best faculties, since they will be liberated
by seeing the true meaning—the nondual, naturally present maṇḍala.” Lus dkyil dbang
chog, folio 5b, p. 394.5–7. Similarly, for Bsod nams rtse mo the skillful ones in this verse are
those endowed with the best faculties who will be liberated by seeing the true meaning of
the natural maṇḍala, ibid., folio 119a, p. 408.3.6.

For use by the Author only | © 2017 Koninklijke Brill NV

Tibetan Interpretations of the Opening Verses of Vajraghaṇṭa 245

On the basis the Herukābhyudaya50 and the Abhidhānottara-Tantra,51 the

master Vajraghaṇṭa teaches in terms of a person endowed with superior
mental faculties. Such a person forsakes the two constructed maṇḍalas
and relies on the naturally present maṇḍala (De’i dbang gi bya ba mdor
bsdus, folio 117b, p. 407.3.4–5).52

Once more we find Rang byung rdo rje53 and Bu ston54 in sync. Just like Bsod
nams rtse mo these masters maintain that Vajraghaṇṭa taught the body
maṇḍala for disciples endowed with the sharpest faculties, those no longer in
need of the constructed maṇḍalas of paintings and colored powder.

B Some Later Tibetan Interpretations

The notion that the three types of maṇḍala are meant for disciples with three
levels of ability has held its own in Tibetan writings. For example in the 19th
century in his Treasury of Knowledge, Kong sprul Blo gros mtha’ yas (1813–1899)
explains:55

Vajraghaṇṭa taught three types of maṇḍalas: painted maṇḍalas for disci-

ples endowed with lesser faculties, colored powder maṇḍalas for those of
intermediate faculties, and body maṇḍalas for disciples of sharp faculties
(Shes bya kun khyab 1982, vol. 2, 652).

Nevertheless, after citing this verse by Vajraghaṇṭa, Dwags po Bkra shis rnam
rgyal (1512–1587) states:56

50 Dpal khrag ’thung mngon par ’byung ba, Tōh. 374.

51 Mngon par brjod pa’i rgyud bla ma, Tōh. 369, mentioned above.
52 On forsaking the constructed maṇḍalas, see also Kṛṣṇācārya, Vasanta-tilakā, Dpyid kyi
thig le, Tōh. 1448, D. vol. 21 [wa], folio 298b, p. 596.6, Samdhong Rinpoche and Dwivedi
1990, 7&13; and Kālacakrapāda, Dus ’khor zhabs, Sekoddeśa-ṭīkā, Dbang mdor bstan pa’i
rgya cher ’grel pa, Tōh. 1353, D. vol. 14 [pa], folio 10b, p. 20.1–2.
53 Rang byung rdo rje explains that Vajraghaṇṭa composed his Cakrasaṃvara-sādhana (Tōh.
1432), for disciples endowed with the very best faculties, who forsake the constructed
maṇḍalas of lines and colored powder, and attain the stage of Vajradhara in one lifetime
through the unconstructed body maṇḍala. Lus kyi dkyil ’khor gyi ’thad pa lung sbyor, folio
24a-b, pp. 518.5–519.1.
54 Bu ston as well opens his own commentary on the Seka with the following words: “Here
the master Vajraghaṇṭa teaches that disciples endowed with the very best faculties have
the good fortune of abandoning their clinging to the two constructed maṇḍalas and
entering into the naturally present maṇḍala.” Lus dkyil dbang chog, folio 1b, p. 386.3.
55 For an English translation, see Guarisco and McLeod 2005, 211.
56 For an English translation, see Roberts 2011, 469.

For use by the Author only | © 2017 Koninklijke Brill NV

246 Bentor

Some explain that disciples endowed with lesser faculties are conferred
initiation into painted maṇḍalas, those with intermediate faculties into
mental maṇḍalas, and those with sharp faculties into body maṇḍalas.
But this is not right (Gsang sngags rdo rje theg pa’i spyi don mdor bsdus pa
legs bshad nor bu’i ’od zer, folio 59a, p. 117.1–2).

Thus, some of the later masters followed the position of the earlier masters in
this matter, while some did not.
Tsong kha pa emphasizes that the distinctions made between the three
maṇḍalas do not refer to the initial initiation into the maṇḍala at the start of
the practice. In his own commentary on Vajraghaṇṭa’s Seka Tsong kha pa states:

The meaning of this verse is not that disciples provided with lesser and
intermediate faculties are conferred initiation into the painted and the
colored powder maṇḍalas, while the ‘skillful’ disciples, that is to say those
endowed with the best faculties, are conferred initiation into the body
maṇḍala. This is because it was copiously taught that also when persons
with the best faculties first enter [into the maṇḍala], they are conferred
initiation into the colored powder maṇḍala and also into the painted
maṇḍala (Dril dbang, folio 3b1–2, p. 61.1–2).

Tsong kha pa agrees with Bsod nams rtse mo and Bu ston that the naturally
present body maṇḍala is superior and thus meant for ‘the skillful’, but he
accounts for its superiority differently. He continues:

The meaning of the line: produced for the skillful, is that the maṇḍala
in which the vital points in the body are penetrated must be the body
maṇḍala and not the two constructed maṇḍalas. This is because pene-
trating the vital points in the two contrived maṇḍalas will not bring
liberation, while through penetrating the vital points in their body
maṇḍala, [the yogis] will see the true meaning, and will be liberated.
For this reason initiations into the body maṇḍala are superior over
initiations into the two outer maṇḍalas alone (Dril dbang, folio 3b,
p. 61.2–4).57

57 Similarly in his ’Dod ’jo, Tsong kha pa explains: “Thus [Vajraghaṇṭa] taught that the
maṇḍala achieved from the body is superior to the maṇḍala achieved from painting and
colored powder … . During the path by penetrating the vital points in the two other [con-
structed] maṇḍalas, [the yogi] will not be liberated. However by penetrating the vital

For use by the Author only | © 2017 Koninklijke Brill NV

Tibetan Interpretations of the Opening Verses of Vajraghaṇṭa 247

Though Tsong kha pa emphasizes that “on the path” the body maṇḍala is
exalted over the two other maṇḍalas, the path he has in mind here is not medi-
tation on the body maṇḍala during the creation stage, but meditation on the
subtle body during the completion stage, whereby the subtle mind that can see
the real meaning is activated. According to Tsong kha pa:

This meditation on the creation stage with its practice of the body
maṇḍala is in order to ripen the meditator’s mental continuum (’Dod ’jo,
folio 126a, p. 449.1).

For Tsong kha pa, at the level of the ground, that is to say before the disciple
embarks on the path, the bases for generating the body maṇḍala are naturally
present in the body. Yet the body maṇḍala meditated upon during the path of
the creation stage is contrived and thus incapable of actual effect. Tsong kha
pa certainly regards meditation on the body maṇḍala during the creation stage
as indispensable for reaching awakening, but he does not take it as the sub-
stantial cause for Buddhahood. For him meditation on the body maṇḍala
serves to ripen the yogi towards the practice in which the subtle body serves a
major role—activating the subtle body for the path of the completion stage,
which is the substantial cause for enlightenment, the fruit of the path.
Coming full circle, then, for Tsong kha pa the meditation on the body
maṇḍala during the creation stage cannot bring about divinization. None-
theless, attainment of full Buddhahood requires that one meditate on the body
as a divine palace inhabited by divine beings. Hence according to Tsong kha
pa, meditation on the body maṇḍala indeed has a transformative effect.

C The Position of Ngor chen Kun dga’ bzang po

Where does Ngor chen Kun dga’ bzang po stand with respect to classification
by quality of meditator’s faculties? Ngor chen does not distinguish between
types of maṇḍala, but rather between types of Cakrasaṃvara Sādhanas:

Lūyīpā’s Sādhana is an elaborate yoga, while Vajraghaṇṭa’s Sādhana is

most concise. It is explained that elaborate yoga is the path for those
endowed with lesser and intermediate faculties, while concise yoga
is the path for those endowed with sharp faculties. This path taught
to those endowed with sharp faculties is most profound, in the same
way the Great Vehicle is profound in comparison to the scriptural

points in the channels and elements of the body, [the yogi] will see the real meaning and
will be liberated. Therefore, the body maṇḍala is superior.” ’Dod ’jo, folio 137b, p. 472.1–4.

For use by the Author only | © 2017 Koninklijke Brill NV

248 Bentor

collection of the Hearers (Dril bu pa’i lus dkyil gyi bshad pa, folio 379a-b,
p. 404.1.6–2.1).

Ngor chen is here referring to the Sādhanas of Cakrasaṃvara by Lūyīpā, the

Bhagavad-abhisamaya,58 and one of Vajraghaṇṭa’s Cakrasaṃvara Sādhanas.59
Once more Ngor chen follows the conventional trifold method of refuting
the position of others, presenting one’s own view and dispelling further objec-
tion. The subject here is how to meditate on the body maṇḍala.

1 Refuting the Position of the Antagonist

1.1 The Position of the Antagonist60
Ngor chen recounts how on the basis of Vajraghaṇṭa’s Sādhana, the opponent61
outlines a gradual visualization of the celestial mansion with Heruka at its cen-
ter, after which the body maṇḍala is set on Heruka in the following way:

From now on, the continuum of your earlier visualization of the physical
elements stacked up one upon the other, Mt. Meru and the celestial man-
sion proceeds without the celestial mansion being gathered. Therefore
when you begin your meditation on your body as the celestial mansion,
the way you meditate is that on the basis of each former similar moment
and each part of your body, the subsequent similar moment arises. This is
your position.

The above-cited description corresponds verbatim to that of Tsong kha pa’s

commentary in ’Dod ’jo.62

1.2 Explaining how these Positions are Refuted by Scripture and Logic
Ngor chen’s refutations take several forms such as the general and the specific.
He begins the former as follows:

The Sādhana of the body maṇḍala according to the tradition of

Vajraghaṇṭa is intended for a person endowed with sharp faculties and
excellent mental capacity. You yourself explain that the body maṇḍala

58 Sakurai 1998, Tōh. 1427.

59 Tōh. 1432, the so-called the ‘Longer Sādhana’. Vajraghaṇṭa composed another Sādhana of
Cakrasaṃvara, Tōh. 1434, the so-called the ‘Shorter Sādhana’.
60 Dril bu pa’i lus dkyil gyi bshad pa, folio 375a-b, p. 402.1.3–2.2.
61 Tib. phyogs snga ma, Skt. pūrva-pakṣin.
62 ’Dod ’jo, folio 122b, p. 442.5–6.

For use by the Author only | © 2017 Koninklijke Brill NV

Tibetan Interpretations of the Opening Verses of Vajraghaṇṭa 249

taught for the best disciples is superior (Dril bu pa’i lus dkyil gyi bshad pa,
folio 375b, p. 402.2.3–6).

Here again Ngor chen is addressing Tsong kha pa and perhaps also Mkhas grub
rje, who follows his teacher’s position. As we have seen, Tsong kha pa main-
tains that the body maṇḍala is superior to the two other maṇḍalas and thus is
practiced by the skillful, namely the superior disciple.63 According to Ngor
chen, however, superior disciples endowed with sharp faculties should prac-
tice the concise Sādhana by Vajraghaṇṭa with its instantaneous generation of
the body maṇḍala.
Ngor chen’s specific refutation takes a somewhat different tack. In it, he
argues that Tsong kha pa is at variance not only with Vajragha, but also with
Tantras and other Indian scholars.64 Let us now turn to Vajragha’s instructions
at the opening of his Cakrasaṃvara Sādhana. As before, I do not aim to uncover
some ‘true authorial intent’ but rather to discuss how the text was understood
in Tibet. As such, the translation follows the Tibetan works discussed here:

[The yogi] emanates a light ray to invite those who are already in
existence,65
makes prostrations to them, and then they enter [the yogi’s] body.
The legs, crotch, navel and heart
are the wind, fire, water and earth.
Likewise the spine is the king of mountains [Mt. Meru],
the four equal [sides of] the body are the four gates.
[This] is asserted as the body maṇḍala
of Heruka beautified by ornaments (Cakrasaṃvara-sādhana, Tōh. 1432,
folio 223a, p. 445.1–2).

The first two lines set forth a very concise practice. The deities are invited and
prostrations are made to them, after which they enter the yogi’s body. The
remaining six lines describe the body maṇḍala, beginning with the four

63 Tsong kha pa, Dril dbang, folio 3b, p. 61.4. However, as we saw, Tsong kha pa did not agree
with Bsod nams rtse mo’s classification of maṇḍala according to the disciple’s faculties.
64 Ngor chen, Dril bu pa’i lus dkyil gyi bshad pa, folios 375b–378b, pp. 402.2.6–403.4.4, cites
the Abhidhānottara Tantra and works of Vajraghaṇṭa, Dārikapa (Tōh. 1429), Maṇikaśrī
(Tōh. 1536), Prajñārakṣita (Tōh. 1465) and others.
65 While D. has dngos grub pa, Tsong kha pa ’Dod ’jo, folio 125a, p. 447.4, has sngon grub pa,
and Ngor chen, Dril bu pa’i lus dkyil gyi bshad pa, folio 378b, p. 403.4.4–5, has mngon grub
pa. The Abhidhānottara Tantra, Tōh. 369, D. vol. 77 [ka], folio 254a2, p. 507.2, has in the
parallel context sngon grub pa rnams.

For use by the Author only | © 2017 Koninklijke Brill NV

250 Bentor

maṇḍalas of the physical elements formed from four parts of the body: the
legs, crotch, navel and heart, after which arises the celestial mansion with its
four gates.66 This description parallels precisely Ngor chen’s own view of the
meditation.
Nevertheless, according to Tsong kha pa, the maṇḍala with its deities are
visualized through the steps known as “the five awakenings into manifesta-
tion.”67 Ngor chen critique of Tsong kha pa, that his position runs counter to
the instructions of Vajraghaṭa, thus comes as no surprise:

The generation of the celestial mansion and the deities residing in the
maṇḍala through ‘the five awakenings into manifestation’68 is not ex-
plicitly taught in the works of the master Vajraghaṇṭa (Dril bu pa’i lus
dkyil gyi bshad pa, folio 378b, p. 403.4.4).

Ngor chen next quotes two Cakrasaṃvara-Sādhanas by Vajraghaṇṭa, the so-

called “longer Sādhana,” mentioned just above,69 and the so-called “shorter
Sādhana,”70 which similarly instructs the yogi to visualize the maṇḍala within
the body precisely as the worshipped deities enter him or her, and concludes:

Thus Vajraghaṇṭa explains that after ‘the field for accumulating merit’71
dissolves into you, you meditate on the body maṇḍala. But he did not
teach a separate generation ritual (Dril bu pa’i lus dkyil gyi bshad pa, folio
378b, p. 403.4.5–6).

Having thus demonstrated that the gradual generation of the maṇḍala through
‘the five awakenings into manifestation’ is not explicitly taught by Vajraghaṇṭa,
Ngor chen continues:

66 These four parts of the body are taken to be in the shape of the four maṇḍalas of the
physical elements. The two legs, spread apart forming the shape of a bow, transform into
the bow-shaped wind maṇḍala. The triangular part of the crotch becomes the triangular
fire maṇḍala. The round belly turns into the circular water maṇḍala. And the square
heart, or rather the chest, becomes the square earth maṇḍala.
67 Bde mchog dril bu lugs kyi mngon rtogs dgongs pa rab gsal, folio 7a, p. 14.5.
68 Skt. abhisambodhi, Tib. mngon par byang chub pa, or mngon byang.
69 Tōh. 1432, folio 223a, p. 445.1.
70 The ’Khor lo sdom pa’i lus dkyil gyi mngon rtogs, Tōh. 1434, folio 227a, p. 453.4. This Sādhana
specifies that these deities enter the yogi’s body through the circle of hair between the
eyebrows.
71 Tib. tshogs zhing.

For use by the Author only | © 2017 Koninklijke Brill NV

Tibetan Interpretations of the Opening Verses of Vajraghaṇṭa 251

Likewise Vajraghaṇṭa did not teach this implicitly; implicit teachings are
not actually expressed through the wording of the text, but must be
derived from context, and cannot run counter to the given explicit teach-
ing. But this system of yours contradicts the explicit teaching of the
scripture (Dril bu pa’i lus dkyil gyi bshad pa, folio 379a, p. 404.1.2–3).

While Ngor chen takes issue with Tsong kha pa on additional points,72 we will
now consider his own view.

2 Presenting the Position of the Protagonist

Ngor chen describes the concise meditation on the body maṇḍala as follows:

After the maṇḍala of ‘the field for the accumulation of merit’ dissolves
into you, instantly generate the celestial mansion of the [body] maṇḍala
and the deities residing there (Dril bu pa’i lus dkyil gyi bshad pa, folio
379b, p. 404.2.4–5).

Here Ngor chen adduces support for his position by citing the teachings of
Bsod nams rtse mo and Bu ston Rin chen grub.73 The deities enter the yogi
through the point between the eyes and thereby instantly the yogi’s body is
transformed into a complete maṇḍala. In his commentary on Vajraghaṇṭa,
Tsong kha pa’s disciple Mkhas grub rje points out to additional lamas of the Sa
skya who followed this system—of which he disapproves:

In their Sādhanas of the body maṇḍala according to Vajraghaṇṭa,

Venerable Sa skya Lamas, [Chos rgyal] ’Phags pa,74 Na bza’ brag phug pa75
and Glorious Lamas instruct that ‘the field for accumulating merit’ enters
at the point in between your eyes, whereby you arise as Heruka and then
engage in meditation on the body maṇḍala (Bde dril bskyed rim, folio 4a,
p. 771.5–6).

72 These include the sequence of the meditation on the outer and body maṇḍalas—con-
nected also to the question of the relations of these two maṇḍala, the deities surrounding
the principal deities of the maṇḍala, and more.
73 Ibid., folios 379b & 380a, pp. 404.2.5–6 & 404.3.2–3. See Bsod nams rtse mo, ’Khor lo bde
mchog dril bu pa’i gzhung gi mngon par rtogs pa, folio 113a, pp. 405.2.3&4, and Bu ston, Dril
bu lus dkyil, folio 3b, p. 358.6–7.
74 Dril bu lus dkyil gyi sgrub thabs, folios 395b–396a, p. 196.2.6–3.3.
75 Na bza’ brag phug pa (1277–1350). According to D. Jackson 1990, 273, he is a master of the
Sa skya lam ’bras and one of the teachers of Bla ma dam pa Bsod nams rgyal mtshan.

For use by the Author only | © 2017 Koninklijke Brill NV

252 Bentor

3 Dispelling Objections
Ngor chen76 cites scriptural authorities such as the Saṃvarodaya77 in an effort to
dispel any further objections. He concludes that disciples endowed with lesser
and intermediate faculties ought to meditate gradually on the creation stage,
while those endowed with sharp faculties are to meditate instantaneously.

D The Position of Tsong kha pa

Tsong kha pa, for his part, articulates a position that differs both from that of
Ngor chen and several earlier Sa skya scholars. According to Tsong kha pa, the
body maṇḍala is to be visualized in stages. Moreover Tsong kha pa criticizes
those who maintain that the body maṇḍala is generated instantly:

With respect to this tradition of Vajraghaṇṭa, Tibetan lamas seem to think

that the blessed deities of ‘the field for accumulating merit’ enter at the
point in between the yogis’ eyes, and dissolve [into their bodies], and as a
consequence [these yogis] generate instantly the entire body maṇḍala
with its celestial mansion and deities. They think that in the case of the
body maṇḍala it is not appropriate to generate the four physical elements
and Mt. Meru from their respective seed syllables,78 the celestial mansion
from the syllable Bhrūṃ and so forth, and Basic Heruka through ‘the five
awakenings into manifestation’. They also seem to think that when
[Vajraghaṇṭa] said in his Sādhana:79 “[The yogi] emanates a light ray to
invite those who are already in existence,80 makes prostrations to them,
and then they enter [the yogi’s] body,” he explained that ‘the field for
accumulating merit’ enters into oneself, for he did not teach any separate
generation ritual. I do not find these proper (’Dod ’jo, folio 125a, p.
447.2–4).

Tsong kha pa here offers a lengthy explanation for his rejection of this position,
ending with:

Neither the Cakrasaṃvara Tantra nor its commentaries nor authentic

Indian works taught a generation ritual whereby the celestial mansion

76 Dril bu pa’i lus dkyil gyi bshad pa, folios 380a–381a, pp. 404.3.5–405.1.2.
77 Sdom ’byung, Tōh. 373, folio 267a, p. 533.3–4, ch. 3, v. 3.
78 The disks of wind, fire, water and earth are generated from the seed syllables Yaṃ, Raṃ,
Baṃ, and Laṃ respectively, while Mt. Meru is generated from the seed syllable Suṃ.
79 Cakrasaṃvara-sādhana, Tōh. 1432, folio 223a, p. 445.1.
80 As we saw, Tsong kha pa [as the Abhidhānottara Tantra] has sngon grub pa, while the
Tengyur editions have dngos grub pa, and Ngor chen, Dril bu pa’i lus dkyil gyi bshad pa,
folio 378b, p. 403.4.4–5, has mngon grub pa.

For use by the Author only | © 2017 Koninklijke Brill NV

Tibetan Interpretations of the Opening Verses of Vajraghaṇṭa 253

and the deities residing in the maṇḍala are generated through the disso-
lution of ‘the field for accumulating merit’. Such an interpretation was
never clearly expressed, and there are no indications for understanding
this to be so in this context (’Dod ’jo, folio 125b, p. 448.3).

Notably, Ngor chen cites this and other passages found in Tsong kha pa’s ’Dod
’jo at length.81
The generation of the body maṇḍala as an immediate result of the entry of
‘the field for accumulating merit’ into the yogi is unique to Vajraghaṇṭa. It is
found neither in the main Tantras of the Cakrasaṃvara cycle nor in the works
of the main commentators on this cycle, Lūyīpā and Nag po pa. Tsong kha pa’s
choice to generate the body maṇḍala in stages through “the five awakenings
into manifestation” is not based on Vajraghaṇṭa’s Sādhana, but on other
sādhanas of Cakrasaṃvara cycle.
Nonetheless, Ngor chen and the earlier lamas do seem to capture the
straightforward sense of Vajraghaṇṭa’s instructions. Tsong kha pa again reads
Vajraghaṇṭa in a non-literal way, but this time Ngor chen disagrees with him.
As Vajraghaṇṭa’s Sādhana is incompatible with other Sādhanas of the highest
tantra, Tsong kha pa followed instead the method of other great Siddhas as
well as the Tantras. Tsong kha pa tended to rely on the past lineage masters of
each tradition, but he evinced a superordinate concern for a uniform structure
of tantric practices. Thus while Sa skya scholars were inclined to follow closely
the Sādhana of Vajraghaṇṭa, Tsong kha pa chose to maintain consistency.
We have sampled the different approaches taken by these two scholars to
scriptural authorities and the use of logic. Ngor chen adhered to the master
Vajraghaṇṭa, despite the fact that the latter’s teachings were not in line with
other, similar Sādhanas. Moreover Ngor chen follows here not only Vajra-
ghaṇṭa, but also Tibetan lamas, including the forefathers of his own school.
What he does not do is adopt the instructions of similar Sādhanas, such as
Lūyīpā’s Sādhana of Cakrasaṃvara, for practicing Vajraghaṇṭa’s Sādhana.
Tsong kha pa, for his part, holds systematic practice to be of cardinal impor-
tance. Hence, relying on reason and analogous practices, he privileges con-
sistency of practice over the specific instructions of past lineage masters.

Conclusions

In this paper I examined several exchanges that took place between Tibetan
scholars during a formative period in the history of Tantric Buddhism in Tibet.

81 Dril bu pa’i lus dkyil gyi bshad pa, folios 378b–379a, pp. 403.4.6–404.1.1.

For use by the Author only | © 2017 Koninklijke Brill NV

254 Bentor

We encountered different ideas about the outcome of meditation on the body

maṇḍala, especially with regard to the extent of divinization achieved through
this practice. We further considered the question whether or not the body is
already divinized, with respect to the role of visualization during tantric medi-
tations. Moreover, we saw diverse views on the transformative power of
meditation on the body maṇḍala and learned that even those who maintain
that the actual fruit of the practice is achieved only during later stages on the
path nonetheless do ascribe transformative powers to this meditation during
the creation stage. Furthermore, we looked at some implications of the view
that the yogi’s body is naturally divinized even when he or she is unaware of
such divination, and the impact of such a stance on the Buddhist path and its
soteriological goals.
Honing in on Vajraghaṇṭa’s notion of maṇḍalas naturally present in the
bodies of sentient beings, we saw how Tibetan authors interpreted this in light
of their stance on Tathāgata-garbha theory. We identified that for Ngor chen
Kun dga’ bzang po, adherence to the interpretation of Tathāgata-garbha cur-
rent in his circle outweighed either a literal interpretation of Vajraghaṇṭa or
conformance with the perspectives of early important members of his own
tradition. Like many other Tibetan masters, Ngor chen disagreed with earlier
members of his lineage, although, following tradition, he generally did not
name them.
Most importantly, we learned about the systemization of tantric traditions.
Tibetan lamas, in spite of their differences, engaged in fruitful exchanges. Yet
there were different systems for establishing traditions, and choices had to be
made when interpretations and traditions clashed. For certain masters, such as
Tsong kha pa, a coherent and comprehensive system was of crucial impor-
tance while for others, including Ngor chen, the scriptural authority of Indian
masters remained paramount.

References

Tantras
Abhidhānottara-Tantra, Mngon par brjod pa’i rgyud bla ma, Tōh. 369, D. vol. 77 [ka], folios
247a–370a, 493.1–739.7. For a partial Sanskrit edition and English translation, see
Kalff 1979.
Guhyasamāja Tantra = Sarva-tathāgata-kāya-vāk-citta-rahasya-guhyasamāja-nāma-
mahā-kalpa-rāja, Gsang ba ’dus pa = De bzhin gshegs pa thams cad kyi sku gsung thugs
kyi gsang chen gsang ba ’dus pa zhes bya ba brtag pa’i rgyal po chen po. Tōh. 442, Derge
vol. 81 [ca], 181.1–295.6, folios 90a1–148a6. For an edition of the Sanskrit, see Matsunaga
1978.

For use by the Author only | © 2017 Koninklijke Brill NV

Tibetan Interpretations of the Opening Verses of Vajraghaṇṭa 255

Herukābhyudaya, Dpal he ru ka mngon par ’byung ba or Dpal khrag ’thung mngon par
’byung ba, Tōh. 374, D. rgyud, vol. 79, [ga], folios 1b–33b, pp. 2.1–66.7.
Hevajra-tantra-rāja, Brtag gnyis: Part I, Kye’i rdo rje zhes bya ba rgyud kyi rgyal po, Tōh.
417, D. vol. 80 [nga], folios 1b1–13b5, 2.1–26.5. Part II, Kye’i rdo rje mkha’ ’gro ma dra
ba’i sdom pa’i rgyud kyi rgyal po, Tōh. 418, D. vol. 80 [nga], folios 13b5–30a3, 26.5–59.3.
For Sanskrit and Tibetan editions and English translation, see Snellgrove 1959.
Saṃpuṭa-nāma-mahā-tantra, Yang dag par sbyor ba zhes bya ba’i rgyud chen po, Tōh.
381, D. vol. 79 [ga], folios 73b1–158b7, 146.1–316.7.
Saṃvarodaya-Tantra = Mahā-saṃvarodaya-tantra-rāja, Sdom ’byung or Dpal bde mchog
’byung ba zhes bya ba’i rgyud kyi rgyal po chen po, Tōh. 373, D. vol. 78 [kha], 529.1–621.6,
folios 265a1–311a6; for partial Sanskrit and Tibetan editions and English translation,
see Tsuda 1974.
Vajramālā = Vajra-mālābhidhāna-mahā-yoga-tantra-sarva-tantra-hṛdaya-rahasya-
vibhaṅga, Rdo rje phreng ba = Rnal ’byor chen po’i rgyud dpal rdo rje phreng ba mngon
par brjod pa rgyud thams cad kyi snying po gsang ba rnam par phye ba, Tōh. 445, D.
vol. 81 [ca], 415.1–554.3, folios 208a1–277b3. Translated into English by Kittay 2011.

Tengyur
Abhayākaragupta = Abhaya, ’Jigs med ’byung gnas sbas pa, Sampuṭa-tantra-rāja-
ṭīkāmnāya-mañjarī, Man ngag snye ma or Man snye = Dpal yang dag par sbyor ba’i
rgyud kyi rgyal po’i rgya cher ’grel pa man ngag gi snye ma. Tōh. 1198, D. vol. 7 [cha],
folios 1b–316a, 2.1–631.7.
Abhayākaragupta = Abhaya, Buddha-kapāla-mahā-tantra-rāja-ṭīkābhayapaddhati, Dpal
sangs rgyas thod pa’i rgyud kyi rgyal po chen po’i rgya cher ’grel pa ’jigs pa med pa’i
gzhung ’grel. Tōh. 1654, D. vol. 26 [ra], folios 166b–225b, 332.1–450.3. For Sanskrit and
Tibetan editions, see Chog Dorje 2009.
Dārikapa, Cakrasaṃvara-sādhana-tattva-saṅgraha, Dpal ’khor lo sdom pa’i sgrub thabs
de kho na nyid kyis bsdus pa, Tōh. 1429, D. vol. 21 [wa], folios 197b–203b, 394.1–406.5.
Kālacakrapāda, Dus ’khor zhabs, Sekoddeśa-ṭīkā, Dbang mdor bstan pa’i rgya cher ’grel
pa. Tōh. 1353, D. vol. 14 [pa], folios 1b–27b, 2.1–54.7.
Kṛṣṇācārya, Nag po spyod pa, Vasanta-tilakā, Dpyid kyi thig le. Tōh. 1448, D. vol. 21 [wa],
folios 298b–306b, 596.2–612.4. For Sanskrit and Tibetan editions see, Samdhong
Rinpoche and Vrajvallabh Dwivedi 1990.
Lūyīpā, Bhagavad-abhisamaya, Dpal bcom ldan ’das mngon par rtogs pa. Tōh. 1427, D.
vol. 21 [wa], folios 186b–193a, 372.3–385.1. For a Sanskrit edition see, Munenobu
Sakurai, Cakrasaṃvarābhisamaya no genten kenkyū. Chizan Gakuho, vol. 47, 1998:
1–32.
Maṇikaśrī, Śrī-cakrasaṃvaraika-vīra-sādhana, Dpal ’khor lo sdom pa dpa’ bo gcig pa’i
sgrub thabs, Tōh. 1536, D. vol. 23 [za], folios 108b–112a, 216.3–223.1.
Prajñārakṣita, Abhisamaya-nāma-pañjikā, Dpal mngon par rtogs pa zhes bya ba’i dka’
’grel, Tōh. 1465, D. vol. 22. [zha], folios 34a–45b, 67.2–90.7.

For use by the Author only | © 2017 Koninklijke Brill NV

256 Bentor

Vajraghaṇṭa, Rdo rje dril bu or Dril bu pa, Seka = Cakrasaṃvara-ṣeka-prakriyopadeśa,

Dpal ’khor lo sdom pa’i dbang gi bya ba mdor bsdus pa. Tōh. 1431, D. vol. 21 [wa], folios
219b–222b, 438.3–444.5. For a Sanskrit edition and French translation of the first
three and a half verses, see Finot 1934, 61–62 & 78.
Vajraghaṇṭa, [The longer work =] Cakrasaṃvara-sādhana, Dpal ’khor lo sdom pa’i sgrub
pa’i thabs. Tōh. 1432, D. vol. 21 [wa], folios 222b–224b, 444.5–448.5.
Vajraghaṇṭa, [The shorter work =] *Cakrasaṃvara-kāya-maṇḍala-abhisamaya, ’Khor lo
sdom pa’i lus dkyil gyi mngon rtogs. Tōh. 1434, D. vol. 21 [wa], folio 227a–227b,
453.1–454.3.

Tibetan Works
Bsod nams rtse mo (1142–1182). 1968. Sa skya pa’i bka’ ’bum, vol. 2. Tokyo: The Toyo Bunko:
Bsod nams rtse mo (1142–1182). ’Khor lo bde mchog dril bu pa’i gzhung gi mngon par rtogs
pa, folios 112a–117b, 404.4.1–407.3.1.
Bsod nams rtse mo (1142–1182). De’i dbang gi bya ba mdor bsdus, folios 117b–140a,
407.3.1–418.4.6.
Bu ston Rin chen grub (1290–1364). 1967. Collected Works, vol. 7 [Ja]. New Delhi:
International Academy of Indian Culture:
Bu ston Rin chen grub (1290–1364). Dril bu lus dkyil = Dpal ’khor lo sdom pa’i lus kyi dkyil
’khor gyi mngon par rtogs pa lhun gyis grub pa’i dkyil ’khor gsal bar byed pa. 16 folios,
353.1–384.3.
Bu ston Rin chen grub (1290–1364). Lus dkyil dbang chog = Dpal ’khor lo sdom pa’i rang
bzhin gyis grub pa’i dkyil ’khor du dbang bskur ba’i cho ga zab don gsal ba. 20 folios,
385.1–424.6.
Dwags po Bkra shis rnam rgyal (1512–1587). 2004. Gsang sngags rdo rje theg pa’i spyi don
mdor bsdus pa legs bshad nor bu’i ’od zer. Delhi: Drikung Kagyu Publications. 368pp.
TBRC: W29340. Translated into English by Roberts 2011, 401–620.
Kong sprul Yon tan rgya mtsho (1813–1899). 1982. Shes bya kun khyab. Beijing: Mi rigs
dpe skrun khang. The relevant part is translated into English by Guarisco and McLeod
2005.
Mkhas grub rje Dge legs dpal bzang po (1385–1438). 1982. Collected Works. New Delhi:
Gurudeva. Reproduced from the 1897 Old Zhol blocks:
Mkhas grub rje Dge legs dpal bzang po (1385–1438). Bskyed rim dngos grub rgya
mtsho = Rgyud thams cad kyi rgyal po dpal gsang ba ’dus pa’i bskyed rim dngos grub
rgya mtsho, vol. 7, folios 1a–190b, 3–381.
Mkhas grub rje Dge legs dpal bzang po (1385–1438). Bde dril bskyed rim = Bde mchog dril
bu lus dkyil gyi dbang du byas pa’i bskyed rim gyi dka’ gnas, vol. 6, folios 1a–12a,
765.1–787.2.
Mkhas grub rje Dge legs dpal bzang po (1385–1438). Dge bshes kon ting gu[g] śrī ba la
phul ba, work no. 43 in the Gsung thor bu, vol. 9, folios 153a–169b, 775.1–808.1.

For use by the Author only | © 2017 Koninklijke Brill NV

Tibetan Interpretations of the Opening Verses of Vajraghaṇṭa 257

Ngor chen Kun dga’ bzang po (1382–1456). Dril bu pa’i lus dkyil gyi bshad pa. Sa skya pa’i
bka’ ’bum. Tokyo: The Toyo Bunko, 1969, vol. 10, folios 117b–140a, 398.1.1–405.4.1.
’Phags pa Blo gros rgyal mtshan (1235–1280). Dril bu lus dkyil gyi sgrub thabs. Sa skya pa’i
bka’ ’bum. Tokyo: The Toyo Bunko, 1968, vol. 6, folios 394b–401a, 195.4.6–199.1.6.
Rang byung rdo rje, Third Karmapa (1284–1339). 2006. Lus kyi dkyil ’khor gyi ’thad pa
lung sbyor, included in: Shes rab snang ba’i sgrub thabs sogs. Collected Works. Zi ling,
vol. 9, folios 23b–28a, 517.4–526.4. W30541.
Tshe tan zhabs drung. 1982. Bstan rtsis kun las btus pa. Xining: Qinghai People’s Publishing
House.
Tsong kha pa Blo bzang grags pa (1357–1419). 1975–1979. Collected Works. New Delhi:
Ngawang Gelek Demo, 27 vols. Old Bkra shis lhun po redaction. Listed here in the
order found in this edition of his Collected Works:
Tsong kha pa Blo bzang grags pa (1357–1419). Sngags rim chen mo = Rgyal ba khyab bdag
rdo rje ’chang chen po’i lam gyi rim pa gsang ba kun gyi gnad rnam par phye ba, vol.
4, 512 folios, 1–494 and vol. 5, 1–530. My notes refer to the edition published in Xining:
Mtsho sngon mi rigs dpe skrun khang, 1995.
Tsong kha pa Blo bzang grags pa (1357–1419). ’Dod ’jo = Bcom ldan ’das dpal ’khor lo bde
mchog gi mngon par rtogs pa’i rgya cher bshad pa ’dod pa ’jo ba, vol. 14, 195 folios,
72–460.
Tsong kha pa Blo bzang grags pa (1357–1419). Lus dkyil = Bde mchog dril bu lugs kyi mngon
rtogs dgongs pa rab gsal, vol. 15, 25 folios, 2–55.
Tsong kha pa Blo bzang grags pa (1357–1419). Dril dbang = Rnal ’byor dbang phyug dril
bu lugs bde mchog lus dkyil gyi dbang chog rin po che’i bang mdzod, vol. 15, 27 folios,
56–109.

Sanskrit Editions [and Translations]

Chog Dorje. ed. 2009. Abhaya-paddhati of Abhayākaragupta: Commentary on the
Buddha-kapāla-mahā-tantra. Sarnath: Central Institute of Higher Tibetan Studies.
Finot, Louis. 1934. “Manuscrits Sanskrits de sādhana’s retrouvés en Chine.” Journal
Asiatique 225: 1–86.
Kalff, Martin M. 1979. Selected Chapters from the Abhidhānottara-Tantra: The Union of
Female and Male Deities. PhD diss., New York, Columbia University.
Matsunaga Yukei. 1978. The Guhyasamāja Tantra: A New Critical Edition. Osaka: Toho
Shuppan.
Sakurai, Munenobu. 1998. “Cakrasaṃvarābhisamaya no genten kenkyū.” Chizan Gakuho
47: 1–32.
Samdhong Rinpoche and Vrajvallabh Dwivedi. 1990. Vasanta-tilakā of Caryā-vratī-śrī-
kṛṣṇācārya with commentary Rahasya-dīpikā by Vanaratna. Sarnath: Central Institute
of Higher Tibetan Studies.

For use by the Author only | © 2017 Koninklijke Brill NV

258 Bentor

Snellgrove, David L. 1959. Hevajra Tantra: A Critical Study. London: Oxford University
Press.
Tsuda, Shinichi. 1974. The Saṃvarodaya-Tantra: Selected Chapters. Tokyo: Hokuseido
Press.

Non-Tibetan Works
Bentor, Yael. 1996. Consecration of Images and Stūpas in Indo Tibetan Tantric Buddhism.
Leiden: E.J. Brill.
Cabezón, José Ignacio and Geshe Lobsang Dargyay. 2007. Freedom from Extremes:
Gorampa’s “Distinguishing the Views” and the Polemics of Emptiness. Boston: Wisdom
Publications.
Davidson, Ronald M. 1991. “Reflections on the Maheśvara Subjugation Myth: Indic
Materials, Sa-skya-pa apologetics, and the birth of Heruka.” The Journal of the
International Association of Buddhist Studies 14 (2): 197–235.
Flood, Gavin. 2006. The Tantric Body: The Secret Tradition of Hindu Religion. London:
I.B. Tauris.
Flood, Gavin. 2012. “Fashioning Human Bodies in the Divine Likeness: Divinizing the
Body in Tantric Traditions.” Unpublished paper read at the AAR 2012 in Chicago.
Guarisco, Elio and McLeod, Ingrid. 2005. Systems of Buddhist Tantra. Ithaca: Snow Lion
Publications.
Gray, David. 2011. “Experiencing the Single Savior: Divinizing the Body and the Senses
in Tantric Buddhist Meditation.” In Perceiving the Divine through the Human Body:
Mystical Sensuality, edited by Thomas Cattoi and June McDaniel, 45–65. New York:
Palgrave Macmillan.
Jackson, David. 1990. “The Identification of Individual Teachers in Paintings of Sa-skya-
pa Lineages.” In Indo-Tibetan Studies: Papers in Honour and Appreciation of David
L. Snellgrove’s Contributions to Indo-Tibetan Studies, edited by Tadeusz Skorupski,
129–144. Tring: Institute of Buddhist Studies.
Jackson, David. 2007. “Rong ston bKa’ bcu pa: Notes on the Title and Travels of a Great
Tibetan Scholastic.” In Pramāṇakīrtiḥ: Papers Dedicated to Ernst Steinkellner on the
Occasion of his 70th Birthday, edited by Birgit Kellner et al., 345–360. Wien: Arbeitskreis
für Tibetische und Buddhistische Studien, Universität Wien.
Kittay, David R. 2011. Interpreting the Vajra Rosary: Truth and Method Meets Wisdom and
Method. Ph.D. diss., Columbia University, New York.
van der Kuijp, Leonard. 1985a. “Apropos of a Recent Contribution to the History of
Central Way Philosophy in Tibet: Tsong Khapa’s Speech of Gold.” Berliner Indologische
Studien 1: 47–74.
van der Kuijp, Leonard. 1985b. “A Text-Historical Note on Hevajratantra II: v: 1–2.” The
Journal of the International Association of Buddhist Studies 8 (1): 83–89.

For use by the Author only | © 2017 Koninklijke Brill NV

 Tibetan Interpretations of the Opening Verses of Vajraghaṇṭa 259

Matsunaga Yukei. 1977. “Some Problems of the Guhyasamāja-tantra.” Studies in Indo-

Asian Art & Culture 5: 109–120.
Roberts, Peter Alan. 2011. Mahāmudrā and Related Instructions: Core Teachings of the
Kagyu Schools. Boston: Wisdom Publications and the Institute of Tibetan Classics.
Ruegg, David Seyfort. 1968. “On the Dge-lugs-pa Theory of the Tathāgatagarbha.” In
Pratidānam, edited by J.C. Heesterman et al., 500–509. The Hague: Mouton.

For use by the Author only | © 2017 Koninklijke Brill NV

View publication stats