Body Mandala

UC Berkeley
UC Berkeley Electronic Theses and Dissertations
Title
The Body Mandala Debate: Knowing the Body through a Network of Fifteenth-Century
Tibetan Buddhist Texts
Permalink
https://escholarship.org/uc/item/2850d3rf
Author
Dachille, Rae Erin
Publication Date
2015
Peer reviewed|Thesis/dissertation
eScholarship.org Powered by the California Digital Library

University of California
by
Rae Erin Dachille
A dissertation submitted in partial satisfaction of
Doctor of Philosophy
in
Buddhist Studies
in the
Graduate Division
of the
University of California at Berkeley
Committee in charge:
Professor Alexander von Rospatt, Co-chair

Professor Patricia Berger, Co-chair
Professor Jacob P. Dalton
Professor Mark Csikszentmihalyi
Fall 2015
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
©2015, Rae Erin Dachille
All Rights Reserved.
!
Abstract
by
Rae Erin Dachille
Doctor of Philosophy
in
Buddhist Studies
University of California at Berkeley
Professor Alexander von Rospatt, Co-chair
Professor Patricia Berger, Co-chair
Buddhist texts perpetually remind readers to realize the pervasive nature of suffering by
reflecting upon the impermanent and even putrid nature of the human form. However,
they also proclaim birth in a human body to be the ideal condition for liberating oneself
from that suffering. How can the body be both a tool for transcendence and an obstacle to
be overcome? Within tantric Buddhism, the body mandala is a ritual process of imagining
parts of the human body as parts of the mandala, a cosmic palace inhabited by Buddhas
and attendant deities. In examining a network of texts by scholar-monks Mkhas grub rje
(1385-1438) and Ngorchen Kun dga' bzang po (1382-1456) concerning body mandala,
this dissertation brings to light complex attitudes towards the role of the body in tantric
practice and contextualizes esoteric conceptions of the body in terms of larger social,
religious, and political dynamics circulating in fifteenth-century Tibet. In bringing the
esoteric into conversation with the humanistic, this dissertation demonstrates the value of
studying ritual technologies of the body within their historical contexts as well as in
relation to discourses on the body across disciplines and cultures.
! 1!
!
!
!
!
!
!
!
!
!
!
!
!
!
For my loving mother and father,
With gratitude for this fortunate birth as your daughter
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
! i!
!
!
!
Dedication________________________________________p.i
Table of Contents__ pp.ii-iv
List of Illustrations_________________________pp.v-ix
Introduction pp.x-xxxv
I. Writing on the Body: The Body Mandala Debate as History of the Body
II. Portraits of Mkhas grub and Ngor chen: Guru, Lineage, and Mandala in Tibetan
Buddhism
III. State of Research: Sources for Making Sense of the Body Mandala Debate
IV. Challenges and Support
V. Chapter Overview
Acknowledgements__________________________________pp.xxxvi
Chapter One: Setting the Stage for the Body Mandala Debate: Polemics,
Apologetics, and Expertise in the Lives and Times of Mkhas grub rje and
Ngor chen kun dga’ bzang po pp.1-35
I. The Virtues and Limitations of Debate as a Tool for Interpretation

II. Mkhas grub rje’s Contributions to the Body Mandala Debate
III. Ngor chen’s Contributions to the Body Mandala Debate
IV. Additional Resources: Colophons and Collected Works
V. The Fifteenth-Century Context: Monastic Patronage, Scholasticism, and Charisma
Chapter Two: What is the Body Mandala? pp.36-56

I. Parsing and Purifying the Body: The Early Buddhist Context
II. Purifying and Deifying the Body in Non-Buddhist Tantra
III. Proto- Body Mandala: Inscribing the Body in a Text from Dunhuang
IV. Conclusion
Chapter Three: Mapping the Body: Locating Deities within the Bodily
Landscape in the Guhyasamåja Body Mandala Practice pp.57-103
Introduction
I. An Outline of the Guhyasamåja Body Mandala Practice
II. Issues Regarding the Mapping of Deities onto the body in Mkhas grub’s Ocean of
Attainment
IIA. Mapping the Five Buddha Families onto the Body
IIB. Locating and Correlating the Four Goddesses
! ii!
Chapter Four: Imagined or Real?: The Use of the Category of
“Fabrication”[ bcos ma ] in Mkhas grub’s Body Mandala Chapter
pp.104-135
I. The Positions of others

IA. Category confusion
IB. The Causal Link
IC. Pramåña
ID. Buddha Nature and Buddha Bodies
IE. ‘Grinding the Axe’: Mkhas grub’s Approach to Sütra and Tantra
Part II: Mkhas grub’s Own Perspective

IIA. The Mandalas of Support and Supported
IIB. Piercing to the Pith and the Superior Body Mandala
IIC. Transformation and the Ritual Context
Chapter Five: The Body Mandala Debate: Body as Explanatory Tantra

pp.135-173
I. Ngor chen’s Introduction

II. The Opponent’s View
IIA. The Protective Circle and the Body Mandala of the Support
IIB. Bu ston on Nag po pa
IIC. What Really Qualifies as a Body Mandala?
III. Ngor chen’s Reply Part One: On Bu ston and the Mandala of the Support
IIIA. A Diagram of the Mandala of the Support?: Cultivating “Iconophilia” in
Approaching Sådhana as Bodily Discourse
IIIB. Ngor chen and the Sampu†a Tantra
IV. Ngor chen’s Reply Part Two: On the Mandala of the Supported
V. Transforming the Basis
VI. Explanatory Continuum/ Explanatory Tantra [bshad rgyud]: Corporeality and

Textuality in the Body Mandala Debate
Chapter Six: In Defense of the Hevajra Body Mandala pp.174-202
I. Ngor chen’s Reply Part Three: Textual Authority

IA. On the Hevajra Tantras and Oral Instructions
IB. On the Commentaries of the other Mahåsiddhas
IBi. Nag po pa’s Position
! iii!
IBii. Jetari’s position
IBiii. Indrabhüti’s position
IBiv. Vajragarbha’s position
IC. The Vajramålå
II. Ngor chen’s Conclusion
III. Dispelling Evil Views vs. Destroyer of the Proponents of Evil : A
Comparison
IV. Reflections on the Body Mandala Debate as Bodily Discourse
Conclusion: Iconoclasm and the Subtle Body: Art, Ritual, and the Body in
a Single Painting of the Cakrasaµvara Body Mañ∂ala ” pp.203-231
I. Modes of Representing the Subtle Body: The Microcosm-Macrocosm correlation

II. Methods for Interpreting Representations of the Subtle Body
III. Visualizing the Invisible: Tantric and Medical Imagery in Tibet
IV. The Nepalese Body Mañ∂ala Painting: Challenges and Insights
A. Situating the body
B. Locating the Mañ∂ala deities
V. The Problem of Fabrication: Body and Image in the Body Mañ∂ala Debate
VI. Anomaly and Patronage: Further Reflections on the Body Mandala Painting
VII. Body, Image, and the Problem of Matter
VIII. Representation and Reality
IX. Piercing to the Pith of the Body Mandala Debate
Illustrations pp.232-284
Bibliography pp. 285-310
! iv!
List of Illustrations
[HAR #’s correspond to the Himalayan Art Resource www.himalayanart.org]
Fig.1 Mkhas grug dge legs dpal bzang po

Nineteenth-century
40.64 x 68.58 cm (16 x 27 in)
Collection of Shelley and Donald Rubin P1994.8.4
himalayanart.org HAR#56
!
Fig.2
Mkhas grub rje and the bodhisattva Manju"rî
Eighteenth-century
30 in x 20 in, 76.2 cm x 50.1 cm (image); 58 1/2 in x 33 in, 148.6 cm x 83.8 cm (overall)
The Avery Brundage Collection, Asian Art Museum Chong-Moon Lee Center for Asian
Art and Culture, San Francisco B62D33
Fig.3
Mkhas grub rje and the great adept Dombhi Heruka
Eighteenth-century
29 3/4 in x W. 20 in, 75.6 cm x W. 50.1 cm (image); 58 in x W. 34 in. 147.3 cm x W.
86.4 cm (overall)
Fig.4
Ngor chen kun dga’ bzang po with two lineages
1430’s-1460
34 1/16 x 28 3/8 in. (86.5 x 72 cm)
Michael Henss Collection, Zurich
Published in Jackson 2010 himalayanart.org HAR#88708
!
Fig.5 Mandalas from theVajråvalî Cycle, Painting Five in the Set of Fourteen
Ngor Monastery, 1429-56
35 1/4 x 29 in (89.5 x 73.7 cm); Mount: 53 x 33 in (134.6 x 83.8 cm)
Philadephia Museum of Art, Stella Kramrisch Collection 1994-138-635
Published in Thurman and Rhie 1997 [Fig.21], Kossak and Singer 1998 [Fig.47c]
!
Fig.6 and detail
Kålacakra Mandala from the Vajråvalî cycle, Painting 11 in the set of Fourteen
And painting detail of sådhaka
35 5/8 x 29 in (90.5 x 75.5 cm)
Published in Kossak and Singer 1998 [Fig.47b]
Fig.7 and detail
! v!
Mandalas of the Vajråvalî and Kriya-samuccaya cycles and detail of patron, final
painting in the set of Fourteen
35 x 29 in (88.9 x 73.7 cm)
Kimbell Art Museum AP 2000.01
Published in Thurman and Rhie 1991 [Fig.73]
Fig.8. The Maitreya Temple, Glo smon thang [Exterior and interior of middle floor
restoration]
Mustang, Nepal
Photos by Luigi Fieni, published in Lo Bue 2010
!
Fig.9 mapping of deities and tattvas onto the body according to the Trikasadbhåva
Tantra
Published in Sanderson 1986
!
Fig.10 tri"ulåbja-mañ∂ala according to the Trikasadbhåva Tantra
Published in Sanderson 1986
Fig.11 diagram of Proto-body mandala in IOL Tib J 576 accompanied by Deity List
based on IOL Tib J 576
Fig.12. Diagram & deity list of the Vajradhåtu-mañ∂ala from Giebel 2001
Fig.13 Scroll painting of the Vajradhåtu-mañ∂ala from Dunhuang

MG 17780 Recto
Institute & Copyright:
Le Musée Guimet
Site: Dunhuang Mogao Form: painting, liturgical Materials: ink and colours on silk
Size (h x w) cm: 101.5 x 61
Fig.14 Mandala painting from Dunhuang with consorts

EO 1148 Recto
Le Musée Guimet
Site: Dunhuang Mogao
Form: mandala
Materials: ink and colours on silk
Size (h x w) cm: 67.2 x 68
Fig.15 Chart of deities for cultivating the Guhysamåja-mañ∂ala palace according to the
Piñ∂î-k®ta. Appendix A in Wright 2010.
Fig.16 Chart of deities together with bodily locations for generating the body mandala
of the father deity according to the Piñ∂î-k®ta. Appendix B.2 in Wright 2010.
! vi!
Fig.17 Chart of deities together with bodily locations for generating the body mandala
of the consort according to the Piñ∂î-k®ta. Appendix B.4 in Wright 2010.
Fig.18 Main assembly hall in relation to the Vajradhåtu-mañ∂ala, Tabo monastery,

Luczanits 2004
Fig.19
Dharañî/mañ∂ala
1919,0101,0.18
Institute & Copyright: British Museum
(Ch.xxii.0015)
Form: mandala, painting
Size (h x w) cm:
58.5 x 56.3
image: http://idp.bl.uk
Fig.20
Dharañî
Pelliot tibétain 4216
La Bibliothèque nationale de France
Site: Dunhuang Mogao Materials: ink on paper
Fig.21
Dharañî/mañ∂ala
Pelliot Tibetain 389
Materials: ink on paper Size (h x w) cm: 31 x 40
Fig.22
Mandala of Guhyasamåja body mandala based on Piñ∂ik®†a sådhana
Fig.23
“Diagram for Cakra meditations” (recto & verso)
published in Heller 2010 and Pal 2007
Fig.24
Painting of the Cakrasaµvara body mandala
HAR# 5968
Nepal
Nineteenth or twentieth century
! vii!
Ground Mineral pigment on cotton
Private Collection
www.himalayanart.org
Fig.25
HAR#100001
Nepal, Eighteenth century
91.44 x 210.82 cm [36 x 83 in]
Ground mineral pigment on cotton
Collection of Shelley and Donald Rubin
Fig.26
Vishnu Vi"varüpa.
India, Rajasthan, Jaipur, ca. 1800-20.
Opaque watercolor and gold on paper, 38.5 x 28cm.
Victoria and Albert Museum, London, Given by Mrs. Gerald Clark, IS.33-2006
Published in Diamond 2013, fig. 10b
Fig.27
Painting from the Blue Beryl series illustrating section of chapters three and four of the
“Explanatory Tantra” [bshad rgyud] section of Desi Sangye Gyatso’s Blue Beryl Treatise
[Vaidurya sngon po]
Twentieth-century copy
Collection of the Buryiat Museum, Ulan-Ude
Published in Fernand Meyer, Parfionovich & Dorje 1992, Plate 6.
Fig.28 “Cosmological Buddha from a fresco in Cave 428 at Tun-Huang. Dated to the
Northern Wei dynasty, circa 525 (After Tonkø Makkøkutsu, vol.1, pl.162.)”
Published in Howard 1986, fig. 2
Fig.29 “Cosmological Buddha. Dated to the Sui Dynasty (581-618). Stone, 176.5 x
64.2 cm. Courtesy, Freer Gallery of Art, Smithsonian Institution, Washington, D.C.”
Published in Howard 1986, fig.1
Fig.30
Folio 4 from the Siddha Siddhanta Paddhati
Bulaki
India, Rajasthan, Jodhpur, dated 1824 (Samvat 1881)
Opaque watercolor and gold on paper, 122 x 46 cm
Mehrangarh Museum Trust, RJS 2376
Published in Diamond 2013 Fig.11B,
Fig.31
Bulaki
! viii!
Published in Diamond 2013 Fig.10D
Fig.32
Jain image of a cosmic man
Published in Granoff 2009-10. Victorious Ones: Jain Images of Perfection 2009-2010
Fig 3.5
fifteenth-seventeenth century
Fig.33
“The Jain Universe in the Shape of a Cosmic Man or lokåpurußa”
Folia from loose-leaf manuscript
Gujarat or Rajasthan, early seventeenth century
Ink and opaque water color on paper
Collection of Bina and Navin Kumar Jain
Published in Granoff Victorious Ones: Jain Images of Perfection 2009-2010 Fig.2.1
Fig.34
Painting from the Blue Beryl series illustrating sections of chapter four of the
“Explanatory Tantra” [bshad rgyud] section of Desi Sangye Gyatso’s Blue Beryl Treatise
[Vaidurya sngon po]
HAR# 81836;
Tibet, eighteenth-century copy
Ground Mineral Pigment on Cotton
Collection of the Rubin Museum of Art
Fig.35
Body mandala of Vajrayoginî
Published in English 2002 fig.33
Fig.36
Cakrasaµvara-mañ∂ala
HAR #85813;
Nepal, 1490 Buddhist Lineage;
Ground Mineral Pigment on Cotton; Collection of the LA County Museum of Art;
Published in Huntington and Bangdel 2003 Fig.70; www.himalayanart.org
Fig.37
numbered diagram of the Cakrasaµvara body mandala painting (Fig. 24 HAR# 5968)
with accompanying list of deities and inscriptions
! ix!
!
!Introduction
Within tantric Buddhism, the body mandala is a ritual process of imagining the
human body as a mandala, a cosmic palace inhabited by buddhas and attendant deities.
This dissertation examines a network of texts by two Tibetan scholar-monks, Mkhas grub
rje (1385-1438), a champion of the Gelukpa tradition, and Ngorchen Kun dga' bzang po
(1382-1456), a hero of the Sakyapa legacy, concerning body mandala. In the process, it
brings to light complex attitudes towards the role of the body in tantric practice. It also
contextualizes esoteric conceptions of the body in terms of larger social, religious and
political dynamics circulating in fifteenth-century Tibet.
The dissertation will consider how Mkhas grub and Ngor chen’s respective
varieties of polemical, ritual, philosophical, and exegetical expertise inform their
approaches to body mandala. It is built upon the conviction that the body mandala texts
provide insight into the authorial personas of Mkhas grub and Ngor chen as well as into
the polemical and exegetical cultures of fifteenth-century Tibetan Buddhism. These
insights include but are not limited to their potential involvement in sectarian formation
and “sectarian differentiation.”1 Analyzing the body mandala debate texts and their later
interpretations therefore provides an opportunity to work towards a better understanding
of how Geluk and Sakya identities came to be regarded as distinct. Therefore, it takes
into account how institutional and socio-political relationships of patronage, lineage, and
abbatial succession factored in the creation and interpretation of this network of texts.
These texts are classified as tantric polemics, a genre that by definition experiments with
the boundaries between at least two Buddhist discourses, tantra and of philosophy.
Therefore, this project also illuminates the subtleties of how Tibetan scholastics
reinforced or recreated relationships to the past as well as relationships between texts,
between different modes of Buddhist discourses, and between different varieties of
Buddhist practice. Finally, the dissertation creates a space to explore broader questions
about the interpretation of ritual and the body’s role in both soteriological advancement
and exegetical practice.
I. Writing on the Body: The Body Mandala Debate as History of the Body
Whether regarded as a tool or as an obstacle, the human body is the ultimate

framework situating us in time and space, the backdrop against which our stories are
woven and the altar upon which our rituals are enacted. It may be the ground for
contestation of socio-political tensions as well as a powerful agent for change. Broadly,
this dissertation is motivated by questions surrounding the paradoxical status of the
human body as limitation and potentiality as reflected in visual and textual
representations of Himalayan ritual life.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
1
Cabezón 2007 creates an important distinction between “sectarian differentiation” and
sectarianism; this distinction will be discussed below. Cabezón, José Ignacio and Geshe
Lobsang Dargyay. 2007. Freedom from Extremes: Gorampa’s ‘Distinguishing the
Views’ and the Polemics of Emptiness. Boston: Wisdom Publications. See p.7.
! x!
The body mandala is not a monolithic practice, but rather varies across Buddhist
tantric cycles and transmissions. Comparing different ways of mapping deities onto or
within the body may reveal important transformations, exchanges, and influences that
will contribute to a more complex and engaging portrait of the development of the role of
the body in tantric ritual. For example, such comparisons may allow us to observe how
conceptions of the body changed at the intersection of religious, medical, socio-political
and perhaps even artistic domains. This dissertation gestures toward this larger nexus of
ideas of the body in which tantric ritual participates. In part, it is an attempt to bring
tantric bodies out of the shadows and to demonstrate the larger relevance of tantric
conceptions of the body to the history of the body across temporal and geographical
boundaries. This project takes a step towards engaging tantric corporeality in a dialogue
with other ritual technologies focused upon the body, technologies that form the basis of
study for scholars of religious studies, anthropology, and even the history of medicine. In
addition, it builds upon a trend in historical studies, the “New Historicism” of the 1980’s
and the resultant new “cultural history” whereby:
“history was coming to be seen primarily as a set of changing representations of the past.
Situating bodies historically in their appropriate ‘representational regimes’ was part and
parcel of the re-thinking of meaning, purpose and shape of history. Increasingly,
therefore, history (as in the history of the body) was approached as a text: authored,
discursive, and malleable in every respect...Thus did the new cultural history render the
body and historical epistemology privileged sites for literary and cultural analysis.”2
In engaging with the body mandala debate texts as examples of bodily

representations, we must first define what we mean by “representation.” Within
philosophy, “representationalism asserts that we know not external things, but the
representations, or copies, of external things in our own minds.”3 While many strands of
Buddhist philosophical discourse espouse skepticism towards the relationship between
representation and reality, tantric practices embrace representations as soteriological
tools. In particular, deity yoga, of which body mandala practice is a form, employs a
variety of representations of deities, evoked through ritualized mediums such as sound,
recitation, and image, to recreate as the practitioner’s sense of self. Ritual is therefore, in
a sense, a creative act that employs categories of representation, imagination, and
metaphor to transform the relationship of self and world.
Many definitions of representation suggest a transformation of an “original”
through acts of replication, imagination, performance, and re-creation. In her 2008 study
of Tibetan embryological narratives, Frances Garrett remarks: “The understanding of
metaphoric language puts emphasis on an aesthetic appreciation of the world, which may
be experienced in ritual, symbolic and artistic activity, and which serves to help recreate
one’s own life-world, a central goal of much religious activity.”4 The boundaries of the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2
Roger Cooter, “The Turn of the Body: History and the Politics of the Corporeal.”
ARBOR CLXXXVI 743 may-june (2010): 397.
3
C.E.M. Joad, Guide to Philosophy, I.ii.41 (1936) [OED online]
4
Garrett, Frances. 2008, Religion, Medicine and the Human Embryo in Tibet, New York:
Routledge, p.124.
! xi!
Buddhist “life-world” are fixed within saµsåra, a cycle of perpetual rebirth impelled by
desire, hatred, and ignorance. Saµsåra is a framework governed by perpetual replication
and representation. Replication applies in its sense of repetition; rebirth is a form of
replication, therefore, in the earlier musical definition of the word, rather than its later
definition as a copy, reproduction, or likeness of an “original.” The infinity of rebirth
confounds any attempt to locate such an original. The continuity of karma and subtle
mental habits and attachments takes form in different representations or bodies across
limitless lifetimes. Only the most realized beings, buddhas and bodhisattvas, have the
ability to recognize the connections between these representations, to perceive how they
are linked by a chain of cause and effect and a profound level of human inter-
relationship. Only these beings can control their rebirth process. Therefore, from a
Buddhist cosmological perspective, representation and embodiment are intrinsically
linked.
Tantric Buddhism reinforces this link between representation and embodiment
through ritual technologies of control. Through controlling the body, one learns to control
the mind. In his study of Tibetan ritual, Stephen Beyer defines tantra in terms of control,
and specifically, the control of representations of the deity: “Tantra is thus the ‘quick
path’ wherein control is synonymous with power; to control the divine appearance,
mantra and ego is to act with the deity’s body, speech and mind, and to control the mind
and body is to own the world.”5 Tantric ritual acts of imagination involve the repeated
creation and destruction of representations. In identifying with the deity in deity yoga,
the practitioner, by analogy creates and destroys the ego, the sense of self that confines
them. Therefore, from one perspective, controlling one’s sense of the body through ritual
allows one to transform the sense of self. Like a ritual effigy, the body as representation
is a substitute for the self, one with which it is deeply connected. It is the exclusive
medium and vehicle through which an individual can strive for liberation from the
confines of saµsåra.
Like an image or other variety of representation, the body functions as a support
[Tib. rten] for tantric ritual practice. Bodies are a kind of representation linked to notions
of being inhabited, derived from an original, or the outcome of a creative process. The
relationship between “representation” and body is expressed in the overlap in the
semantic range of “image” and “body” in Sanskrit and Tibetan languages as “support
[rten],” “reflection [gzugs brnyan Skt.prati-bimba]” “or something that is fashioned or
molded [Skt. deha].” The link between representations and bodies is reinforced by ritual
practice. Both must be consecrated in order to be suitable to contain a divine presence or
“represent” the deity.6 Bentor 1997 emphasizes this dimension of the concept of rten by
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
5
Beyer 1973, p.94. Stephen Beyer, The Cult of Tåra: Magic and Ritual in Tibet
(Berkeley and Los Angeles: University of California Press, 1978).
6
In his 2014 study of Newar Old Age rituals, Von Rospatt highlights this correlation
between the consecration of images and persons, detailing rites in which the two
activities even occur side by side.
von Rospatt, Alexander. 2014. “Negotiating the Passage Beyond a Full Span of Life: old
Age Rituals Among the Newars.” South Asia: Journal of South Asian Studies. Vol.37.
No.1, 104-129.
! xii!
translating the term as “receptacle.”7 This translation is favorable in also lending itself to
microcosm-macrocosmic correlations of bodies and representations with the world as a
container inhabited by sentient beings. Such correlations are certainly relevant to the
discussion of body mandala as composed of the abode, the mandala of the support [rten]
and the supported [brten pa], the deities inhabiting it. rten may function as containers of
divine presence, providing a site for ritualized veneration. Certainly in the case of mchod
rten [Skt. stüpa], reliquaries housing the remains of buddhas and enlightened masters (as
well as a variety of other representations), the element of containment is key. Close
studies of consecration rites for stüpas and images demonstrate the ways in which ritual is
used to navigate the complex relationships between container and contained.8 The
interplay of these elements is suggested by the relationship of Tibetan paintings [thangka]
to the consecratory inscriptions and forms that appear on the reverse side. 9 The reverse
provides an essentialized view of the forms from the painting, sometimes marked by the
seed syllables and mantras that generate the deities on the front side, making them
present in the container of the image. Traditionally the syllables om, ah, and hum are
inscribed on the rear at the points corresponding to the crown, throat, and heart of the
image of the deity on the front together with the consecration verse. Some paintings even
bear the outline of a stüpa on their reverse side in the place where the body of the main
deity appears on the front. Belonging to the highest class of forms of representation,
those of the mind, stüpa are embodiments of the dharmakåya. Paintings, one the other
hand, are nirmåñakåya representations, compassionate emanations for the benefit of
beings, but more coarse by nature, and correspond to the body of the buddha. The
complex interplay of support and supported, container and contained, brought to light by
such ‘double images’ provoke some interesting questions about the status of images or
representations both as bodies and as containers.
Tantric practice predicates that the practitioner fashion oneself into the perfect
container for receiving powerful teachings. The famous Tibetan tantric master Milarepa
explained this metaphor as follows:
“The milk of a white lion must have a special container. It cannot be put in any ordinary
one. If it is put in a clay pot for instance, as soon as the milk touches the clay pot, the pot
cracks. For these vast and profound teachings of this lineage, there must be a special
kind of practitioner. I refuse to teach the tradition to anyone who comes to receive my
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
7!In her 1997 monograph, Bentor provides a detailed look at the connection between
tantric sådhana practice and the consecration rites of Tibetan reliquary stüpas.!Bentor,
Yael. 1996.Consecration of Images and Stüpas in Indo-Tibetan Tantric Buddhism.
Leiden: E.J. Brill.
!!
8
See Bentor 1995 & 1997, von Rospatt 2010, and Owens 1995.
9!“Flip Side,” an exhibition curated by Christian Luczanits for the Rubin Museum of Art
(3/2013-2/2014), displayed a choice selection of paintings together with their ‘flip sides,’
bringing attention to this important and long overlooked aspect of Tibetan art and its links
with ritual practice.
!
! xiii!
teachings who is not ready for them. I will only teach it to persons who are completely
developed and suitable, who are ready for this teaching and the practice of it.”10
However, in translating rten as “support” rather than “container” or “receptacle,” we are

emphasizing that rten are representations that act as the foundations for a creative
process, a process of reimagining and thereby recreating the self. 11
As a technology of inversion, tantra applies the poisons that obstruct liberation
from saµsåra, particularly desire and violence, as tools for over-coming those self-same
poisons. In this light, the violent or sexualized imagery of tantric deities, ‘re-presents’
the very ties that bind; through multiple acts of representation, one moves closer to
overcoming these fetters. This logic of inversion governs the principle of representation
itself. From a Buddhist perspective, one of the fatal flaws of the mind is its tendency to
proliferate [prapa•ca] ideas and images and to falsely ascribe [vikalpa] meaning to them.
Yet the tantric ritual framework of deity yoga demands that the practitioner generate a
host of images and connect them with particular ideas and qualities. The purpose is to
acquaint him/her with the divine nature of buddhahood and to habituate him/her in
cultivating these divine qualities, in representing the deity. These techniques aid the
practitioner in “overcoming the tyranny of ordinary appearances,” and specifically in
cutting attachment to a fixed notion of self and world.12 Tantric ritual acknowledges the
deep-rooted nature of the human mind’s tendencies to fixate upon ideas and images,
however refined or well-intentioned. This awareness is reflected in practices such as the
common preparatory ritual of dissolving ordinary appearances into emptiness before
cultivating or consecrating a divine form. Likewise, the images generated by the mind in
tantric sådhana practice are dissolved into emptiness. Through this repeated process of
creating and destroying representations, the practitioner progresses toward the ultimate
goal of buddhahood.
The emphasis upon this process of continuously creating and destroying
representations suggests that the theory of representation embraced by tantric Buddhism
resists a one-to-one correlation between representation and reality. Bruno Latour has
provided useful tools for better understanding how representation works in his study of
iconoclasm through the lense of science, art, and religion.13 He has observed the ways in
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
10
Geshe Ngabang Dargyey. Oral translation by Lobsang Gyaltsen. (Dharamsala, India:
1979, edited by Samaya Hart and Alexander Berzin, November 2003).
www.berzinarchives.com
11
See Owens’ 1995 study of Newar consecration rites of images for an interesting
exploration of the problem of agency in interpreting these rituals. Owens, Bruce McCoy.
1995. “Human Agency and Divine Power: Transforming Images and Recreating Gods
Among the Newar.” History of Religions 34 (3). University of Chicago Press: 201-40.
In translating rten as “support,” we are, like Owens, promoting the sense of implied
agency of human ritual agents rather than exclusively divine ones.
12
Thubten Yeshe, and Jonathan Landaw. 2001. Introduction to Tantra: the transformation
of desire. Boston: Wisdom Publications: p.44.
13
Latour, Bruno, “How to be Iconophile in Science, Art and Religion” in Picturing
Science, Producing Art. Ed. Caroline A. Jones & Peter Galison. New York & London:
Routledge, 1998. 418-440: 420-421.
! xiv!
which not only have images themselves been fetishized but their denial and destruction
has been fetishized as well. We might use Latour’s term, “iconophilia” to better
understand the ideal understanding of representations and by extension, even bodies,
from the tantric perspective. Latour coined the term to describe the process of privileging
paths of transition, “the movement of images,” in the meaning-making process. A vital
aspect of this principle is to dispense with the misconception that there is a direct
correspondence between meaning and representation. Latour describes the way that
meaning flows through form as a process that eliminates the possibility of a one-to-one
correspondence between a message and delivered statement. The message is inevitably
transformed. The process of continuously generating and destroying images in tantric
practice is thereby an iconophilic process. The two-fold structure of tantric sådhana
practice as composed of a generation and completion or perfection process supports this
interpretation. Sådhana is the core practice of tantric ritual “accomplishment” whereby a
practitioner who has been initiated into the cult of a particular deity intensifies their
relationship to that deity and thereby to the ultimate goal of buddhahood through the
daily cultivation of those divine forms and qualities. The generation stage focuses upon
the creation of divine forms and the initial correlation of self and buddha. To ‘complete’
or ‘perfect’ the representation of self as deity is to let go of it by degrees of increasing
subtlety.
The role of the body in this process of creation and destruction is complex. The
body is the context, the stage for the ritual drama, the foundation for spiritual practice,
and also a reminder of the liminality of the human condition. The body is also the
prototype or measure for the generation of divine bodies. It is the point of departure for
the correlation of self and buddha. Tantric ritual recognizes additional capacities of the
human body, subtle aspects of human potentiality that are central to the completion or
perfection process of tantric sådhana. The term “subtle body” is often used to describe
these more elusive aspects of human potentiality. In his recent edited volume on the
“subtle body,” Geoffrey Samuel has traced the Western usage of the term back to a
translation of the Vedantic term sukßma-"arîra employed by members of the
Theosophical Society.14 Samuel accounts for the challenges posed by the history of the
term while preserving it as a workable category for a complex network of concepts and
practices suggested by early Upanißadic, late Vedic, and classical Vedantic literature in
addition to their more explicit and familiar development in yogic as well as Buddhist and
Hindu tantric literature. The specificity and diversity of conceptualization and practical
application varied across traditions.
Bodies, like images, also assume both material and non-material forms. The
broader Mahåyåna Buddhist tradition describes the Buddha’s own embodiment through a
variety of frameworks of corporeal classification by degrees of subtlety. The most
widely discussed is the three-fold structure of emanation body [nirmåñakåya], a flesh and
blood form, enjoyment body [sambhogakåya], a body composed of light as seen in
dreams, visions, and other liminal states, and a dharma body [dharmakåya], a formless
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
14
Samuel, Geoffrey, and Jay Johnston. 2013. Religion and the subtle body in Asia and
the West: between mind and body. Abingdon, Oxon: Routledge: p.2.
! xv!
body that is the truth of the buddha’s teachings themselves. These three types of Buddha
body are therefore organized by subtlety of form and of perception. The first is a
material body, one that can be touched as well as seen. In the case of the Buddha,
Siddhartha Gautama, this form was ultimately consumed by fire upon the event of his
ultimate enlightenment or paranirvåña. Unlike the event of initial awakening or nirvåña,
this ultimate event is one described as “without remainder,” in other words without a
body. The second type of body is immaterial and cannot be perceived under ordinary
conditions. It is the body of illusions, dreams, visions, or the body teaching to disciples
in another dimension, like one of the Buddhist Pure Lands. While the emanation and
enjoyment bodies are both identified as “form bodies” [rüpa-kåyå], the third type of body
is both immaterial and formless; it cannot be seen but only known.
Representations are also often organized along a similar trajectory of gross to
more subtle varieties. One familiar framework for classifying representations of the
Buddha is according to their expression of the Buddha’s body, speech, and mind [sku,
gsung, thugs Skt. kåya, våc, citta]. Over time, Buddhist scholars have theorized these
classifications in different ways, ranking and stratifying representations accordingly.
Dagyab describes the framework as follows:
‘rten’ in the religious sense means an aid to memory, an aide memoire or reminder of the
real thing which the object stands for (hence a ‘support’). For example all religious
statues portraying a buddha, deity or holy being belong to ‘sku-rten’ (physical
reminders); all religious written works belong to ‘gsung-rten’(verbal reminders); and all
objects directly related to religious practices, such as mchod-rten, mandala and attributes,
belong to ‘thugs-rten’ (spiritual reminders).15
Dagyab’s emphasis upon the property of recollection in defining rten invites the question
of what precisely is being recollected. Some of the perspectives discussed within the
body mandala debate suggest that the human body functions as a support for recollection
of one’s own divinity. The ontological implications of this notion of recollection for the
body of the tantric practitioner will be examined more deeply within this dissertation.
One of the distinguishing features of tantra is that it provides the practitioner with
means for attaining a buddha body, a body of enlightened form.16 The Tibetan tantric
perspective describes many kinds of subtle bodies, like the illusory body [sgyu lus], the
rainbow body ['ja lus], and most importantly for the purposes of this dissertation, the
vajra body [rdo rje'i lus]. The vajra body is the body of the tantric practitioner
characterized by the network of channels, winds, and drops [Tib. rtsa rlung thig le].
Although theoretically these elements are present in all human bodies, they are only
perceived by accomplished practitioners. Through the completion or perfection stage of
sådhana, the tantric practitioner comes to see and manipulate these subtle elements as a
means of realizing one’s own buddhahood. The subtle body is another variety of
representation, another kind of body. This dissertation will work with a provisional
definition of the subtle body as a body defined by invisible structures and processes
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
15!See!DaDagyab, Loden Sherab. 1977. Tibetan religious art. Wiesbaden: Harrassowitz:
p.25
16!See!Dagyab!1977,!p.15!
! xvi!
realized exclusively by the advanced tantric practitioner through sustained ritual practice.
Samuel regards the category of subtle body as a particularly useful tool for exploring the
space between “materialistic and idealistic extremes.”17 We might add that the subtle
body re-presents corporeality itself, that it is crucial to the practitioner’s iconophillic
meaning-making process. As such, it facilitates a transition between different kinds of
representations, flesh and blood matter, immaterial mental images, and subtle states that
hover somewhere between the mental and the somatic. The key to buddhahood lies not in
any one particular representation but rather in the process of continual creation and
destruction and the resultant transformation of one’s mind state and one’s relationship to
self and world.
In devising a method for interpreting the body mandala debate texts as bodily
representations, representations that may be related to a larger history of the body, we
might begin with Foucault’s notion of the body as a “cultural text.” In this light, the body
may be regarded as a text upon which society maps its norms and desires as well as the
ground for regulation through “discipline.”18 Thus, the body mandala would be one
variety of writing on the body, one that might be considered alongside laws regulating
bodily conduct, medical prescriptions for diet and behavior, and reproductive politics, to
name a few. Of particular interest are ritual technologies that “discipline” the body as
well as those that reveal soteriological goals. In all instances, we must be cautious in
correlating representations and reality and abandon the notion of a perfect replica. We
must strive instead to become iconophiles, to grapple with the transitions between
representations in order to make meaning of them.
Like representation, ritual presents us with a means to re-frame our perception of
reality, to see it in a new way. Sharf and Bell both suggest ways in which ritual is
embodied. 19 Sharf remarks upon the way ritual reshapes the ways in which we
experience the world: “Participation in a living ritual tradition reaches beyond the
vagaries of the intellect to one’s somatic being; ritual habituation indelibly inscribes the
self with a set of perceptual orientations, affective dispositions, and autonomic responses
that are, in effect, pre-cognitive.”20 In proposing we consider the possibility that ritual is a
form of play, Sharf has built upon Bateson’s work to suggest an analogy between the
manner in which we “re-create” out life worlds through ritual and the nature of
“recreation” itself. Bell uses “ritualization,” a term coined by Gluckman, to describe the
way in which ritual is “embedded within the dynamics of the body defined within a
symbolically structured environment.”21 Drawing upon Bordieu’s “dialectic of
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
17!Samuel 2013, 4.!
18
See Susan Hekman, “Material Bodies,” in Body and Flesh: A Philosophical Reader,
edited by Donn Welton (Malden, MA & Oxford: Blackwell Publishers, Ltd., 1998), 61-
70. Hekman provides an interesting comparison of Judith Butler and Susan’s Bordo’s
approaches to Foucault’s writings on the body.
19
See Bell, Catherine. 1992. Ritual Theory, Ritual Practice. Oxford University Press.
& Sharf, Robert H. 2005. "Ritual," in Critical Terms for the Study of Buddhism, edited
by Donald S. Lopez, Jr. (Chicago: University of Chicago Press), pp. 245-269.
20
Sharf 2005, p.249
21
Bell 2009, 93. For Gluckman’s introduction of the term, see Gluckman 1962.
! xvii!
objectification and embodiment,” Bell describes the generation of this ritual body as a
circular process whereby ritualization sets the conditions for a “restructuring” of human
agents through their interaction with a hierarchically crafted environment.22 Individuals
thereby internalize the designations of social and cosmic order laid out in ritual and
reproduce them in spontaneous ways. They make them their own by using them as tools
for realizing their own vision of how they fit into a wider realm of human relationships.
They are thereby both transformed by and transforming the parameters of the world
around them, creating shifts in dynamics of power and subordination.
We will return to reexamine some of these larger questions of ritual,
representation, and the body in a new light in the conclusion of the dissertation. There,
we will explore an anomalous body mandala painting, considering why it appears to be
one of the only of its kind. We will observe the solutions the artist produced in response
to the challenges of representing the body mandala practice without an identifiable
prototype and consider other varieties of representations of the subtle body from India
and the Himalayas. If the tantric tradition indeed embraces representation in many
different senses, as replication, imagination, performance, and re-creation, if it
encourages us to be iconophilic, why not represent body mandala in visual and material
form? We will consider this question in dialogue with controversies regarding the
fabrication of mental and material images as well as different varieties of bodies, some
more subtle than others, that arise within the body mandala debate texts. In the process,
we will uncover hidden aspects of the relationship of representations and bodies as
‘supports’ for achieving ritual goals.
We will begin by becoming acquainted with the two authors of our body mandala
debate texts, Mkhas grub and Ngor chen, and the broader Tibetan Buddhist context in
which they participate. The role of the mandala as a support for tantric practice will also
be contextualized within a framework of ritual and institutional associations. These
associations will be useful in progressing towards engaging with a particular form of
mandala, body mandala, as represented within the body mandala debate texts.
II. Portraits of Mkhas grub and Ngor chen: Guru, Lineage, and Mandala in
Tibetan Buddhism
The mandala is a cosmic palace inhabited by Buddhas and attendant deities. To offer the
mandala to one’s teacher, is to offer oneself and the world to them. Several portraits of
Mkhas grub rje dge legs dpal bzang po (1385-1438), or Mkhas grub rje, famed fifteenth-
century Gelukpa scholar and second abbot of Ganden monastery, depict him making a
ritual offering of mandala to his teacher, Tsong kha pa (1357-1419). [Fig.1]23 The
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
22
Bell 2009, 97. See Pierre Bordieu 1977, p.87.!
23
Examples include this one from the Collection of Shelly and Donald Rubin. Himalayan
Art Resource [HAR] #56, as well as HAR#23391 from a Private Collection, #71928 from
Tibet House, Delhi. I am grateful to Wenshing Chou for bringing an example from the
Freer and Sackler Galleries [Acc. #F1905.74] to my attention as well as the example from
the Asian Art Museum of San Francisco referenced below. Dr. Chou has explored such
images in her work on Wutaishan, the Chinese mountain known as the abode of
Manju"rî.
! xviii!
abbacy of Ganden monastery, founded by Tsong kha pa’s students for their master,
became a central institutional role for the emerging Geluk tradition. In this nineteenth-
century portrait from the Collection of Shelly and Donald Rubin, Tsong kha pa appears to
Mkhas grub as a vision in the clouds performing the dharmacakra mudrå or the gesture of
teaching.
Ary 2007 has highlighted the significance of such representations of Mkhas grub
and Tsong kha pa’s encounter in visual and textual sources, linking them to an event from
Mkhas grub’s hagiography. After his teacher’s death, during his time at Mdangs chen,
Mkhas grub is alleged by his biographers to have experienced such visions of Tsong kha
pa. These visions solidified Mkhas grub’s status as heir to Tsong kha pa’s spiritual
legacy. They also promoted Tsong kha pa’s divine identity, an identity understood to
have been perfected through tantric practice. For example, Tsong kha pa’s identity with
the bodhisattva Manju"rî is expressed in visual form in a painting from the collection of
the Asian Art Museum in San Francisco.24 [Fig.2] In another painting from the same set,
also in the collection of the Asian Art Museum, Tsong kha pa appears to Mkhas grub in
the guise of the great Indian realized tantric master, Dombhi Heruka.25 [Fig.3]
In all of these paintings, Mkhas grub makes the mandala offering to his teacher.
In a sense, they demonstrate both Tsong kha pa’s ability to become a buddha through
ritual practice as well as Mkhas grub’s ability to regard him as such. In performing any
initiation ritual, the guru must first assume this divine status, to act as the deity, what is
commonly referred to as “deity yoga.” Learning to regard the guru as buddha shapes the
disciple into a suitable vessel for receiving the tantric teachings; the process begins in
ritual acts of initiation and develops further through daily practice. The ritual practice of
adopting this mind frame is termed “guru yoga.” Through daily rituals like mandala
offering, the disciple cultivates this proximity to the guru and thereby to buddhahood.
The act of mandala offering in these portraits evokes these tantric ritual relationships; it
provides a context in which the guru acts as the deity, eliciting the student’s allegiance
and bestowing permission upon the student to also act as the deity. In this way, the
portraits reinforce the authority invested in Mkhas grub by Tsong kha pa as well as the
intensity of their relationship.
Historically, lineage representations have played an important role in creating and
supporting institutional identities for the new schools of Tibetan Buddhism (Sakya,
Kagyu, Kadam/Geluk) emerging after the so-called dark age. Such representations
emphasized the bond between guru and disciple and connected Tibetan traditions to
Indian origins. Luczanits has highlighted the significance of representing the relationship
of guru and disciple through lineage portraiture within the larger progression of Tibetan
Buddhist history:
“The notion of the direct succession of a certain teaching tradition from person to person
has its root in the Tantric tradition, which prescribes initiation into a certain type of
teaching. However, the systematic emphasis on such a derivation by means of a teacher’s
lineage appears to have become prominent in Tibet only during the 12th century with the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
24
Collection of the Asian Art Museum, San Francisco: Avery Brundage Collection
B62D33.
25
Asian Art Museum, Avery Brundage Collection B62D37.!
! xix!
new schools, and became extremely influential. Whatever the social and political
circumstances that supported such a move, the need to justify a teaching by its link to the
Indian tradition, thus demonstrating its authoritative derivation, is evidenced by the
prominent position given to the lineage in the literature and painting of the time.” 26
The relationship of Tibetan gurus to Indian masters assumed an iconic status symbolizing
the legitimacy of the second wave of Tibetan Buddhism; to represent a succession of
Tibetan masters perpetuating the teachings of these Indian masters was to affirm the
legitimacy of these traditions, an unbroken link between past and present. As somewhat
of a latecomer among the new schools, the Geluk tradition was required to innovate in its
representations of its teaching lineages. In particular, artists and biographers creatively
reworked the relationship of the tradition’s founder, Tsong kha pa, to his own teachers
and disciples. Not entirely unlike the trope of reincarnation (another tool for the Gelukpa
identity formation well-exemplified in the institution of the Dalai Lamas), the trope of the
vision provided a creative framework for reinventing the past in imagining the future.
Although Tsong kha pa lived in the fourteenth century, the Geluk tradition was not
immediately conceived of as an independent tradition. Over time, the views and
initiatives of Tsong kha pa came to be regarded as different from the Sakya and Kagyu
traditions of his teachers while consistent with the early teachings of the Kadampa. The
Kadampa tradition, associated with the pioneering efforts the Indian teacher Ati"a (982-
1054) in Tibet in the wake of the dark age, projected an aura of monastic reform
consistent with Tsong kha pa’s own efforts. Shared emphases upon themes such as the
graded approach to Buddhist practice, the necessity of carefully designating the role of
tantric practice in monastic life, and an overarching concern with restoring the ethical
integrity of monastic institutions allowed for natural parallels between the two masters.
This portrait of Mkhas grub and Tsong kha pa’s ritual and visionary relationship
shifts the focus from chronological transmission of the teachings from one generation to
the next to a moment out of time. Mkhas grub offers mandala as self and world to his
teacher, and in exchange, he receives unmediated and renewable access to the teachings.
The “symbolic capital” of the mandala offering, laden with homologies of the form of the
mandala with that of both the self and the cosmos, is empowered by its potential to be
repeated ad infinitum.27 The offering is a ritual action that reinforces the bond between
guru and disciple and invests Mkhas grub with the authority to act on Tsong kha pa’s
behalf.
All of theses portraits of Tsong kha pa and Mkhas grub evoke these aspects of
tantric practice, of habituating oneself in regarding guru as buddha and reaffirming one’s
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
26
The essay by Luczanits, “Siddhas, Hierarchs, and Lineages: Three Examples for
Dating Tibetan Art” appears in Jackson’s 2011 catalogue on Tibetan portraiture. Both
authors have made significant contributions to contemporary understandings of the
structure of lineage paintings as well as to their role in the evolution of Tibetan
Buddhism. Jackson 2005 provides key insights into different modes of representing
lineage. I was fortunate to attend a workshop lead by Luczanits and Jackson on the
subject at UC Berkeley in Spring 2011. !
27
On the use of the category of “symbolic capital” by Bourdieu, see for example,
Bourdieu 1990 pp. 112-121 and Bourdieu 1998 pp. 47-52.
! xx!
conviction of this attitude through repeated ritual acts such as the mandala offering.
However, the painting from the Rubin Collection [Fig.1] adds another dimension to the
nature of the authority Tsong kha pa has invested in Mkhas grub. The inscription reads:
“The Venerable Dharmaråjå Tsong kha pa offers the empowerment and vows of
Vajrabhairava to Mkhas grub dge legs dpal who clarifies interpolations from the
scripture, ‘Offering and service of the Six-armed protector.” 28 The figure of Mkhas grub
on the left has been interpreted as expressive of the student’s despair at his teacher’s
absence. Mkhas grub turns to the ritual altar seeking answers; his despair is ultimately
remedied by the appearance of the vision of Tsong kha pa. However, Mkhas grub
appears once again on the lower right, hard at work sifting through piles of ritual texts.
From the inscription we know that Mkhas grub is “clarifying interpolations” from the
tantric ritual texts for propitiating the featured deities. This figure therefore alerts us to
another kind of authority Tsong kha pa has conferred upon Mkhas grub, the power to
revise the tantric teachings to perfect their meaning. This involves rooting out spurious
incursions and restoring the “original” meaning of the texts. This particular aspect of
Mkas grub’s identity factors prominently in this dissertation; it helps us to make sense of
what’s at stake for a scholar like Mkhas grub, renowned for his expertise in philosophical
commentary and debate, in a controversy over tantric practice and a special variety of
mandala practice at that. The body mandala debate presents tensions surrounding a
practice in which the correlation of human being, cosmos, and mandala becomes ritually
explicit, bring human being and buddhahood into even closer proximity.
To better understand the qualities that distinguish the mode of representing Mhas
grub’s visions from some other conventions of lineage portraiture and to better acquaint
ourselves with the other main protagonist of this dissertation, we turn to a fifteenth-
century portrait of Ngorchen Kun dga' bzang po (1382-1456). [Fig.4] The inscription
reads, “Homage to the venerable Great Vajradhara, Kun dga’ bzang po,” and the vajra
and bell resting atop the lotuses alongside him reinforce his identity as a tantric master.29
Vajradhara is the blue Buddha pictured above Ngor chen’s head, a figure deeply
associated with the source of tantric initiation and, particularly (though not exclusively),
with the Hevajra Tantra. A common epithet for Ngor chen, “Vajradhara” expresses the
network of associations of Buddha, guru, and disciple enacted through the tantric
initiation rituals described above. Unlike the visionary portraits of Tsong kha pa and
Mkhas grub, snapshots of a moment out of time, this painting focuses upon tracing a long
chronological succession of gurus that culminate in Ngor chen. These gurus have
conferred initiations into particular forms of tantric practices passed down through select
lines of disciples together with requisite vows of secrecy and pledges to perfect these
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
28#rje btsun chos kyi rgyal po tsong kha pas/ rdo rje ‘jigs byed dbang dang gdams pa
gnang/ phyag drug dgon po’i bsnyen sgrub be bum la/ lhad zhugs gsal mdzad mkhas grub
dge legs dpal
The verso bears an inscription referring to Mkhas grub’s status as a previous incarnation
in the line of Panchen Lamas. See Jeff Watt’s description of this painting on
www.himalayanart.org. His translation of lhad zhugs gsal mdzad as “clarifying the
interpolations” conveys the spirit of the inscription nicely.!
29
See Jackson 2010 Figure 8.2 and discussion pp.179-181. For other studies of Sakyapa
lineage portraiture in particular, see Jackson 1986 and 1990.
! xxi!
practices. As a record of the unadulterated transmission of tantric teachings, this portrait
attests to Ngor chen’s status as a tantric master infused with the wisdom of buddhahood
as well to his role in preserving the integrity of the Sakya tradition.
Jackson 2010 has identified two lineages depicted in this painting. The first is the
Lam ‘bras or “Path and Fruit” tradition, one of the core signature tantric practices of the
Sakya tradition; the ritual practices of the Hevajra cycle that are Ngor chen’s main
concern in the body mandala debate texts described in this dissertation are part of the
Lam ‘bras tradition. The other lineage is for the initiation into the cult of the Goddess
Nairatmya, consort to the tantric deity Hevajra; Nairatmya is famous for transmitting the
Lam ‘bras to the great realized Indian tantric master, the mahåsiddha Virüpa, in a vision.
Both lineages begin at the top center with Vajradhara, branching out in both directions;
both also present the blue Nairatmya and the dark-skinned Virüpa next in the lineage.
The latter’s skin color is a marker of his Indian-ness, and thereby a trademark of the
authority that lies at the root of the traditions. Both lineages proceed from buddha to
divine consort on to Indian masters and futher on through a series of Tibetan masters of
the Sakya tradition.
To be a Sakya is to have been ordained in the tradition understood to begin with
‘Khon dkon mchog rgyal po (1024/34-1102), the eleventh-century founder of Sakya [Sa
skya] monastery in Gtsang. The five patriarchs of the Sakya tradition are a standard
identifying feature of Sakya lineage portraiture (appearing in rows three through five of
this painting). Sa chen kun dga’ snying po (1092-1158) is shown with his two sons, Bsod
nam rtse mo (1142-1182) and Grags pa rgyal mtshan (1147-1216). The white robes these
three figures (each depicted twice) wear indicate their status as sngags pa, or non-celibate
tantric practitioners. The next of the great Sakya patriarchs, Grags pa rgyal mtshan’s
nephew, the famous scholar monk Sa skya Pan di ta kun dga’ rgyal mtshan or Sa pan
(1182-1251), is depicted in a very similar fashion to Ngor chen, with a red hat and full
monastic garb. The fifth of the five Sakya forefathers, Sa pan’s nephew ‘Phag pa blo
gros rgyal mtshan (1235-1280), appears here as well. The pattern of spiritual inheritance
descending from uncle to nephew was a common one among the Sakya; this pattern
allowed the tradition to maintain an inviolate tradition of celibate monasticism together
with a clan-based institutional structure. Sakya Pandita and his nephew ‘Phag pa are
especially significant to the history of the Sakya. They solidified a relationship with the
Mongol Yuan (1271-1368) dynasty that became an important political prototype, the
mchod yon or “patron-priest” relationship. This alliance invested the Sakyapas with
authority over all the Buddhist institutions of Tibet until the fall of the Yuan in the mid-
fourteenth century. This dissertation focuses upon the fifteenth century, a period in
which the political patronage of Sakya institutions was somewhat less secure.
Both of the lineages depicted in this painting end with Ngor chen, whom Jackson
2010 has astutely identified as appearing twice below the main representation, as the two
central figures immediately below the throne. The inscriptions marking these figures
provide Ngor chen’s Sanskrit name, Anantabhadra, perhaps to reinforce his connection to
an Indian Buddhist legacy. In both cases, he faces the gurus from whom he received the
teachings, Buddha"rî for the Nairatmya initiation and both Buddha"rî and Ye shes rgyal
mtshan for the Lam ‘bras. The latter is the guru to whom Ngor chen’s body mandala
debate text is dedicated.
! xxii!
Jackson has shown how the identity of the final master in a lineage can be used to
date a painting; according to this logic, the painting was likely created during Ngor
chen’s own lifetime, and the patron, shown in the bottom row below Ngor chen was one
of his students.30 It is possible that this student received the Lam ‘bras and Nairatmya
initiations from Ngor chen. If so, this painting is a material representation of a bond
created between the student and Ngor chen through ritual; this bond further connects that
student to Ngor chen’s teachers, to the great Sakya forefathers, to the great Indian masters
through whom the tantric teachings came to Tibet, and to the very source of tantric
knowledge itself, the buddha Vajradhara. The inscription, together with the attributes of
vajra and bell, correlate Ngor chen and Vajradhara. Therefore, like the visionary portraits
of Mkhas grub and Tsong kha pa, this portrait of Ngor chen reinforces the connection
between guru and disciple as well as between guru and Buddha. The portraits also
reinforce the role of ritual in the preservation of tradition. Through the mandala offering,
Mkhas grub re-establishes the link to the teachings through his deceased master, a link
originally instantiated through ritual transmission. However, the elaboration of spiritual
ancestry distinguishes Ngor chen’s portrait from that of Mkhas grub and Tsong kha pa.
The latter relies instead upon the power of the ritualized visionary moment to reinvent
tradition. Moreover, the reference to “eliminating imperfections in the scripture” implies
shortcomings in the transmission of the tantric tradition among their contemporaries and
invests Mkhas grub via Tsong kha pa with the unique responsibility of restoring the
integrity of the tradition.
As we proceed to examine the network of texts composed by Mkhas grub and
Ngor chen referred to as the body mandala debate, we will investigate the subtleties of
what precisely it means to “clarify interpolations” or “eliminate imperfections” from
tantric ritual texts. What sorts of imperfections might be perceived in the interpretation of
particular versions of mandala practice and of the relationship of human body and
mandala? Whose integrity is at stake in the challenges posed to these interpretations? The
focus upon spiritual succession exemplified by Ngor chen’s portrait clearly
communicates the message that to question the transmission of a ritual practice is to gnaw
at the very thread connecting generations of accomplished masters. As a trope common
to the polemical and exegetical genres of Tibetan textuality, “eliminating imperfections”
or “clarifying interpolations” is a quintessential Tibetan way of framing innovation; it
creates space for modifying practices and institutional identities while simultaneously
claiming allegiance to tradition.
In describing the portraits of Mkhas grub, we foregrounded the symbolic capital
of the mandala offering and the web of relationships it evokes: of human being and
cosmos and of buddha, guru, and disciple. The mandala functions as a key feature of
tantric ritual practice, as a signature element in rites of initiation, a pedagogical tool, and
a framework for rekindling and enhancing ritual moments in daily practice. This range of
associations might be expanded even further to include some of the institutional
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
30
For Jackson’s approach to dating Tibetan paintings, see Jackson, David. 2003. “The
Dating of Tibetan Paintings is Perfectly Possible though not always perfectly exact.” in
Dating Tibetan Art: Essays on the Possibilities and Impossibilities of Chronology from
the Lempertz Symposium, Cologne, edited by Ingrid Kreide-Damani. Wiesbaden:
Ludwig Reichert Verlag. 91-103.
! xxiii!
dimensions of mandala. For example, the mastery and unadulterated transmission of
mandala technologies through lineages such as those depicted in Ngor chen’s portrait
bestowed prestige upon Tibetan masters and their traditions. To account for these
dimensions of mandala, we will consider another painting, one commissioned by Ngor
chen as part of a larger set of fourteen.31 [Fig.5] The series as a whole was executed by
Newar painters that had travelled to Tibet and attests to the cosmopolitan nature of
Himalayan art and ritual. Ngor chen created the set, according to biographical sources as
well as the inscription, to “fulfill the intention” [dgongs rdzogs] of his deceased master,
Sa bzang ‘phag pa gzhon nu blo gros (1358-1412/24). Therefore, in a sense, the
paintings themselves function as a mandala offering to the guru.
The paintings depict mandalas from a series Sa bzang transmitted to Ngor chen,
mandalas described in Abbhayåkåragupta’s Vajråvalî together with Darpanåcårya’s
Kriya-samuccaya.32 These are Indian tantric ritual compendia dating between the
eleventh and thirteenth centuries detailing tantric mandala initiation rites. In order to
assure the continued transmission of the Kriya-samuccaya, Sa bzang travelled to the
Kathmandu valley to receive the initiation. This painting, the fifth in the set, is one of
two in the series that fuses the genres of lineage portrait and mandala painting. The
lineage gurus appear in the center surrounded by four mandalas. Forms of Hevajra
dominate all four, appearing in the center together with a consort and surrounded by
attendant deities. The cremation grounds that typically form the periphery of the
individual mandala structure are shared by all four mandala..33 In “fulfilling the
intention” of his guru through the patronage of works of art, Ngor chen commemorates
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
31
The location of only seven or eight of these original fourteen paintings is known. In his
brief 2008 article, Mori has presented the most detailed research on this mandala set to
date. On Ngor chen’s commissions of lineage portraiture as well as of the Vajråvalî
mandala set, see Jackson 2010 pp.182-90 and Jackson 1996 pp.77-82.
32
Abhayåkaragupta’s Vajråvali-nåma- mañ∂alopayika is a cycle of three texts, the
Phreng ba ‘khor gsum. The Vajråvalî [VA] forms the nucleus. The other two texts in the
cycle are the Nißpanna-yogåvali, a sådhana manual listing the iconographic descriptions
of the deities of the same mandalas discussed in the Vajråvalî (in the very same order)
and a homa text, the Jyotirma÷jari. These two accompanying or supplementary texts to
the VA are known as parikaras. As a cycle, the texts describe complementary processes
of consecration, initiation, visualization, and offering which should be contemplated as a
holistic system. The texts appear to have been composed in the late eleventh or early
twelfth century. For studies of the Vajråvalî cycle, see Lee’s 2003 dissertation and 2004
study. See also Bhattacarya’s 1972 edition and Buhnemann and Tachikawa’s 1991
edition. I am grateful to have participated in a seminar on the Nißpannayogåvali lead by
Dr. Tachikawa at UC Berkeley in Fall 2012.
Ian Alsop is currently working on a study of the Kriya-samuccaya, a popular text among
the Vajråcårya priests of the Kathmandu valley. See Buhnemann 1992/3: “Some
Remarks on the Text of the Nispannayogavali as Found in Jagaddarpana's
Kriyasamuccaya.: Zentralasiatische Studien 23. 1992/1993. 18-21.
33
Mori 2008 has identified the central deities of these mandalas as Garbha-Hevajra,
Citta-Hevajra, Våk-Hevajra and Kåya-Hevajra from mandalas five through eight in the
Vajråvali text. !
! xxiv!
his relationship to Sa bzang as well as to a lineage of masters that preceded him. He also
asserts his ritual expertise in continuing to transmit these mandala teachings to the next
generation of disciples. Finally, he lays claim to the prestige surrounding artistic
patronage, prestige for himself, the Sakya tradition at large, and the monastery he has
founded.
Ngor chen is believed to appear at least twice within the set of paintings; these
instances reinforce the themes of relationship to the guru and the assertion of ritual
expertise and prestige. The first instance is in the eleventh painting in the set, the
Kålacakra mandala [Fig.6]; Ngor chen appears here as the sådhaka or ritual specialist in
the bottom right together with a host of offerings. Ngor chen appears once again in the
same position in the final painting in the series. together with an inscription reinforcing
his connection to Sa bzang. [Fig.7] We might continue to explore further dimensions of
Ngor chen’s career through his reception and transmission of mandala rites, his
composition of texts on mandala, and his patronage and consecration of artistic
representations of mandala. For example, Ngor chen was involved in organizing a group
of Newar artists to complete the murals for the Byams pa temple at Glo smon thang.34
[Fig.8] Ngor chen consecrated this temple, conferring his authority upon its foundation
and ensuring the continuation of his Sakya spiritual legacy in Mustang’s soil.
The examples discussed above introduce multiple aspects of the mandala’s
symbolic capital in Himalayan art and ritual. These include, but are not limited to their
role in orchestrating relationships between guru, disciple and buddha, introducing tantric
teachings and preserving their transmission, and maintaining and modifying institutional
and socio-political relationships.
If indeed to offer the mandala is to offer oneself and the world, if the center of the
mandala is the center of the Buddhist universe and, as found in some varieties of body
mandala practice, the very backbone of the human form, then the perfection of this ritual
act is the perfection at the very heart of tantric practice. We have begun to get a sense
that Mkhas grub and Ngor chen’s endeavors to “eliminate imperfections” in ritual texts or
to “fulfill the intentions” of a deceased master are projects deeply imbued with both
explicit and implicit levels of meaning. As Mkhas grub and Ngor chen assert their
different approaches to this process of perfecting self and world in their writings on body
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
34
The iconographic layout of the temples at Glo remains unresolved in scholarship to
date. Much research remains to be done on both the artistic and historical aspects of the
creation of the mandala murals. The recent preservation projects undertaken at the site
have attracted international attention. Lo Bue’s 2010 monograph is the most recent
contribution within the field; Luczanits’s 2013 review addresses some of the
controversies surrounding the restoration of the murals. Kramer 2008, Jackson 1984 and
Dhungel 2002 have laid the groundwork for historical study while Matthiesson 1996 and
http://dl.lib.brown.edu/BuddhistTempleArt/history2.html provide some photographs of
the artwork. In my next project, I intend to publish the research I conducted during field
work in Nepal in May 2012 with the support of the Fulbright IIE and the UC Berkeley
Graduate Division. I am grateful to Luigi Fieni and Samantha Ezeiza, Christian Luczanits
and Kimiaki Tanaka for their support in pursuing that research.
!
! xxv!
mandala, they create a portal into fifteenth-century Tibetan ritual, institutional, and
scholastic life. This dissertation is an invitation to explore that portal.
III. State of Research: Sources for Making Sense of the Body Mandala
Debate
An exhaustive study of biographical materials featuring both authors is beyond
the scope of the present work. However, we are prepared to acknowledge that particular
aspects of their expertise have been emphasized by certain biographers at the expense of
others. Mkhas grub and Ngor chen, too, take an active role in constructing their own
authorial identities by stressing particular dimensions of the debate and invoking different
varieties of Buddhist discourse for support. We begin by contextualizing the body
mandala texts within their authors’ respective spheres of activity through reference to
biographical materials, colophons, and secondary scholarship.
Cabezón, Dreyfus, van der Kuijp, and most recently, Ary have been influential in
enhancing understanding of Mkhas grub rje’s life and work, with particular attention to
his philosophical accomplishments.35 Lessing and Wayman’s 1968 translation of Mkhas
grub’s Fundamentals of Buddhist Tantra has made the Gelukpa scholar’s work a
fundamental part of the canon of Western scholarship on tantric Buddhism as well. Yael
Bentor continues to contribute to our knowledge of Mkhas grub’s role in shaping and
transmitting the ritual tradition of the Årya Guhyasamåja system. In her 2006 and 2015
articles, Bentor has engaged in brief but meaningful ways with the relationship of Mkas
grub and Tsong kha pa’s views on body mandala practice. Her forthcoming translation of
the entirety of Mkhas grub’s Ocean of Attainment of the Guhyasamåja Generation Stage
[Gsang 'dus bskyed rim dngos grub rgya mtsho] (henceforth referred to as Ocean of
Attainment ) of which this dissertation engages exclusively with the portion on body
mandala, promises to further illuminate the study of Mkhas grub’s tantric endeavors.36
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
35!See!Cabezón 1992, Dreyfus 1997, van der Kuijp 1985b and Ary’s 2015 publication
based upon his 2007 doctoral dissertation.

Ary, Elijah Sacvan. “Logic and lineage: Jetsun Chokyi Gyaltsen’s Ascension and the
Secret Biography of Khedrup Geleg Pelzang.” Ph.D. Diss., Harvard University, 2007.
2015. Authorized lives: biography and the early formation of Geluk identity
Boston : Wisdom Publications, 2015.
Mkhas-grub Dge-legs-dpal-bzang-po, and José Ignacio Cabezón. 1992. A dose of
emptiness: an annotated translation of the sTong thun chen mo of mKhas-grub dGe-legs-
dpal-bzang. Albany: State University of New York Press.
Dreyfus, Georges. 1997. Recognizing Reality: Dharmakîrti’s Pholosophy and Its Tibetan
Interpretations. Albany: SUNY Press, 1997.
van der Kuijp, Leonard W. J. "Studies in the Life and Thought of Mkhas grub rje I:
Mkhas grub rje's Epistemological Oeuvre and His Philological Remarks on Dignåga's
Pramåñasamuccaya," Berliner Indologische Studien 1 (1985b), 75-105. (1985b).
!
36
Bentor, Yael. 2006. “Identifying the Unnamed Opponents of Tsong kha pa and Mkhas
grub rje Concerning the Transformation of Ordinary Birth, Death and the Intermediary
State into the Three Bodies.” Tibetan Buddhist Literature and Praxis: Studies in its
! xxvi!
For many years, the Sakya tradition has received comparatively sparse attention
within the field of Tibetan studies, when considered alongside the proliferation of works
on Geluk interpretations of Madhyamaka philosophy and works on Nyingma esotericism.
In recent years, the Sakyapas have been receiving more scholarly attention. Tibetology
has, for example, been enriched by scholars broadening the view of Tibetan approaches
to the Madhyamaka philosophical tradition beyond the Gelukpa. Cabezón and, most
recently, Kassor have illuminated important Sakyapa contributions to the study of
Madhyamaka with their in-depth analyses of one of Ngor chen’s most prolific successors,
Go rams pa bsod nams seng ge (1429-89).37
The growing interest in ritual within religious studies and Tibetan studies may
have attracted attention to the rich Sakya tantric ritual and exegetical tradition, in
particular the elaboration upon practices based in the Hevajra Tantra such as the Lam
‘bras or “Path and Fruit” tradition. This tradition is lauded as transmitted from divinely-
inspired Indian masters to Tibetan disciples by which the perfected meaning of the
Hevajra Tantra is brought to life through the teachings of the tantric guru. Cyrus Stearns
has made some of the most significant contributions to the cataloguing and translation of
Sakya literature together with the documentation of the history of its transmission. Most
recently, Stearns has translated key texts of the Lam ‘bras tradition that provide insight
into the mechanics and interpretation of the Hevajra body mandala practice.38
Sobisch 2007 and 2008 has furthered the research of the Sakyapa tantric
exegetical tradition.39 Much of the work presented in Chapter Six of this dissertation is
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Formative Period, 900-1400. Ed. Ronald M. Davidson and Christian Wedemeyer.
Leiden: Brill: 185-200.
2014. “The Body in Buddhist Tantric Meditations” conference paper presented at
“The Evolution of Tantric Ritual” held at UC Berkeley in March 2014.
2015."Interpreting the Body Mandala: Tsongkhapa versus Later Geluk Scholars"
Revue d’Etudes Tibétaines. no.31, Feb 2015 pp 63-74.
!
37
See:!Cabezón, José Ignacio and Geshe Lobsang Dargyay. 2007. Freedom from
Extremes: Gorampa’s ‘Distinguishing the Views’ and the Polemics of Emptiness.
Boston: Wisdom Publications.
Kassor, Constance. “Thinking the Unthinkable / Unthinking the Thinkable: Conceptual
thought, nonconceptuality, and Gorampa Sonam Senge's Synopsis of Madhyamaka.”
Ph.D. Diss., Emory University, 2014.
!
38
Stearns, Cyrus. Luminous Lives: The Story of the Masters of the Lam’Bras Tradition
in Tibet. Boston: Wisdom Publications, 2001.
-------------------,Trans. Taking the Result as the Path: Core Teachings of the Sakya
Lamdré Tradition. Boston: Wisdom Publications, 2006.
39!Sobisch, Jan-Ulrich, and A-mes-zhabs. 2007. Life, transmissions, and works of A-mes-
zhabs Ngag-dbang-kun-dga'-bsod-nams, the great 17th century Sa-skya-pa bibliophile.
Stuttgart: F. Steiner.
Sobisch, Jan-Ulrich. 2008. Hevajra and Lam 'bras literature of India and Tibet as seen
through the eyes of A-mes-zhabs. Wiesbaden: Reichert.
! xxvii!
based upon his study and classification of the Lam ‘bras literature. In addition to
contributing to the study of the early Sakya philosphical tradition, David Jackson has
opened up the field of Sakya art history, with his detailed studies of systems for depicting
lineage masters within paintings associated with the Sakya tradition.40 More recently, in
2010, in an exhibition catalogue for the Rubin Museum of Art, Jackson has traced the
Newar style of Tibetan painting, devoting significant attention to commissions executed
at the behest of Ngor chen and his successors.41 Ngor chen is among the subjects
featured in these paintings; in some cases, he is even the patron or ritual consecrator.
These artworks provide an alternative resource for approaching modes of representing
Ngor chen as a transmitter of the Sakya tradition. They reinforce his connection to gurus
of the past as well as to later disciples and foreground his role as a tantric specialist as
well as a vinaya master.
Heimbel 2011 has published a preliminary overview of biographical materials on
Ngor chen. Like Ary, he has made valuable methodological suggestions for
contextualizing and interpreting biographical sources.42 Heimbel’s forthcoming
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
40
See: Jackson, David P. 1986 “A Painting of Sa skya pa Masers from an Old Ngor pa
Series of Lam ‘bras Thangkas,” in Berliner Indologische Studien, (2), pp.181-91.
1989. The Early Abbots of ‘Phan-po Ne-len-dra: The Vicissitudes of a Great
Tibetan Monastery in the 15th Century. Wien: Arbeitskreis für Tibetische und
Buddhistische Studien, Universität Wien.
1986 “The Identification of Individual Teachers in Paintings of Sa skya pa
Lineages,” in T. Skorupski, ed., Indo-Tibetan Studies, Spring, pp. 129-44.
A History of Tibetan Painting: The Great Tibetan Painters and Their
Traditions. Wien: Verlag der Osterreichischen Akademie der Wissenschaften,
1996. Chapter 2.
2003. “The Dating of Tibetan Paintings is Perfectly Possible – though not always
perfectly exact.” in Dating Tibetan Art: Essays on the Possibilities and Impossibilities of
Chronology from the Lempertz Symposium, Cologne, edited by Ingrid Kreide-Damani.
Wiesbaden: Ludwig Reichert Verlag. 91-103.
2005. “Lineages and Structure in Tibetan Buddhist Painting: Principles and
Practice of an Ancient Sacred Choreography. “ Journal of the International
Association of Tibetan Studies 1: 1-40.
2010.The Nepalese Legacy in Tibetan Painting. Seattle: University of
Washington Press.
41
Jackson 2010, especially Chapter Eight: “Ngor and Its Beri Paintings,” pp.176-215.
42
Heimbel proposes that “an important approach to biography-based research is the
investigation of the background of the biography’s author, the author’s relation to his
biographical subject, the sources he or she employed, and his or her methods of
composing and editing.” See Heimbel 2011, p. 80. He uses Sangs rgyas phun tshog’s
(1649-1705) biography of Ngor chen (upon which this dissertation relies as an important
source on Ngor chen’s life and attitudes towards it) as a rich example for demonstrating
the efficacy of this method. He also examines a biography composed by Ngor chen’s
student, Mus chen sems dpa’ chenpo dkon mchog rgyal mtshan (1388-1469). Heimbel
! xxviii!
dissertation on Ngor chen promises to bring further depth to the study of the Sakya
tradition more broadly together with Ngor chen’s unique contributions to it. Studies of
later masters of the Sakya tradition such as Kramer’s 2008 monograph on A ma dpal
(1456-1532) have also provided useful clues to Ngor chen’s role within the Sakya
tradition as it was transmitted and evolved over time.
Davidson’s studies of the ritual and institutional dimensions of the Ngor pa
tradition have been crucial to this project from its inception. His 1981 and 1991 articles
provide background on Ngor chen’s life and activities as well as the intellectual climate
in which the Sakyapas participated. In assessing Ngor chen’s overall contribution to the
Sakya tradition, Davidson emphasizes his institutional role, his “willingness to accept a
modification of monastic policy, based on observation and brought about through careful
planning.”43 His engagement with sources relevant to the body mandala debate in his
1992 essay on the Hevajra abhisamaya tradition has laid the groundwork for the present
study.
Davidson’s 1981 article appears to contain the first, albeit brief, reference to the
body mandala debate in Western scholarship.44 In 1985, Van der Kuijp, was the next to
refer to Ngor chen’s polemical writings on body mandala.45 Van der Kuijp describes the
two works by Ngor chen to be evaluated in the present study, Destroyer of the Proponents
of Evil by Eliminating Objections to the Hevajra Body Mandala [Kye'i rdo rje'i lus kyi
dkyil 'khor la rtsod spong smra ba ngan 'joms] (henceforth referenced as Destroyer of
the Proponents of Evil or N1) and Dispelling Evil Views by Eliminating Objections
to the Hevajra Body Mandala [Kye rdo rje'i lus kyi dkyil 'khor la rtsod spong lta ba ngan
sel] (henceforth referenced as Dispelling Evil Views or N2), as follows:
“Written in the first half of 1426, these are two prints of the same text, with some
interesting variant readings. It is a polemical work dealing with the mandala of Hevajra,
conceived as a reply to and criticism of Mkhas-grub-rje’s aside on the same in his Gsang
'dus bskyed rim dngos grub rgya mtsho.”46
Van der Kuijp also mentions Mkhas grub’s reply to Ngor chen (not addressed by
this dissertation) as well as other relevant polemical texts by Ngor chen and his
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
astutely observes Sangs rgyas phun tshog’s skillful incorporation of Mus chen’s
biography of Ngor chen into his own text. See p.69 of this article for an example of
Sangs rgyas phun tshog’s method.
Heimbel, Jorg. 2011. “Biographical Sources for Researching the Life of Ngor chen Kun
dga’ bzang po (1382-1456).” Revue d’Etudes Tibétaines 22, pp. 47-91.
43
Davidson, R.M. “The Ngor-pa Tradition.” Wind Horse. Vol.1, 1981, pp.79-98. See
p.93.
44
Davidson 1981, p.88.
45!van der Kuijp, Leonard W. J."A Text- Historical Note on Hevajra Tantra II: v:1-2."
Journal of the International Association for Buddhist Studies, Vol. 8, No. 1 1985, pp. 83-
89. (1985a).
!
46
Van der Kuijp 1985a, p.88.
! xxix!
disciples.47 Standing upon the shoulders of these trailblazers, we proceed to acquaint
ourselves with the authors and the evolving climate of creation and interpretation of the
body mandala texts.
IV. Challenges and Support

The greatest technical challenge posed by the body mandala debate, beyond the highly
esoteric nature of the materials, is the vast mastery the texts assume of three tantric ritual
cycles: Guhyasamåja, Cakrasaµvara and Hevajra. Wayman, Gray, and Snellgrove have
laid the groundwork for the study of these three root texts respectively.48
Kilty’s 2013 translation of Tsong kha pa’s Lamp to Illuminate the Five Stages
Teachings on the Guhyasamåja Tantra [Rgyud kyi rgal po dpal gsang ba ‘dus pa’i man
ngag rim pa lnga rab tu gsal ba’i sgron me ] and Geshe Lobsang Tsephel’s translation
and commentary on eighteenth-century Gelukpa Yangchen Galo’s text, Paths and
Grounds of Guhyasamåja According to Arya Någårjuna, have been excellent resources
for the study of the Guhysamåja. Kittay’s 2011 translation and study of the Vajramålå,
an explanatory tantra of the Årya Guhyasamåja tradition (a tradition claiming Någårjuna
as its progenitor) has been an important source for interpreting citations from that text
within the body mandala debate. Likewise, Wright’s 2010 translation of Någårjuna’s
Piñ∂ik®ta-sådhana, a ritual text designed to develop the generation stage of the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
47
For Mkhas grub’s reply, see Mkhas grub rje dge legs dpal bzang. Ngor lan gnam lcags
'khor lo. Dgag lan phyogs bsgrigs . Ch'eng tu: Si khron Mi rigs Dpe skrun khan, 1997.
Chinese colophon title: Pien lun wen hsuan pien. pp. 1-68. Van der Kuijp classifies Ngor
chen’s Rgyud gsum gnod ‘joms as explicitly polemical and his dPal kye rdo rje’i sgrub
thabs kyi rgya cher b"ad pa bskyed rim gnad kyi zla zer (mentioned below) as “frequently
polemical.” Both of these texts were also cited in Davidson 1981 and will be discussed
briefly below.
!
48
Wayman, Alex. 2005 (Reprint of 1977 publication). Yoga of the Guhyasamåja Tantra:
the arcane lore of forty verses : a Buddhist tantric commentary. Delhi: Motilal
Banarsidass.
Gray, David B., and Thomas F. Yarnall. 2007. The Cakrasaµvara Tantra: the discourse
of Sri Heruka (Sri heruka abhidhana). New York: American Institute of Buddhist Studies
at Columbia University.
Snellgrove, David L. 1959 (2010 Reprint). Hevajra Tantra: a critical study. Hong Kong:
Orchid.
Fremantle’s 1971 Dissertation on the Guhyasamåja also provides some useful orientation.
Gray also also attempts to situate body mandala practice within the larger context of
religious studies concerns with selfhood. See David B. Gray, “Mandala of the Self:
Embodiment, Practice, and Identity Construction in the Cakrasamvara Tradition,” The
Journal of Religious History Vol. 30, No. 3 (October 2006): 294-310. I am grateful to
David Gray and Yael Bentor for bringing my attention to this article as well as for
presenting their own research on body mandala at the “Evolution of Tantra” conference
at UC Berkeley in March 2014.
! xxx!
Guhyasamåja-based practice, has been instructive. I am grateful to Jacob Dalton for his
guidance in reading the Guhyasamåja sådhanas as well as for the opportunity to attend a
workshop on this tantra held in March 2014 at UC Berkeley. Khenpo Choying Dorje of
Dzongsar Monastery as well as Khenpo Yeshe provided essential feedback in the
translation and interpretation of Mkhas grub’s body mandala text.
Sugiki’s work on the diversity of approaches to the Cakrasaµvara Tantra
exhibited within Indian sådhanas has been an excellent resource for engaging with the
complexities of that tantric cycle.49 English’s 2002 work on Vajrayoginî has been
essential to understanding the body mandala within the Cakrasaµvara ritual cycle.50
English’s philological and methodological approaches have also served as guides in
unpacking the many layers and modifications of body mandala as a ritual practice.
As mentioned above, with the endorsement of His Holiness Sakya Trizin and the
guidance of the deceased eminent Chogay Trichen Rinpoche, Cyrus Stearns has
pioneered the study of the Sakya “Path and Fruit” tradition elaborating upon the Hevajra
Tantra. I am grateful to have also participated in a portion of a workshop on the Hevajra
Tantra with Harunaga Isaacson in February 2013 at UC Berkeley. Finally, the expertise
of Drapa Gyatso, my teacher at the Sakya International Buddhist Academy in Kathmandu
has deeply informed my interpretation of citations from the Hevajra ritual manuals within
Ngor chen’s body mandala text. Szanto’s and Elder’s preliminary studies of the Sampu†a
Explanatory Tantra, a tantra shared by both the Cakrasaµvara and Hevajra cycles, have
further enhanced the approach to the Hevajra tradition.
V. Chapter Overview
Whether regarded as a tool or as an obstacle, the human body is the ultimate framework
situating us in time and space, the backdrop against which our stories are woven and the
altar upon which our rituals are enacted. It may be the ground for contestation of socio-
political tensions as well as a powerful agent for change. Broadly, this dissertation is
motivated by questions surrounding the paradoxical status of the human body as
limitation and potentiality as reflected in visual and textual representations of Himalayan
ritual life.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
49!Sugiki, Tsunehiko, and 2003. Five Types of Internal Mandala Described in the
Cakrasamvara Buddhist Literature: Somatic Representations of One's Innate Sacredness.

144, 2003-12, P. 276-202.
2009. “The Structure and Traditions of the Systems of Holy Sites in the Buddhist
Saµvara Cycle and its Related Scriptural Cycles in Early Medieval South Asia: The
Geography of Esoteric Buddhism in the Eyes of the Compilers of the Scriptures,”
Genesis and Development of Tantrism. edited by Shingo Einoo. Tokyo: Institute of
Oriental Culture, University of Tokyo, Institute of Oriental Culture Special Series, 23:
515-562.!
Review of The Cakrasaµvara Tantra (The Discourse of ¸rî Heruka): A Study and
Annotated Translation, by David Gray, Journal of the International Association of
Buddhist Studies, Vol. 31 No.1-2 (2010): 505-542.
!
50!English, Elizabeth. 2002. Vajrayoginî: Her Visualizations, Rituals, and Forms.
Boston: Wisdom Publications.
! xxxi!
Within tantric Buddhism, the body mandala is a ritual process of imagining the
human body as a mandala, a cosmic palace inhabited by Buddhas and attendant deities.
This dissertation examines a network of texts by two Tibetan scholar-monks, Mkhas grub
rje (1385-1438), a champion of the Gelukpa tradition, and Ngorchen Kun dga' bzang po
(1382-1456), a hero of the Sakyapa legacy, concerning body mandala. In the process, it
brings to light complex attitudes towards the role of the body in tantric practice. It also
contextualizes esoteric conceptions of the body in terms of larger social, religious, and
political dynamics circulating in fifteenth-century Tibet.
Why is the body mandala debate important, and how much does it actually have
to do with the body? In the process of interpreting the body mandala debate, the body is
revealed as site for experimenting with the boundaries of tantric exegesis. Through the
technical details of the mechanics of visualization and commentarial method, authorial
and institutional identities are concretized, authenticated, and reinvented. The human
body provides the arena for this debate.
Focusing upon themes of ritual, embodiment, and representation, the introduction

suggests a method for engaging with the body mandala debate within Buddhist, tantric
Buddhist, and Western theoretical frameworks. It considers how the body mandala debate
texts may be interpreted alongside other corporeal representations to contribute to the
study of the history of the body across geographic and disciplinary boundaries. This goal
requires an articulation of theories of representation as defined in terms of dynamics of
replication, imagination, performance, and re-creation.
The next section introduces the two main proponents in the debate, Mkhas grub
rje (1385-1438) and Ngorchen Kun dga' bzang po (1382-1456) through a comparison of
lineage paintings. This comparison provides an opportunity to examine these authors’
associations with two different traditions of Tibetan Buddhism, the Geluk and Sakya,
respectively, in a preliminary way. This section also elaborates upon some of the key
features of Tibetan tantric Buddhism such as the centrality of lineage, the guru-disciple
relationship, and ritual consecration. Most importantly, the introduction prepares us to
work with body mandala by demonstrating the symbolic capital of the category of
mandala as evident in Tibetan ritual, institutional, and socio-political life. It concludes
with literary review of important translations and studies upon which the dissertation
builds and to which it responds.
Chapter One situates the body mandala debate within the historical context and
intellectual climate of fifteenth-century Tibet. This overview includes an account of
biographical sources on Mkhas grub and Ngor chen’s lives and activities that inform the
interpretation of their writings on body mandala. In addition, it includes an introduction
to the scholastic and ritual contexts in which these two scholar monks partake.
Chapter Two explores the contours of body mandala practice through comparison
with other Buddhist ritual techniques for knowing the body through acts of mental
deconstruction and recollection. The comparison is extended to non-Buddhist tantric
practices for purifying and deifying the body; these practices connect visualized forms to
the human body through the manipulation of breath and the recitation of mantra. This
comparison establishes a shared ritual discourse with body mandala, a means by which
human bodies are transformed into divine bodies in accordance with the logic of tantra: to
worship a god, one must, in a sense, become a god. The comparison also demonstrates
! xxxii!
different approaches to organizing the body and establishing bodily hierarchies.
Furthermore, through analyzing a ‘proto-body mandala’ text recovered from the
Dunhuang caves, a connection is suggested between the spirit of experimentation found
in this early ritual manual and that found in other (non-body) mandala representations. In
considering inscribing the body and inscribing the text as parallel technologies, the
chapter highights a parallel between corporeality and textuality informing this
dissertation.
Inscribing and mapping the body through ritual practice are themes elaborated in
Chapter Three. This chapter considers the role of acts of mapping the body as mandala
within the ritual system of the Årya Guhyasamåja tradition. It lays the groundwork for
approaching Mkhas grub’s chapter on body mandala from the Ocean of Attainment
[Rgyud thams cad kyi rgyal po dpal Gsang ba ‘dus pa’i bskyed rim dngos grub rgya
mtsho ] with a basic outline for the Årya Guhyasamåja body mandala. This outline is
based upon Geshe Lobsang Tsephel’s contemporary commentary upon an eighteenth-
century text by Yangchen Galo. Building upon this foundation, two specific issues from
Mkhas grub’s chapter are addressed: mapping the five Buddha families and mapping four
associated goddesses onto the body. Conflicts between varying modes of mapping
deities onto the body suggest compelling implications for the continued spirit of
experimentation beyond early mandala manuals and visual representations into the
fifteenth-century scholastic context. In investigating Mkhas grub’s skillful negotiation of
competing versions of the practice suggested by different authoritative texts of the Årya
tradition, this chapter introduces the connection between tantric exegesis and polemics
framing the body mandala debate.
Chapter Four focuses upon Mkhas grub’s use of the notion of “fabrication” [Tib.
bcos ma Skt. k®trima] in establishing the superiority of body mandala over other mandala
practices. The term bcos ma itself bears connotations of artifice and a substantial lack in
authenticity. Within Buddhist philosophical discourse, “fabrication” has a decidedly
negative valence, associated with the mind’s problematic tendency to superimpose false
structures upon reality; these superimpositions are understood to obstruct our ability to
perceive things as they truly are. In bringing both ritual and philosophical perspectives to
bear upon tantric acts of imagination, Mkhas grub interrogates the relationship of the
soteriological approaches of the mantranaya and påramitånaya (roughly defined,
respectively, as the tantric method emplying mantra and the method of cultivating the
perfections [påramitå] which characterizes mainstream Mahåyåna practice outside the
tantric fold). The relationship of enlightened and unenlightened bodies, the logic of
causality, and questions of “valid cognition” [Tib. tshad ma Skt. pramåña] are among the
topics Mkhas grub evaluates in his refutation of the views of his “unnamed opponents.”
His engagement with valid cognition, in particular, in the context of mandala ritual raises
problems of relating representation to reality that are revisited in the conclusion of the
dissertation.
The second part of the chapter continues the investigation of “fabrication” [bcos
ma] through Mkhas grub’s own views on two aspects of body mandala practice: the
mandalas of “support” [rten] and “supported” [brten pa]. The mandala of the support is
the body as the celestial palace of the mandala. The mandala of the supported is the
collection of deities inhabiting that bodily palace, and in some instances, the psycho-
physical elements of the subtle body. In examining Mkhas grub’s creative engagement
! xxxiii!
with different varieties of Buddhist discourse in the body mandala debate, this chapter
adds dimension to his literary persona. It also suggests that tantric ritual uses of
imagination and of the body provided a context for challenging the relationship between
competing approaches to Buddhist theory and practice current in fifteenth-century Tibet.
In Chapter Five, the explicit encounter of Mkhas grub and Ngor chen in Mkhas
grub’s chapter from the Ocean of Attainment and Ngor chen’s reply, Destroyer of
the Proponents of Evil [ Kye'i rdo rje'i lus kyi dkyil 'khor la rtsod spong smra ba ngan
'joms] takes center stage. It is organized according to three main topics introduced by
Mkhas grub together with Ngor chen’s responses to them: the mandala of the support, the
mandala of the supported, and the generation of seed syllables on the body. Building
upon the observations on Mkhas grub’s method and style made in the previous chapter,
Chapter Five further elaborates upon Mkhas grub’s authorial identity as expressed in his
polemics and develops a portrait of Ngor chen based upon his responses. Particular
attention is devoted to Ngor chen’s use of Sampu†a Tantra, a explanatory tantra often
applied to interpreting both the Cakrasaµvara and Hevajra tantras. Its application within
the body mandala debate suggests that tantric polemics may have played an instrumental
role in setting the bounds of exegetical method. The chapter concludes with an analysis
of Ngor chen’s views on the soteriological role of the body as grounded within a broader
Sakyapa perspective on tantric practice. In the process, corporeality and textuality are
revealed to be parallel sites within the meaning-making process.
Chapter Six focuses upon the large portion of Ngor chen’s text specifically
devoted to defending against charges that a version of body mandala resembling the
Hevajra transmission lacks a basis in Indian sources. Here, Ngor chen reformulates the
debate in his own terms, exercising his prowess as a tantric commentator to defend the
Hevajra body mandala practice, and by extension, the Sakya tradition with which it is
intimately associated. The chapter explores the two aspects of his defense. The first is
based in the Hevajra commentarial tradition as composed of the three core Hevajra
tantras (the Hevajra root, Sampu†a and Vajrapa•jara), the Indian commentaries, and the
oral instructions. The second deals with the Vajramålå, an explanatory tantra of the Årya
Guhyasamåja tradition cited extensively by Mkhas grub and discussed in that context in
Chapter Four. Finally, the chapter compares two versions of Ngor chen’s text,
Destroyer of the Proponents of Evil and Dispelling Evil Views [Kye rdo rje'i
lus kyi dkyil 'khor la rtsod spong lta ba ngan sel] , examining key differences in citation
strategies, polemical tone, and syncretic emphases. This comparison enriches the portrait
of Ngor chen and highlights the ways in which his identity as a tantric commentator is
harmonized with his polemics.
The conclusion of the dissertation produces a dialogue between textual and visual
representations of the body through exploring the modes of representation in an
anomalous nineteenth or twentieth-century body mandala painting from Nepal. It calls
into question assumptions about the relationship of embodiment to materiality as well as
about the use of the body, visualized images, and their material representations as tools
for liberation. In doing so, it brings to light significant aspects of Tibetan Buddhist ritual
and philosophical understandings of material creation that may enrich art historical
approaches. Finally, it suggests questions raised by the body mandala debate regarding
the relationship of body, image, and matter relevant to the study of religion and ritual and
bodily discourses more broadly.
! xxxiv!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
Acknowledgements
To my teachers at UC Berkeley: To my advisor, Alexander von Rospatt, for giving me

this wonderful and challenging opportunity and for always encouraging me to pursue my
intellectual passions. To Patricia Berger, for all of your faith in me and in my abilities to
work with images in a meaningful way. To Jake Dalton, for your generous feedback in
developing this project. I have learned so much from all of you.
! xxxv!
To Christian Luczanits, for encouraging my love of mandala and listening
patiently to my ideas. To Charles Hallisey who first worked with me in developing my
theories of representation. To all of my teachers of Tibetan and Sanskrit language, who
provide me the tools to know the world in a new way.
To Kurt Keutzer, who first introduced me to body mandala. To Lama Kunga
Thartse Rinpoche, Dong Sung Shabdrung Rinpoche , Garshab Rinpoche, Khenpo
Choying Dorje, and Khenpo Yeshe. Without their expertise, this project would not have
been possible. With particular thanks to Khenpo Dr. Ngawang Jorden, Khenpo Tashi
Dorje, Drapa Gyatso of the International Buddhist Academy in Nepal. I am forever
indebted to you.
I am grateful for the support I received from the Group in Buddhist Studies, the
Graduate Division, the Townsend Center for the Humanities, and the Institute of
International Studies at UC Berkeley as well as from the Fulbright IIE and the Title VI
FLAS Program.
Many thanks to my cohort at UC Berkeley and in particular to Meghan Howard,
Lauren Bausch, Trent Walker, and Janet Um, and Sung Ha Yun for supporting me
through this final phase of my time here. To Nancy Lin, thank you for the moments you
lead the way and those you walked by my side. To Emily McRae, the universe has
blessed me with your friendship; your open heart and mind never cease to astound me. To
Dan McNamara, who first embarked into the Tibetan world with me and encouraged me
all along the way.
To Luigi Fieni, Samantha Ezeiza, Christian Luczanits and Kimiaki Tanaka for
making my time among the mandala murals in Mustang the experience of a lifetime.
Finally, I want to express my utter gratitude to my parents for always encouraging
my love of adventure (despite the fact it was hard to let me go) and of learning and to all
my family for their support during this long journey. I only wish that those who have
passed were here to rejoice with us in this moment of completion, especially my dear
Aunt Ann and Uncle Nino and my Aunt Gloria. Many thanks for this auspicious birth.
! xxxvi!
Chapter One: Setting the Stage for the Body Mandala Debate: Polemics,
Apologetics, and Expertise in the Lives and Times of Mkhas Grub rje and
Ngor chen kun dga’ bzang po
The following chapter provides the necessary foundation to prepare us to engage

with the body mandala debate texts. It begins by reflecting upon the complexities of
framing this network of texts as a “debate.” Next, it presents a context for situating the
debate within Mkhas grub and Ngor chen’s respective careers and thereby furnishes a
more intimate portrait of their lives and accomplishments. It also introduces some critical
tools for evaluating representations of Mkhas grub and Ngor chen and how the body
mandala debate may be intertwined with those representations. Finally, it suggests some
broader institutional and socio-political dynamics at play in fifteenth-century Tibetan
monastic life that may be useful in interpreting the significance of the body mandala
debate.
I. The Virtues and Limitations of ‘Debate’ as a Tool for Interpretation

The “Samye Debate,” the alleged historical encounter between Kamala!ila and Mohoyen,
is a prime example of the iconic status that a debate may hold in constructing a tradition.1
That debate functions as the origin story of sorts for Tibetan scholasticism; it is Tibetan
Buddhism’s testament to its transmission of the Indian “gradualist” vs. the Chinese
“subitist” approach to enlightenment. However, the actual contents of the debate and its
historical veracity remain unclear. Therefore, the category of “debate” in Tibetan
Buddhism may be understood as charged not merely with doctrinal significance but also
with the socio-political aspects of the invention, preservation, revision and revival of
tradition. Although Mkhas grub and Ngor chen’s encounter in the “body mandala
debate” is textual rather than performative, there is a sense in which they are performing
their traditions. There are also ways in which their performances were received and
reinvented that may have exceeded their intentions. The Tibetan term for polemics, dgag
lan, translated as “answers to refutations” expresses dialectical and performative aspects
of the spirit of intellectual engagement engendered by texts like those of the body
mandala debate.2
This dissertation focuses only upon what we might call “round one” of the body
mandala debate, identified as the charges leveled by Mkhas grub in the chapter of body
mandala from his Ocean of Attainment and the defense they provoked in Ngor chen’s
Destroyer of the Proponents of Evil and another “version” of the same text,
Dispelling Evil Views. The relationship between these two “versions” authored by
Ngor chen will be addressed in Chapter Six. In them, Ngor chen strategically engages
with texts and oral teachings from centuries past; these issues have, moreover, continued
to be taken up in debates for generations. For example, a generation later, Ngor chen’s
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
1
The imagined debate between Manju!ri and Vimalakîrti based on the Vimalakîrti-
nirde!a-sütra assumes a similar iconic status as attested by its popularity in Chinese art.
See Berger 2003’s introduction and Dunhuang cave 103 for some examples.
2
Lopez 1996 provides this translation of the term in his study of the genre of polemical
literature through the Gelukpa scholar Se ra rje tsun pa’s replies to his Sakyapa and
Kagyupa contemporaries.
! 1!
student Go rams pa (1429-1489), was still defending, clarifying, and elaborating upon
Ngor chen’s views on visualization practice within the Hevajra mandala system.3
Although the dissertation acknowledges the reverberations of the issues and
dynamics initiated in these texts in other texts by these same authors, their disciples, and
later members of their traditions, it focuses on just this limited range. The purpose is to
lay the foundation for scholarship of other body mandala debate texts like the chapter
from Mkas grub’s Ocean of Attainment and Ngor chen’s Destroyer of the
Proponents of Evil and Dispelling Evil Views . It does so by familiarizing the
reader with the three tantric cycles (Guhyasamåja, Cakrasaµvara, and Hevajra) and the
different ritual approaches to body mandala they provide. It also builds upon previous
scholarship on Tibetan scholasticism to contextualize Ngor chen and Mkhas grub’s
writings and to suggest that there is more than just bodies or mandala at stake in their
exchange. Moreover, it investigates their individual styles of argumentation and reflects
more broadly upon how they connect Tibetan polemics and tantric exegesis. Even to
unpack all the details of just “round one” of the debate while satisfying these goals is
beyond the boundaries of this dissertation.
Moreover, even within our limited range of texts, there are potentially many more
interlocutors or “unnamed opponents” than Mkhas grub, Ngor chen, and their immediate
affiliates. Bentor 2006 has provided the most directly relevant insight in this regard,
helping us to better understand both Mkhas grub and Tsong kha pa’ s writings on the
ritual of the Guhyasamåja tradition. Davidson 1992 has cautioned:
“Modern Tibetan religious folklore often reifies all Saskya-pa critics into dGe-lugs-pa
monks, and in the case of Ngor-chen, into Mkhas grub dge-legs dpal-bzang-po (1385-
1438)...However, the circumstances were more complex than reification into a single
protagonist...Moreover, Tibetan proclivity towards oral exaggeration certainly
exacerbated the problem, some members of the clergy assuming that the refutation of a
facet of a practice indicates a wholesale condemnation of the tradition.”4
Bearing these observations in mind, it is important that we adopt a self-conscious
attitude in referring to this network of texts as indicative of a ‘debate’ and remain open to
the possibility of multiple simultaneous interlocutors. Scholars of religion such as JZ
Smith have given us good reason to question the tendency to reduce the import of such
debates to conflicts in philosophical perspectives defined by sectarianism.5 Cabezón 2007
wisely observes a difference between “sectarian differentiation” and “sectarianism.” He
distinguishes the two phenomena as follows:
“The former is simply an inevitable historical development that arises out of human
beings’ desire to create and nurture social and institutional structures of belonging-
intellectual and spiritual homes, places where we share common goals and a common
language-in a word, traditions. Sectarianism, by contrast, is a pathological outgrowth of
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
3
!See!Davidson!1981,!p88!&!fn25.!!Go!rams!pa’s!text,!Illuminating*the*Pith:*Dispelling*
Objections*to*the*Moonrays*of*the*Pith![Gnad gyi zla zer la rtsod pa spong ba gnad kyi
gsal byed], is a!response!to!critiques!of!Ngor!chen’s!positions!in!the!Moonrays of the
Pith [!dPal*kye*rdo*rje’i*sgrub*thabs*kyi*rgya*cher*bshad*pa*bskyed*rim*gnad*kyi*zla*
zer].!
4
Davidson 1992, 20. See the author’s note 58.
5
For one influential example, see J.Z. Smith 1998.
! 2!
sectarian differentiation wherein traditions become static and reified, and wherein
dogmatism prevails.” 6
We must, therefore, also take into consideration how potentially reductionist
views of the subjects and objects of these controversies formulated by later scholars,
within both Tibetan scholastic and Western academic frameworks, may have been
distorted our current understanding of “sectarian differentiation” in fifteenth-century
Tibet.
In reducing the body mandala debate to a sectarian conflict between two opposing
factions, we run the risk of missing some of the more exciting dynamics it suggests for
describing the fifteenth-century monastic context. These may include dynamics of
patronage, prestige, and identity construction as well as previously neglected dimensions
of exegetical and polemical practice. At the same time, we must also consider the
possibility that tradition formation is one, if not the only, factor motivating Mkhas grub
and Ngor chen. So while their engagement is not just about Gelukpas vs. Sakyapas, it
may be understood as contributing to the terms that have even suggested that
interpretation. Finally, the multiple dimensions of debate itself must be considered.
Debate functions as a pedagogical tool for clarifying ideas and attaining convictions
through challenging assumptions.7 This aspect of debate must be evaluated alongside
its more antagonistic associations to determine its utility as a framework for
understanding the body mandala texts.
II. Mkhas grubs rje’s Contributions to the ‘Body Mandala Debate’

The colophon for Mkhas grub’s Ocean of Attainment does not provide a precise date
for its composition. The only clue it includes for the circumstances of its creation is that
it was written “at the Great Temple of Dpal ‘khor sde chen, the place which is the source
of reasoning (in) upper Nyang [nyang stod], in Gtsang.”8 Mkhas grub was involved in
the foundation of several monasteries such as Nyang stod lcangs ra monastery in Gtsang
[resulting in the epithet lCang ra bKa’ bcu pa, “lCang ra Master of Ten Treatises”],
Mdangs chen , and Dpal ‘khor sde chen in Gyantse [Rgyal rtse]. 9 His patron for the latter
two projects was local ruler, Rab btan kun bzang (1389-1442).10 Wylie 1980 has
identified Mkhas grub’s involvement at Dpal ‘khor sde chen as indicative of a larger
pattern of patronage in the early to mid-fifteenth century. This pattern is characterized by
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
6
Cabezón 2007, p.7.
7
As I was reminded by Phachok Rinpoche, current head of the Taklung Kagyu order of
Tibetan Buddhism, the term ‘debate’ implies that there is a residual absence of
understanding on the part of the participants. ‘If there were clarity, there would be no
need for debate.’ [Personal communication, 8/31/2011]
8
Mkhas grub rje 380.6: dge legs dpal bzang pos; gtsang nyang stod rigs pa'i 'byung gnas
kyi sa'i cha; chos grwa chen po legs bshad sgrogs pa'i tshul; dpal 'khor sde chen gyi gtsug
la 'khang chen por lag par sbyar ba'i yi ge pa ni
9
Cabezón 1992, p.15. fn36.
10
For more on Rab btan, see Ary 2007 fn 267 and I1 21 2 2F1 1 BA 1
Ahmad 1995, p.149 & Wylie 1980 p.484 as referenced there.
! 3!
patronage of the emerging Gelukpas by “local rulers appointed by the Fifth Lha-tsun of
Phag-mo-gru, Gong-ma Grags-pa rgyal-mtshan, who was known for his patronage of
Tsong kha pa and his disciples.”11 None of these monastic founders had occupied
monastic seats before, and all of the monasteries were located in close proximity to the
“lay patron’s” political stronghold.12 However, Ehrhard 2000 seems to temper Wylie’s
generalization in pointing out that “while the princes of rgyal [mKhar] rtse acted as
ministers of the Phag mo gru, by the fifteenth century, they had achieved a quite
independent position against the dominance of the Phags-mo-gru dynasty.”13
After Tsong kha pa’s death in 1419, Mkhas grub spent time at Mdangs chen. In
1424, five years after the death of his master, at age thirty-nine, Mkhas grub founded
Dpal ‘khor sde chen in Gyantse.14 He is alleged to have spent four years there.15 It seems
likely, therefore, that Mkhas grub composed the Ocean of Attainment between
approximately 1424 and 1428.16 Jackson posits 1425 as the year in which Mkhas grub
composed his “anti-Sakya tantric polemics.”17
The conflicting accounts of the circumstances of his departure from Gyantse
invite further investigation. Specifically, they raise questions about the nature and
potential transformation of Mkhas grub’s relationship to his patron, local ruler, Rab btan
kun bzang (1389-1442). They also raise questions about how best to interpret accounts
of ‘debates’ within biographical materials. For instance, what role might debate play in
forming a very particular vision of an individual and distinguishing them from other
individuals and traditions?
While several primary and secondary sources suggest that Mkhas grub rje left his
seat at Gyantse as a result of a scheduled debate between himself and the Sakyapa Rong
ston ¸akya rgyal mtshan (1367-1449), the conditions of the debate and of his departure
remain unclear. Cabezón 1992 conducted the initial comparison of biographical accounts
of this event.18 He observed that while many accounts devote minimal attention to
Mkhas grub’s activities in establishing a monastery at Gyantse, two biographies, Rje
brtsun chos kyi rgyal mtshan’s (1469-1544/46) Secret Biography [Gsang ba’i rnam thar]
(SNT) and Gnas rnying ‘Jam dbyangs kun dga’ dge legs rin chen rgyal mtshan’s (1446-
1496) Rnam mthar mkhas pa’i yid ‘phrog (KYP), are quite concerned with depicting his
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
11
Wylie 1980, p.485.
12
Wylie 1980, p.485.
13
Ehrhard 2000, p.249. Ehrhard goes on to point out that Rab btan kun bzang possessed
titles from both Gong ma Grags pa rgyal mtshan and the Yongle emperor. However, we
have reason, based in Wylie’s argument to reconsider the degree of “independence” that a
Chinese title might indicate during the Ming (vs. the Yuan).
14
Wylie, however, dates the founding of dPal ‘khor sde chen to 1418. For his sources,
see fn10 in Wylie 1980.
15
These biographical details are gleaned from Cabezón 1992, p.16.
16
The Stong thun chen mo [TTC], the philosophical text that is the focus of Cabezón’s
1992 study, was also composed during this time.
17
Jackson 2010, p.178. This seems to be a reference to the body mandala debate.
18!Cabezón 1992 p.6 fn29 suggests that Rong ston is among the scholars whose
approaches to Madhyamaka thought Mkhas grub “may have very well found anathema.”!
! 4!
departure, albeit in different ways.19 Cabezón describes the latter account as emphasizing
the conflict between Mkhas grub and his patron:
“According to KYP and an oral tradition of this monastery itself, (fn43) a disagreement
arose between mKhas grub rje and the monastery’s sponsor, the local monarch Rab brtan
kun bzang, over a debate that the latter wished to organize, one that would pit mKhas
grub rje against one of the other great scholars of the day, Rong ston ¸akya rgyal mtshan
(1367-1449).(fn44) According to KYP, mKhas grub rje agreed to participate in the
debate(fn45). Scholars were invited to serve as judges, and the event was scheduled to
take place on a specific date...”20
It seems that this biographer envisions Mkhas grub’s patron as deliberately

attempting to orchestrate an encounter between Mkhas grub and Rong ston. The author
claims that Rong ston evaded the encounter, leaving Gyantse together with his patron;
Mkhas grub was thus provoked, according to this account, to post a letter to Rong ston on
the door to the monastery in retaliation.21 Cabezón translated the verses of this letter as
cited within the KYP itself. Curiously, a few lines of the letter actually suggest that the
patron, Rab brtan kun bzang, enabled Rong ston in his cowardice (though the KYP itself
credits Rab brtan as initiating the debate):
“...Finding refuge in the words of your benefactor
Who has said that in this district
Conferences of scholars are not allowed,
You tell him what a great kindness he has done you.
Though many hooded någas, the scholars,
Have come together in this place as judges,
From their respective oceans, the great seats of learning...
The word has come from him who is your protector
That, having returned to their respective abodes,
The oceans, which are the treasuries of water,
The mandala of disputation will not take place...”22
Such disparities in the accounts regarding Rab brtan’s role, in particular, suggest that
patronage may have played an important part in the construction of this ‘debate,’ whether
it be historical or solely narrative or some combination of the two.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
19
Both of these biographies appear in Mkhas rgub rje’s Collected Works:
Rje brtsun chos kyi rgyal mtshan. Gsang ba’i rnam mthar (SNT) Collected works, vol. a
pp.421-493.
Gnas rnying ‘Jam dbyangs kun dga’ dge legs rin chen rgyal mtshan.Rnam mthar mkhas
pa’i yid ‘phrog (KYP), Collected works, Vol.ka, pp.1-22.
20
Cabezón 1992, p.17.
21
See Cabezón 1992 p.17.
22
See Cabezón 1992 , Appendix I which includes a translation of the verses of this letter
cited from KYP 6a-7a.
! 5!
Cabezón describes the Secret Biography (SNT) by Rje brtsun chos kyi rgyal
mtshan (or Rje tsun pa) as a more “partisan” account than the KYP; in this case Rong
ston is painted as an antagonist of Tsong kha pa’s followers who creates schisms within
the Sakya tradition:
“The text states that Rong ston declared Tsong kha pa’s enterprise to be one of refuting
the Sa skya tradition and that this created an atmosphere of tremendous hostility toward
Tsong kha pa and his disciples, to the point where certain Sa skya pas even resorted to
black magic (gtor ma ‘phen pa) against them.”23
Ary translates the relevant section from the biography as follows:
“While at Nyang stod, Rong ston continuously nagged Mkhas grub rje, saying ‘Come, let
us debate together.’ At that time, all the fools at Sakya who were skilled in magical
incantations, blind to what are dharma activities and what are not, spread the rumor out of
competitiveness, attachment and aversion, that Tsong kha pa and his followers had
criticized the tenets of the Sakya school. They even performed rituals such as casting out
ritual cakes to send evil their way...”24
This account suggests that Rong ston as one among many Sakyas internally
differentiating their tradition from that of Tsong kha pa and his disciples. It also
characterizes the intent of these Sakyapas as malevolent and un-Buddhist. Rje tsun pa
thereby frees Mkhas grub of blame, guarding him against charges of the sin of causing
schisms in the sangha or of betraying his lineage of Sakyapa ancestors. In absolving
Mkhas grub in this way, the biographer justifies Mkhas grub’s polemical activities (and
potentially, his own). Rje tsun pa continues with what appears to be a reference to Ngor
chen:
“Also, Ngorpa Kunzangpa, who was treated like Vajradhara himself by the Sakyapas, out
of attachment to worldly fame and riches and disregarding his precepts and promises,
slandered Tsong kha pa despite having received Vajrayåna teachings from Tsong kha pa
[himself]. The scathing rejoinders to the self-damaging (rang tshang ston pa) argument
texts (rtsod yig) composed by the so-called Kunting Gushri geshe of Sakya are clearly
given in [Tsong kha pa’s] Collected Works.25 (8a) Thus, indeed, how could the Sakyapas
be anything but idiots who had never heard the essence of the teachings?”26
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
23
24
Translation of Secret Biography (SNT) by Rje brtsun chos kyi rgyal mtshan (or Rje
tsun pa) 7b by Ary 2015, p.128 (2007, p.226).!
25
No work fitting this description has turned up in a preliminary scan of Tsong kha pa’s
collected works. The connection between this “Gushri” and the “Kong ting gug shri ba”
addressed in the text from Mkhas grub’s thor bu will be discussed below.
26
tsun pa) 7b-8a by Ary 2015, p.128.
! 6!
This vehement denunciation of the Sakyapas explicitly names both Rong stong and Ngor
chen as culprits in a larger phenomenon of malicious attacks upon Tsong kha pa and his
follower by Sakyapas jealous and threatened by their prowess and success.
In his 2007 dissertation, Ary has suggested new ways of approaching Mkhas
grub’s biographical materials. Ary has demonstrated how Mkhas grub’s biographers
crafted and augmented his identity as a “defender of the faith.” Ary brings attention to the
ways in which Rje tsun pa, the author of Mkhas grub's Secret Biography, elevates Mkhas
grub's persona as a “great debater” and “defender of Tsong kha pa's tradition” (and
praises Tsong kha pa as holding similar skills). Some other biographies of Mkhas grub
apparently completely neglect this dimension of his persona. Rje tsun pa is the same
scholar who Ary argues promoted Mkhas grub’s image as one of Tsong kha pa's "main
disciples"; he also replaced the current philosophical textbook in use at Sera monastery
with one promoting Mkhas grub's interpretations. In fact, Rje tsun pa himself is also
famed for some of his intellectual encounters with Sakyapa scholars, including two of
Ngor chen’s students, ¸åkya mchog ldan (1428-1507) and Go ram pa.27 Ary’s study
presents the possibility that Mkhas grub may not have envisioned himself as a “defender
of the faith” to the same degree that his descendents did.
In his translation and study of Rje tsun pa’s Secret Biography of Mkhas grub, Ary
identifies the “debate” between Mkhas grub and Rong ston as the cause for the
deterioration of Mkhas grub’s relationship with his patron, Rab btan kun bzang.28 In Rje
tsun pa’s account, Mkhas grub initiates the debate, and there is an elaborate portrayal of
Rong ston’s agitation in his various attempts to circumvent the encounter. Mkhas grub
responds: “If you truly cannot debate [me], then you must cease denigrating Tsong kha
pa and the tenets of Någårjuna and his disciples!”29 None of the content of the debate is
even mentioned, only Rong ston’s acute humiliation, so profound apparently that he was
rumored to have been driven to take a vow of silence. Therefore, the Secret Biography
paints Mkhas grub as the victor and upholder of the integrity of Tsong kha pa’s legacy
while Rong ston is made the fool; however, there is no mention of how the encounter
impacted Mkhas grub’s relationship with his patron. Moreover, the Secret Biography
implies that Rong ston is creating the schism in his refusal to properly credit Tsong kha
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Although Ary had tentatively identified “Ngor pa Kun zang pa” as Ngor pa dkon mchog
seng ge in his 2007 dissertation (p227 fn451), he has emended this identification to Ngor
chen in his 2015 publication (p128 fn335). !
27!Cabezón 1992, p.6 fn26. Cabezón adds that: “The two Sakya scholars Go and ¸åk, as
they were known to the Gelukpas, were also fond of polemics, many of their criticisms
being directed against Tsong kha pa’s interpretation of the Madhyamaka.” !
28
For Ary’s discussion of the event, see Ary 2007 pp.120-121 and fn 267. For his
translation of the encounter within Rje tsun pa’s biography of Mkhas grub, see pp. 226 &
230-3. Cabezón 1992’s comment on p.6 fn29 confirms that he too identified this figure
with Ngor chen.
29
Translation of Rje tsun pa’s Secret Biography of Mkhas grub 11b by
Ary 2007, pp.232-3.
! 7!
pa’s contribution to the Sakya legacy as a faithful transmitter of Någårjuna’s spiritual
lineage.
In his own brief description of the encounter, Jackson 2010 presents the patron as
the obstructer of the debate:
“...in 1427, Khedrubje tried to engineer a public doctrinal confrontation with Rongton
Sheja Kunrik (1367-1449), another preeminent Sakyapa luminary...When at the last
minute the Prince of Gyantse stepped in and prevented a planned public debate, the
disappointed Khedrubje resigned the abbacy of Gyantse monastery.”30
Van der Kuijp even attributes Mkhas grub’s departure to his patron’s disapproval
of Mkhas grub’s engagement with Ngor chen and other Sakyapas.31 Cabezón 1992 makes
note of another theory, by which Mkhas grub left Gyantse for different reasons,
connected to “the status of the dGe lugs pa colleges at Dpal ‘khor sde chen.”32 Cabezón’s
theory has the most potential for informing our understanding of Mkahs grub’s motives
for critiquing the Sakyapas in his writings on body mandala during this time. It is
possible that the institutional dynamics at Gyantse and his relationship to his patron did
not assure Mkhas grub of a dominant position for Tsong kha pa’s tradition among the
other traditions represented there. In that case, one way of interpreting his writings on
body mandala, and perhaps even his ‘debate’ with Rong ston, is as an attempt to secure
that dominance.
The variety in these accounts of the debate with Rong ston, in terms of how and
by whom the debate was initiated and interrupted or completed, suggests that the
phenomenon of “debate” served as a platform for both real and imagined encounters in
the Tibetan context. These encounters are often framed as instrumental in tradition
formation and “sectarian differentiation.” The implications for the relationship of
patronage dynamics and polemical exchange in fifteenth-century Tibet will be addressed
further below.
After his departure from Gyantse, Mkhas grub returned to Mdangs chen, where he
allegedly had encountered visions of his deceased master. In 1431, Mkhas proceeded to
Ganden [dga’ ldan] where he assumed the throne and the responsibilities of the newly
formed order.33 According to its colophon, Mkhas grub composed another body mandala
debate text at Ganden. This text is a reply to Ngor chen’s rebuttal to the charges leveled
by Mkhas grub in the relevant chapter of the Ocean of Attainment. This text, The
Thunderbolt Wheel of Reply to Ngor [Ngor lan gnam lcags 'khor lo], is composed largely
of citations from the first round of the debate (including both Mkhas grub and Ngor
chen’s contributions).34 As no date is indicated, it must be assumed that it was created
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
30
Jackson 2010, p.178-9
31!Van der Kuijp 1985b p.98, n.18. Cabezón 1992, p.17 fn44 also refers to Van der
Kuijp’s theory. As will be discussed below, Ngor chen and Mkhas grub also engaged in
another controversy on deity yoga in the four tantric classes.!
32!Cabezón 1992, p. 17 fn 44. Cabezón cites the BE p.122 in this capacity.!
33
Cabezón 1992, p.18. See fn55.
34
Phyin ci log gig tam gyi sbyor ba la ‘jug pa’i smra ba ngan pa rnam par ‘thag pa bstan
bcos gnam lcags ‘khor lo. Collected Works [zhol] Vol. 2, pp7-100. Or see modern
! 8!
between Mkhas grub’s arrival at Ganden in 1431 and his death in 1438. Rje tsun pa’s
Secret Biography of Mkhas grub frames this text as a response to Ngor chen’s attack
against Tsong kha pa’s legacy:
Around that time, Ngorpa Kunsang was the bearer of the Sakya tenets, and he sent
a letter to both Rongton and Chojé Sonam Lodro saying ‘Since Khedrup Jé has criticized
the Sakya tenets, I will debate him on Mantra! Rongton, you are known for debating
Praj•åpåramitå, and Sonam Lo for debating pramåña. It is not good that you were
ineffective against him!’
Then, relying on the Sakya tenets, Ngorpa composed a critique of the views of
Tsong kha pa and his followers and had it delivered to Khedrup Jé. In this text, he
misunderstood the teachings, writing that, ‘Earth, water, fire, and wind are the five types
of form’! Furthermore, having misunderstood J•ånagarbha’s discussion of the two truths,
he [misquoted the texts] saying, ‘According to the self-commentary of ¸åntarakßita on
the two truths,’ and so on. In short, he did not even understand the general language of
the texts and succeeded only in shaming himself. Nevertheless, to refute this nonsensical
argument, Khedrup Jé composed a rejoinder entitled Wheel of Thunderbolts and
disproved all his erroneous views.35
This passage suggests a few valuable points for contemplation. First, it affirms the
portrait of Ngor chen as an expert in the field of tantra, a depiction we find in Ngor
chen’s own biographies and reinforced by his display of exegetical skill within the body
mandala debate texts. Mkhas grub is once again depicted as “defender of the faith,” and
specifically of Tsong kha pa’s tradition. Unlike the passage on Ngor chen from the
Secret Biography cited above, however, this one at least mentions a perceived attack
upon the Sakyapa tradition; the previous passage, on the other hand, painted Ngor chen as
merely spiteful, traitorous, and slandering. Furthermore, the present passage creates a
more substantial link between Ngor chen’s engagement with Mkhas grub on tantric
subject matter and the alleged encounter of Mkhas grub and Rong ston at Gyantse; it
even presents an additional figure who may have been involved in similar contentious
interchanges with Mkhas grub.36 However, while the passage appears to refer to the
circumstances of the body mandala debate, the references to Ngor chen’s erroneous
statements are not familiar from his body mandala texts. This suggests that there may be
further relevant materials within Ngor chen’s and Mkhas grubs’ collected writings or lost
correspondence between the two figures. This dissertation lays the foundation for
scholarship of such texts.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
publication: Dgag lan phyogs bsgrigs . Ch'eng tu: Si khron Mi rigs Dpe skrun khan,
1997. Chinese colophon title: Pien lun wen hsuan pien. pp. 1-68. Ary 2007 has described
this collection as “a compilation of polemical works refuting non-Gelukpa criticisms of
Tsong kha pa’s views on Madhyamaka philosophy.” (p.164)
35
tsun pa) 17b-18a in Ary 2015, p. 140.
36
Further research is needed to determine the identity of Chos rje bsod nams blo gros.
Ary 2015 (p.140 fn352) suggests that this may be a “Khewang Sonam Lodro.”!
! 9!
For the purposes of this dissertation, the Thunderbolt Wheel of Reply to Ngor,
sheds light on how the dynamics of the body mandala continued to evolve.
It also, provides an important clue for resolving the relationship between Ngorchen’s two
texts, Destroyer of the Proponents of Evil [N1] and the other “version,”
Dispelling Evil Views [N2]. In the Thunderbolt Wheel of Reply to Ngor ,
Mkhas grub cites selected points from Ngor chen’s argument, and upon preliminary
investigation, these selections appear to be from the longer “version,” Dispelling Evil
Views. Later in the dissertation, we will have the opportunity to compare the two
versions of Ngor chen’s text and to consider the implications of this discovery.
There is another text by Mkhas grub deserving of mention in light of the
complexity of attitudes it brings to our interpretation of the body mandala debate. This
letter from his miscellaneous writings [thor bu], may be as close as Mkhas grub comes to
a retraction of his polemical statements on body mandala. Davidson 1992 summarizes the
import of Mkhas grub’s apparent change in attitude as reflected in the letter from the thor
bu as follows:
“...he had generally refuted the Lam ‘bras ideas of the physical mandala (lus-dkyil) and
the reception of consecration during meditation (lam dus kyi dbang), without citing the
system by name. He complains that everyone jumped to conclusions. Given the
inflammatory language mKhas-grub was wont to use, it is easy to see how such an
impression developed.”37
Of course in Mkhas grub’s case, his inherently polemical style as expressed by

this “inflammatory language” seems to make a true retraction impossible. However even
the sense of ‘your misunderstood me,’ suggests that there were real consequences for his
writings on body mandala. These consequences may have impacted his reputation as
well as his relationships to his contemporaries (patrons and other superiors as well as
peers) and reflected poorly upon the public image of Tsong kha pa’s followers.
According to its colophon, the text was composed at Nyang stod.38 However, the
introduction informs the reader that the text summarizes Mkhas grub’s reply to a message
from Kon ting gug shri ba Mkhas grub received when he was at Gyantse.
To review, Mkhas grub’s Ocean of Attainment was composed at Gyantse
where Mkhas grub resided between 1424 and 1427/8.39 Ngor chen’s reply was composed
in 1426. Mkhas grub’s thor bu describes an event that also occurred during Mkhas grub’s
time at Gyantse. It does not specify whether it precedes or follows Ngor chen’s response
to the Ocean of Attainment or Mkhas grub’s Thunderbolt Wheel of Reply to
Ngor . Therefore, it is possible that it provides insight into the circumstances that
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
37
Davidson 1992, p.21 note 64 refers to this letter included in Mkhas grub’s Thor-bu
Collected Works, Vol. 9, 775-808 esp.776-7. Bentor 2006 also includes this text in her
bibliography.
38
The colophon also describes Mkhas grub as mkhas pa’i mdun sar mi ‘jigs pa’i slob pa
thob pa and rgyal khams kyi spangs btsun. Possible translations of these epithets include
“Learned and fearless before scholars” and “venerable dispeller,” suggesting a potential
reference to his skill in debate.!
39
Jackson 2010 dates Mkhas grub’s “anti-Sakya tantric polemics” to 1425.
! 10!
motivated Ngor chen’s reply to Mkhas grub on behalf of the Sakyapas. Alternatively,
since it was composed at Nyang stod and not specifically at Gyantse (though reporting
upon the circumstances of correspondence that occurred during Mkhas grub’s residence
there from 1424-8), it may have been written later, perhaps even after his Thunderbolt
Wheel of Reply to Ngor. Mkhas grub describes the circumstances that inspired this
text from his miscellaneous writings as follows:
“Kon ting gug shri ba, the kalyåñamitra of the great monastic seat, hoarded up a
few choice comments I made in the context of body mandala in my commentary on the
Guhyasamåja sådhana. A request to expand upon them arrived before my eyes at
Gyantse.”40
It is a challenge to determine the identity of Kon ting gug shri ba. Rockhill 1891
translates the phrase Kon ting gug shri, a Tibetan transliteration of a Chinese title, as
“Holy Anointed Adviser (Preceptor) of the realm.”41 Roerich 1988 indicates that the title
was conferred upon the abbots of ‘Tshur pu in the form of a seal by the Yongle/Ming
emperor.42 Cabezón 1992 does, however, make note of a reference to a Sakyapa “Kan
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
40
gdan sa chen po'i yongs kyi dge ba’i bshes gnyen chen po kon ting gug shri bas
bdag gi gsang pa 'dus pa'i sgrub thabs kyi bshad pa'i lus dkyil gyi skabs su bris pa'i tshig
'ga' zhig zur du bkol ba la lan btab pa'i rtsom ba rgyas par mdzad nas rgyal mkhar rtser
bskur ba kho bo'i mig lam du bab par bgyis
41
Rockhill 1891, p199 fn1.
42
The first to receive the title was the Sakypa abbot mkhas btsun Nam mkha’ legs pa’i
rgyal mtshan (1305-43) from the Yuan emperor Yesun Temur (r.1323-8). The title was
conferred upon laymen as well, for example, the Phag mo gru administrators, by the
Ming emperors. See Schweiger 2009/10 p315 together with reference to Petech 1990,
p.82 in fn9. Schweiger refers to the work of Shen Weirong 2007 in considering how the
conferral of these titles during the Ming may have been less politically significant than
during the Yuan. As such, these titles functioned more like gifts offered to visitors to the
Chinese court to assure Tibetan allegiance than as an actual political position (as it had in
the Yuan). See Schwieger 2009/10, p.314.
It is possible that Kon ting gug shri ba was a Kagyupa named Chos gyi rgyal
mtshan (1377-1448), a student of the Fifth Karmapa, renowned also for his transmission
of the title of Ta’i Si tu. See Roerich 1988, p.520, TBRC P3572 and Treasury of Lives
entry [treasuryoflives.org] by Thinlay Gyatso, 2014. Gyatso, however, makes no
mention of ‘Tshur pu but associates this figure with Karma Gon. While somewhat
unlikely, considering a possible Kagyupa identity for Kon ting gug shri ba prevents us
from automatically reducing the significance of the debate to merely a sectarian conflict
between Sakyapas and a lineage of Tsong kha pa’s descendents that eventually became
known as Gelukpas.
Knowing that Ngor chen assumed responsibility at Sakya after the passing of
Theg chen chos rje (1349-1425), we may consider the possibility that that figure is the
very Kunting Gushri we seek. This is a compelling possibility strengthened by the fact
that Theg chen chos rje was invited to court and conferred various titles by the Ming
! 11!
ting gu !rî” as one of Mkhas grub’s “opponents” cited in the Secret Biography.43 The
relevant passage from Rje tsun pa’s text, translated by Ary, was cited above: “The
scathing rejoinders to the self-damaging (rang tshang ston pa) argument texts (rtsod yig)
composed by the so-called Kunting Gushri geshe of Sakya are clearly given in [Tsong
kha pa’s] Collected Works.”44 Ary 2015 has identified Kunting Gu!ri as “Kunting
Gushri Namkha Sangpo,” but no additional information is provided.45 Moreover, no
work fitting this description has turned up in a preliminary scan of Tsong kha pa’s
collected works. However, even without identifying this figure definitively, this
statement suggests a connection between the body mandala debate between Mkhas grub
and Ngor chen and an earlier exchange between Tsong kha pa and a Sakyapa Geshe.
In comparing the tone of Mkhas grub’s description of the circumstances of
creating the Thunderbolt Wheel of Reply to Ngor with the one found in this
Reply to the Questions of the Kalyånamîtra Kon ting gug !rî ba , we find the
latter to be more extreme. Mkhas grub is reacting to public opinion of his body mandala
writings and a general sense that he went too far in his critique of the Sakyapas and their
Lam ‘bras tradition. Unlike the Ocean of Attainment or the Thunderbolt Wheel
of Reply to Ngor, Mkhas grub explicitly mentions the Sakyapas in setting the stage
for his statements here. In this text, Mkhas grub begins by denying the accusation that he
had denounced the Lam ‘bras tradition in his writings on body mandala:
“Which Lam ‘bras exactly do I refute? How, before whom, and in which text do I refute
it? Having carefully sought the answer to these questions, seek the unchanging source
resembling the speech of the rainbow in the sky of the ultimate. Remaining faithful [yid
ches kyi gnas su] to the Lam ‘bras rdo rje'i tshig rkang and the three tantras (Hevajra,
Sampu†a & Vajrapa•jara) without distinction, though I’ve already explained this material
more than once, I will do it again. There is no reason to think otherwise. The venerable
great Sakya fathers and sons, the Dharmaråjas and mahåpanditas expand the knowledge
of all the Buddha’s scriptures, Sütra and tantra, without obstacles.”46
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
emperor. It is not, however, clear that the title of Kon ting gug shri was among them. See
TBRC P3565.
43
Cabezón 1992, p.6 fn29. However, he does not provide an exact reference, and I was
unable to locate any such reference within the SNT.
44!Translation of Secret Biography (SNT) by Rje brtsun chos kyi rgyal mtshan (or Rje
tsun pa) 7b-8a by Ary 2015, p.128.!

45!If we consider the possibility that this figure may have been a contemporary of Tsong
kha pa but not of Mkhas grub himself, some additional possibilities present themselves.
For example, Kun dga’ rin chen (1339-1399)(TBRC P1862), the sixteenth throne holder
of bzhi thog bla brang at Sakya, was also granted the title Kon ting gug shri ba by the
Ming emperor.
!
46
Mkhas grub’s Thor-bu , Collected Works, Vol. 7, 775-808. See 775-6:
kho bos lam ‘bras gang du bkag; gang gi tshe bkag; tshul ji ltar du bkag; su zhig gi mdun
du bkag; yi ger bris pa ji 'dra ba zhig gi nang du bkag ces zhib tu rnam par dbyad nas;
mthar gtugs pa nam kha'i 'ja' tshon gsung ba ltar mi 'gyur ba'i khungs btsal bar gyis shig;
lam ‘bras rdo rje'i tshig rkang rgyud gsum dang khyad par med par yid ches kyi gnas su
kho bsos rkang tshugs pa nyid du sngon chad kyang lan cig ma yin par dbyangs su
! 12!
In “remaining faithful,” he establishes the requisite respect for his Sakya forefathers.
Mkhas grub then proceeds to praise the Sakpas as “lords of yoga” with mastery of the
generation and completion stages, letting his interlocutor know that his own awareness of
the integrity of the Sakya tradition is one of “firm faith, not merely words” [tshig tsam
ma yin pa'i dad pa brtan po thob zin pas]. [776] He then attempts to correct the mistaken
impressions his writings on body mandala have produced: “Some have mistakenly
assumed that the little bit of affirming and negating [dgag sgrub] I’ve done on body
mandala is a refutation [bkag pa] of the Lam ‘bras and empowerment at the time of the
path [lam dus].”47
In this text, Mkhas grub’s polemical tone seems to have escalated, and his
statement that he’s already explained his position on the Hevajra tantras and Lam ‘bras
materials “more than once” implies that this text may indeed post-date both the Ocean
of Attainment and the Thunderbolt Wheel of Reply to Ngor. He accuses his
interlocutor of being a liar obsessed with his own opinions and also critiques the
interlocutor’s method of argumentation:
“You’ve told me lots of things I already know about practicing mantra, like the necessity
of empowerment and vows. When it comes to the main issue, you totally lack any proof
[sgrub byed] regarding body mandala. You work so hard to articulate an attitude while
lacking familiarity with (the principles of) affirmation and negation. All scholars can see
that you are a fool. Yet with little hope of classifying the general teachings (sütra and
tantra), you diminish your own reputation (by writing as your have.)”48
Mkhas grub first problematizes the way in which his interlocutor has used the
classification of four philosophical schools [grub mtha' bzhi'i rnam gzhag]. He also
clarifies the goals of the !råvakas and pratyeka-buddhas as inferior to ultimate
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
bsgrags shing; da dung de la gzhan su kyang bsgrub dgos pa mi bdog; rje sa skya pa chen
po yab sras dang; chos rje ban chen ; bde bar gshegs [776.1] pa'i gsungs rab mdo rgyud
mtha' dag la mkhyen ba thogs med du rgyas shing
I am grateful to Khenpo Tashi Dorje for his guidance in reading this text in November
2011. I have focused on communicating the tone of the text faithfully. Due to the many
colloquialisms it employs, my translation of these excerpts remains somewhat
unpolished.
47
Mkhas grub’s Thor-bu , Collected Works, Vol. 7, 775-808. See 776:
da lan gyi lus dkyil gyi dgag sgrub zur tsam 'di lam ‘bras dang lam dus kyi dbang bkag pa
yin snyam pa'i go ba nor brgyud la skyes pa 'ga' zhig gis
48
Mkhas grub’s Thor-bu , Collected Works, Vol. 7, 775-808. See 776-7: sngags la 'jugs
pa la dbang dang dam tshig dgos pa sogs grub zin du ma zhig bsgrubs; skabs don lus
dkyil bsgrub ba la sgrub byed kyis shin tu phongs shing dgag sgrub kyi brda la ma byang
ba'i rnam 'gyur ches gsal bar sgrogs pa lhur byed pa'i [3] bzhad gad kyi gnas su mkhas pa
mtha' dag gis go bar nus pa [777.1] zhig bkod gda' na'ang khyed la gsung rab spyi'i rnam
gzhag re ba shas chung bas de'i cha nas mtshad chung mchi mod
!!
! 13!
buddhahood. His other critiques range from the wrongful denigration of Sa pan’s
astrological calculations in favor of the Kålacakra49 to the confusion of !amatha with
vipa!yana. Some of Mkhas grub’s concerns with defining the relationship of sütra and
tantra also appear here, a concern we will later examine as it emerges in his chapter on
body mandala. The articulation of this relationship is a common theme in Mkhas grub’s
polemics. After describing a few particular points of contention on the consecration
within the Lam ‘bras tradition, about half way into the text Mkhas grub begins to
specifically engage with the interpretation of his own writings on body mandala within
the Ocean of Attainment :
“Now, I must address the main topic. These few things I say below are a response to
the distortion of the relevant explanation of Guhyasamåja body mandala. I was afraid of
saying too much; I did not arrange an extensive account since it was not the appropriate
time to get into the establishment and refutation of the classification of other body
mandala. I did not realize there would be any proponent who would probe to the depths,
analyzing closely, clinging to what little I’ve said. Because of the utterance in the
courtyard [gya tshoms su], first I will ascertain my reply to those objections.”50
Mkhas grub’s use of the term gya tshoms su seems to imply that an actual debate
took place in the monastic compound or, at the very least, that his writings on body
mandala were a source of controversy. He then begins his assessment of his own
positions on the technical details of the practice, beginning with a statement he made in
the Ocean of Attainment . We will not address the contents of this portion of the text
from the thor bu here. Rather, we will leave it open as an avenue to potentially return to
together with the Thunderbolt Wheel of Reply to Ngor after working closely
through the chapter from the Ocean of Attainment and Ngor chen’s reply. Whether
regarded as a retraction, apologetic, or self-reflection of sorts, this text from Mkhas
grub’s miscellaneous writings provides a glimpse of how the body mandala debate
continued to evolve.
Did Mkhas grub change his mind about some of his more extreme critiques? Is it
perhaps that he regarded the genre of polemical writing, like debate in the courtyard, as a
liminal zone, a kind of safe space in which stronger, harsher declarations were allowed
without long-lasting consequences? What kinds of expectations did readers bring to the
their encounters with polemical literature? Were people who read Mkhas grub’s writings
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
49
The calculations may be those concerning the date of the Buddha’s passing, which
according to Sa pan, predates Rje btsun Grags pa rgyal mtshan’s death in 1216 by 3350
years. See Kramer 2008, p. 148 fn 29 together with her reference to Ruegg 1992, 272f. !
50
Mkhas grub’s Thor-bu , Collected Works, Vol. 7, 775-808. See 782-3:
da ni skabs don brjod par bya ste; de kho bos smras pa'i tshig tshan pa 'og ma 'di; gsang
'dus kyi lus dkyil 'chad pa skabs su bab pa'i dkyus yin pas; de'i skabs su lus dkyil gzhan
gyi rnam gzhag rgyas par dgag sgrub byed pa skabs ma yin zhing yi ge mangs pas 'jigs te
rgyas par ma bkod par zur tsam zhig smras pa la; zhib tu dpyad nas gting phyin pa'i 'dod
pa gang yin mi 'tshol bar gya tshoms su smra bar snang bas; thog mar 'di'i brgal lan gtan
la [783.1] dbab par bya'o
! 14!
on body mandala overreacting or was he attempting to produce precisely the response he
did? Is it possible, moreover, that changes in the socio-political and economic climate of
monastic life inspired Mkhas grub to reformulate his approach?
At the outset, we exercised caution against moving too quickly towards sectarian
explanations for framing the body mandala debate. However, even this preliminary
enagement with Mkhas grub’s writings on body mandala beyond the Ocean of
Attainment, has demonstrated that Mkhas grub passed the point of no return in his
writings on body mandala. Whether he intentionally set out to divorce his tradition from
its Sakyapa roots or not, once he had set the wheel in motion, there was no turning back.
Mkhas grub’s polemics played a formative role in distinguishing the project of defending
Tsong kha pa’s tradition from “remaining faithful” to its Sakyapa roots. This is a
distinction that continued to evolve over time. Ary observes:
“...while Tsong kha pa’s followers may have not have necessarily seen themselves in the
very beginning as members of a new and distinct religious order, over time they came to
distinguish themselves more and more from the Sakyas, with whom they initially shared
many common traits. Eventually, Tsong kha pa was perceived as having diverged from
the interpretations accepted by the Sakyas, for which he was greatly criticized by some of
the latter tradition’s foremost scholars, and the Gelukpas’self-identity as a distinct
religious order centered upon the acceptance of Tsong kha pa’s particular interpretations
began to solidify, yielding the new phenomenon of what we can aptly label Geluk textual
community.”51
This dissertation adds to Ary’s work with biographical materials. It also makes a
stronger claim for Mkhas grub’s own agency in authoring his fate. While he may have
not ‘intended’ to burn his bridges with the Sakyapas and their patrons and allies, his
actions aggravated already existing tensions to produce irreversible effects, effects
demonstrated clearly in his own nebulous departure from Gyantse.
Having considered some of the dynamics informing Mkhas grub’s polemics and
his relationship to an emerging Gelukpa identity, we now proceed to examine the identity
of the monk who chose to reply to his objections, Ngor chen kun dga’ bzang po. In the
process we will evaluate Ngor chen’s role as a Sakyapa, how that role is understood from
both an institutional and intellectual perspective, and the ways in which it characterizes
his engagement in the debate. We will also revisit the trope of “defender of the faith” to
determine how it may impact representations of Ngor chen’s accomplishments and in
particular, of his polemics.
III. Ngor chen’s Contributions to the Body Mandala Debate

During his time as abbot of Ngor E wam chos ldan, Sangs rgyas phun tsogs [1649-1705]
composed a biography of the monastery’s founder. The biographer contextualizes Ngor
chen’s contributions to the body mandala debate within a series of polemical writings
intended to correct mistaken views:
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
51
Ary 2007, p.162.
! 15!
“Through debate, he reversed mistaken views. At an earlier time, some said that the great
Dharma protector (Virüpa) was a Cittamåtrin pandit and that the intention of his three
tantras together with oral instructions was to spread the Cittamåtrin perspective. (In
response,) Ngor chen composed the great treatise that defends through scripture and
reasoning, the Root and Commentary for Overcoming Objections to the Three Tantras
[rgyud gsum gnod 'joms rtsa 'grel]. Later on, there was a terrible misconception of
imagining the Hevajra body mandala not to be explained anywhere in the Indian
tantric system. Through scripture and reasoning and the oral instructions, he (Ngor chen)
thoroughly refuted that circumstance of the Hevajra initiate admitting wrong [lam dus
blangs pa mthol bshags byed pa'i skabs byung ba]. He composed the great treatise that
establishes the unsurpassable intention of the tantric system called Destroyer of the
Proponents of Evil through Eliminating Objections to the Body Mandala and Dispelling
Evil View(s).52
The two latter texts, together with Mkhas grub’s chapter on body mandala from
the Ocean of Attainment form “round one” of the body mandala debate; they are the
focus of this dissertation. The colophons of these two texts by Ngor chen identify their
date of composition as 1426 [zil gnon], three years before he founded Ngor monastery [E
wam chos ldan]. Therefore, the period of composition of this network of texts coincides
with institution-building activities for both Ngor chen and Mkhas grub at Ngor monastery
and Gyantse, respectively.
One particular phrase in Sangs rgyas phun tsogs’s overview of Ngor chen’s
polemical activities is of note: lam dus blangs pa mthol bshags byed pa'i skabs byung ba.
Roughly translated here as “that circumstance of the Hevajra initiate admitting wrong,”
the implications of the phrase are compelling but difficult to pin down. One way of
interpreting this remark about Ngor chen’s unnamed opponent in the body mandala
debate is that he is a Sakyapa who has turned his back on his own tradition. Khenpo
Tashi Dorje of the International Buddhist Academy (IBA) in Kathmandu suggested that it
refers to Mkhas grub as someone who received the Lam ‘bras initiation but later critiqued
the legitimacy of the transmission.53 Before ever studying with Tsong kha pa, Mkhas
grub indeed received the Hevajra initiation and Lam’ bras teachings from Ye shes dpal
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
52
Sangs rgyas phun tshogs 1688, p.546.2-.5:
rtsod pa’i sgo nas gzhan gyi log par rtog pa bzlog pa yang; kha cig sngon gyi dus su dpal
ldan chos skyong [546.3] sems tsam gyi pan di ta yin pas; bir wa pa'i rgyud gsum man
ngag dang bcas pa'i dgongs pa sems tsam du bkral ba yin no zhes pa'i log par rtog pa
byung ba; lung rigs kyis sun 'byin par byed pa'i bstan bcos chen po rgyud gsum gnod
'joms rtsa 'grel mdzad; yang phyis kyis kye [546.4] rdo rje'i lus dkyil rgyud rgya gzhung
gang nas kyang ma bshad pa'i rtog brtags yin no zhes pa'i log rtog 'jigs su rung ba lam
dus blangs pa mthol bshags byed pa'i skabs byung ba de lung rigs man ngag gi sgo nas
legs par sun phyung nas; rgyud gzhung gi dgongs pa bla na med pa sgrub par byed
[546.5] pa'i bstan bcos chen po'i lus dkyil rtsod spong smra ba mngan 'jom zhes bya ba
dang; lta ba ngan sel zhes bya ba gnyis mdzdad do
53
Personal communication, Fall 2011.
! 16!
and studied with esteemed Sakya teachers like Bsod rnams rgyal mtshan and Na bza’ ba.
Two years after his full ordination, Ren mda’ ba sent him to study with Tsong kha pa.54
This aspect of Mkhas grub’s biography reinforces his identity as a Sakyapa, inculcated in
the primary tantric ritual tradition of the lam’ bras including the Hevajra body mandala
practice.
In the previous section of this introduction, we investigated the contours of
representations of Mkhas grub as a debater and “defender of the faith.” We also observed
how has status as defender is construed specifically as a defense of the teachings of
Tsong kha pa and his disciples. We encountered a self-consciousness in some excerpts
from Mkhas grub’s biographical materials to avoid associations of divergence or betrayal
in distinguishing Tsong kha pa’s views from those of his teachers. One solution was to
depict the Sakypas who chose to challenge his views as ignorant of and therefore
themselves divorced from the true tradition. Both Tsong kha pa and Mkhas grub
possessed strong ties to the Sakya tradition; both were students of great Sakya masters
like Ren mda’ ba. Locating the precise historical moment when defending Tsong kha pa’s
views became formally distinguished from defending the Sakya tradition is a challenge.
By analyzing the body mandala debate texts and their later interpretations, we may work
towards a better understanding of how this distinction was realized.
Sangs rgyas phun tshogs undeniably portrays Ngor chen as a defender of the
Sakya tradition. A verse playing upon Ngor chen’s name concludes the section on Ngor
chen’s polemical writings: “The master of scripture and reasoning, the glorious one
beloved by all [Kun dga’] expels the demigods’ misconceptions with the vajra. The
excellent [bzang po] author bestows wonders through increasing the three conditions of
the Buddha’s teachings.”55 An earlier section of the biography describes how Ngor chen
was obliged to decline the ruler A ma dpal’s initial invitation to come to teach in Mustang
due to ideological attacks leveled against the Sakyapas at the time.56 The author of these
attacks is not identified: “It was said that ‘the view of the three tantras of the former
Sakyapa hierarchs is Cittamåtrin’ and the refutation arose saying, ‘the Sakyapa body
mandala is not explained from the tantric (textual) system.”57
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
54
55
dpal ldan kun dga'i lung rigs lha dbang gis; rdo rje log rtog lha min kun bcil nas; thub
bstan skabs gsum rgyas par mdzad [547.1] pa yis; mdzad pa bzang po e ma ngo mtshar
phul
I am grateful to Khenpo Tashi Dorje for his guidance in reading selections from Ngor
chen’s biography. He suggested that skabs gsum may be a reference to the three vows.
[Personal communication, Fall 2011] The three vows are the pråtimokßa vows of
monastic conduct, the bodhisattva vows of altruistic intention, and the samaya vows
guarding tantric practice.
56
Ngor chen ultimately made three journeys to Mustang and was instrumental in
establishing and perpetuating the Sakya tradition there through bringing the Tibetan
compilation of the Buddha’s teachings, the Kanjur, establishing and consecrating
monasteries and temples, and conferring teachings, ordinations, and tantric initiations.
See Sangs rgyas phun tshogs 536.5-539.2.
57
snga ma sa skya pa'i rgyud gsum man ngag dang bcas pa'i lta ba sems tsam yin zer ba
dang ; sa skya pa'i lus dkyil gyud gzhung nas ma bshad zer ba’i rtsod pa byung nas
! 17!
As we know from exploring the account of Ngor chen’s polemical activities
above, Ngor chen composed the Root and Commentary for Overcoming Objections to the
Three Tantras [rgyud gsum gnod 'joms rtsa 'grel ] in response to the first of these charges.
Van der Kuijp dates this text to 1406 and identifies it as: “a polemical text against
Ratnåkara!ånti and his Tibetan followers, who maintained that the Hevajra Tantra was
mentalisitic (sems-tsam-pa) in philosophical persuasion. These Tibetan followers
included Ren-mda’-ba Gzhon-ni blo-gos (1348/9-1412) and Bo-dong Pañ-chen.”58 As
suggested above, the issue of Virüpa, the progenitor of the Sakya Lam’bras tradition,
being a Cittamåtrin, apparently cast a shadow upon the legitimacy the Sakya tradition at
large. However, the notion that the Hevajra literary tradition might itself be tainted with a
Cittamåtrin approach, a philosophical approach commonly devalued by Tibetan
commentators in favor of the Madhyamaka, posed an even greater threat. It implied that
the Sakyapa interpretation of the tantric texts was imperfect; further, it suggested that the
particular tantric canonical texts upon which the Sakya ritual tradition was based were
not necessarily to be regarded as the ultimate means for accessing soteriological truth. Of
course this was not a critique so radical as to claim that the Hevajra Tantra was not the
word of the Buddha. Instead, the critique operated on the basis of established
frameworks of Buddhist doxography such as the “three turnings of the wheel.” One
version of this framework propounds the Madhyamaka as the pinnacle of all the
Buddha’s teachings, as a more profound and accurate teaching than the Cittamåtrin view.
The subtlety of the critique allowed by this classificatory schema plays upon the
assumption that the Buddha taught in accord with the needs and capacities of his
disciples. Therefore, according to one version of the “three turnings” embraced by the
Tibetans, only the most sophisticated students would be capable of understanding the
Madhyamaka perspective. Thus, to claim that the Hevajra Tantra itself as well as the
Indian master from whom the lineage of ritual transmission of the Sakya interpretation of
this tantra emanated were both Cittamåtrin, was to say that, in a sense, the Sakyapas
(including Tsong kha pa’s own teacher, Ren mda’ ba) were not among the Buddha’s most
gifted disciples.
Tibetan Buddhist scholasticism encouraged charismatic individuals (and the
institutions they belonged to) to seek prestige (and power) through the skillful
performance of the Prasangika Madhyamaka method. This translated into the
composition of polemical texts and the practice of philosophical debate. The polemical
claim that the Sakyapa interpretation of the Hevajra Tantra was Cittamåtra reveals a great
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
58
van der Kuijp 1985, p.87. Khenpo Tashi Dorje of the IBA suggested that Ren mda’
was the one to demand a reply on this count and that Tsong kha pa supported him in this
demand. Khenpo Tashi Dorje also referred to a relevant conversation between Khenpo
Gyatso of Sakya College and His Holiness the Fourteenth Dalai Lama during a
conference in Varanasi in 2004. Khenpo Gyatso claimed that the faulty identification of
Virüpa as a Cittamåtrin was the result of confusing two different historical masters;
apparently His Holiness accepted the claim. That this issue would continue to be regarded
as relevant in the present age attests to the enduring significance of the tantric polemics
that are the focus of this dissertation. [Personal communication, Fall 2011]
! 18!
deal about what was at stake in the body mandala debate. In the Introduction to this
dissertation, we observed the fundamental tension between the mind's tendencies to
proliferate and generate ideas and images, and to superimpose false frameworks upon
reality, on the one hand, and the tantric approach to harness the mind's faculty of imaging
to the soteriological practice of generating oneself (or external objects) as Buddhist
deities, on the other hand. That is to say, in a fashion characteristic for tantra, the mind’s
normally negative compulsion to proliferate is employed as a means (upåya) to overcome
the limitations of the mind and to liberate it from this very compulsion. This
soteriological employment of the mind's faculty of imagining is at odds with the
Madhyamaka identification of discursive language-based thinking as the principal source
of ignorance that is to be overcome in the quest for liberation and a more accurate
experience of reality. By contrast, the Cittamåtra position of the Yogåcåra school with its
emphasis that perceptual data are only mental productions lends itself well to the tantric
project. Mental acts of imagining are not in themselves- as in the Madhyamaka- the
reason for our state of bondage, but rather our way of engaging them. This means that the
Cittamåtra stance of the Yogåcåras allows more easily for the employment of tantric
techniques of visualization and mental fabrications. The Cittamåtra stance is also closely
associated with the principle of tathågatagarbha or innate Buddhahood, a hotly contested
teaching in fourteenth century Tibet. The next chapter will engage with the tensions
between natural and fabricated, enlightened and defiled objects-tensions that factor
prominently in the treatment of tathågatagarbha and Cittamåtra in the Tibetan tradition.
As Chapter Four will show, the teachings of tathågatagarbha and Cittamåtra, and the
bearing they have on the efficacy of tantric ritual action, are subjects that feature
prominently in tantric exegetical treatises such as Mkhas grub’s writings on the body
mandala.
Twenty years after producing the Root and Commentary for Overcoming
Objections to the Three Tantras, Ngor chen composed the body mandala debate texts,
Destroyer of the Proponents of Evil and Dispelling Evil Views, in response to
the latter charge, that “the Sakyapa body mandala is not explained from the tantric
(textual) system.” However, Sangs rgyas phun tshogs provides an interesting if
enigmatic detail here. Apparently, a rule [bca' khrims ] was made preventing Sakyapa
Geshes from travelling outside of Sakya as a direct response to the threats.59 This rule is
the reason Ngor chen could not accept his first invitation to Mustang. It is unclear how
Ngor chen was elected to rise to the challenge of fending off these charges or, for that
matter, if there was any precedent for such rules to be made in response to offensive
views. The comment does, however, imply that the critique of Sakyapa body mandala
we assume to have been leveled by Mkhas grub, was perceived as a serious threat to the
tradition, a threat demanding of a reply. Jackson 2010 points out that the death of one of
the most important members of the Sakya Khon family, Theg chen chos rje, in 1425
prompted Ngorchen “to assume much responsibility.”60 He dates Mkhas grub’s “anti-
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
59
[537.4]sa skya pa'i dge bshes rnams 'thor sa med pa'i bca' khrims byas nas 'byon stabs
ma byung
60
Jackson 2010, p.178.
! 19!
Sakya tantric polemics” to the same year, suggesting that Ngor chen’s response was part
of this newfound responsibility.61
Sangs rgyas phun tshogs refers to additional details surrounding Ngor chen’s
invitation to Mustang in Glob bo mkhan chen’s (1456-1532) [rje mkhan chen pa]
biography.62 Kramer’s 2008 critical edition and translation of Glo bo’s autobiography
features the relevant section.63 Kramer has noted the interesting way in which the details
of Ngor chen’s visits to Mustang assume such a prominent role (approximately forty
percent of the text) in this figure’s autobiography, despite the fact that the two masters
never met. She describes this prominence as testament to Glo bo’s “strong obligation...to
this great master of the Ngor pa tradition.”64
Glo bo’s account contradicts Sangs rgyas phun tshog’s in suggesting that the
events surrounding the body mandala debate and the consequent delay of Ngor chen’s
travels to Mustang concern the second rather than the first visit. It also identifies Mkhas
grub as the antagonist, referring to him as lCang ra bKa’ bcu (as he had founded lCang ra
monastery):
“Later, [A-med-dpal] thought to invite Ngor chen for a second time. After he had sent
one petition [zhu yig] after another, he received [this] reply in a first answering letter:
‘Although I was planning to come this time, due to a letter written by lCang ra bKa’ bcu
pa which says that the Hevajra body mandala is not a correct teaching, all the monks of
the monastic seat got angry, and therefore an order has been enacted for the religious
scholars not to go anywhere as long as this [matter] has not been resolved. So, I too,
have no possibility of going.”65
He then asks A ma dpal, if he’s intent upon his visit, to send a letter to prevent him from
receiving invitations to lots of neighboring monasteries and guaranteeing that his stay
will be brief, “for the confidence of those here.”
If we apply the skepticism of the biographer’s intentions we have acquired from
Ary’s study of Je tsun pa’s Secret Biography of Mkhas grub, we might be inclined to
consider the possibility that Sangs rgyas phun tshogs is retroactively investing Ngor chen
with a comparable identity of “defender of the faith.” It is possible that the body mandala
debate acquired more significance in ensuing generations as the divide between Sakyapas
and Gelukpas [or dga’ ldan pas] became more pronounced. Davidson, for example, has
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
61
Heimbel 2012 further enriches our knowledge of Ngor chen’s activities during this
period of time. Through recourse to several biographies of Ngor chen’s student and
eventual abbatial successor, Mus chen sems dpa’ chen po dkon mkhog rgyal mtshan
(1388-1469), Heimbel points out that Ngor chen fell quite ill in 1426. After he
recuperated, Ngor chen bestowed the Lam ‘bras initiation upon Mus chen. See Hembel
20212 p.54 and fn 23.!
62
See reference in Sangs rgyas phun tshogs 537.5.
63
For the version translated by Kramer, see Tøyø Bunko 41-683.
64
Kramer 2008, p.56.!
65
Kramer 2008, p.146-7. Glob bo mkhan chen bsod nams lhun grub. (1456-1532) Rje
btsun bsod nams lhun grub legs pa’i ‘byung gnas rgyal mtshan dpal bzang po’i rnam par
mthar pa zhus lan. Manuscript 6a-b.!
! 20!
also noted a disjuncture between the perceived terms of the ‘debate’ between the two
figures and the details of their biographies.66 Bearing Ary’s assessment of the
biographical genre as “at the heart of Tibetan Buddhist sectarian formation itself,” it
becomes necessary to consider whether the debate was a sectarian issue or rather was
later interpreted one.67
Heimbel 2012 interprets descriptions of the founding of Ngor E wam chos ldan in
biographical accounts of Ngor chen in a manner suggesting that Ngor chen was partially
motivated by sectarian concerns like those commonly understood to characterize the
body mandala debate:
“Withdrawing from sectarian conflicts with the Dge lugs school on one hand, and from
the worldly distractions of the bustling town of Sa skya on the other hand, Ngor chen
founded E waµ chos lan in the remote Ngor valley, located around 20 km southwest of
Gzhis ka rtse, hoping to go back to traditional Sa skya teaching and practice in a more
supportive environment.”68
Heimbel’s statement suggests that the atmosphere at Sakya at the time was contentious
and not entirely conducive to intense spiritual training and, further, that Sakya-Geluk
tensions were responsible, in part, for the situation.69 It also frames Ngor chen’s project
as a return to the roots of the tradition through moving away from the institutional center.
Therefore, the body mandala debate might be read as one instance of a broader conflict
used and potentially even amplified in textual sources in order to validate Ngor chen’s
monastic project.
Jackson 2010 not only affirms the sectarian dimension of Ngor chen’s monastic
foundation but also links the conflict between Ngor chen and Mkhas grub to the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
66
Davidson 1991, 20
67
Ary 2007, 13.
68
Heimbel 2012, p.48.!
69
The other biography of Mkhas grub translated in Ary 2007, 30a, uses a similar trope of
the degeneration of Sakya. Of course the description is more severe and the intention
somewhat different. In that account. Mahåkåla of the Tent appears to Mkhas Grub,
imploring him to remain at Sakya. Mkhas grub replies, “If there is one person here with
pure views and conduct, I may stay. However, the tenets of the spiritual forefathers no
longer exist, and their descendants have gone too. The temples are filled with women,
donkeys, cows and barmaids...” (Strangely enough, Sangs rgyas phun tshogs also
mentions barmaids [chang ma ] in his description of the distractions at Sakya in Ngor
chen’s biography. Although Khenpo Tashi Dorje suggested emending the reading to
“bar” or “drinking place”[chang sa] in our readings of that text, Heimbel 2012 points out
that all versions of the text indeed read chang ma. See Sangs rgyas phun tshogs 524.3
and Hembel 2012 p49 fn 5.) Mahåkåla actually commiserates with Mkhas grub as to the
state of the place. However, he attempts to order Mkhas grub to stay on the grounds that
he too, would have gone to study with Tsong kha pa were in not for the tragic state of
affairs. He even confesses: “The Sakyapas definitely don’t like you. They gave me ritual
cakes and told me to harm Tsong kha pa and his spiritual heirs, particularly you!” 30a in
Chos ldan rab ‘byor’s Short Biography of Mkhas grub translated in Ary 2015 p.112.
! 21!
scheduled “public doctrinal confrontation” between Mkhas grub and Rong ston. As
mentioned above, Jackson’s version of the encounter of the latter figures posits the
patron’s intervention and the “disappointed” Mkhas grub’s departure from Gyantse. He
adds further dimension to the story by explaining Rong ston’s involvement with an
institutional project he regards as conflicting with Mkhas grub’s tradition-building
initiatives:
“Rongton had recently helped the Rinpung lords found a multisectarian monastic
complex at the site of the Rinpung Great Maitreya Statue and Temple where scholarship
of the Sakya order was strongly represented (but which also included Geluk and Bodong
colleges). Khedrubje, on the other hand, was still exerting himself to achieve doctrinal
hegemony for his new Geluk School through conversions, polemics, and other means.
He had little patience with prominent representatives of the other schools or broad-
minded patrons.”70
Davidson’s interpretation frames both Mkhas grub’s encounter with Rong ston (or
lack thereof) and the former’s ensuing departure from Gyantse in terms of what Cabezón
has termed “sectarian differentiation.” He also suggests that the events at Gyantse built
upon tensions that had already escalated in ‘round one’ of the body mandala debate
between Mkhas grub and Ngor chen. Furthermore, the situation at Gyantse appears to be
embroiled in a dense web of patronage dynamics involving both the Rin spungs and
Phags mo gru lords. It may also be connected with diverging views on institutional
organization and competing bids for patronage. Decades after round one of the body
mandala debate and the events at Gyantse, Ngor chen himself is also alleged, by some
accounts, to have resisted multi-sectarian patronage projects. Cabezón 2007 recounts
such an instance from Bsod nams grag pa’s (1478-1554) Deb ther mar po gsar ma. Ngor
chen apparently demanded a more exclusive allegiance from the Rin spungs lord Nor bu
bzang po [d.1466], who expressed devotion to both the Sakyapas and Kagyupas but had
“also looked kindly upon the dGe ldan pas.”71 As a preconditon for conferring teachings
upon the ruler, Ngor chen allegedly unsuccessfully attempted to make the following
demands: that “all the dGe ldan pas under his [Nor bzang pa’s] rule were converted to Sa
skya pas” and the prevention of “bKa’ bcu pa dGe ‘dun ‘grub from building his
monastery.”72 He also demanded the ruler’s patronage of Ngor monastery. Cabezón uses
the account to set the stage for one of Ngor chen’s later disciples, Go ram pa’s (1429-
1489), polemical writings challenging Tsong kha pa’s views. He describes the climate of
the times as one in which “in the wake of the loss of Sakyapa political hegemony in
Tibet, in a period in which rival schools were vying with one another for the support of
patrons, and at a time of great political instability... an institution’s affiliation with one
political faction could cause retaliation from others.”73 Cabezón does note that (much
like the events at Gyantse involving Mkhas grub and Rong ston and the body mandala
polemics of Mkhas grub and Ngor chen) this exchange between Ngor chen and Nor
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
70
Jackson 2010, p. 179
71
72
Tucci, Deb ther, pp.99aa, 239-40 as cited in Cabezón 2007, p.44.!
73
! 22!
bzang pa has been contested.74 Yet, while the precise circumstances of the encounters are
not always clear, a palpable quest for “sectarian differentiation” was indeed part of the
world Ngor chen and Mkhas grub’s inhabited and participated in.
Another method for situating the body mandala debate texts within Ngor chen
and Mkhas grub’s larger spheres of activity is by consulting colophons and their
compilations of collected works [gsum ‘bum]. The next section will employ these
resources to determine how the body mandala debate texts may function as an extension
or synthesis of or even departure from these authors’ overall literary trajectories.
IV. Additional Resources: Colophons and Collected Works

Before proceeding to focus upon the fifteenth-century polemical context within Tibetan
scholasticism, we will pause to make a few observations based upon reviewing the
overall contents of Ngor chen and Mkhas grub’s collected works as well as the colophons
for individual texts. Many of Ngor chen’s works are dated and were composed either at
Sakya between approximately 1405 and 1429 or at E Wam chos ldan between 1429 and
1456. Ngor chen’s tantric ritual expertise is confirmed by the contents of his collected
works; the majority are ritual or exegetical texts devoted to tantric subject matter. In
addition to his tantric ritual and exegetical texts, he composed a number of praises [bstod
pa] extolling teachers, gurus of the lineage [bla brgyud gsol’debs] and deities, as well as
biographies [rnam mthar] of Sa pan, masters of the Lam ‘bras tradition, and his teacher
Buddha!rî. Several of these seem to have been written earlier in his career, during his
time at Sakya (and therefore before the founding of Ngor monastery in 1429). Other
genres represented include replies to questions [dris len], teachings for students [gdams
ngag], as well as letters of correspondence [spring yig]. We have a number of letters sent
from the hermitage at E Wam chos ldan. Ngor chen also made significant contributions to
the cataloguing [dkar chag] of canonical texts. He created catalogues of the Kanjur and
Tanjur as well as of tantric commentaries and the collected writings of the Sakyapa
patriarchs. His cataloguing activities were particularly important to his institution-
building activities in the region of Mustang.75 During his time at Sakya, Ngor chen
produced a partial translation of the Hevajra Tantra, consulting both Sanskrit manuscripts
and previous translations.76 Ngor chen also catalogued the commentaries upon the
Hevajra Tantra.77 Therefore, he had some experience with the translation of the root text
as well as with the scope of the commentarial tradition.
In taking stock of Ngor chen’s collected works, we can see how between 1406
and 1419 Ngor chen developed his expertise on the interpretation and ritual application of
the Hevajra Tantra through ritual, tantric exegetical, and polemical compositions. The
latter, in particular, lay the groundwork for his engagement in the body mandala debate in
defense of the integrity of the Hevajra tradition as maintained by the Sakyapas. Ngor
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
74
See Cabezón 2007 p44 fn215 for details.!
75
The colophon to his Mustang Tenjur is dated 1447. See Eimer’s 1999 edition of Ngor
chen’s Mustang Kanjur catalogue.
76
Rtsa rgyud brtag gnyis kyi ‘gyur. Collected Works Vol.2 pp.445-6. Van der Kuijp’s
1985 article focuses upon Ngor chen’s translation of two Verses [II. V.1-2] not present in
several other translations as presented by this short text.
77
This text immediately follows the translation in the Collected Works, Vol. II pp.446-7.
! 23!
chen composed his first major polemical text, Root and Commentary for
Overcoming Objections to the Three Tantras at Sakya in 1406.78 As discussed
above, this text was a response to the idea that the Hevajra tantras were expressive of the
Cittamåtrin perspective. Just a year earlier, Ngor chen composed a work embedding the
transmission of the Hevajra Tantra within the larger scope of tantric history.79 In 1407,
he completed a Hevajra sådhana as well as another Hevajra-based ritual text and in 1412
yet another Hevajra sådhana.80 In 1419, Ngor chen produced another relevant text to the
interpretation of the body mandala debate, the Moonrays of the Pith.81 As referenced
above, scholars continued to respond to the views Ngor chen articulated in this
comprehensive interpretation of the Hevajra sådhana practice for generations to come.
Between 1423 and 1425, Ngor chen composed multiple texts on the Guhyasamåja
system. The most significant for our interpretation of his writings on body mandala is
The Ocean of Attainment of the Sådhana of the Guhyasamåja Mandala.82 Not only does
this text bear a very similar title to Mkhas grub’s Ocean of Attainment, it likewise
appears to focus upon the generation stage phase practice of the Guhyasamåja. The
colophon indicates that Ngor chen implemented some of the very same core texts of the
Årya Guhyasamåja tradition employed by Mkhas grub: the Mdor byas (Piñ∂î-k®ta), Mdo
dang bsres pa, Rnam gzhag rim pa [Samåja sådhana vyavasthole [sthåli ], and Sgron ma
gsal ba.83 Over the course of this dissertation, the significance of Mkhas grub and Ngor
chen’s choices for citation in the body mandala debate texts will become apparent. These
choices are expressive of their command of a vast array of tantric literature and their
strategic endorsement of particular lineages of transmission. These choices also
contribute to their style of argumentation. Knowing that Ngor chen was equipped to
engage with the very same Årya Guhyasamåja texts chosen by Mkhas grub but chose
instead to introduce sources from the Cakrasaµvara and Hevajra traditions in the body
mandala debate raises questions for further exploration. It also prompts us to look more
closely at his choice to isolate the interpretation of one particular Guhyasamåja-related
text, the Vajramålå Explanatory Tantra. In another text Ngor chen composed on the
Guhyasamåja, undated but composed at Sakya and therefore likely before 1429, the
colophon specifies that he relied upon the Vajramålå Explanatory Tantra in addition to
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
78
rgyud gsum gnod 'joms rtsa 'grel
79
The Extraordinary Ocean of Biographies of Lineage lamas and the Manner of Arising
of the Hevajra Tantra [kye rdo rje’i byung tshul dang brgyad pa’I bla ma’I rnam thar ngor
mtshar rgya mtsho] See van der Kuijp 1985, Appendix p.87.
80
In1412, Ngor chen also composed several texts on the rituals of the Vajradhåtu:
Rdo rje dbyings kyi dkyil 'khor sgrub thabs dngos grub snye ma; Rdo rje dbyings
kyi dkyil 'khor mchod pa'i cho ga tshogs gnyis snye ma; Rdo rje dbyings kyi dkyil
'khor gyi cho ga dbang gi snye ma; Rdo rje dbyings kyi sbyin sreg gi cho ga phrin
las snye ma; Rdo rje dbyings kyi dkyil 'khor gyi bkra shis kyi tshigs bcad dge legs
snye ma.
81
Dpal kye rdo rje’i sgrub pa’i thabs gyi rgya cher bshad pa bskyed rim gnad kyi zla zer.
Davidson 1992 includes this text in his study of the Hevajra abhisamaya.
82
Gsang ‘dus dkyil ‘khor gyi sgrub thabs dngos grub rgya mtsho.
83
See Toh 1796, Toh 1797, Toh 1809, and Toh 1785, respectively.
! 24!
the root tantra.84 The colophon adds that he consolidated the branches of the Årya
tradition and supplemented them in accord with the views of father tantra, the class of
tantra to which the Guhyasamåja belongs. The later chapters of the dissertation will
focus more prominently upon the themes of citation, consolidation, and supplementation
within the tantric polemics of the body mandala debate.
In 1426, three years before founding Ngor monastery and one year before his first
trip to Mustang, Ngor chen composed Destroyer of the Proponents of Evil and
Dispelling Evil Views . These two texts or versions of a singular text, form the core
of Ngor chen’s engagement in the body mandala debate and therefore of this dissertation.
Dated texts produced at E Wam chos ldan include three texts from 1434 on the
Dåkårna√ava-mahåyoginî-tantra as well as two texts on the tradition of Saravavid
Vairocana from 1442. Rituals propitiating this form of Vairocana are commonly
associated with funerary rites.85 The title of one of these Vairocana texts, Eliminating
Impurities in the Sådhana of Completing the Saravavid-Mañ∂ala, suggests a tantric
polemical tone.86
Beyond the body mandala debate texts, Ngor chen composed one other text
explicitly on the topic of body mandala, his Commentary on the Ghantapa Body Mandala
Practice.87 The text is undated; therefore it is hard to say whether it builds upon or lays
the foundation for the body mandala debate. Nonetheless, it provides an example of
Ngor chen’s skill in navigating the relationships between different interpretations of this
tantric ritual practice. It also confirms his facility in working within the Cakrasaµvara
materials.
Unfortunately, very few of Mkhas grub’s writings are dated. The place of
composition tends to be the most specific data available for determining how a particular
text fits within the author’s career; most of the texts that do include a place of
composition were written either at Ri bo mdangs chen or at Ganden. Mkhas grub founded
Mdangs chen in 1412 and spent time there after Tsong kha pa’s death in 1419.88
Therefore we can roughly deduce that these texts were composed between 1412/1419 and
1424 when Mkhas grub arrived at Gyantse. We have devoted attention to the contested
circumstances surrounding Mkhas grub’s departure from Gyantse above and the
ambiguous implications they hold for his relationship to his patron as well as to his
alleged opponent Rong ston. As referenced above, he was involved in founding Dpal
‘khor sde chen under the patronage of Rab brtan kun bzang. During his time in Gyantse,
between approximately 1424 and 1428, Mkhas grub composed the text that is the focus
of this dissertation, The Ocean of Attainment . During this same period, he also
composed the polemical work referred to as the “Thousand Topics” [stong thun chen mo]
(TTC), translated and interpreted in Cabezón’s 1992 study, A Dose of Emptiness.
Cabezón explicitly identifies that work as polemical, as a defense of Tsong ka pa’s
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
84
See Gsang ‘dus dkyil ‘khor gyi lha tshogs rnam mchod pa’i cho ga mchod sprin rgya
mtsho.
85
See Pakhoutova 2014 on Vairocana’s role in funerary rituals.!
86
Kun rig kyi dkyil ‘khor yongs rdzogs kyi sgrub thabs sgrib pa rnam sel
87
Dril bu pa’i lus dkyil gyi bshad pa, collected Works Vol.4, pp. 733-64.
88
Cabezón 1992 p16 fn38 suggests that Mdangs chen may have served as a home-base
for Mkhas grub during his travels in western Tibet.
! 25!
interpretation of the Madhyamaka philosophical perspective. On this basis, we can
deduce that during his time at Gyantse, Mkhas grub composed at least two significant
polemical works, based in the tantric and philosophical genres respectively.
Most of the remaining texts whose colophons provide a site of composition were
composed at Ganden and therefore between 1431/2 when Mkhas assumed the abbacy and
his death in 1438. As referened above, Mkhas grub’s reply to Ngor chen’s text, the
Thunderbolt Wheel of Reply to Ngor , was composed at Ganden.89 Two other
potentially relevant works for the interpretation of Mkhas grub’s body mandala debate
texts were also produced at Ganden. The first deals with Some Difficult Points in the
Generation Stage of the Ghañ†apa Body Mandala.90 As remarked above, Ngor chen also
composed a text based upon this version of the body mandala practice. Mkhas grub’s
chapter on body mandala from his Ocean of Attainment begins with a quotation from
Ghantapa’s text. This quotation provides Mkhas grub with the opportunity to engage his
opponents on the implications of tantric ritual practices like body mandala for the status
of the human body in the world. In the process, he experiments with the relationships
between multiple varieties of Buddhist discourse. Therefore, his explorations of
Ghantapa’s transmission of the body mandala practice are part of this larger project.
Another text composed at Ganden, The Illuminating Lamp for Traversing the Paths and
Grounds of Mantra(naya) and Påramitå(naya), promises to provide insight into how
Mkhas grub articulates these relationships between discourses and more specifically
between the tantric and non-tantric approaches to Buddhist practice.91
There are a number of texts composed at “Upper Nyang in Gtsang” province
[gtsang nyang stod],” a somewhat imprecise location that might refer to almost any of the
monasteries he spent substantial time at. It is, therefore, more difficult to estimate a date
for these materials. Among them is the Reply to the Questions of the
Kalyånamîtra Kon ting gug !rî ba , the text from Mkhas grub’s Miscellaneous
Writings [thor bu] introduced above.92 This particular text seems to have been composed
after the Thunderbolt Wheel of Reply to Ngor , and therefore later in the 1431-38
span. Also among his undated compositions from “nyang stod” is one of many texts
Mkhas grub authored on the Hevajra system. The title, Dispelling Delusions regarding
the Hevajra Sådhana, indicates a tantric polemical tone.93 The colophon pays tribute to
his Lam ‘bras masters, Ye shes dpal and Buddha!rî, and specifically references the oral
instructions of the Sakyapa masters on the generation and completion stages of Hevajra
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
89
Phyin ci log gig tam gyi sbyor ba la ‘jug pa’i smra ba ngan pa rnam par ‘thag pa bstan
bcos gnam lcags ‘khor lo. Collected Works [zhol] Vol. 2, pp7-100. Or see modern
publication: Dgag lan phyogs bsgrigs . Ch'eng tu: Si khron Mi rigs Dpe skrun khan,
1997. Chinese colophon title: Pien lun wen hsuan pien. pp. 1-68.
90
Dril bu lus dkyil gyi byang du byas pa’i bskyed rim gyi bka’ gnas ‘ga’ zhig. Collected
Works [zhol] Vol. 6, pp.765-787.
91
Sngags dang pha rol tu phyin pa’i sa lam bsgrod tshul gsal sgron dang sbyar ba.
Collected Works [zhol] Vol 8, pp.533-554.
92
Dge ba’i bshes gnyen kon ting gug shri ba’i dris lan. Collected Works [zhol] Vol. 9,
pp. 775-808.!
93
Kyai rdo rje’i sgrub thabs ‘khrul spong. Collected Works [zhol] Vol. 8, pp.89-135
! 26!
practice as well as the three principal versions of the sådhana compiled by the Sakyapa
patriarchs. It also refers to the Vanjrapa•jara and Sampu†a tantras, the other two tantras
embaced by the Sakyapas as part of the Hevajra cycle. This colophon reminds us of
Mkhas grub’s Sakyapa roots while also asserting his purpose to fulfill Tsong kha pa’s
intent.
Cabezón 1992 has highlighted Mkhas grub’s talent for “synthetic” compositions
on the varied topics of valid cognition [pramåña], Madhyamaka philosophy, and tantra.94
Mkhas grub’s Fundamentals of Buddhist Tantra, translated by Lessing and Wayman in
1968, is one example of such a text, one that has played a formative role in shaping
Western understandings of Tibetan Buddhist tantra.95 He was also particularly prolific on
the Kålacakra system, with several volumes of commentary on the Vimålaprabhå.
However, despite the strong representation of tantric subject matter in Mkhas grub’s
collected writings, they do not serve as the main focus in the way that they do for Ngor
chen. Moreover, he appears to have been more concerned with propitiating Tsong kha
pa’s work than with cataloguing the major works of the Tibetan Buddhist canon.
V. The Fifteenth-Century Context: Monastic Patronage, Scholasticism,

and Charisma
For the purpose of establishing a context for the study which follows, the most essential
step is to outline the institutional climate of fifteenth-century Tibet together with the
scholastic and ritual contexts in which Ngor chen and Mkhas grub participated. The
fifteenth century has been somewhat neglected in historical scholarship on Tibet, perhaps
in part because it marks a period of intensive growth and change in the relationship of
political, monastic, and intellectual institutions.96 As a result of its transitional nature, the
fifteenth century has generally been characterized by the more easily defined periods it
follows and precedes. The previous century and a half is understood in terms of the
reform of the monastic system, the fall of the Mongol-Yuan dynasty, and the
solidification of the Tibetan Buddhist canon.97 On the other hand, the seventeenth
century is regarded as a time in which political power was centralized, the Geluk sect
rose to power, and encyclopedic projects in all domains of Tibetan life were initiated by
the Fifth Dalai Lama.
In the mid- fourteenth-century, the political hegemony of the Sakya sect was
disrupted by the ascent of Phag mo grug pa Ta’i sit u to power. “Phag mo gru was one of
the altogether thirteen myriarchies of the Yuan times’ administrative division of central
Tibet. Beginning with Byang chub rgyal mtshan, (1302-64, reign since 1342), its
administrator had inherited the power of the Sakyapa in central Tibet.”98 Ehrhard 2004
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
94
Cabezón 1992, p. 4
95
Rgyud sde spyi’i rnam par bzhag pa rgyas par bshad pa. Collected Works [zhol], Vol.8,
pp.443-630.!
96
Kapstein 2006 and Davidson 2005 are among the most useful historical references for
this period in Tibetan history. Others are Ary 2007/2015, Aris 1989, Wylie 2003 &
Jackson 1989.
97
It was also a time marked by attempts at organizing and standardizing Buddhist
iconography to reflect the connection of deities with particular texts in that canon.!
98
Schweiger 2009, p.315. For more on the Phag mo gru, see Petech 1990, p. 85-138.
! 27!
points out the generous patronage of Tsong kha pa’s monasteries by the fifth Phag mo
gru ruler, Grags pa rgyal mtshan.99 We have already discussed the complexities
surrounding Mkhas grub’s relationship to his Phag mo gru patron, Rab btan kun bzang,
ruler of Gyantse and its connection with larger patterns of patronage of emerging
Gelukpa institutions by Phag mo gru pa rulers. In 1434, strife arose amongst the
successors to the Phag mo gru legacy, destabilizing power relations in Dbus and Gtsang
for the next century, with the gradual rise of the Rin spungs family.100
Foreign relations continued to impact Tibetan religious life in meaningful ways in
the fifteenth century. Indian masters continued to travel to Tibet at this time, although
certainly with less regularity than in the early Gsar ma era.101 Ehrhard 2004 documents
the travels of one such master, Vanaratna (1384-1468). He suggests Ngor chen was
actually the first master Vanaratna encountered on his first trip to Tibet in 1426. This is
the very same year that Ngor chen composed his body mandala texts. It is hard to accept
as coincidental that Vanaratna proceeded on to Gyantse where he established a
relationship with Rong ston, Mkhas grub’s ‘opponent’ in the elusive debate referenced in
Chapter One of this dissertation. Moreover, although the nature of Tibetan-Chinese
relations shifted during the Ming Dynasty (1368-1644), important diplomatic connections
between religious masters and the court continued to be forged. For example, the Fifth
Karmapa visited the court of the Ming emperor Ch’eng-tsu in 1407-1408 to perform
funerary rites for the emperor’s parents. Sperling 2003 describes the visit as multi-
faceted:
“Political, religious and commercial activities all played a part his mission to the court of
Ming Ch’eng-tsu, and all were important in the relationship of Tibet and China during
this period. This new Ming-Tibetan relationship accorded with the fresh circumstances
of both countries following the collapse of Mongol power. It was also clearly distinct
from their previous relations. Thus, although the Ming circles harked back to the T’ang
for their theoretical lessons in Sino-Tibetan affairs, and although Ch’eng-tsu attempted to
imitate the Yuan dynasty’s ties with the Sa-skya-pa, early Ming Tibetan relations existed
upon a footing of their own amidst new circumstances for Tibet and China.”102
This brief outline attests to the complex socio-political circumstances framing the
body mandala debate. While the Sakyas retained some influence, the fifteenth and
sixteenth centuries were a period of uncertainty and change for the Sakyas in particular. It
was a period in which the internal organization of their institutions transformed and
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
99
Ehrhard 2004, p.247.
100
Ehrhard 2004, p.249.
101
See Ehrhard 2004, p.246. For another excellent resource on the visitation of Indian
masters to Tibet (and China) in the fifteenth century, see Mckeown’s 2010 dissertation on
the life of ¸ariputra (1335-1426).!
102
Sperling in McKay 2003 Vol.2, p.478. See also Sperling 2000, referenced by Dan
Martin on Tiblical [ https://sites.google.com/site/tiblical/home ], proper name index 1b
under “Go shri”. See Berger 2001 for a study of the miraculous events surrounding this
visit and their representation in visual sources.
! 28!
expanded and the dynamics of their relationship to other traditions shifted. Jose
Cabezón’s work on the polemical writings of Ngor chen’s student Go rams pa (1429-89)
provides a model for contextualizing Ngor chen and Mkhas grub’s ’s writings on the
body mandala practice within the larger history of Tibetan polemical literature and for
understanding its place within the very complex Sakya- Geluk relations of the fifteenth
century.103 For example, Mkhas grub’s offensive may be interpreted within the broader
context of the Geluk movement’s attempts at distinguishing themselves from the Sakyas,
to whom they were historically indebted. It was well known that Re mda' ba, one of
Tshong kha pa’s main teachers, was himself a prominent Sakyapa. Determining the tone
of such self-definition will be a guiding theme in approaching the texts of both Mkhas
grub and Ngor chen and orienting them within the fifteenth-century monastic climate.
Cabezón paints the following picture of this climate:
“In Tibet, as elsewhere, the success of new religious institutions depended upon a variety
of factors: spiritual, intellectual, economic, and of course, political. The financial support
of patrons was essential, but this, in turn, depended upon other factors: the charisma and
vision of the founding figure; the commitment, persistence, and intellectual abilities of
his successors; the public perception of the order’s monks; their perceived ability to enact
rituals that brought about the goals of patrons, and so forth. These were some of the
factors that attracted not only patrons but also prospective monks to newly founded
monasteries.”104
The relevance of these institutional dynamics to the primary texts upon which this
dissertation is based is one line of historical inquiry for this study.
As founder of Ngor, one of the main monasteries of the Sakya tradition, Ngor
chen certainly qualifies as such a charismatic individual. Through various diplomatic,
scholastic, and ritual means, Ngor chen solidified the connection between religious and
political institutions. He was also revered for spreading that tradition to Western Tibet
[Ngari]. For example, during his time in Mustang [Glo smon thang], he allegedly
increased the number of monks there from less than four to one thousand.105 He ordained
the ruler of Glo smon thang, A ma dpal, as well as the king of Gu ge.106 Glob bo mkhan
chen’s autobiography describes many examples of important institutional relationships
forged by Ngor chen at Mustang. For example: “Ngor chen erected the great
Cakrasaµvara palace...and consecrated it extensively. In connection with a ceremony on
that occasion, the granting from the great bla brang of the title (las ka) of “chief” [chen
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
103 !Cabezón 2007.!
104
Cabezón 2007, 42
105
Sangs rgyas phun tshog, 540.6. As mentioned above, for a description of Ngor chen’s
activities at Mustang, see the biography 536.5-539.2. For important background on the
region through the lens of oratory see Jackson 1984.
106
On Ngor chen’s ordination of the A ma dpal, see Kramer’s 2208 translation of Glo bo
mkhan chen’s autobiography, pp.145-6 as well as Sangs rgyas phun tshogs’ biography of
Ngor chen, 538.4-.5. On his ordination of the king of Guge, see Kramer 2008 p.148
together with her reference in fn27 to Vitali 1996, 508f.
! 29!
po] to the religious king [A ma dpal]...”107 Historical and biographical sources detail the
extensive support the royals of Mustang, A ma dpal and his sons, pledged to the sangha
there.108 This support was understood as mutually beneficial. For example, after
conferring a tantric empowerment upon A ma dpal’s sons, Ngor chen declared, “If we all,
donors and teachers, manage to remain in this commitment without contradicting it, then
[here], in the region of Glob bo, all religious and worldly deeds will be increased.”109
Ngor chen also ensured the continued political and religious relationship of Mustang and
Sakya by instituting a practice of sending monks from Mustang to Sakya and its affiliated
monasteries in Dbus and Gtsang for training.
In addition to his role as ambassador of the Sakya tradition, Ngor chen was an
accomplished ritual master, tantric commentator, and esteemed interpreter of the vinaya.
It is important to consider how polemical writings such as the body mandala debate,
executed just three years before he founded Ngor E wam chos ldan, may have further
increased Ngor chen’s prestige in the eyes of prospective donors and adherents. Heimbel
2012 reveals an important source of patronage for Ngor chen within the confusing details
of his father’s identity. Alhough a layman, Dpon tshang Grub pa yon tan, was popularly
regarded as his father, Ngor’s true father was known to be the chief lama at Sakya, Ta
dben kun dga’ rin chen (1339-1399) of the Gzhi thog blab rang. Heimbel remarks: “This
family relation was of great importance for Ngor chen’s monastic education and future
activities, since members of the Gzhi thog blab rang and one of its branches, the masters
residing at Chu mig estate, would later patronize his activities and act as the main donors
for the first abbots of Ngor.”110 The land on which Ngor chen built E wam chos ldan was
part of the Chu mig estate and virtually gifted to him by his half-brother, though Ngor
chen was determined to make the transaction official with a “formal payment.”111 Even a
surface evaluation of the situation suggests that patterns of patronage supporting Ngor
chen and Mkhas grub’s activities, in particular as monastic founders, differed. For
example, while Ngor chen’s main source of patronage stemmed from a familial
affiliation, Mkhas grub’s was based in the socio-political network of the Phag mo gru.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
107
Kramer 2008 translation, p147. Kramer refers the reader to Petech’s explanation of
this title, las ka, used during the Phag mo gru period, as the translation of a Mongolian
term indicating greatness, “eke.” See Kramer 147 fn 22 and Petech 1990, p. 117, fn 131.
108
See, for example, Kramer 2008 p.151 &153 and Sangs rgyas phun tshogs 539.1.
109
Kramer 2008 translation of Glo bo mkhan chen, p.153.!
110
Heimbel 2012, pp.47-8.
111
Jackson 2010, p.178. See also Sangs rgyas phun tshog, 527.6:
“Although Bdag chen chu mig pa had offered the land, when he (Ngor chen) arrived at
Chu mig he requested a letter of permission and offered a crystal bowl and so forth as
payment.”
bdag chen chu mig pas sa thams cad bde spyod du phul ba yin na'ang chu mig tu byon te
bka' shog zhus; shel gyi phor pa sogs sa rin du 'bul ba mdzad
Sangs rgyas phun tshogs [530.6] lists Ngor chen’s half-brother, Bdag chen chu mig pa
grags pa blo gros, first in his enumeration of Ngor chen’s disciples and patron.
! 30!
It’s also important to situate Ngor chen’s accomplishments within the broader
context of diverse charismatic Sakyapas of the fourteenth and fifteenth centuries. For
example, Bu ston [Bu ston rin chen grub] (1290-1364), Ren mda’ ba [Ren mda’ ba gzhon
nu blo gros] (1349-1412), Bo dong pa [Bo dong Pan chen phyogs las rnam rgyal] (1375-
1451), Rong ston [Rong ston shes bya kun rig] (1367-1449), and Ngor chen have all been
described as maintainers of their own traditions or rang rkang bzo pa.112 Sangs rgyas
phun tshogs specifies in his account of A ma dpal’s invitation of Ngor chen to Mustang,
that initially Bo dong Pan chen had been invited; however, Bo dong pa’s teachings were
ultimately deemed less efficacious.113 Such comments may be read as confirming Ngor
chen’s superior status among the great contemporary masters of the Sakya tradition
through illustrating his unique ability to transmit the tradition to a new region and secure
its continued flourishing there.114 Sangs rgyas phun tshogs makes an even more
significant effort to explain Rong ston’s respect for Ngor chen and to substantiate some
kind of relationship between them, a kind of “melding of minds” [thugs yid gcig tu 'dres
pa yin]. According to the biographer, Rong ston said, “Here in my monastery Nalendra,
there is the din and roaring (of ) the explanation of scripture and reasoning. If you strive
for the meaning of the mantric perspective, one goes to the new Ngor monastery.”115
This statement both asserts Ngor chen’s reputation for tantric mastery and explains his
relationship to one of the Sakya masters of the age through a division of areas of
expertise.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
112
I am grateful to Khenpo Tashi Dorje for illuminating this complex background of the
Sakypa tradition in our readings of Ngor chen’s biography in the Fall of 2011 at the IBA.
113
Sangs rgyas phun tshogs, 537.3.
Bo dong Pan chen phyogs las rnam rgyal is an interesting figure to consider in relation to
both Mkhas grub and Ngor chen. Rje tsun pa’s Secret Biography of Mkhas grub (6a-6b)
describes a debate between Mkhas grub and this master held at Ngam ring when the latter
was only 16. According to the account, Mkhas grub emerged victorious. See Ary 2007,
p. 225. Van der Kuijp 1985a mentions Bo dong together with Mkhas grub in the context
of Ngor chen’s polemical writings. See van der Kuijp 1985a, p.87. The context will be
addressed further below. According to Adams 2007 (treasuryoflives.org) Bo dong was
one of Mkhas grub’s teachers in Buddhist logic and philosophy. For more on the tradition
of Bo dong Pan chen phyogs las rnam rgyal, see Smith and schaeffer 2001 Chapter 14,
pp.179-208.!
114
In addition to the spread of the Ngor tradition to Western Tibet during Ngor chen’s
own time, ties were established between Ngor and the Sde dge royal family in the
fifteenth century; by the seventeenth century, these ties flourished. Heimbel 2012, pp.49-
50 and fn 9.
115
Sangs rgyas phun tshogs 531.4-.5:
rong ston thams cad mkhyen pa sha kya rgyal mtshan gyi zhal snga [531.5] nas; nga'i
dgon pa na len dra 'dir lung rigs kyi bshad pa'ur chem chem pa gyis; sngags phyogs don
du gnyer na ngor dgon gsar du song
For more on the Rong ston in Ngor chen’s biography, see 531.4-532.1.
! 31!
Davidson’s interprets Ngor chen’s more polemical writings within the framework
of fifteenth-century “Sakyapa apologetics.”116 This framework of interpretation helps to
elucidate the significance of Ngor chen and Mkhas grub’s exchanges as public
intellectuals within the context of fifteenth-century Tibet. As noted above, the Sakya
tradition was on somewhat shaky ground in the fifteenth century relative to its previous
flourishing in alliance with the Mongol-Yuan Dynasty (1271-1381). Cabezón, van der
Kuijp, Davidson, and Stearns have highlighted instances from as early as the twelfth
century in which the validity of the Sakya Hevajra Tantra-based teachings was called into
question.117 The long-term oral transmission of the Hevajra-based Lam ‘bras tradition
was often taken as cause to challenge its legitimate basis in Indian sources. One way of
combating such claims was to locate references to these teachings within the works of
Indian authors.118 Another was to produce representations verifying an unbroken lineage
of transmission of these teachings. Ngor chen produced both textual and visual
representations of this nature; he wrote texts documenting the transmission of the Lam
‘bras teachings and commissioned portraits of the Lam ‘bras lineage masters for
display.119
Cabezón places the challenges to Sakya authority within the larger history of
tantra in Tibet, backlashes against tantric practice in the later dissemination of Buddhism
there and in particular, repeated assaults on Nyingma tantric practices. The author also
shows how Ngor chen was not simply a defender of his own tradition, but also a critic of
others practices, namely Geluk (referred to in this period as Dga' ldan pa) practices
associated with Yamåntaka.120 Cabezón speculates that in that instance, “what was at
stake was not so much the authenticity of texts as their interpretations and ritual
enactment.”121 In other words, the controversy was focused upon the actual manner of
carrying out the practices described in the texts; the validity of the texts as testified in
Indian sources, commonly understood as the ultimate measure of authenticity in the
classification of Tibetan texts, was not at issue. This observation provides a model for
the kind of issues to be weighed in approaching Ngor chen and Mkhas grub’s body
mandala texts.
The Sakya tradition itself makes a distinction between the transmission of the
Lam ‘bras according to the “explanatory system” ['grel lugs] of scriptural exegesis based
in the Hevajra Tantra and the “oral instructions system” [man ngag lugs] whereby the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
116
Davidson 1991 has used this term most explicitly.
117
Cabezón 2007, 25; van der Kuijp 1985; Davidson 1991&1992; Stearns 2001.
118
Darpaña Åcårya was, for example, credited as such by Ngor chen for citing the
mahåsiddha Virüpa’s Rdo rje tshig rkang, the core text of the Lam ‘bras, in his
Kriyåsamuccaya. Stearns refers both to Ngor chen’s Lam ‘bras bu dang bcas pa’i man
ngag gi byung tshul gsung ngag rin po che bstan pa rgyas pa’i nyis ‘od, 110.3 and Mus
chen’s Dkon mchog rgyal mtshan’s Lam ‘bras bu dang bcas pa’i gnad kyi gsung sgros
zin bris, 448. (Stearns 2001, p 10 & fn15)
119
On Ngor chen’s artistic commissions, see Jackson 2010 pp.179-90.
120
Cabezón 2007, 25 and note 16. Cabezón cites Gu ru bkra shis’s Ngag dbang blo gros
Stag sgang mkhas mchog in Gu bkra’ichos ‘byung, p .992
121
! 32!
practices themselves are explained.122 Ngor chen’s polemical texts on visualization
practice seem to potentially bridge the emic gap created by this distinction, bringing
discourses on textual authority and on practice into the same textual arena.
Chapter Six of this dissertation focuses specifically upon issues of textual
authority. In his defense of the validity of the Hevajra body mandala practice in the
second part of his text, Ngor chen demonstrates the complementarity and intrinsic worth
of the genre of oral instruction [man ngag]. Demonstrating his skill in tantric exegesis,
Ngor chen synthesizes the tantras, their Indian commentaries, and the oral instructions of
the great Indian realized tantric masters [mahåsiddhas]. He depicts the relationship of
these genres as composite, forming a total system for interpreting the rites of the Hevajra
cycle. In analyzing Ngor chen’s defense, we will find that the body mandala debate is,
for him, just as much, if not more, concerned with working through such issues of textual
authority as its is with the mechanics of ritual practice or the status of the body itself. In
his study of Mkhas grub’s philosophical treatise, the Stong thun chen mo [TTC] Cabezón
observes a significant distinction in Mkhas grub’s perspective on textual authority:
“Mkhas grub rje is paradigmatic of the dGe lugs pa exegetes in having great disdain for
short and pithy teachings known as man ngag. The point he makes here he will make
again and again throughout the TTC, namely, that the way to a true understanding of
Buddhism is not through mystical oral tradition, passed down in secret from master to
disciple, but through long and arduous study and analysis of scriptures...”123
The genre of oral instructions plays such a definitive role in the Lam ‘bras
tradition, it is not surprising that this aspect of Mkhas grub’s perspective may have been
construed as a threat to the Sakya tradition at large or even a disavowal of his spiritual
heritage.
Davidson suggests that Ngor chen and Mkhas grub were also involved in another
‘debate’ over tantric materials. This exchange called into question whether the practice
of generating oneself as a deity [bdag bskyed] had a place in the kriyå and caryå tantric
systems.124 Davidson contextualizes this controversy within Ngor chen’s introductory
writings on these two lower classes of tantra. He summarizes the terms of the
controversy as a clash of perspectives in which Ngor chen supported the use of the lower
tantras for promoting the ethical and devotional qualities conducive to monasticism.125
Ngor chen did not acknowledge the existence of the practice of generating oneself as the
deity within the krîyå system; moreover, he regarded its implementation within the next
tantric class, the caryå, as useful exclusively for its role in promoting monastic discipline.
He regarded krîyå and caryå as suitable for monastic discipline because they lack the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
122
Davidson 1991, p.12.!
123
Cabezón 1992, p.418 fn53.
124
Davidson 1981, 86. In Note 17 he directs the reader to points in another edition of the
biography of Ngor chen by Sangs rgyas phun tshogs: Bsod nams rgya mtsho-Complete
Works, Vol 10, pp 149.4.4, 251.1.4, 267.4.6. He also refers to Wayman’68 pp163-71 for
more on this debate.
125
Davidson 1981, 86.
! 33!
sexualized content of the higher yoga tantras and promote an “emphasis on personal
purification, ritual endeavor, and devotion.”126 This was an important issue for Tsong kha
pa and, following his lead, for Mkhas grub.127 Mkhas grub (in what became the standard
Gelukpa position) included meditation upon oneself as a deity within all divisions of
tantra, including kriyå.128 Sangs rgyas phun tshogs’s biography of Ngor chen mentions
this controversy within the larger context of his polemical writings or “activities aimed at
clarifying the Buddha’s teachings through debate which corrects the erroneous views of
others”: 129
He (Ngor chen) disputed the sådhana of self- generation in the dharma terminology of
krîyå tantra itself; thus, he clearly taught the intention of the various manner of
commentary, increasing the teaching of the krîyå tantras. (Therefore) it is called the
Ocean of Excellent Explanation of the Establishment of Kriyå Tantra. Likewise, he
refuted the entrance of the wisdom beings in the caryå tantra, he clarified the teachings of
caryå tantra.130
Ngor chen wrote his treatises on the lower tantras in 1420.131 Therefore, they
post-date Tsong kha pa’s refusal of Ngor chen’s request for teachings during Ngor chen’s
visit to Ganden in 1413.132 Ngor chen wrote these treatises six years before his writings
on body mandala.133
This controversy over the nature of ritual practice based in the lower classes of
tantra highlights the ways in which concerns with the mechanics of visualization practice,
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
126
Sangs rgyas phun tshog, 547.1-2
127 !Hopkins 2009, pp. 303-318; Wayman 1968, pp.163-171.!
128
See Davidson 1981, p.86.
129
Sangs rgyas phun tshog, 546.2. See 546.2-547.2 on Ngor chen’s polemical writings.
Portions of this section have also been translated above.
130
Sangs rgyas phun tshog 547.1-.2: gzhan yang legs par bshad pa rtsom pa'i skor yang;
bla ma yi dam chos skyong rnams kyi bstod pa'i skor dang; bya ba'i rgyud rang rkang gi
chos skad la bdag bkyed sgrub pa sogs bkag nas 'grel tshul sna tshogs kyi [547.2] dgongs
pa gsal bar ston cing bya rgyud bstan pa rgyas par bya ba'i phyir; bya rgyud spyi'i rnam
par gzhag pa legs par bshad pa'i rgya mtsho zhes bya ba dang de bzhin du spyod rgyud la
ye shes pa 'jugs pa sogs bkag nas spyod rgyud bstan pa'i gsal byed
This excerpt is derived from the translations of portions of Ngor chen’s biography I
executed under the tutelage of Khenpo Tashi Dorje at the IBA in Kathmandu in the fall of
2011.
131
See Davidson 1981, 86. Davidson dates the introductory writings to 1420. Although
his dates seem more viable, in my own readings of the colophons of the Lamp of
Eloquent Explanation for Classifying the Krîyå Tantras and the Ocean of Eloquent
Explanation of the Caryå Tantras, I initially understood them to have been composed in
1396.
132
I am grateful to Khenpo Tashi Dorje for bringing my attention to this point. [Personal
communication, Fall 2011] See also Davidson 1981, p.84.
133
Davidson 1981, p.86
! 34!
doctrine, and monastic discipline intermingle in tantric polemics. This controversy is also
relevant to this dissertation because it presents another potential example of a “debate”
within the careers of these same two authors in which the practice of generating oneself
as a deity, a practice so central to definitions of tantra as a fast track to attaining
buddhahood in this lifetime, assumes a primary position.
Apparently, both Mkhas grub and Bo dong Pan chen phyogs las rnam rgyal had
taken issue with aspects of Ngor chen’s elaborations on the Hevajra sådhana practice in
his Moonrays of the Pith of the Generation Stage: The Extensive Explanation of the
Hevajra Sådhana [Dpal kyai rdo rje’i sgrub pa’i thabs gyi rgya cher bshad pa bskyed rim
gnad kyi zla zer], composed in 1419 (seven years before the body mandala texts).134 The
dynamics set in motion by these exchanges may have laid some of the groundwork for
the texts of the body mandala debate. Ngor chen’s student Go rams pa later composed a
text to dispel these particular objections.135
One final potential conflict between Ngor chen and Mkhas grub’s tantric
orientations is worthy of mention here. This conflict concerns the power of the tantric
path itself. Allegedly, while for Ngor chen, the mantranaya lead to a more potent
experience of buddhahood than the påramitånaya, for Mkhas grub both paths produced
the same result.136 The ramifications of this sort of ideological controversy are treated in
Chapter Four of this dissertation, with particular attention to the manner in which they
impact Mkhas grub’s strategy of argumentation.
We have evaluated the ways in which the body mandala debate fits within a
broader domain of polemical literature in fifteenth-century Tibet, and more specifically
polemical literature focused upon tantric subject matter. In the process, we have
considered the potential value of categories such as “sectarian differentiation” and
“Sakya apologetics” for better understanding the dynamics of the debate. We have also
investigated the ways in which this debate and its interpretation may have influenced the
patronage and prestige of its authors, Mkhas grub and Ngor chen, two charismatic
scholar-monks of their era. Remaining open to the possibility that sectarian conflict is just
one among many possible factors informing the body mandala debate, a factor that may
have been amplified over time, is one of the great challenges this dissertation grapples
with. We now proceed with a key question for approaching Ngor chen and Mkhas grub’s
texts, namely: “What is a Body Mandala?”
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
134
van der Kuijp 1985, p.87. Ngor chen’s Moonrays of the Pith is one of Davidson’s key
sources for his 1992 study of the Hevajra abhisamaya tradition.
135
This text is referenced by van der Kuijp 1985, p.87, Davidson 1981, p.88, Davidson
1991 fn60, and Davidson 1992 fn8. Mkhas grub and Go rams pa’s apparent philosophical
differences may have informed this round of tantric polemics as well. See Go rams pa’s
Illuminating the Pith: Dispelling Objections to the Moonrays of the Pith. Gnad gyi zla zer
la rtsod pa spong ba gnad kyi gsal byed.
136
Davidson 1991, p.20 and fn59. Davidson refers the reader to Mkhas grub’s Secret
Treasury of the Vajra-Dåkinîs Explanation of the Two Part (Hevajra Tantra). Dpal brtag
pa gnyis pa’i rnam par bshad pa rdo rje mkha’ ‘gro ma rnams kyi gsang ba’i mdzod,
especially pp.481-515. !
! 35!
Chapter Two: What is the Body Mandala?
I. Parsing and Purifying the Body: the Early Buddhist context
Buddhaghosa’s Visuddhimagga (The Path of Purification) provides valuable insight into

how learning was understood in early Buddhist contexts and, in particular, how
information was parsed and recited through enumeration, repetition, association, and re-
organization of details. While Buddhaghosa himself lived a millennium after
¸åkyamuni’s passing, his synthesis of sources and practices of the early tradition in his
writings provides insight into fundamental Buddhist approaches to knowing. The
Visuddhimagga, in particular, presents practices that suggest that to know the body is to
know the world. It shows how the memorization of details is regarded as producing a
spontaneous and radical change in one’s view of body and world as well as of their
relationship. Comparing these practices from the Visuddhimagga with themes found in
body mandala practice allows us to consider how such techniques for organizing and
internalizing information worked. Through this comparison, we can begin to see how
the toolbox of techniques for imaginatively breaking apart and re-asssembling the body
included powerful tools for achieving the soteriological goal of liberation.
In the Visuddhimagga, such practices of parsing the body are regarded as the
basis for all of the five paths of purification and therefore fundamental to soteriological
progress:
Now, the things classed as aggregates, bases, elements, faculties, truths, dependent
origination, etc., are the soil of this understanding, and the [first] two purifications,
namely, purification of virtue and purification of consciousness, are its roots, while the
five purifications, namely, purification of view, purification by overcoming doubt,
purification by knowledge and vision of what is the path and what is not the path,
purification by knowledge and vision of the way, and purification by knowledge and
vision, are the trunk. Consequently, one who is perfecting these should first fortify his
knowledge by learning and questioning about those things that are the “soil” after he has
perfected the two purifications that are the “roots,” then he can develop the five
purifications that are the “trunk.”137
One key theme linking this text to the body mandala is the correlation of microcosm and
macrocosm reinforced through practices of purifying the bodily elements. The trope of
construction by, identification with, and dissolution of the elements can be found in early
Buddhist literature such as Abhidharmic descriptions of the formation and destruction of
the cosmos. For example, Chapter Eleven of theVisuddhimagga describes such
destruction:
102. (d) Great alteration: the unclung-to and the clung-to are the [basis of] great
alterations. Herein, the great alteration of the unclung-to evidences itself in the
emergence of an aeon (see XIII.34), and that of the clung-to in the disturbance of
the elements [in the body]. For accordingly:
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
137 !Nåñamoli 2010 translation, Visuddhimagga 14.32.!
! 36!
The conflagration’s flame bursts up
Out of the ground and races higher
And higher, right to the Brahmå heaven,
When the world is burnt up by fire.
A whole world system measuring
One hundred thousand millions wide
Subsides, as with its furious waters
The flood dissolves the world beside.
One hundred thousand million leagues,
A whole world system’s broad extent
Is rent and scattered, when the world
Succumbs to the air element.
The bite of wooden-mouths can make
The body stiff; to all intent,
When roused is its earth element,
It might be gripped by such a snake.
The bite of rotten-mouths can make
The body rot; to all intent,
When roused its water element,
It might be gripped by such a snake. [368]
The bite of fiery-mouths can make
The body burn; to all intent,
When roused is its fire element,
It might be gripped by such a snake.
So they are great primaries (mahåbhüta) because they have become (bhüta) [the
basis of] great (mahant) alteration.138
Read in tandem with body mandala practice, we find a parallel between such
descriptions of the breakdown of the body in relation to cosmic destruction and tantric
descriptions of the dissolution of the body at death. The body mandala practice, in
particular its manifestation within the Guhyasamåja cycle, participates in ritual
technologies of harnessing the forces of nature present in the human body through an
understanding of how they function in relation to birth and death. Through the repeated
reenactment of these processes through visualization, the tantric practitioner strives to
transcend ordinary birth and death and to attain a Buddha body.
Chapter Eight of the Visuddhimagga describes practices of mindfulness of the
body; these practices include use techniques of enumerating parts and parsing the body
that may be understood in relation to body mandala practice. The goal of these practices
is to dis-aggregate one’s perception of one’s own body and to eliminate attachment both
to one’s own body and to the bodies of others. The meditator reflects on the thrity-two
parts of the human body through the sevenfold skills in learning: verbal recitation, mental
recitation, color, shape, direction, location, delimitation (i.e similarity and difference) and
then tenfold skills in attention. The sevenfold skills in learning are of particular interest
in that they employ mnemonic techniques of breaking down and organizing information
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
138
Nåñamoli 2010, Visuddhimagga 11.102
! 37!
in a very similar way to those found in mandala construction.139 The result of this
practice is described as follows:
“Then just as when a man with good sight is observing a garland of flowers of thirty-two
colours knotted on a single string and all the flowers become evident to him
simultaneously, so too, when the meditator observes the body thus, ‘There are in this
body head hairs,’ then all these things become evident to him, as it were,
simultaneously.”140
These practices use recitation and visualization to enumerate and parse the parts
of the human body and thereby alter one’s perception of one’s own body and its
relationship to the world. Through repetition of these techniques, one attains a certain
kind of mental flexibility by which such parsing of the body can begin to occur
automatically and even be applied as a strategy for combating attachment. Similarly,
within body mandala practice, the repeated visualization of the individual parts of the
body as parts of the celestial palace of the buddhas, of the body itself as a container
housing deities whose true nature is understood as that of the aggregates, elements, and
so forth, results in the ability to instantaneously generate and dissolve these forms. This
ability prepares the practitioner to acquire a Buddha body naturally or even automatically
and to ultimately abandon attachment to rebirth within saµsåra. Chapter Eleven of the
Visuddhimagga enhances the description of the body by introducing the forty-two aspects
of the elements used to parse the body, to break up and make it intelligible. Once again,
the focus is to reveal the true nature of things:
And just as the great creatures known as female spirits (yakkhinì) conceal
their own fearfulness with a pleasing colour, shape and gesture to deceive beings,
so too, these elements conceal each their own characteristic and function classed
as hardness, etc., by means of a pleasing skin colour of women’s and men’s
bodies, etc., and pleasing shapes of limbs and pleasing gestures of fingers, toes
and eyebrows, and they deceive simple people by concealing their own functions
and characteristics beginning with hardness and do not allow their individual
essences to be seen. Thus they are great primaries (mahåbhüta) in being equal to
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
139
Nåñamoli 2010, Visuddhimagga 8.60: “When the teacher tells the skill in learning in
seven ways thus, he should do so knowing that in certain suttas this meditation!subject!is!
expounded!from!the!point!of!view!of!repulsiveness!and!in!certain!suttas!from!the!
point!of!view!of!elements.!!For!in!the!Mahåsatipa††åna Sutta (DN 22) it is expounded
only as repulsiveness. In the Mahå Hatthipadopama Sutta (MN 28), in the Mahå
Råhulovåda Sutta (MN 62), and the Dhåtuvibhañga (MN 140, also Vibh 82), it is
expounded as elements. In the Kåyagatåsati Sutta (MN 119), however, four jhånas are
expounded with reference to one to whom it has appeared as a colour [kaßina] (see
III.107). Herein, it is an insight meditation subject that is expounded as elements and a
serenity meditation subject that is expounded as repulsiveness. Consequently, it is only
the serenity meditation subject [that is relevant] here.” (Nåñamoli 2010, pp239-40)
140 !Nåñamoli 2010, Visuddhimagga 8.139.!
! 38!
the great creatures (mahåbhüta), the female spirits, since they are deceivers.141
Therefore, this Buddhist practice allows the practitioner to probe beneath the
surface of the body to see how the great elements function within it. In breaking down the
nature of an independent self, one also revises one’s perception of the relationship of
internal and external worlds.142
II. Purifying and Deifying the Body in Non-Buddhist Tantra
Non-Buddhist tantric sources provide the most direct correlate for body mandala practice.
A key feature of this resemblance occurs in ritual techniques for purifying the elements of
the human body known as bhüta!uddhi. In his discussion of the Vaißñava tantric
tradition of Påncaråtra, Gavin Flood describes the bhüta!uddhi and its role in forging the
microcosm-macrocosm relation, articulating levels of gross and subtle that provide the
hierarchy for processes of creation and destruction. He relates these practices of mapping
cosmic structures onto the body within the larger domain of tantric practice:
“...cosmology has a primarily ritual function in these traditions. This can be illustrated
particularly well in the bhüta!uddhi sequence where the cosmos is mapped on to the body
and dissolved, as the lower levels of the cosmos are dissolved into the higher during the
cosmic dissolution (pralaya). The terminology here is that of the tattvas of Såmkhya in
which the gross elements (bhüta) that comprise the physical world are dissolved into the
subtle elements (tanmåtra) that are their source. The purification of the body through
dissolving its constituent elements into their cause would seem to be a characteristically
tantric practice.”143
The correlation of body and cosmos and the parsing and ranking of the qualities
of existence we saw within the Abhidharmic genre are present here. However, tantric
techniques more explicitly harness the dynamics of creation and destruction, locating
them within the human body itself through visualizing the location and dissolution of the
elements. Flood looks closely at the structure of ritual practice in the Jayåkhya saµhita, a
text produced sometime between the seventh and tenth centuries in Kashmir as organized
by a fourfold ritual sequence he sees reiterated across tantric systems: purification of the
place and body, divinization of the body, inner or mental worship followed by external
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
141 !Nåñamoli 2010, 11.100
142 !It is interesting that the elusive and deceptive nature of the elements is compared to
that of the yakßînis or female spirits. Certainly there are numerous accounts of the
deceptive nature of the female form in Buddhist literature. In such instances, the female
body often becomes the signifier for embodiment more largely as well as for the lust
associated with it. [See Wilson 1996] In some of the accounts of the tantric body mandala
practice to be addressed below, the goddesses are correlated with the elements. It is
possible that this correlation draws upon such earlier notions of female spirits as
concealing the form of the elements.
!
143 !Flood 2006, p109, see fn 42.!
! 39!
worship.144 Ideally, this process is part of the daily practice of an initiate who has
previously undergone branding with Vißñu’s symbol, the discus (cakra) and received a
special ritual name and a mantra. Buddhist tantra shares a similar system of pre-requisite
initiations and practices.
Chapter Ten of this text describes the purification of the body, preceded by
purification of the ritual space as the prerequisite for worship of the deity:
“Through symbolically destroying the physical or gross body, the adept can create a pure,
divinized body (divyadeha) with which to offer worship to the deities of his system. He
does this first only in imagination and second in the physical world, for as in all tantric
systems-only a god can worship a god...” 145
The tantric ritual logic of becoming a god to worship a god is, likewise, fundamental to
understanding interpretations of body mandala practice as a means of breaking down the
gross body to access its most subtle potentialities. Themes of purification and protection
are instrumental in facilitating this transition. In the Vaißñava tradition presented by
Flood, the process of purification involves associating each of the five elements (earth,
water, fire, air and space) with a shape, inhaling the element into a particular region of
the body, “dissolving it in its mantra, then into its subtle cause, and exhaling it... ”146 The
exhalation is accompanied by association with a sense faculty. So in the case of the earth
element, one inhales a yellow four-sided shape marked with a thunder emblem and linked
with the five sense objects into the space from the knees to the soles of the feet, dissolves
it into its mantra, and exhales the sense of smell. After completing a similar practice for
the other elements, the practitioner imagines the body burning from the bottom up and
the ashes scattering in the four directions. In this way, the practitioner has aligned himself
with the subtle body and begins the process of becoming divine.
The main technique for divinizing the body is a process known as nyåsa; the
practitioner applyies mantras to the body through touch and recitation focused on
particular bodily locations beginning with the hands. The initial purification of the
hands, often identified as hasta-püjå-vidhi, can be found in many tantric rites; for
example, tantric Buddhist ritual specialist, the vajråcårya, must consecrate his hands as
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
144
Flood 2006, 121.
145 !Flood 2006, 110. Sanderson 1986 fn 30 provides some textual sources for this
concept of becoming a god to worship a god: JS 12.1; MTV II (M®gendratantrav®††i) 29,

9-10; SvT (Svacchandratantra) 2.55 ab.
This notion of divine transformation is not necessarily exclusive to tantric ritual
contexts. The idea of the interpenetration of divine and human bodies in the Vedic
context was explored in a paper entitled “Ritual Devices to Become a God in Vedic and
Post-Vedic Rituals” presented by Shingo Einoo at the conference on “The Evolution of
Tantric Ritual” held at UC Berkeley in March 2014. The Aiteryopanißad 1.2.4 describes
deities entering the body throught the mouth, nostrils and eyes. I am grateful to
Alexander von Rospatt for this reference.
! !
146
Flood 2006, 112.
! 40!
ritual implements before engaging in ritual action.147 The Jayåkhya describes how the
practitioner then redistributes the same mantras associated with the same deities “on the
head, eyes, ears, mouth, shoulders, hands (again), buttocks, heart, back, navel, hips, knees
and feet.”148 The bodily points articulated here seem oriented around external points,
with the exception of the heart and perhaps the navel, as well as around the sensory
orifices. The process is one of armouring the body with mantra, culminating in the
distribution of the mantra of Nåråyaña across the entire span of the body. Touch, sound
and vision all figure in this process. Below, we will see how such practices of
purification and protection as described in these first two phases of the Jayåkhya appear
in early Buddhist tantric literature and are elaborated in the generation stage practices of
the body mandala.
In the next phase, that of inner, or, more appropriately, “mental worship”
[manasayåga], the practitioner visualizes the deities of the universe located in the
practitioner’s own body in the region between the genitals and the heart.149 These levels,
with the earth at its basis, located on the penis, and the culmination, the throne of Vißnu
at the heart, are correlated with the tattvas of Såµkhya philosophy. This dual process of
correlation effected through mapping of structures onto the body and then associating
those structures with philosophical ideals or aspects of existence or liberation is a trend
we find repeated in Buddhist tantric literature. These dense acts of encoding are what
Flood describes as “entextualisation of the body.” Such acts seem to reach their pinnacle
in body mandala practice, in which the structures of ritual, cosmos, and even society are
inscribed upon the body. Similar acts of encoding can be found even in the literature of
stüpa construction; each tier and form of the stüpa may be correlated with the layers of
the cosmos, a portion of the body of the Buddha as well as with a philosophical principal
of enlightenment.150 The final phase of the rite described by Flood is termed “external
worship” [Skt. båhyayåga]. Having appropriated the divine and soteriological structure
through the medium of the body, and thereby in a sense, ‘become a god’ in the
manasayåga, the “internal or mental worship,” the practitioner is now truly prepared to
worship. Body and environment have been purified or reconstructed, and the practitioner
constructs a mandala to house the deity. The repeated navigation and purification of
imaginary and real/material environments plays out in different ways in body mandala
practice. The body mandala debate, in particular, probes these boundaries between
varieties of internal and external mandala, between mandala that are imagined and those
that are represented in material form. As the dissertation progresses, we will have the
opportunity to think critically about the relationship between representation, reality, and
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
147
On this ritual purification of the hands, see Meisezahl 1985. !
148 !Flood 2006, 114.!
149
Flood 2006, 116
150
For earlier Indian examples, see Snodgrass 1992. For somewhat later examples
connected with tantric ritual, see two ninth to eleventh century Tibetan manuscripts from
the library cave at Dunhuang, IOL TIB J 338 and IOL TIB J 435, described by Dalton
and van Schaik 2006 pp67-68 and 178-179 respectively. In regard to the correlation
between stupa and Buddha body, Brauen 1997 (p.127 fn45) refers to Bénisti 1960 & Mus
1933. See also Snodgrass 1992, p.361.
! 41!
materiality. Like “inner” and “outer,” “real” and “imagined” are not straightforward
dualities in the ritual and philosophical contexts this dissertation engages with. Despite
the differences with the Vaißñava practice described by Flood, the comparison suggests
that body mandala shares an emphasis upon prerequisite deification and purification of
the body with non-Buddhist tantric ritual.
Alexis Sanderson describes a similar practice from the Kashmiri Trika tradition as
elaborated in the Trikasadbhåva Tantra.151 In this text, the structure imposed upon or
realized within the body during the phase of “installation of the mandala in internal
sensation (pråña˙)” is one of trident as throne extending from the genital region to a site
above the top of the head known as the tri!ulåbjamañ∂ala. [See Fig.9] All of the
structural correlations are focused upon the head and the area just above it. The mapping
of the mandala onto the body in this practice emphasizes vertical hierarchy and the
primacy of the central channel. The trunk of the body is left virtually vacant in this
account, marked only by the axis of the central channel. Most notable, perhaps, is the
lack of any demarcation of the heart center; in the Vaißñava tantric practice described by
Flood, the heart served as the pinnacle, the throne of Vißnu. Moreover, this Trika
practice elaborates upon subtle body practices of manipulation of the breath and inner fire
that within the Buddhist tantric context would sometimes be classified within the
repertory of the completion stage. The mandala is installed within “the level of internal
sensation” [pråña˙]:
“Internal sensation is reached when this oscillation of the breath has become so faint that
its two movements are fused in a subtle, pulsating point of quintessential vitality
(samåna˙), in the ‘I’ as it subsists in dreamless but blissful sleep (savedyaµ saußuptam,
pråñasaußuptam). If the practice is sustained at this stage, then awareness penetrates the
sensationless void (apavedyam saußuptam. !ünyasaußuptam) in its core and passing
through this final barrier enters the pulsation of autonomous consciousness (!akta˙
spanda˙). The fused breath (samåna˙) is totally fused into the ‘fire’ of the ‘rising breath’
(agni˙, udåna˙) blazes up from below the navel. Devouring all duality it ascends through
a central, vertical channel (sußiramårgena), penetrating the cranial ‘aperture of Brahmå’
(brahmarandhram, kakham) to culminate as ¸iva consciousness (=vyåna˙) at a point
twelve finger spaces (c. 20 to 25 cm) directly above it (dvåda!åntam, ¨rdhvakuñ∂dalinî,
nådhyådhåra˙).” 152
In beginning to approach the body mandala practices of the Guhyasamåja and in

particular the evolution of completion stage practices of higher Buddhist tantra,
comparison with such ritual acts of breath manipulation and “penetration of the void” as
presented here in the first part of Sanderson’s description is instructive. Such practices
resemble Buddhist tantric practices of drawing the winds into the central channel and
consequently immersing the subtle body in various degrees of clear light (correlated with
the attainment of respective degrees of wisdom). In the Guhyasamåja practices, the site of
the heart functions as the key orifice by which emptiness is experienced. Unlike terms
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
151
See Sanderson 1986. For another account of ¸aivite ritual deification, see Davis 2000,
especially pp.83-111.
152 !Sanderson 1986, pp. 177-8.!
! 42!
like “vital point” or “energy center,” terms like “orifice” or “aperture” help to emphasize
the heart’s role as a point of exchange in a key ritual moment of emission or dissolution.
The overall goal of many Guhyasamåja practices is to untie the karmic knots obstructing
the heart, and to ultimately dissolve the bodily winds there. However, in the ¸aiva
context described here by Sanderson, the crown of the head is the aperture through which
liberation is accessed in the manipulation of subtle energies.153
In the mapping of structures onto the body in these two examples, one non-
Buddhist and one Buddhist, subtle distinctions may be made in perceptions of hierarchy,
centrality, and the location of vital points for enacting transformation. Themes of
verticality and centrality are hierarchies shared by both examples, however, this
particular rendition of Trika leaves the central axis bare, unmarked by any vital point at
the heart. Verticality and centrality are two hierarchical standards that the mandala is
capable of translating simultaneously. In other words, as a two-dimensional
representation of a three dimensional form, zenith, center, and nadir are all positioned
along a central axis.154 While in the case of the mandala palace, one might say centrality
is the dominant hierarchical standard, with deities organized according to the relationship
of center and periphery, introducing the body into the equation perhaps introduces a
vertical standard of purity. Comparison of the tri!ulåbja-mandala used in internal
worship with that used in external worship supports this as well. [See Fig. 10] In
comparing different elaborations of the practice of mapping the mandala onto the body,
we will find variations in the way the key points of the body are understood. In looking at
different versions of this practice in the tantric Buddhist context, we will make note of
different expressions of hierarchy and relationship we find. We will begin with the
curious case of a body mandala described in a text from Dunhuang.
III. Proto- Body Mandala: Inscribing the body in a text from Dunhuang
IOL Tib J 576 is a Tibetan tantric text discovered in Cave 17, otherwise known as the
“library cave,” at Dunhuang. It dates no later than the eleventh century, the time when the
caves were sealed; it is among the manuscripts rediscovered in the twentieth century,
procured by Aurel Stein, and eventually transferred to the collection of the British
Library. This text maps the deities of the Vajradhåtu mandala, a mandala based upon the
tantric root text, the Sarva-tathågatha-tattva-saµgraha [STTS], onto points on the surface
of the human body.
As with many Dunhuang texts, the beginning of IOL Tib J 576 is missing, and the
order of the folia has been a bit confused. The manuscript is made up of several short
texts, most of which Dalton has discerned to be at least marginally related to the
Vajradhåtu mandala from the STTS.155 The text described here, IOL Tib J 576/1 is the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
153 !This is not to deny the central symbolic importance of the heart in the ¸aivite context,
as demonstrated by Muller-Ortega 1989.!

154
I am grateful to Steve Jenkins for reminding me of this aspect of the mandala’s
structure.
155
Dalton and van Schaik, 2006, pp.275-8. Dalton suggests that the five texts that make
up this manuscript may be a ritual manual for initiation practices in the yoga tantras.
However, one particular detail he mentions in describing the second text in the
! 43!
first of these. The folia discussed here are to be read in the following order:
1,3,4,2,5,6,7,8.156
Of course, while themes such as armoring and protection provide us with clues as
to what the precise ritual context for this text is, several aspects remain ambiguous. It is
likely that this text forms a part of a larger mandala initiation rite, in which the ritual
specialist empowers the practitioner within mantras and practices of the Vajradhåtu
mandala. This text presents an opportunity to reflect upon how the envisaged mandala
may be understood in relation to the body. As we will see in examining the Guhyasamåja
body mandala in the next chapter, the mode of mapping we find in IOL Tib J 576 is far
from standardized and therefore suggests an experimental or transitional phase in the
evolution of both the body mandala practice specifically and mandala rites more
generally. Perhaps most importantly, it presents the question of what makes a body
mandala.
In its surviving form, this first text within IOL Tib J 576 can be divided into three
parts: folia 1.1-.5 & 3.1-.2, 3.2-.5, devoid of interlinear commentary and mostly
providing protective mantras and allusions to emptiness; folia 2.1-.5 & 4.1-.3 folios, our
prime focus as it maps different deities of the Vajradhåtu mandala (and then some) onto
the form of the human body and is filled with interlinear commentary; and finally folia
sides 6-8 where the visualization of wrathful deities is described augmented, once again,
by interlinear commentary.
In its surviving form, IOL Tib J 576 begins with a description of a particular kind
of armor, an armor which “conquers all the armies of signs in battle, resting victorious
among all.” [1.1] The Sanskrit mantra which follows, “va jra ka va ca ka va ca ya man
hang”[1.1-1.2] reminds the reader that this armor is the vajra kavaca or “vajra armor.”
As we will see throughout this text, the Sanskrit mantra is transliterated into Tibetan
script. Mantras in this portion of the text, such as this one, are transliterated but remain
untranslated. Yet, we will continue to see ways in which the mantras are connected with
the ritual functions of protection the text describes. We can therefore begin to imagine
how ritual gestures may have been performed in this context in conjunction with the
recitation of mantra and how this mutual performance would trigger the memory of such
protective functions in the mind of the practitioner in future. In short, we can imagine
how even a ritual practitioner with no knowledge of Sanskrit would hear “va jra ka va ca”
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
manuscript (not described here), the reference to the “supreme bliss of union” [sbyor ba’i
bde ba mchog rnams], together with some of the details from our text suggests some
elements associated with higher yoga tantra.
Dalton suggests further comparison of the text with IOL Tib J 495 and Pelliot
tibétain 292. (Dalton and van Schaik 2006, p276).
156
Dalton and van Schaik 2006 [xx] draw attention to the “disordered state” of the
manuscripts as one of the major deterrents for scholars wishing to studying the materials
from the library cave. Although the catalogue provides helpful insights into where each
text begins and ends, I am grateful to Jake Dalton for helping me to make sense of their
internal order in our readings together in 2009-2010. The numbers I use here refer to the
folia sides for simplicity rather than the recto and verso numbers.!
! 44!
and recall a gesture, context, or accompanying oral explanation of the function of
protection or armoring.157
The text then continues in this vein to describe the qualities of this vajra armor:
“Untouched by marks, not abiding, undivided, unconditioned; because (it is) protected
through the dharma-dhåtu, (it is) not even perceived by the Buddhas.” [1.2-1.3] Another
mantra follows: “aoum va jra ra kßa ka ra kßa ka man hung bhang hang,” evoking the
“vajra rakßaka” or “vajra protector.” Finally, the nature of the protection being offered is
clarified: “In the momentary sign, phenomenon, reality does not abide. The momentary
wisdom cakra dispels the obstacles, the corrupting influence of discursive thought.” [1.3-
1.4] The reader then encounters yet another mantra, “aoum vajra pra ka ra pra ka ra hung
phat” eliciting the creation of the “vajra fence,” the “vajra pråkåra,” as a protective
boundary. The mantra, “aoum pra pa ÷ca ra ya; pa ÷ca ra ya hung phat,” follows.
“Prapa÷ca” may be used here in the sense of appearance or false imaginings, expansion
or elaboration. This is followed by what appears to be another mantra calling upon the
“vajra wrath,” [vajra krodha] to perform a series of tasks (to slay, to run, to cook and to
go), all issued in the imperative form. [1.5 & 3.1]
This first portion of the text we have clearly conveys the vajra armor to be the
power of mantra itself to defeat the imputations of conceptual thought, thereby
associating it with the nature of emptiness. Armoring, protection, fencing off, and illusion
or imagination are all evoked through the ritual performance of mantras. The mantra
becomes the sign that defeats all other signs. By partaking in the semiotic world, it
defeats it. Invested with this deep sense of paradox, the mantra, the momentary sign, is
the hallmark of impermanence.158 As such, it functions as the ultimate apotropaic,
driving away all obstacles.
The core part of the text is prefaced by the following statement: “The so-called
Buddhas of the five families are the image159 of the five wisdoms of the Buddhas of the
ten directions and the three times. They continuously abide, not passing away into
nirvana.” [3.1-.2] At this point, at the end of line two on folio 3, the text changes modes.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
157
For important discussions of the connections between mantra and ritual performance,
see Patton 2005 and Staal 2008 pp.191-221.
158
Janet Gyatso’s 1992 essay provides a Peircean reading of a late nineteenth/early
twentieth-century Tibetan text on the use of “literal formulas” known as dhårañî in
Buddhist visualization practice. Gyatso presents a fruitful assessment of the somewhat
paradoxical role of syllables and phrases in Buddhist practice as a means of defeating the
cognitive bonds of language. It is difficult to differentiate definitions of dhårañî from
those of mantra as they generally share many of the same qualities of unmodified form,
resistance to translation and emphasis upon use vs. semantic meaning. Themes of their
role in imprinting texts and experiences upon the memory through ‘holding them fast’ in
a visible and oral form are integral to definitions of dhårañî, and their protective functions
are key. Later in this chapter, we will return briefly to the dhårañî and their apotropaic
functions to consider some inscribed example from Dunhuang. See Fig. 19-21.
159
Gzugs bsnyan can also be translated as “reflection.”
! 45!
This shift is manifested visually by an increased “density”160 in the text, filled as it is by
interlinear commentary and mantra. For the next two folia or so, the primary text is
composed of pairs of Sanskrit mantras, transliterated into Tibetan script.
At this juncture, we should pause for a moment to reinforce the connection
between inscribing the text and inscribing the body on a larger level. Tantric ritual
manuals such as this one are themselves a form of commentary, a ritual commentary that
augments and transforms the tantric root text (here the STTS) upon which it is loosely
based. The addition of mantras for recitation and of visual and performative strategies all
work together to orally, visually, and physically relocate the narrative action of the tantra
to the body of the practitioner, or rather, the bodies of initiate and ritual specialist.
Therefore, on a certain level, the interlinear commentary we see in IOL Tib J 576 is
merely another layer of inscription upon the root text of the tantra itself (not seen here).
In the root text, each of the bodhisattvas is summoned into the mandala and consecrated
through the recitation of their name; the consecration is finalized through their recitation
of a mantra. However, in IOL Tib J 576, the manner in which the mandala is built is
somewhat different. The drama of the tantra is transposed to the human body.
Here, there are five sets of five pairs of mantras that make up the mandala. See Fig.11
for a visual map of this portion of together with a numbered deity list based on IOL Tib J
576; compare this with Fig. 12, a more standardized representation of the Vajradhåtu
mandala. Each pair of Sanskrit mantras (transliterated into Tibetan script) is
accompanied by one line of interlinear text that supplies a location on the body where
two deities are positioned. The names of these deities are translated into Tibetan. The
five Buddha families are mapped onto the head, two hands and two feet in a manner akin
to the five Buddha-families of the Vajradhåtu mandala. The resemblance is due primarily
to the order in which the deities are presented and their grouping into retinues centered on
the five jinas or main Buddhas. The connection of this mandala with the Vajradhåtu is
further substantiated by the mantra appearing on folio 2.5: aoum ba dzra dha tu ma 'da la
si ti hung; the corrected Sanskrit version would be: om va jra dhå tu mañ ∂a la sid dhi
hung. Thus the reader is encouraged to recall the siddhis or powers to be attained in
connection with the practices related to this mandala.
The five jinas are associated with the head, the middle fingers of the hands, and
the middle toes of the feet; each is accompanied by a goddess. Their association with the
directions in the Vajradhåtu mandala is not referenced. Therefore, Vairocana, who
resides in the centre of the standard Vajradhåtu mandala, here inhabits the head, likely at
the crown, together with Vajradhåtvî!varî. Akßobhya, typically in eastern direction of the
mandala, resides on the middle finger of the right hand, together with his Buddha Locanå.
Ratnasambhava, usually an inhabitant of the Southern direction of the mandala is
associated here with the middle finger of the left hand, together with Måmakî; Amitabha,
usually in the West, with the middle toe of the left foot, together with Påñ∂aravåsinî;
Amoghasiddhi, typically in the north, with the middle toe of the left foot together with his
consort Samayatårå. So something of the centrality of the main Buddhas of the mandala
is communicated through their association with these respective parts of the body,
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
160
Goodman in his account of ‘notational systems’ and ‘qualities of the aesthetic’
describes as increased “density” in which a visual surface can be characterized by a “lack
of differentiation” amongst its parts. See Goodman, 1976.!
! 46!
namely aligning the main Buddha of each family with a central appendage. By
‘centrality,’ I am referring to the logic by which the mandala’s form evolves over time
into a space which negotiates between a core and a periphery as articulated by the
arraying of lesser deities around a central Buddha image.161 Therefore, the association of
centrality with primacy which comes to typify the structure of the mandala is asserted
here in the context of the body, albeit in a different form.
However, this is the only instance I have encountered of mapping the five families
onto the five appendages of the body. We will see in the next chapter how there are
many instances of mapping other deities onto the limbs, fingers, sense orifices, and so
forth. In many cases, the crown is the primary site and maintains its associations with
Vairocana as head of the tathågata family. Some scholars identify the STTS as the tantric
system in which the shift from a three-Buddha system to a five-Buddha system occurs
(Ratnasambhava and Amoghasiddhi being the additional jinas).162 In establishing the
deities within the mandala, the root text of the STTS itself does not appear to explicitly
set forth the colors, directions, or attributes of the Buddhas and bodhisattvas. The
canonization of these associations is itself likely produced over time through
experimentation with various ritual commentaries and visual forms. In the pivotal
transition from a three to five Buddha family structure that begins to emerge in the STTS,
we encounter many non-standardized and experimental visual representations of
mandala. For example, the frequently referenced tenth-century scroll painting of the five
jinas from Dunhuang shows a yellow Vairocana (usually white) in the center with the
four Buddhas/jinas in the corners surrounding him [Fig.13]. The colors, attributes, and
positions of these Budhhas all contain non-standard elements.163 The mandala itself is
not round, but rather fits the definition of mandala as an assembly of deities in the
relation of central deity and retinue as described by Christian Luczanits.164 Perhaps we
may think about the model presented here in a similar way, as a mode of experimenting
with a five family structure in relation to the body where the most obvious pentad is that
of the protuberances of head and limbs.
An even more relevant comparison might be made with another painted mandala
from Dunhuang in the collection of the Musée Guimet; in this painting, the five buddhas
sit alongside their consorts. [See Fig.14] The inclusion of goddesses or consorts to
accompany the buddhas in IOL Tib J 576 is anomalous; it indicates that it is a transitional
text, a text absorbing details and practices from other yoga and even higher yoga tantric
practices. To my knowledge, the goddesses accompanying the main Buddhas here do not
appear in the Sarva-tathågatha-tattva-saµgraha. Locanå, Måmakî, Påñ∂aravåsinî, and
Tårå are the four goddesses of the intermediary directions in both the Sarva-durgati-
pari!odhana mandala and Guhyasamåja mandala. There are also some inconsistencies in
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
161
See Luczanits 2008 for an important introduction to the evolution of the form and
structure of early mandala. Luczanits’ essay deals specifically with many of the drawings
and paintings from the library cave at Dunhuang for examples that challenge the standard
definitions of mandala, definitions based in later fixed iconographic standards.
162
See Snellgrove 1987 (2002 Reprint) pp198-213 and Luczanits 2013.
163
I am grateful to have studied this mandala painting under the tutelage of Patricia
Berger and Christian Luczanits in Spring 2011 at UC Berkeley.!
164
See Luczanits 2008, especially p.113-115. !
! 47!
IOL Tib J 576 in the rendering of their names from Sanskrit into Tibetan. Most
significantly perhaps, the Sanskrit Måmakî is translated within the interlinear
commentary as “jewel-eye”(rin cen spyan), what appears to be a reference to the
association with the jewel family and to the goddess Locanå.165
The case of the relationship of Måmakî and Locanå has been assessed in detail by
Roger Wright 2010 in the context of Någårjuna’s Piñ∂ik®ta-sådhana on the
Guhyasamåja.166 In this sådhana, the position of the two goddesses is reversed. Wright
discounts Tsuda’s theory that the authors of the sådhana misunderstood the tantra and
questions his assertion of the relevance of links between the Sarva-tathågata-tattva-
saµgraha and the Guhyasamåja. Instead, he points to the inclusion of the goddesses
Locanå, Måmakî, Påñ∂aravåsinî, and Tårå in an eighth-century translation of the Sarva-
durgati-pari!odhana-tantra in place of Vajralåsyå and the other inner offering goddesses
(shared by STTS & SDPT).167 The inner offering goddesses are then relegated to the
next level of the mandala where they reside alongside the outer offering goddesses. The
author concludes that the source for this set of four goddesses (Locanå, Måmakî,
Påñ∂aravåsinî and Tårå) is therefore the SDPT rather than the STTS. As for the reversal
of the positions of Locanå and Måmakî, Wright suggests that the “promotion of
Akßobhya” to the center of the mandala occurring in the Årya Guhyasamåja mandala is
the reason, thereby reversing the positions of Akßobhya and Vairocana.168 What all this
suggests is that ambiguities in both names and positions of the goddesses may be one
sign of the transitional and experimental nature of IOL Tib J 576 . We will see how the
transitions and influences operating here suggest not only a dialogue between the
practices of different yoga tantras, but also with those of higher yoga tantra as well.
Vairocana’s retinue is organized within the space of the head with the four
påramitås, goddesses embodying spiritual perfections, placed on the eyes, nose, ears and
tongue. We will see how in the more developed versions of body mandala practice, the
sense doors are key sites around which single and even pairs of deities may be positioned.
Locating deities at these sites indicates the need for protection of the doors to the senses;
it may even represent the transformation of sense perception itself into a divine and non-
dual form that casts off the distinction of subject and object binding us to the cycle of
rebirth. One anomaly in the enumeration and distribution of the påramitås here in IOL
Tib J 576, however, is that of gender. Namely, there is no obvious marker of female
gender in either the transliteration of the Sanskrit (less surprising) or in its Tibetan
translation. Furthermore, the Tibetan commentary pairs these figures with four
seemingly un-standardized goddesses. The primary lines of text in IOL TIB J 576 pair the
mantras of the four påramitås with the same mantra: “Aoum vajra muß ti hum.” The
vajra mußti is the vajra fist, a particular position of the hands. The reason for introducing
the term here is unclear. It seems likely that it refers either to a ritual gesture and/or pose
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
165 !Mallmann states that Måmakî can be associated with either Akßobhya or
Ratnasambhava.!
166
See Wright 2010, pp.38-41
167
That particular manuscript is referred to in Skorupski 1983 as Source A.
168
One might even see places within the STTS and related practices where the vajra
family is emerging as primary. However, in the case of the body mandala described in
IOL Tib J 576, Vairocana maintains the central position.!
! 48!
in which the påramitås are imagined to be engaged or which the practitioner him/herself
performs.169 However, the Tibetan text ‘translates’ or identifies vajra mußti in the
commentary with four discrete names of unfamiliar goddesses: brtan ma rdo rje, 'bar ma
rdo rje, dud ma rdo rje and bskyod ma rdo rje 170 So some form of gender confusion or
manipulation appears to be at play, with the påramitås becoming male and accompanied
by the seemingly female figures.
As for the limbs, beginning with the right hand proceeding to the left, then to the
left foot and finally to the right foot, for each the order of enumeration is middle (the jina
or chief buddha of the family), index finger /toe, ring, pinky and thumb. The four
attendant bodhisattvas of each buddha are invoked through mantra and laid out in this
order from index finger through thumb. They are each accompanied by an inner offering
goddess, an outer offering goddess, what appears to be a door protector and, somewhat
mysteriously, by yet another goddess. In three of the five instances, this latter set of
goddesses placed on the thumb/big toe may identified as bearing names related to the
goddess accompanying the main buddha on the middle finger/toe of the same limb. They
furthermore may be seen to symbolize the respective poisons that each buddha family is
identified with bearing the capacity to purify. For example, Dveßaratî, related to
Akßobhya’s consort Måmakî, represents “rejoicing in enmity.”
In the external Guhyasamåja mandala and the body mandala of the father deity as
described in the Piñ∂ik®t-sådhana, the four goddesses Locanå, Måmakî, Pañ∂aravåsinî,
and Tårå are referred to as Moharatî, Dveßaratî, Rågaratî and Vajraratî respectively. In
the external mandala, they inhabit the intermediary directions of the central palace, while
in the body mandala, they are located on the elements. [See Fig. 15-17] As several
manuscripts related to the Guhyasamåja Tantra were also found in the library cave, there
is the possibility of influence of this textual cycle upon IOL Tib J, 576.171 However, just
when we think we have pinpointed the logic of including these goddesses, the Tibetan
translation of their names in the interlinear commentary confounds. In IOL Tib J 576, the
names given for these goddesses on the thumb/big toe of the body mandala are:
Teja-ratî/Dveßaratî Tibetan translates as : Rdo rje sgril ma
Vajra-ratnaratî Rdo rje bde ma
Råga-ratî Gsal bkra ma
Vajra-Ratî Rdo rje’i ‘bebs ma
In the Sarva-durgati-pari!odhana mandala, there is a Rdo rje dril bu ma, one of the four
female gate guardians. Perhaps there is a connection to Rdo rje sgril ma; confusion of
these homophones sgril and dril is an error easily made by a scribe. This does not of
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
169
It is less likely that it is the goddess known as Vajramußti mentioned in the SDPT .
See Mallmann, 411.
170
The standard Vajradhåtu mandala does not include any goddesses in Vairocana’s
portion of the mandala except for the four påramitås. It is possible that since the Sanskrit
mantras which appear here do not bear any obvious marker of the feminine or the
påramitås, the Tibetans thought they were supplying consorts for male deities here.
!
171
For example, under the tutelage of Jacob Dalton I have read portions of ITJ 438/1,
IOL T J 331/2 and 464/1. Dalton and van Schaik, 2006 also identify other versions found
at Dunhuang: 419/1,/6, /9 &/10; 437/1& 2; 454/1; 481/1; 508/1; 565/1.
! 49!
course explain the connection to Teja-ratî/Dveßaratî. Vajra-Ratî, translated here as Rdo
rje’i ‘bebs ma appearing on the big toe of the right foot, and thereby associated with
Amoghasiddhi and Tårå, presents a similar and perhaps related conundrum. Rdo rje’i
‘bebs pa [Vajråve!a], a male form, is a gate guardian of the Vajradhåtu mandala
associated with Amoghasiddhi and the female form, rdo rje’i ‘bebs ma, is a gate guardian
of the Sarva-durgati-pari!odhana mandala.172 We should bear in mind that these are both
yoga tantras in which the main mandala locates Vairocana at the center and of which
there are several related manuscripts and ritual diagrams at Dunhuang.173 These tantric
systems may therefore be regarded as compatible on a couple of levels. However, in the
pattern that seems to emerge in IOl Tib J 576,174 we would expect a door guardian to
appear instead on the pinky or little toe. Nonetheless, on the little toe of the right foot,
instead of Vajråve!a, we find Vajra-ghañ†a, 175 translated in the Tibetan commentary as
Rdo rje dril 'sgrol ma. Now, the 'sgrol may be read in connection with Tårå [Sgrol ma],
paired with Amoghasiddhi. However, the standard Tibetan translation of Vajra-ghañ†a
would be rdo rje dril bu. Furthermore, the door guardian associated with
Amoghasiddhi’s retinue in the STTS mandala176 is once again Vajråve!a, or in Tibetan,
Rdo rje‘bebs pa.
So we can observe in just these few examples how in the overall arrangement of
deities on the body in IOL Tib J 576, the identity and position of many of the goddesses
is a source of confusion. It seems that they are being introduced here merely to produce a
set of five pairs in each region of the five regions of the body/mandala. The fact that the
goddess and accompanying bodhisattva placed on the thumb/big toe of each hand/finger
are explicitly referred to as “resting in union” [sbyor ba’i tshul du bzhugs]
presents yet another anomaly. Note that this phrase occurs with the last pair of deities of
each Buddha family (and bodily protuberance) with the exception of Vairocana’s and
might therefore be read as applying to all of the deities in that Buddha family. One
possibility is that this phrase refers to a position of the hands.
English 2002 described a rite of hastapüjå- vidhi within the ‘external worship’
portion of the Vajravåråhî sådhana that is the subject of her study. In this rite, the six
‘armour gods’ associated with Cakrasaµvara tradition are installed on the fingers and
nails of the left hand as the six buddhas. Sanderson pointed out that the ¸aiva
“prototypes” of the rite install a mantra on the thumb with the index finger and then in
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
172
There is also a female gate guardian by this name in the Vajradhåtu mandala based in
the Sampu†a Tantra (rgyud sde kun btus #111).
173
On the SDPT at Dunhuang see Dalton and van Schaik, 2006 on IOL Tib J 318/1,
384/1, 384/4 & 420/1.
174
By this pattern I am referring to the fact that the right hand, left hand and left foot, on
the little finger or toe we find respectively: A÷ku!a translated as rdo rje gyo ga ma;
Vajra-på!a translated as rdo rje zhags pa; Vajra-spho†a translated as rdo rje lcags sgrog
ma. The Sanskrit refers to the door guardians with the associations with Buddha families
we would expect to find in the Vajradhåtu mandala. Two of the Tibetan names do as
well, rdo rje rdo rje gyo ga [‘vajra dishonesty’] would need to be emended to rdo rje
lcags kyu in order for all of the Tibetan translations to match.
175
It appears here as ba dzra gan ta in Tibetan transliteration.
176
See diagram Giebel 2001.!
! 50!
turn use the thumb to install mantras on the others. 177 Perhaps a similar ritual gesture is
occurring in IOL Tib J 576. However, ‘rnams sbyor ba'i tshul du bzhugs’ would most
commonly be translated as ‘abide in union,’ suggesting that the bodhisattvas (and perhaps
the buddhas as well) are imagined to be in union with their consorts. However, one key
feature of yoga tantric literature (to which the STTS belongs) is that, unlike that of higher
yoga tantra, it does not depict deities in union. Therefore, there is a distinct possibility
that IOL Tib J 576 is drawing upon a larger inventory of tantric texts to complete this
mandala, potentially including those of higher yoga tantra.178
It is essential to recognize that the enigmas that arise in observing the manner in
which this text inscribes the portions of the mandala on the human body is not exclusive
to the medium of the body. The translation of the mandala into the architectural
framework in the eleventh century expresses the same kind of experimental spirit we find
in IOL TIB J 576’s translation of the form of the mandala onto the body. Within a
hundred years or so of the composition/inscription of IOL TIB J 576, the Vajradhåtu
mandala was mapped onto the assembly hall at Tabo monastery in Western Tibet [See
Fig. 18 and compare with Fig. 11 & 12] through the arrangement of clay
sculptures of the deities of the mandala around the periphery of the hall.179 Only
Vairocana and his immediate retinue are positioned centrally, at the ‘head.’ The ‘extra’
pråj•ås (found on the thumbs of the body mandala) are not found in this model, nor are
the main consorts of the five jinas. Perhaps the most potent aspect of this arrangement is
the placement of the door guardians (the ‘pinkies’ of the body mandala) flanking the
actual entryways into the hall. As indicated in this comparison with imposing the
mandala upon an architectural space, as well as with more general comparisons to the
standardized artistic representations of the mandala, we see how the arrangement of its
component parts in space and on the surface of the body emphasizes fundamental aspects
of the logic of its design.
The necessity of guarding the doorways points to an important aspect of the
model found in IOL TIB J 576. Namely, virtually all of the points specified on the body
are connected with the sense faculties (eyes- sight; nose-smell; ears- hearing; tongue-
taste; fingers and toes-touch).180 As points of vulnerability in the Buddhist construction of
the person, the armoring, and in the tantric case, deification of the sense doors is essential
to insuring ritual purity and a pristine condition of awareness. Considering corporeality
and textuality side by side, the protection of the holes for binding together pothi
maniscruscripts with dhårañî may be regarded as an analogical process. Inscribing the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
177
See English 2002, pp218-220; on p.219 English references this 1999 personal
communication with Alexis Sanderson.
178
I am grateful to Kris Anderson for reminding me of the frequent addition of goddesses
in the higher and yogini tantras.!
179
On the he possibility that the ninth-century Indonesian stüpa complex at Borobudur
expresses the order of the Vajradhåtu mandala and forms a pair with Candi Mendut, a
nearby complex mapping the Garbhadhåtu mandala consider Kemper 1959 and Chandra
1980, referenced in Kim 2007 p141 fn44. See also Mus 1998 and Huntington 1994.
180 !Though of course the fingers and toes are not the only sites on the body to absorb
sensation, they are more obvious instruments of touch. !
! 51!
body and inscribing the text are both acts of protection of volatile openings.181 In the
model of the Guhyasamåja body mandala presented in the Piñ∂ik®ta-sådhana (to be
discussed in the next chapter and frequently cited by Mkhas grub rje in the context of the
body mandala debate), bodhisattvas are positioned on the sense faculties, the heart,
totality of the body, head, and the joints in the body mandala of the father deity.182 On
the body mandala of the consort, on the other hand, the bodhisattvas appear on the sense
faculties in union with the sense goddesses of form, taste, and so forth.183 Therefore,
reading these examples together with IOL Tib J 576, we can deduce that the association
of apertures of the body with sensory perception is highly charged in the Buddhist
context.
One might also consider the power invested in the apertures of the body as sites of
liminality, not just between interiors and exteriors but between life and death. Alexis
Sanderson discusses Buddhist tantric practices that specify how the consciousness leaves
the body at death through one of nine doors or orifices depending upon one’s karmic
“destiny.” The door at the crown of the head is considered the most auspicious.
Sanderson traces this model back to Brahmanical sources as well as to early non-tantric
Buddhist ones. 184 Furthermore, as mentioned above, related rites such as nyåsa and
hastapüjå-vidhi found in non-Buddhist tantric ritual as well, employ the placement of
deities and their related seed syllables on the body as a mode of protection and
purification.185 Elizabeth English has astutely noted the resemblance of the technologies
of purifying the body of the practitioner through association of the components of the
body with deities,186 of the armoring the body through nyåsa,187 hastapüjå-vidhi188 and
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
181
I am grateful to Alexander von Rospatt for suggesting this comparison. If we continue
to evaluate body and text as “rten,” representations or “supports,” one might ask what
these mutual acts of protection are warding off. In the case of manuscripts, like that of
sculptures in need of consecration, the perceived threat is likely the invasion of demonic
presences. In the case of the body, the nature of the threat may be somewhat different;
the points of vulnerability are sites of perception. The concern is with being distracted or
persuaded by sensory stimuli. Pursuing the comparison further, therefore, would invite
us to engage with Buddhist understandings of the senses as found in the Abhidharma
literature.
182 !See Wright 2010, Appendix B2!
183
See Wright 2010, Appendix B4. According to the Hevajra system, moreover, these
sense goddesses appear like reflections.
184
See Sanderson 2009, note 297. Sanderson uses the term utkrånti˙ in his description.
Sanderson refers to the Abhidharmako!abhåßya 3.43abc where Vasubandhu describes the
cessation of consciousness at various bodily sites and specific case of the arhat for whom
consciousness may cease at the heart or crown. Among the tantric sources he refers to
Bhavabhatta’s commentary of the Catußpî†ha-tantra f.52r2.
185
I have located several manuscripts on nyåsa practice within the catalogues of the
Nepal German Manuscript Preservation Project that may be of interest. These texts
appear to be based in Hindu tantra. A few examples are NAK accession numbers: 5/2284,
4/3206, 1/630, 5/4456, 8/1410, 6/302.
186
See English 2002, pp.114-119.
187
See English 2002, pp. 163-6.
! 52!
body mandala practice.189 We have touched upon the employment of visions of
generation and dissolution, and of mantra and breath manipulation to purify the elements
of the body in the context of the bhüta!uddhi as related by Flood from the Jayåkhya. To
review, after the purification of the body, Flood presented the “divinization” of the body
through nyåsa, installing mantras on points of the body through touch and recitation as a
means of making the practitioner worthy of worshipping a divinity. Mantras were
installed “on the head, eyes, ears, mouth, shoulders, hands (again), buttocks, heart, back,
navel, hips, knees and feet.”190 This stage prepared the practitioner to envision the deities
on his/her own body in the stage of “mental worship” [manasayåga]. The practice then
concluded with external worship of the mandala. The rite of armoring the body through
nyåsa described by English in the context of the Buddhist Vajravåråhî sådhana is
comparable to Flood’s phase two, “divinizing the body.” After imagining oneself as
Vajravåråhî, the practitioner “protects the body of ‘himself-as-goddess’ with an armor
(kavaca) of mantra syllables, and then infuses it with transcendental knowledge.”191
Through reference to these practices we see how inscribing deities on the body
through mantra serves an apotropaic function fundamental to achieving both immediate
ritual and far-reaching soteriological goals. We are reminded of the mantra of the vajra
kavaca presented in the first surviving portion of IOL Tib J 576. The apotropaic potential
of inscription may be reinforced through reference to Buddhist practicing of inscribing
such dhårañî upon amulets worn for protective purposes as well as their enshrinement in
memorial sites such as stüpas. In fact, examples of such dhårañî are profuse at
Dunhuang.192 [See Fig. 19-21]
In IOL TIB J 576, the main deities of the mandala are inscribed upon the template
of the human body through mantras that are attached to points on the body together with
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
188
See English 2002 pp.218-20.
189
On body mandala, see English 2002, pp. 197-203. For observations of the similarity
of these ritual technologies existing side-by-side the Vajravåråhi sådhana that is the focus
of her study, see for example pp. 116, 166 & 197.!
190
Flood 2006, 114
191
English 2002, p163
192
See, for example, Tsiang 2010. Tsiang presents numerous correlations between
employment of dhårañî for protecting the body and creating a sacred ritual space:
“As a group, the dhårañî sheets may be seen to have functioned and to have been
understood on various levels. One one level, they were regarded as talismans and charms
having magical power for protection. On another, they are representations of ritual
performance and sacred spaces...While as sacred enclosures, their symbolic spatial
configurations can be considered as mandalas, as they have been identified by Ma
Shichang, they are of a different type from the hierarchical groupings of deities that are
common in later periods. The objects and hand gestures depicted on the dhårañî sheets
are representations of ritual performance that are closely related to those in esoteric
practice manuals, yigui, used in the Tang period and refer to practice rather than
cosmology.” (Tsiang 2010, p.246) For other examples of the apotropaic uses of
inscription in the Chinese religious context, see James Robson 2008 & Paul F. Copp
2014.
!
! 53!
the names of the deities to be imagined there in Tibetan. There are however, no
descriptions of the colors or attributes of these deities, no visual cues, only mantras. It is
beyond the scope of the present work to fully examine the structure of these mantras.
However, on a general level we can see that all of the deities associated with Akßobhya
are linked with hung/hüµ, of Ratnasaµbhava with hraµ/traµ, of Amitåbha with hrî˙,
and of Amoghasiddhi with å. Vairocana and Locanå are associated with aoum / oµ but
among the other deities in his retinue and located on the head, (seemingly male forms of )
the påramîtås are all associated with variations upon oµ, while the four iterations of vajra
muß†i paired with them end with: hung, hram, dhi and å respectively. If we were to read
dhi˙ as hri˙, then this set would imply the connection of each of these pairs with the
retinues of the Buddhas distributed across the limbs. At the completion of the retinue of
each Buddha, there is a mantra, transliterated into Tibetan as: sa ma ya sa tvam sa ma ya
stvam [Samayasattvaµ samayasattvaµ]. This mantra indicates that ‘you are the samaya’
[Tib. dam tshig], the “vow”; it seems to refers to the identity of the practitioner’s mind
with that of the Buddha. 193
The transition to the next section of the text is made with the mantra: aoum ba
dzra dha tu ma 'da la si ti hung [Oµ vajradhåtu mañ∂ala siddhi huµ], a reference to the
powers attained through practice of the Vajradhåtu mandala. The final section of the text,
pages 5-8, features mantras for the ten krodhas or wrathful deities together with their
consorts.194 The interlinear commentary supplies their names in Tibetan together with
those of their consorts as well as the directions they inhabit, their colors, and their
attributes. There are two striking elements in this transition. The first is that suddenly the
ritual action has been relocated outside of the human body. There is no placement of
these deities on the body of the sort we will find in the Guhyasamåja body mandala
where they are located on the hands, mouth, vajra, shoulders, knees, top of head and
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
193
Initially, I interpreted this mantra as a reference to the samayasattva, the “pledge
beings” that the practitioner imaginatively merges with the jïånasattva, the “wisdom
beings,” in many tantric sådhana ; such a merger is one technology of ritualized
divination in Buddhist tantric sådhana. However, drawing upon his experience with such
Dunhuang ritual texts, Jake Dalton brought my attention to the subtleties of the context
and to the closer parallel of the “samaya” with the qualities of the jïånasattva here. [Jake
Dalton, Personal communication, December 2015]
English 2002 fn411 remarks: “The appearance of the samayasattva and jïånasattva in
yoga tantra texts is worth further study. Mkhas grub rje (p.235) gives little idea of the use
of these terms in the yogatantra scriptures, citing only the explanatory tantras, the
Paramådya and Vajra!ekhara, rather than the root yoga tantra, the Sarva-tathågata-tattva-
saµgraha. Mention of the samayasattva appears in the Sarva-durgati-pari!odhana-tantra
(19b) where it is described in terms that are associated in our texts with the jnånasattva,
namely, the drawing down of deities into the heart mandala with rays, a process that,
however, is said to complete the samayamandala...” These references to the STTS
explanatory tantras and to the root text of the SDPT are worth consderation in our context
together with other instances of the appearance of these terms in Dunhuang ritual
manuals. See, for example, IOL Tib J 422 as described in Dalton and van Schaik 206, p.
166-7.
194
Aparåjitå appears to be the exception.
! 54!
bottom of feet [see Fig.16 & 22]; the Guhyasamåja model itself closely resembles the
nyåsa ritual for divinizing the body described by Flood in the context of non-Buddhist
tantra in this regard. In the standard external Årya form of the Guhyasamåja mandala, the
krodhas inhabit the doors of the mandala as well as the corners, zenith, and nadir, much
as we find here. [See Fig.15] They are wrathful protectors extraordinaire, and they
guard the boundaries of the mandala.
The other striking feature of this transition in the text is that the descriptions in the
interlinear commentary have suddenly become explicitly visual. Why is so much energy
being devoted to describe these fierce protectors when the buddhas and bodhisattvas
themselves were presented only through names, bodily locations, and mantra? While we
may not be able to resolve this question here, we can observe that the ritual goal of
protection appears to be paramount here, protection not only of the body but also the
space around it.
We will not deal with this portion of the text with the same level of detail as the
preceding section as the body is not explicitly involved, and the krodhas appear to be a
standard set. Only the male krodhas associated with the zenith and nadir, hung ka ra
hung, translated in the commentary as hung mdzad hung gi rgyal po, and ba drza ka ma
li kun da li, translated as pad ma 'khyil pa seem unfamiliar. The consorts, explicitly
referred to as yum here, also require further research. The final section begins with a
transliteration of what appears to be a set of somewhat standard Sanskrit mantras Sarva
tathågata mahå !unyatå jnåna vajra atmako hum...Mandalaja hrung bang ho, likely an
evocation of the jïånasattvas to merge with the samayasattvas. Then we have a formula
for twenty-four deities evoked via mantra ending ja hrung bang ho. 195 There is no
commentary providing any information about their appearance or the translation of their
names.
IV. Conclusion
Over the course of IOL Tib J 576, we see a progression from subtle to gross, from
descriptions of emptiness, to the location of specific deities upon the body through
mantra, the evocation and visualization of wrathful protectors in the directions outside the
body, and then a movement further outward into the realm of mundane worldly deities.
In observing the many anomalies in the naming and arrangement of deities, in particular,
the case of the goddesses, we identified this text as transitional and experimental in
nature, drawing upon a larger inventory of tantric texts from both the yoga and perhaps
even the higher yoga tantric classes. Themes of protection and purification are apparent
in the choice of locations on the body, in the mantras tied to protective aims, and in the
evocation of wrathful protectors to guard the boundaries of space. Through comparison
with similar ritual technologies of protection and purification from Buddhist and non-
Buddhist sources, we were able to see how this proto-body mandala text from Dunhuang
employs the power of inscription as an apotropaic force. One might link this power to
the abundant production of dhårañî at Dunhuang in media that fused both text and image.
By analyzing the form and contents of IOL Tib J 576 through the lense of inscribing the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
195
I am grateful to Professor Jacob Dalton for his help in reading this sådhana and in
particular for helping me to discern how this final portion of the text fits in.
! 55!
body, we created an analogy between corporeality and textuality, an analogy that will
continue to unfold over the course of the dissertation.
In examining the body mandala practice of the Guhysamåja cycle in the next
chapter, we will continue to engage with ritual technologies of relating deities to sites on
the body through visualization, mantra, and the use of syllables in their recited or
envisioned forms. Transitioning from the theme of “inscribing” to “mapping,” we will
explore how these practices of locating deities on the body are employed to protect sites
of vulnerability or even to pierce to the pith of potent energy centers of the human body.
Furthermore, in beginning to address Mkhas grub rje’s writing on such ritual actions in
the text that sparked the body mandala debate, we will explore his citations from the vast
commentarial literature of the Guhysamåja system. In doing so, we hope to catch a
glimpse of how the body mandala practice evolved and how anomalous details like those
brought to our attention in exploring the IOL Tib J 576 were revised, standardized, and
reinterpreted across Indian and Tibetan traditions.
! 56!
Chapter Three: Mapping the Body: Locating Deities within the Bodily
Landscape in the Guhyasamåja Body Mandala Practice
Introduction:
How do we prepare the body as the proper basis for enlightenment? The early Buddhist
practices of mindfulness of the body discussed in the previous chapter suggest that
knowledge of the body, and specifically of its constituent parts, facilitates the qualities of
non-attachment essential for enlightened realization. We have also observed non-
Buddhist tantric practices that demonstrate the fundamental tantric principle that one
must become a god to worship a god. The Vaißñava practices described by Flood
exemplify the kinds of practices of purifying the body that play a key role in this
transformation. The practices of purification [vi!uddhi] correlate microcosmic and
macrocosmic elements and manipulate visualized forms and breath through regions of the
body; this purification concludes with the imagined incineration of the body from the
ground up and its dissemination in the four directions. The subsequent “deification” of
the body involves the placement of mantras onto bodily sites through nyåsa practice,
making the practitioner a god fit to worship a god. The ‘proto-body mandala’ practice
described in the ritual text from Dunhuang (and created during a similar time frame to
Flood’s Vaißñava source), places or ritually ‘inscribes’ deities on the body in a way that
resembles Vaißñava practices’ objectives of purification and protection as well as a
variety of preliminary deification. Based loosely upon the STTS mandala, this text (IOL
Tib J 576) however, arrays deities onto the body’s protuberances rather than either upon
concentrated regions or a total surface area. In translating the form of the mandala onto
the surface of the body, it adapts the radial hierarchy of mandala in a creative way. IOL
Tib J 576’s model of the body reveals a concern with protecting the sensory capacities as
sites of vulnerability in the Buddhist construction of personhood. It also enacts an
abstract correlation of the human body with the mandala. Yet it reveals little else about
the nature of the body itself, its subtle or hidden potentialities. Our Dunhuang text
employs a ritual technology for deifying the body through the placement of deities upon
it, treating the body predominantly as a surface for inscription.
As we begin to engage with the Guhysamåja body mandala practice that is the
focus of Mkhas grub rje’s body mandala text, we will encounter a tension between the
treatment of the body as a surface for inscription and as a basis for enlightenedment. We
will also observe the varieties of ritual transformation required to prepare this basis.
This tension between the body as surface and as basis indicates larger themes in the
evolution of tantric ritual. These include the development of a sophisticated
understanding of the body’s hidden sites of power and the movement of its energies
through these sites in the form of winds [rlung]. In addition, the stratification of sådhana
practice into a two-fold structure of generation and completion stages, a division that
originates within later additions to the Guhysamåja Tantra (in the Eighteenth Chapter, the
Uttaratantra) plays an important role. Both of these themes are central to Mkhas grub’s
and his teacher Tsongkhapa’s role in emphasizing the completion stage as a ritual
technology for accessing the subtle potentialities of the body and for freeing those
potentialites from the ties that bind.
! 57!
We have already encountered the idea that to know the body is to know the world
in early Buddhist practices. The Gelukpa emphasis upon the completion stage of the
Guhyasamåja sådhana promotes an especially potent manifestation of this principle.
Namely, through the completion stage, the practitioner has the opportunity to manipulate
the process of death and of rebirth. Locating deities upon the body in the generation
stage prepares the body as a basis by making its hidden potentialities malleable,
‘simulating’ a deified body in preparing to produce even more subtle varieties of body in
the completion stage.
In transitioning to a metaphor of mapping rather than inscribing, we begin to
consider what lies beneath the surface of the body, the qualities that are not apparent to
the untrained eye. The project of mapping the bodily landscape invites a comparison
with the mapping of the Tibetan landscape as the body of a supine demoness.196
According to legend, the great Indian tantric master Padmasambhava travelled to Tibet to
quell the demonic forces inhabiting it. Through a geomantic enterprise of locating these
hidden forces, Padmasambhava then erected temples to ‘pin down the demons,’ to control
the body of the demoness in order to establish Buddhism within the body of Tibet. He
erected temples upon these bodily points, with the temple at the heart, the Jokhang, as the
most sacred. Therefore to map the bodily landscape is is to see beyond its surface, to
know its hidden points of power and danger as a means of harnessing them.
We will begin by outlining the basic features of the Guhyasamåja body mandala
practice of the Årya tradition, based in the thirty-two deity Akßobhyavajra mandala. For
this description, we will rely upon Geshe Lobsang Tsephel’s contemporary commentary
on Yangchen Galo’s eighteenth-century text, Paths and Grounds of Tantra.197 Beginning
with this accessible and practice-focused account provides us with an opportunity to
familiarize ourselves with the Gelukpa perspective on the Årya Guhyasamåja tradition.
This commentary frequently references Tsong kha pa’s Lamp to Illuminate the Five
Stages Teachings on Guhyasamaja Tantra. [Rgyud kyi rgal po dpal gsang ba ‘dus pa’i
man ngag rim pa lnga rab tu gsal ba’i sgron me], a text we will deal with in more detail
later in the dissertation with the aid of Gavin Kilty’s 2013 translation. It also invokes
Mkhas grub rje’s Ocean of Attainment [Rgyud thams cad kyi rgyal po dpal gsang ba
‘dus pa’i bskyed rim dngos grub rgya mtsho], the very text in which the body mandala
debate is formally initiated. This contemporary account will aid us in forming of holistic
view of the practice as it has evolved within the Årya Guhyasamåja lineage of the
Gelukpa tradition. The commentary suggests that Tsong kha pa and Mkhas grub were
influential in standardizing the form of the body mandala practice for the Gelukpas and
making sense of its connection to a tradition of authoritative Indian texts. It will,
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
196
See Gyasto, Janet. 2003. “Down with the Demoness: Reflections on the Feminine
Ground in Tibet.” in The History of Tibet Vol.1, edited by Alex McKay. Oxford:
Routledge Curzon: 307-321.
197 !Geshe Lobsang Tsephel 1995 (Reprint 2008) is translating and commenting upon
Yangchen Gawai Lodoe’s “Eloquent Explanation- A Port of Entry for the Fortunate Ones
into the Paths and Grounds of Mantra according to the Glorious Guhyasamåhja of the
Årya (Någårjuna) Tradition.” Dpal gsang ba ‘dus pa ‘phags lugs dang mthun pa’i sngags
kyi sa lam rnam gzhag legs bshad skal bzang ‘jug ngogs. Class No Ga-4,34; Acc No-
1043, Tibetan Manuscript Seciton, LTWA, Dharamsala, H.P. India.
! 58!
therefore, prepare us to recognize the role that Mkhas grub’s Ocean of Attainment
and the body mandala debate played in this standardization processs.
In outlining the practice, we will remain attuned to different approaches to
mapping the mandala onto the human form, to correlating macrocosm and microcosm,
and to ordering ritual forms such as the spectrum of subtle and gross phenomena. Having
examined some key themes in the evolution of tantric ritual, namely the development of
more sophisticated knowledge of the subtle body and the two-stage sådhana structure, we
are equipped to recognize the curious manner in which different versions of the practice
exist side beside within the body mandala ritual. These acts of inscribing and mapping,
transforming and even transcending the body suggest co-existent phases in the evolution
of tantric ritual practices oriented around the body. In engaging with various
interpretations of the Guhysamåja body mandala through the citation of carefully-selected
texts, Mkhas grub further contributes to this evolution.
I. An Outline of the Guhyasamåja Body Mandala Practice
In laying the groundwork for the body mandala, the body is generated as a
celestial palace to house the mandala deities. Geshe Lobsang Tsephel translates the term
for this structure, rten pa’i dkyil ‘khor, as “residence mandala.” The introduction to this
dissertation highlighted the nuances of rten in framing Himalayan artistic and ritual
approaches to representation. This dissertation translates rten pa’i dkyil ‘khor as
“mandala of the support,” to emphasize the foundational role of representations and
bodies in the creative soteriological project that is body mandala ritual. The mapping of
the celestial palace onto the body encompasses internal and external aspects of the body,
surfaces and protrusions as well as inner organs together with more subtle aspects of
winds, sensory perception, and consciousness. Geshe Lobsang Tsephel describes this
correlation of the parts of the body with those of the celestial palace as follows:
“Visualize the front, back and the two sides of our body form the four corners of the
walls of the mansion; the mouth, nose, anus and urethra form the four doors; the five
coloured wind energies, as the basis of conceptions, form the five fold layers of the walls
which are white, yellow, red, green and blue; tongue consciousness becomes the precious
molding; intestines become jeweled nets; the sinews and so forth become half nets; a
certain portion of the white drop of the mind of enlightenment becomes the half moon;
eye consciousness becomes the mirrors; nose consciousness becomes garlands of flowers;
tongue sense becomes the bells; body sense becomes the yak tail fans adorning the jewel
nets and half nets; ear and body consciousness become the banner and pendants on the
parapet; the two shins, thighs, forearms and upper arms become the eight pillars; the belly
becomes the interior vases; the ear sense becomes the half moon adorned with vajras at
the four corners; the four physical and mental objects such as form become the five
colours of the mansion-white, yellow, red, green, blue; the secret place, navel, heart and
tip of nose become the four arches; the eye sense becomes the wheel of dharma and the
mental consciousness becomes a buck and a doe on the top depicted over the eastern
door; the nose sense becomes the banners on the four arches and the mental sense
! 59!
becomes the lotus in the center of the mansion. In this way, the different parts of our
body are transformed into the residence mandala.”198
We can see a few interesting points in the means of correlating the body with the
architectural structure of the palace. Some portions are fundamental such as the five
winds as the layered walls. None of the other natural elements appear to be represented
here. In fact, this is not the general wind [rlung] of the great elements, but rather, “the
basis of conception,” the very vehicle for the mind and the precondition for embodiment.
Equating limbs and pillars shows a more straightforward structural correlation. Other
aspects are what we might call “ornamental,” such as the intestines and sinews as nets.
The sense consciousnesses and objects also seem to fit within the class of the ornamental.
The material representation of the sense consciousness and objects prepares us for their
embodiment as deities within the “mandala of the residents” [brten pa’i dkyil ‘khor] to be
discussed next; this representation also reminds us of the symbolism of the sense
offerings through image, gesture, and sound within Tibetan art and ritual. There appears
to be a hierarchy expressed among the senses, with the “mental sense” as the central
lotus. Relating “mouth, nose, anus and urethra” to four doors demonstrates a conception
of orifices as doorways and potentially as sites of vulnerability, as the doors of the
conventional mandala palace are often guarded by fierce protectors. The designation of
the “secret place, navel, heart and tip of nose” as the four arches seems to assign them a
primary role in defining the boundaries of the space. 199
We can now proceed to unpack this mandala, to observe the manner in which it
comes to be populated and to form the brten pa’i dkyil ‘khor, the “mandala of the
supported,” (referred to by Tsephel as the “mandala of the residents”). The generation
stage can be broken down in terms of the three kinds of Buddha bodies: dharmakåya,
sambhogakåya, and nirmåñakåya. These bodies are the three results of the practice with
three corresponding bases of purification: death, the intermediate state, and rebirth. The
actual attainment of these buddha bodies occurs later on in the completion stage, but the
generation stage is regarded as the essential trial run, a repeated fabrication that becomes
a reality.
In producing the dharmakåya, the “mind of clear light” is the path of purification,
whereas for producing the sambhogakåya, the path of purification is the “illusory body.”
Kilty 2013 defines the illusory body and its relationship to the clear light as follows:
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
198 !Geshe Lobsang Tsephel, Paths and Grounds, pps.33-34.!
199
The Guhyasamåja system contains a unique grouping of “nose tips.” Mkhas grub
debates the question of whether they are to be enumerated as three or four in his Ocean of
Attainment (256.5-257.1) through recourse to the Vajramålå explanatory tantra.
TheVajramålå is a text quoted extensively by Mkhas grub and will be examined as such
within this dissertation. The description or the mandala of the support provided here by
Geshe Lobsang Tsephel appears to be in accordance with the description found in
Chapter Sixty-eight of the Vajramålå Sde dge 275a.2-.6 [549.2-.6] as cited in Mkhas grub
255.6-256.1.
!
! 60!
“This body is created from the subtle inner winds and is in the aspect of the resultant
Buddha form that is the goal of the practice. This illusory body is the exclusive cause of
the form body of the Buddha, the rüpakåya. Alongside this practice is the wisdom
development of the mental state of clear light. This is in the nature of a very subtle level
of mind and is the exclusive cause for the enlightened mind, or dharmakåyå.”200
The generation stage itself produces the nirmåñakåya. It is likely that the
aggregates, elements, and so forth of the body have already undergone a basic
purification before this phase begins as part of the preparatory rites; the limbs and sense
spheres are further purified in constructing the body as the celestial palace, the “mandala
of the support” described above. The body mandala practice seems to include multiple
phases of purification, a theme we will attend to in our description; light and sound,
extolled from the time of the Vedas in the Indic context, for example, are two common
purifying factors in tantric ritual manuals. Throughout the process of the body mandala
practice, the body described becomes increasingly more subtle, bathed in the luminance
of clear light.
First, one envisions oneself as Akßobhya (the central deity of the outer mandala of
the Årya tradition, the Akßobhyavajra mandala). Then, one imaginatively installs the
deities of the mandala on key parts of the body. The tathågatas emerge from the
skandhas as Vairocana, Amitåbha, Ratnasaµbhava, and Amoghasiddhi and are
positioned at the crown, throat, navel and “groin” with Akßobhya fused with the main
deity (embodied by the practitioner).201 Four goddesses then emerge from the elements
and unite with these tathågatas: Green Tårå with Vairocana (with Amoghasiddhi in her
crown), red Påñ∂aravåsinî with Amitåbha (with Amitåbha also in her crown), and white
Locanå with Ratnasaµbhava (with Vairocana in her crown). Måmakî resides at the heart
with Akßobhya in her crown.202 As observed through the summary of Wright’s theories
in the previous chapter, Måmakî’s affiliation is complex. We will return to the case of
Måmakî later in the context of the evolution of body mandala materials within the
literature of the Guhyasamåja.
As we have already noted in our exploration of the proto-body mandala from
Dunhuang, the positions and associations of the goddesses exhibit a great deal of slippage
in their evolution through practice manuals and commentarial literature. In Geshe
Lobsang Tsephel’s account of the practice used as a standard here, the goddesses are
associated with respective sites on the body where they unite with a tathågata, a lord of
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
200
Kilty 2013, Translator’s Introduction, pp.2-3.
201
Akßobhya may also simultaneously be envisioned at the heart.
202
The emergence of the buddhas from the skandhas and of the goddesses from the
elements and their subsequent placement upon more specific bodily locales may conflate
two forms in the evolution of the practice. Where precisely are the skandhas and the
elements and how are they to be imagined? Do all texts imagine the goddesses to be in
union with the tathågatas? There is no mention of arraying deities upon the body of a
consort here, unlike the descriptions found in some versions of the practice. What role
does this absence play in this interpretation of the practice? These are the sorts of
questions that loom in the background of Mkhas grub’s body mandala text. We will
examine some related points of controversy later in this chapter.
! 61!
the family who mark their crowns, a body color, a natural element, and a colored seed
syllable. So Locanå is associated with earth and a yellow låµ, Måmakî with water and a
blue måµ, Påñ∂aravåsinî with fire and a red påµ, and Tårå with wind and a green tåµ.
The colors seem to reflect connections both with the symbolism of the elements and of
the buddhas they unite with, though it is hardly a one-to one correlation. For example,
Tårå unites with white Vairocana but is associated with a green seed syllable, the color of
Amoghasiddhi the lord of her family.203 Later in our discussion we will encounter texts
that assign further correlations to these goddesses, with particular winds, the Buddhist
doctrine of the perfections, and so forth. For the time being, we will bracket the
resolution of these issues and merely note their complexity and the need for recourse to a
variety of literature of the Guhyasamåja and other tantric systems across time to attempt
to resolve them. In addition, in light of this complexity, we will pay careful attention to
the citation of passages regarding these goddesses within the texts of the body mandala
debate.
To continue with our basic description of laying out the body mandala according
to Yangchen Galo’s account, eight bodhisattvas are each associated with sites on the
body: eyes, ears, nose, tongue, heart, “vajra organ,” joints, and crown. Five of them are
embraced by the goddesses of the sense objects at the eyes, ears, nose, tongue, and vajra
organ.204 Furthermore, the ten krodhas are positioned on the right thumb, left thumb,
mouth, tip of vajra, joints of the right and left shoulder, the knees, crown of the head, and
soles of feet. Geshe Lobsang Tsephel clarifies the purpose of identifying the deities with
bodily sites as follows: “We must know these deities of the body mandala, their locations
in it, and the respective constituents of our body they are associated with, in order for us
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
203
The evolution of the color symbolism of the elements as found in Abhidharmic
descriptions of the formation of the cosmos, Indian and Tibetan medical accounts, and
non- Buddhist tantric literature on bhüta!uddhi such as the texts described by Gavid
Flood, are potential sources for understanding the relationship of the elements to
particular colors (and often, shapes). Payne refers to Wayman 1977’s proposed
correlation of the shapes of Vedic fire altars with those of the tantric cakras as well as
with “continents of Purañic mythology.” See Payne 2002, p.195 and Wayman 1977
pp66-67.
One might also consider how the designation of the central deity of the mandala
may shift color associations of the directions of the mandala and how a similar logic
might be at play in body mandala practice. In the tradition of the Årya Guhyasamåja,
Åkßobhya replaces Vairocana as the main deity of the mandala. For example, when the
main deity of the mandala is of the Vajra family, as is the case for the Årya
Akßobhyavajra mandala, the central portion of the mandala is depicted as blue while the
eastern portion becomes white, the color associated with the tathågata family; this
adjustment alters the ‘standard’ association of the eastern quadrant with blue and the
center with white (and the vajra and tathågata families respectively).
204
Note that in this account of the practice, the fifth rdo rje ma, Spar!avajra, crowned by
Akßobhya, unites with Sarvanivarañavißkambhin at the opening of the vajra organ. The
role played by this goddess in different accounts of the practice vaies; in some she is
considered to be the consort of the main deity of the mandala, Akßobhya.
! 62!
to meditate on the process of death and take it as a path to actualize the truth body.”205
This interpretation of the body mandala practice as a means to attaining or rehearsing this
kind of knowledge of the body directed towards successful navigation of the death
process is significant. It exemplifies a dimension of the practice that may have become
more prominent over time and may be deeply endebted to Tibetan interpretations. We
will touch upon this dimension of the Årya Guhyasamåja interpretation in the final
portion of this chapter.
The next step is to dissolve the deities and their associated bodily sites. Each
deity is associated with a physical element in this process. The stages accord with the
order of the dissolution of the body at death, a process ordered as a progression from
gross to subtle properties and levels of experience. Geshe Lobsang Tsephel outlines the
stages of dissolution at death as follows:
Earth into water

water into fire
fire into wind
wind into consciousness
consciousness into mind of white appearance
mind of white appearance into radiant red increase
radiant red increase into black near-attainment
dawning of mind of clear light of death206
In his description of the body mandala practice, the first four of these stages, in which the
potency of earth is subsumed by that of water, water by that of fire, fire by that of wind,
and wind by that of consciousness are each associated with a tathågata (Vairocana,
Ratnasaµbhava, Amitåbha, and Amoghasiddhi respectively) as well as with a goddess,
two bodhisattvas, and two krodhas. The pairing of goddess with tathågata here matches
that of the lord of the family and seems to agree with what has become the more
standardized association of elements with directions of the mandala.207 Yet, the dominant
schema for this system is the dissolution of the elements and the concordant signs
experienced at the moment of death.
The final four stages of consciousness being subsumed by the “mind of white
appearance,” and that in turn by the “red increase,” that by “black near attainment,” and
that phenomenon by “clear light” are associated with the deities Ußñîßacakravartî and
Sumbharaja, Ma•ju!rî, and Akßobhya respectively. Note that the lord of the mandala,
Akßobhya is correlated with the final and most subtle stage of this process, that of clear
light.208 Within the outer, or standard Årya Gühyasamåja mandala, Ußnißacakravatin and
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
205 !Paths and Grounds, p.27.!
206 !Paths and Grounds, p.21. For a detailed and compelling account of this process, see
Sogyal1992. !
207 !In understanding the logic of this system, one might consider how the skandhas and
sense spheres fit within this set of correspondences.!

208
We might make sense of the progression between the two krodhas Ußnißacakravatin
and Sumbharaja on to the bodhisattva Ma•ju!rî in relation to tantric subtle body
! 63!
Sumbharaja appear as the zenith and nadir, often shown in front and behind the main
deity, with Ma•ju!rî to the left of the western door of the palace. One might also think
about how Ma•ju!rî in the form of Manjuvajra is the central deity in the Guhyasamåja-
mandala of the J•ånapada tradition. So positioning him as the second most subtle
phenomenon may accord him a somewhat central status, as his position on the heart in
the body mandala certainly does. Read in this way, the celestial palace of the body is
collapsed into its central point, a trope we are familiar with in a variety of descriptions of
the emanation of and dissolution of mandalas and deities. In deity sådhanas, that central
point is often marked by a seed syllable, and in the mandala it may simply be referred to
as the drop (bindu) or the pith.
After the deities have been installed in the mandala, there is the blessing of the
vajra body, speech and mind and the merging of the jïånasattvas, the “wisdom beings” or
in a sense, the true deities, with the samaya-sattvas or “pledge beings,” the imagined form
of the deities the practitioner has just established within the mandala.209 Once again, we
might note the movement towards more subtle forms of emanation as the practice
progresses as well as a sense of what we might hazard to call a more genuine experience
of divine nature. These transformations occur as a result of the repetition of patterns of
ritual activity; through repetition one achieves effortlessness and naturalness in
identifying with enlightened existence. The blessing of body, speech, and mind is
correlated with three sites on the body and three divine couples: Vairocana and Locanå,
as the emissaries of vajra body, direct their activities of emanation toward the crown of
the head; Amitåbha and Påñ∂aravåsiñî, as emissaries of vajra speech, direct theirs
towards the tongue210; Akßobhya and Måmakî, as emissaries of vajra mind, direct theirs
toward the heart. In a sense all three sites are treated as orifices, as porous centers for the
production and reception of divine energy. This energy transforms the nature of one’s
fundamental components of body, speech, and mind into their ideal steadfast and true
“vajra” potential. The transformation is sealed or confirmed by a mantra asserting one’s
identity with these three varieties of vajra potential. This visualization entails the
common theme of emanation and absorption, beginning with an white oµ on a moon disk
from which five rays of light emanate, transforming into a multitude of Locanås, who are
then joined by a multitude of Vairocanas. The practitioner requests their blessing of the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
phenomenon with the descent of the white element, here a light, ascent of the red light
and the dissolution of all the winds or lights into the heart.
209
In her discussion of the Vajrayoginî sådhana, English 2002 clarifies the terms
jnånasattva and samayasattva through reference to Buddhaguhya and Mkhas grub rje:
“The pledge deity is the imaginary form of the goddess created by the ‘pledge-holding’
initiate through the self-generation. In his Tantrårthåvatåra, Buddhaguhya describes the
pledge forms (samayasattva˙...) as‘those [forms] discerned by persons pledged
(*samayin) [to them]...ones imagined as arising from the body of a deity and having the
shape of a deity which the pledge person has generated in conformity with that [body of a
deity], or imagined congruently with the latter’s parts.’ Buddhaguhya describes the
knowledge forms (j•ånasattvam...) as ‘the self-existent (svabhåvin) discerned as deity.’
The knowledge being is said to have both form and ‘inherent nature.’ (Mkhas grub rje:
235, citing the Paramådyatantra).” (English 2002, p.167)!
210 !Often one associates Amitåbha with the throat rather than the tongue specifically.!
! 64!
vajra body and they fuse into one another, with one remaining couple melting and then
diffusing into thirty two drops. The largest of these drops melts into Vairocana at the
crown while the others fuse with the other deities of the body mandala. Then, there is the
recitation of the mantra. The blessing of speech and mind follow and proceed in a similar
manner but with different main deities and colored syllables and slightly varied mantras.
The Supreme Conquerer of the Mandala
This next phase of the body mandala practice presented in Yangchen Galo’s account is
particularly relevant to this dissertation’s approach to the body mandala because it
introduces different varieties of mandala and suggests levels of relationship between
them. A similar process of transformation and identification to that of the fusion of
jnånasattvas and samayasattvas in the previously described phase of the blessing vajra
body, speech, and mind occurs here, but with multiple phases. First, one imagines oneself
as Vajrasattva with a vajra consort emerging from one’s heart. Through mantra, she
dissolves into emptiness, and a syllable marked with a vajra appears in her place. Then,
one places seed syllables of the individual deities of the mandala on her body, and the
associated deities emerge from them. Next, one blesses the vajra organ of oneself as
Vajrasattva and her lotus organ and then unites with her. All the deities of the mandala
dissolve through the intensity of this union. Vajrasattva’s bodhicitta enters her womb,
diffuses into thirty-two drops that become the celestial palace, the mandala of the support
[rten pa’i dkyil ‘khor] and the thirty-two seats within. Then, yet another drop diffuses
into thirty-two parts, and these, in turn, are installed on the seats of the palace; then, they
transform into the syllables and then the symbols of the deities, and finally, into the
deities themselves. The practitioner uses mantra to draw them out of the lotus through
the “vajra path” and to project them in the ten directions, where they act for the benefit of
sentient beings. Akßobhya returns to the heart center and effects the practitioner’s
transformation into Vajradhara and then Dveßavajra. Mkhas grub rje specifies that one
assumes the identity of the latter deity in this process of projecting the deities outwards,
whereas one identifies which each of the mandala deities individually when installing
them in the mandala within the womb of the consort.211 Finally, the deities of the
mandala return from the ten directions and take residence on the seats of the “external
mandala,” the visualized mandala resembling representations painted on cloth or an altar
of painted powders.
Note that in the “Supreme Conquerer of the Mandala,” the sexual organs and the
heart function as the primary orifices or sites of power on the body, and they are
connected energetically through the practice. This process also creates a dialogue
between different layers of interiors and exteriors that lay at the very foundation of
defining mandala, a form based in the understanding of the relation of center and
periphery. There is the body mandala of deities positioned on the body of the envisioned
consort, the celestial palace and residing deities ejected into her womb from the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
211
See Mkhas grub rje, Ocean of Attainment, p.367 (TCPP edition class No Ga-2, 17;
Acc No-4085, Tibetan Manuscript Section, LTWA Dharamsala, H.P., India) as cited by
Geshe Lobsang Tsephel in Paths and Grounds, note 46.
!
! 65!
bodhicitta of the practitioner as Vajrasattva, and finally the outer mandala produced
through the drawing each deity from the mandala in her womb up through the vajra and
out through the heart of Vajrasattva with the aid of mantra.
Mkhas grub rje explains the intention of the ritualized dialogue between inner and
outer mandala as follows:
“Meditating on an inestimable mansion purifies an impure environment. What this
means is: it is not that practitioners can transform all this impure environment into our
inestimable mansion by doing such a meditation but they do it in order to purify
themselves of their potency to utilize the impure environment in the future and also to
ripen their potency by the completion stage for enjoying the inestimable mansion of
exalted wisdom.”212
In this interpretation of the “supreme conqueror of the mandala,” the practitioner

is not changing the world, but rather his/her relation to it.
Completion Stage
There are three main steps during the completion stage that allow the body, speech, and
mind, respectively, of the practitioner to be isolated from their gross identity. Through
these “isolations” [dben Skt. viveka], the practitioner gains access to his/her more subtle
divine identity. The first of these three, body isolation, is most relevant to our discussion
of how the body mandala contributes to the larger ritual project of accessing the subtle
body and preparing the practitioner for the moment of death. Moreover, it bridges both
generation and completion stage practice of the Guhyasamåja. Speech isolation uses the
combination of mantra recitation and breath practice to loosen the knots above and below
the heart center and dissolve their associated winds inside, producing the “wisdom of
appearance.” Mind isolation is characterized by the use of such mantra recitation and
breath practice as well as union with a consort to loosen the knot at the heart center and to
dissolve all the bodily winds there, producing the “four empties.” In the process, the
signs of death appear, ending with clear light. Yangchen Galo clarifies that: “Such a clear
light is the exemplary clear light of the final isolated mind and it is [also] the final
[substantial] basis for accomplishing an illusory body of the third level.” 213 We are now
dealing with a different variety of body, an illusory body, a type that we will see can be
further divided into pure and impure varieties. This body is achieved by once again
reversing the dissolution process:
“As one begins to wake up from the exemplary clear light of the final isolated mind,
which has just been explained, [its] wind is slightly stirred, due to which the mind of
near-attainment of the reversal [process] is accomplished. Along with it, like a fish
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
212
Mkhas grub rje, Ocean of Attainment, p.123 (TCPP edition), as cited by Geshe
Lobsang Tsephel in Paths and Grounds, note 47. See also English 2002’s reference to
Mkhas grub’s parsing of outer rites (båhyakriyå) and inner rites (adhyåtmayoga˙) from
Mkhas grub p219 in Lessing and Wayman, 1978 (1993 Reprint). See English 2002,
p.32.
213
Yangchen Galo, Paths and Grounds, p.63
! 66!
leaping out of water, an illusory body characterized by the noble signs and auspicious
signs is literally accomplished as distinctively separate from the coarse body precipitated
by ripening [actions]. The wind with five rays of light as the mount of the clear light
serves as the substantial cause and the clear light mind itself serves as a cooperative
condition.”214
Apart from the “auspicious signs,” a distinguishing factor of this illusory body seems to
be the fact that it is composed of wind and mind but not of karma. In the system of
causes and conditions that produce form, clear light has replaced karma. Yangchen Galo
invokes Mkhas grub rje, Tsong kha pa, and the Pradîpoddyotana to explain the difference
between this impure illusory body and an ordinary body:
“As Khedrup Je’s Notes on the Five Levels (mkhas grub rje’ rim lnga’i dzin bris) says:
‘An impure illusory body accomplished in this life is distinguished from the coarse body
but it is not accomplished on a separate basis (go sa) as it does not have the ability to do
so.’ The Lamp Illuminating the Five Levels (rim lnga gsal sgron) also says, ‘As stated
earlier, for [an illusory body] to be separated from the old body, it is not absolutely
necessary for it to abandon the basis (go sa) of the old body to exist. The same thing can
also be understood from the Bright Lamp which states that the pure illusory body exists
within the vessel of the old aggregates and so forth.”215
In the conclusion of the completion stage, the pure illusory body, an even more
subtle form of body, is achieved through persistence in practice and consequent
realization of emptiness in the “meaning clear light of the fourth level.” Until one
achieves this pure illusory body in the completion stage, one is unable to attain the
sambhogakåya of a Buddha. Without attaining this fourth level together with an accurate
understanding of emptiness and the practice of taking those three bodies resulting from
the generation stage as paths, one is unable to achieve the dharmakåya either. A system
of practices and realizations together with particular phases of initiation are consequently
required to manifest the three Buddha bodies.
The shared basis [go sa] of ordinary and illusory bodies asserted by Mkhas grub
here signals one of the most complex points of Buddhist tantric practice. What kind of
foundation does the body provide for tantric practice? Moreover, how do the techniques
of body mandala ritual shed light on attitudes toward the liberating and obstructing
potential of the human body in progressing towards a soteriological goal? One of the
goals of this dissertation is to reflect upon how the texts of the body mandala debate may
assert subtle distinctions in Gelukpa and Sakyapa perspectives on tantric practice as
evidenced by attitudes towards the body articulated in terms of the body mandala
practice. How do Gelukpa scholars think about the ordinary body, illusory body, and the
Buddha bodies in relation to one another?
Tsong kha pa clarifies his position on the relation to the different varieties of
bodies as follows:
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
214
Yangchen Galo, Paths and Grounds, p.63!
215 !Yangchen Galo, Paths and Grounds, p.64!
! 67!
“...even when becoming a buddha in a single lifetime on the paths of highest yoga tantra,
it is not taught that from the moment your body is formed it has to be adorned with the
marks and features of enlightenment. If there is no attainment of a body adorned by these
marks and features on the paths of practice, there will no similar-in-type cause, and
consequently there will be no spontaneous transformation of that body. It is necessary to
produce a body adorned with the marks and features of enlightenment while on the
completion-stage paths of practice. This coarse body of ripened karma will not develop
into a body adorned by the marks and features of enlightenment through deity-yoga
meditation, and the body created by just the mind meditating on the body of a deity, as is
done on the generation stage, is not sufficient. You need a very special deity yoga that
will achieve a body distinct from these, and that is adorned by the marks and features of
enlightenment. The substantial cause of a such a body can be none other than the winds.
Therefore a method to achieve an illusory body from the winds is definitely
necessary.”216
This passage touches upon a question central to this dissertation: how do Tibetan
authors explain the distinctions and connections of ordinary and enlightened bodies
effected throuch ritual practice? The first distinction Tsong kha pa makes divides the
coarse ordinary body from a body possessed of the marks of Buddhahood. In order to
transform that coarse body and to manifest those marks, one must practice the completion
stage and therby produce a “similar in type cause.” Tsong kha pa is emphasizing the role
of the completion stage as an exclusive means for attaining the illusory body from the
winds. The significance of this kind of tantric logic of causality for both Tsong kha pa
and Mkhas grub will be discussed in further depth in Chapter Four of this dissertation
together with the distinction of the generation of completion stages.
Bentor 2006 provides some context for better understanding this emphasis upon
the necessity and even superiority of completion stage practice as exemplified by
statements like Tsong kha pa’s. Bentor suggests that the historical evolution of these
practices over time may be at the root of such attitudes towards the distinction of the two
stages of sådhana practice. Bentor describes the Gelukpa perspective on the relation of
the two stages in the Guhyasamåja practice as follows:
“Even though the practice which is centred on the transformation of bardo, death and the
intermediate state into the three bodies of the Buddha is the generation process (bskyed
rim), the actual transformation of these bodies is considered to take place not during the
generation process, but rather at the culmination of the completion process (rdzogs rim).
This apparent contradiction seems to be the result of a historical process in which initially
the generation process may have been an autonomous transcendent process in its own
right, leading to the attainment of complete enlightenment. But later on, the emphasis
was transferred to the completion process, and it received the primary role, while the
generation stage became its preliminary step, in which only similitudes of the true
transformations take place. Still, these similitudes are regarded as eventually enabling
the ripening of the true transformation during the completion process.”217
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
216 !Kilty 2013, pps.118-19.!
217
Bentor 2006, p.186 fn4
! 68!
These categories of generation and completion stages first emerged in the
literature on the Guhyasamåja Tantra, in particular in Chapter Eighteen, the Uttaratantra;
they were later developed in commentaries on the nidåna verses of the Vajramålå. Tsong
kha pa himself dedicated great effort to parsing the elements of sådhana practice into
these two stages. 218 The notion of “similitude” vs “true transformation” proposed by
Bentor will continue to inform our discussion of the two stages in the next chapter.
II. Issues Regarding the Mapping of Deities onto the body in Mkhas grub’s Ocean of
Attainment
Having outlined a basic schematic of the Guhysamåja body mandala practice, we will
now look closely at a few key points from Mkhas grub’s text. Our goal is to draw
attention to the ways in which this fifteenth-century Tibetan scholastic is negotiating
different versions of the practice presented in various Indian texts accepted as
authoritative within the Årya tradition, the texts linked with Någårjuna and his
disciples.219 In doing so, we will learn more about Mkhas grub’s perspective on the
Buddhist tantric tradition at large and the importance of the Guhyasamåja system within
it. We will also observe the ingenuity required of Tibetan commentators in adhering to
the standards of textual authority while struggling to account for discrepancies amongst
authoritative texts. Mkhas grub’s commentarial strategies produce a portrait of his own
identity as a writer, but they also contribute to the formation and distinction of a tradition
of descent from Tsong kha pa that over time became identified as the dGa ldan pa or
Gelukpa [dge lugs pa] tradition. Finally, the citations he provides suggest themes in the
evolution of the Guhyasamåja body mandala practice itself within the Indian context and
its Tibetan reception.
The main points to be addressed in this section of the chapter concern the
principles of mapping and correlation in Mkhas grub’s writings on body mandala. The
featured examples are issues surrounding mapping the Buddha families and the
goddesses Locanå, Måmakî, Påñ∂aravåsinî, and Tårå onto the body as well as the
association of those goddesses with the elements and winds of the body. We will consider
how the details of these aspects of the practice connect with themes in the evolution of
tantric ritual over time.
Some techniques of mapping deities onto the body in association with aspects of
the psycho-physical person reveal a refined understanding of the body’s hidden structures
and processes. These understandings build upon or operate in relation to existing systems
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
218
Tsong kha pa as well as many Indian commentators proposed different methods for
relating these different schema. Tsong kha pa devoted particular attention to relating the
body, speech and mind isolations [dben Skt. viveka] to other schema such as the six
yogas and five stages. See Wayman 1977 (2005 Reprint), pp. 156-180 for a more detailed
discussion.
219
In fn2 of the introduction to their 2009 edition of the Vajrasattva sådhana, Tomabechi
and Hong make the important point that there is no Sanskrit evidence for the term ‘Årya
tradition’ [Tib. ‘Phags lugs] and therefore, we must be cautious about projecting such
terms upon the Indian context. They refer likewise to Tomabechi 2008 or further detail.
! 69!
such as the elements, skandhas, and other lists of bodily constituents. This conjunction of
systems of meaning may require modifications to those existing systems.220 Some
scholars have theorized that the earliest Guhyasamåja materials lack the subtle body
practices that later became a prominent feature of their exegesis and ritual practice.221
Others contest that the systems of winds so profoundly developed with Buddhist tantra
have their origins in the Vedas while others find these origins in the Upanißadic
context.222 Further influences might be traced from non-Buddhist tantra and
Ayurveda.223 However, it is important to consider influences from within the network of
Buddhist tantra as well. For example, some of the commentarial traditions of the
Guhyasamåja coincide with the composition of texts regarded as yoginî tantras, texts
renowned for their sophisticated formulations of the inner workings of the subtle body.
Therefore, Indian commentators on the Guhyasamåja may have been influenced by the
yoginiî tantras. Moreover, the working knowledge of the body informed by both the
fifteenth-century Tibetan tantric and medical discourses further contributed to this
evolution. The competition and prestige involved in presenting a compelling and
efffective ritual means for tapping into the body’s utmost potential motivated the
production of ritual texts. These text employ the principles of mapping and correlation
oriented around the body to navigate the very boundaries of life and death.
Our first example deals with a tension between accounts of mapping the buddha
families onto regions of the body and accounts of mapping them onto more precise
points. In encountering the representations of the body presented by Mkhas grub and his
creative navigation of the gaps between them, it may be fruitful to consider: How specific
is this mode of mapping? Our second example demonstrates how through correlating the
goddesses with elemental winds [rlung], Mkhas grub emphasizes the completion stage of
the Guhyasamåja as a technology for navigating the death process, an imperative of direct
relevance to fifteenth century tantric exegetes.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
220
In the previous chapter, we discussed an example of such a conjuntion in the transition
from a three to five-Buddha family schema and its impact upon the mandala and its
correlations. Once again, see Snellgrove 1987 (2002 Reprint) pp198-213.
221
See Kittay 2011’s synopsis of historical developments in tantra, based upon the work
of Weinberger. Kittay 2011, pp.121-22 and Weinberger 2003, p37.
222
Kittay 2011 cites Hartzell (1995 & 1997) and Samuel 2008 respectively. Kittay adds:
“Hartzell also notes the features of Tantra that do not appear in the Bråmanas ‘in any
germinal form,’ including seed syllable mantras, the aim of liberation as opposed to
immortality through these practices, sexual rites not concerned with procreation, the
central role of elaborate visualizations of deities (although there is an idea of ‘entering
deities’) male/female couples and Buddhist notions of emptiness and the like.” See
Hartzell 1995, p.125-127 and Kittay 2011, p. 128 & fn354.
Patton 2004 (pp.47-8) provides references to the internalization of the sacrifice or
“cosmology of the sacrifice” in the B®ha∂aranyaka-Upanißad, 3.1.5 & 3.1.8-10. Her
reference to the correlation of worldly rivers with rivers inside the body is of parricular
interest for considering the origin of the channels. [see B®ha∂aranyaka-Upanißad, 1.1.1].
I am grateful to Alexander von Rospatt for directing me to this portion of Patton’s work.
223
See Timalsina 2012 and Wujastyk 2009 respectively.!
! 70!
As we proceed to look closely at a few of the issues he raises, his critiques, his
theories, and the sources upon which he relies for support, we begin the slow but
important work of contextualizing his writings on body mandala within a larger project of
tradition-building. In particular, we will consider the manner in which he elevates and
distinguishes what he envisions to be his own tradition in relation to others. Observing
the manner in which he forms his claims, accepting the positions of some writers and
refuting those of others, is a key part of this process.
IIA. Mapping the Five Buddha Families onto the Body
Following Wright’s survey of the scholarship on dating the texts of the Guhyasamåja
cyle, the first seventeen chapters of the root tantra might be dated to the late eighth-
century while Chapter Eighteen, known as the “Subsequent Tantra” [Uttaratantra], to be
discussed a bit later, seems to have been completed by 800 CE. 224 The texts of
Någårjuna, Åryadeva, and Candrakîrti form the corpus of commentaries commonly
referred to as the Årya tradition; thse texts are taken as authoritative by those who, like
Mkhas grub, claim inheritance of that tradition. Many scholars agree that although these
authors share the names of renowned Madhyamaka philosophers, they are distinct from
and lived centuries later than them. Wedemeyer dates the texts of Någårjuna and
Åryadeva to 925-1025 CE and those of Candrakîrti to 950-1057 CE.225 Determining
precise dates for these texts is beyond the scope of this dissertation, but it may be helpful
to keep this general outline in mind.
We begin by examining an excerpt on mapping the five Buddha families onto the
body from the first half of Mkhas grub’s text; this excerpt appears in the context of
refuting the traditions of others. Specifically, Mkhas grub is contesting the assertions of
some proponents of the Guhyasamåja who claim that only seeds syllables are arranged on
the body, not deities. Mkhas grub insists that the authorative Indian sources of the Årya
Guhyasamåja tradition support the location of deities upon the body. We will follow
Mkhas grub in his journey through an elaborate program of textual exegesis, a journey he
undertakes to prove that body mandala ritual rather than mere syllable placement or
nyåsa, is indeed substantiated by the textual sources.
“As for the arrangement of just the seed syllables of the deities, not arranging the deities
on the body, in the Mdor byas [Piñ∂ik®ta ] it says:
‘Having arranged ‘thlim’ on the eye(s),

One should visualize Kßitigarbha.
Arranging ‘aom’ on the ears,
One should visualize Vajrapåñi.’
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
224
See Roger Wright, “A Survey of Explanatory Tantras and Commentaries of the
Guhyasamåja Tantra,” ?Unpublished paper. Accessed via:
http://www.wrighrp.pwp.blueyonder.co.uk/SOASMA2010/GuhyasamajaCommentarySur
vey-corrections-20100718.pdf
225
IBID. See also Wedemeyer 2007.!
! 71!
That (statement by some follower of the Guhyasamåja claiming the arrangement of just
seed syllables and not deities) contradicts the explicit explanation for visualizing the
deities (Piñ∂ik®ta ).226 The Rnam gzhag rim pa says:
‘From the arrangement of the five tathågatas, the five aggregates (skandhas) will become
the cause of enlightenment.’227
And further,
‘The cause and result of arranging the bodhisattvas on the sense spheres, on the eyes and
so forth are taught.’
(This statement) explains arranging the (actual) deities (vs. just syllables).”228
Nåjårjuna’s Mdor byas [Piñ∂ik®ta] and Någabodhi’s Rnam gzhag are both Årya cycle
texts dealing with the generation stage practice of the Guhysamåja. Mkhas grub
continues:
“Moreover, in particular, (there is) that very arrangement of the body mandala, from the
eighth chapter of the [GS] root tantra:
‘from between the breasts up to/in the middle of the crown
by the ritual specialist, moreover between the feet/legs
in the navel-waist-secret, the bodhisattva (lit. son of victor)
The arrangement of the five families is to be performed.’
[nu ma'i dbus bar spyi gtsug mtha' yi bar
cho ga shes pas yang na rkang pa'i bar
lte ba rked pa gsang par rgyal ba'i sras
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
226
See 239.2-240.4 in Mkhas grub’s text for the claims of the opponents.
227
This quote is derived from Chapter Two of the Rnam gzhag. See Tanaka’s partial
edition of this text 2001-2, 2004 & 2009).
gyur ro usually indicates the future tense. rgyu dang 'bras bu may refer to the seed
syllable and body of deity respectively. [Khenpo Choying Dorje, Personal
communication, Spring 2011]
!
228 lus la lha mi 'god par lha'i sa bon 'ba' zhig 'god pa ni mdor byas las; Thi lim mig tu
bkod nas ni ; sa yi [243.2] snying po rnam par bsgom ; aom ni rna ba dag la dgod ; phyag
na rdo rje bsgom par bya ; zhes sogs kyis lha bsgom par dngos su bshad pa dang 'gal
zhing; rnam gzhag rim par yang; de bzhin gshegs pa lnga bkod pa las phung po lnga
byang chub kyi rgyur [243.3] 'gyur ro ; zhes dang; mig la sogs pa'i skye mched rnams la
byang chub sems dpa' rnams dgod pa'i rgyu dang 'bras bu gsungs pa
zhes sogs kyis lha nyid dgod par bshad cing
!See 241.1 for a similar quote from the same text. It may be fruitful to consider the
quality of the different texts of the Guhyasamåja cycle in relation to one another,
contemplating how such comparisons may inform our understanding of an evolving
genre of tantric commentarial literature. For example, can we say that the µdor byas is
more like sådhana and Rnam zhag is more like commentary? [The connections were
initiated in a discussion with Khenpo Yeshe, Spring 2011]
! 72!
rigs lnga rnams ni dgod pa rab tu bya]
It is taught by this quote from the Mdo bsres.” 229
At present I have retained a word by word translation of this verse in order to

leave room for the discussion of different interpretations of its meaning. It would perhaps
be more accurate to say: “the arrangement of the five families (and associated)
bodhisattvas should be performed by the ritual specialist (onto the regions) ranging from
the breasts up to the crown, and moreover, between the feet/legs, in the navel-waist-
secret.” Alternatively, we might read: “The ritual specialist should arrange the five
families between the breasts, ?in the middle of the crown and between the feet and the
bodhisattvas in the (region of) the navel, waist and secret place.” For the time being, we
will refrain from determining a conclusive translation. Instead, we will focus upon what
this quote meant to Mkhas grub, other possible interpretations, and how Mkhas grub goes
about establishing his own interpretation in the face of those alternatives.
The primary citation is derived from Chapter Eight, verse nine of the
Guhyasamåja Root Tantra.230 The quote appears in 203.4 in the sde dge bka’ gyur
edition. However, the form in which it appears there is somewhat different from the form
it takes in Mkhas grub’s text:
nu ma’i dbus gang spyi tsug mtha’ dbus su

cho ga shes pas rkang pa’i bar du yang
lte bar ro smad gsang ba rgyal ba’i sras
rigs lnga rnams ni dgod par rab tu bya
The significance of the appearance of gang rather than bar of the phrase in Mkhas
grub’s text as nu ma’i dbus bar and dbus su vs. yi bar of spyi gtsug mtha' yi bar is
debatable. However, when we look at Candrakîrti’s commentary, we will see how the
meaning of bar as ‘between,’ rather than simply ‘in the middle,’ is fully exploited in
locating Amitåbha in the region of the throat, ‘between’ the crown and the heart center.
The most significant of the differences is the compound lte bar ro smad gsang ba. The
locative has been removed from gsang ba and added to lte ba, and ro smad, the “lower
body,” appears instead of rked pa.231
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
229
khyad par du yang; lus dkyil 'god pa de nyid rtsa rgyud kyi le'u brgyad pa [243.4]
las...zhes pas bstan par mdo bsre las gsungs la
230
See Matsunaga’s 1978 edition of the tantra, p.24 for the Sanskrit text:
stanåntaraµ yåvacchikhåntamadhye carañåntare cåpi nyased vidij•å˙/
nåbhika†iguhye jinåtmajånåµ nyåsaµ prakuryåt kulapa•cakånåm/
Matsunaga 1978 (fn16) states that two manuscripts, BG & BT read valgåntare vs.
carañåntare. Fremantle’s 1971 edition likewise footnotes this an alternative and also
makes note of a manuscript C that reads spar!åntare instead.
231 !IW of thlib [Accessed: 10/23/2013] defines ro smad as “lower body.”
The dpe bdur ma edition of the GST [locate bibliographic info-recent publication] is
identical to the sde dge except that it reads no locative into the compound: lte bar ro smad
gsang ba. See dpe sdur ma publication 317 lines 6-8.!
! 73!
The context in which the verse quoted here appears in the root tantra itself refers
to practice with a consort accompanied by offerings of the impure substances.
However, it is important to remember that Mkhas grub is extracting this quote
from the Guhyasamåja Tantra via Någårjuna’s text, the Mdo bsres, a commentary on
generation stage practice based upon the first seventeen chapters of the Guhyasamåja
Tantra.232 The form of this short text of about ten folia sides is one in which Någårjuna
appears to be extracting quotes from different portions of the root tantra (not in
chronological order) and explaining how they refer to different elements of the generation
stage practice. The sde dge edition of this text reads nu ma’i bar nas klad pa’i rgya bar du
vs. nu ma'i dbus bar spyi gtsug mtha' yi bar. The nas...bar du construction supports an
interpretation of the quote as referring to a span of the body from between the breasts up
until the uppermost portion (of the body).233 The second line adds a verb, with rkang
mthil bar du dgod instead of rkang pa’i bar du yang. It also tells us that the lowermost
portion referred to is the ‘soles of the feet.’ In addition, it is more easily read as “arrange
up until the soles of the feet” than the somewhat confusing designation “between the
feet.” Therefore, this version of the quote from Chapter Eight of the Guhyasamåja Tantra
describes a ritual specialist performing an arrangement of deities spanning from between
the breasts up to the head and then down to the soles of the feet. The emphasis here
appears to be upon covering the entire span of the body rather than targeting specific
sites.
Någårjuna introduces this particular quotation within his text as follows: “Since
beings with awareness, possessed of form, with bodies with the nature of wisdom, taught
in Chapter One, are not able to manifest as sentient beings, Chapter Eight says...”234
Någårjuna seems to be setting up the location of deities within or even as the human body
in terms of the broader theme of compassionate embodiment in a nirmåñakåya form. The
author then concludes discussion of the quotation with: “All the tathågatas enter the body
(with) the nature of wisdom in the nature of the skandhas and so forth through the
mantric consecration [sngags kyis byin gyis brlabs] that is taught in that quote (from the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
232
The quote appears on 24.5-.6 of the sde dge bstan ‘gyur edition of the ¸rî
guhyasamåja-mahåyogatantropattikrama-sådhana-sütra- melåpaka-nåma; Rnal 'byor chen
po'i rgyud dpal gsang ba 'dus pa'i bskyed p’i rim pa'i bsgom pa’i thabs mdo dang bsres pa
zhes bya ba; Translators: Dharma!ribhadra and Rin chen bzang po; P2662, Vol. 60-61;
T1797.
The last two lines of the quote are virtually identical to Mkhas grub’s version.
!
233 !JV, IW and RY of thlib.org all interpret the term klad pa as “the uppermost portion”
or the “brain” or “mind.” (Accessed 10/24/2013) The latter seems inappropriate as within
Indian and Tibetan traditions, the mind is commonly conceived of as abiding in the heart
region. As for the ‘brain,’ it would likely be necessary to consult early Tibetan medical
sources to understand how specific the understanding of its location was. Therefore, for
the time being, it would be best to translate klad pa as referring to the uppermost portion
of the body.!
234
mdo bsres [24.4-.5] : da nas le'u dang po las gsungs pa'i rig pa'i skye bu gzugs can ye
shes kyi rang bzhin gyi lus kyis sems can gyi ngon bya bar mi nus pas na le'u brgyad pa
la
! 74!
tantra).”235 Någårjuna then links our quote to the description of the skandhas as the five
buddhas and of the sense spheres as the “bodhisattva mandala” from Chapter Seventeen
of the tantra.236 Next, he refers to the body mandala explicitly (sku’i dkyil ‘khor):
“Then, after imbuing [zhugs] (the body with) the deities of the body mandala, for the
benefit of disciples, one should cultivate oneself as Vajradhara in his three-faced aspect.
This very point is taught from Chapter One: ‘The form of the being of great awareness,
the lord of all tathågata, consecrates (them) with mantra. As soon as they are
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
235
mdo bsres [24.6]: zhes bya ba la sogs pa gsung pa'i sngags kyis byin gyis brlabs pas
phung po la sogs pa'i ngo bo nyid la de bzhin bshags pa (read gshegs pa) thams cad ye
shes kyi rang bzhin lus la rjes su zhugs par bya'o
236
The Buddha-skandha correlation should not be taken for granted in our approach to
the Guhyasamåja materials; it is a correlative set that evolved over time and in relation to
other sets of syllables, colors, and bodily locations. Though we have the relation of the
five wisdoms and five families emerging in the STTS, the skandhas don’t appear to be
included there. Snellgrove doesn't address this issue directly but rather says that the
scheme of elements and skandhas is "readily adaptable" to the five buddha family
formation. See Snellgrove 1987 (2002 Reprint), p201.
In my preliminary attempts to locate the source of the correlation of Buddhas with
skandhas, the literature of the Guhyasamåja cycle appears to be the context in which the
connection is solidified. As will be discussed below, in his Mdor byas, Någårjuna refers
to Chapter Seventeen of the root tantra for this correlation. However, the specific
correlations are not elaborated there.
Wayman 1977 seems to highlight a rather tenuous link between the skandhas and
Buddha families in Chapter Sixteen of the root tantra; he expects the reader to apply this
connection to the correlation of Buddha families and poisons and in Chapter 18, the
Uttaratantra, which we know to be a later addition. The Piñ∂ik®ta-sådhana and Vajramålå
(and its forty nidåna verses also cited in the in Candrakårti’s Pradîpoddyotana and briefly
in Åryadeva’s Caryåmelåpakapradîpa) are among the texts from the Årya cycle
referenced by Wayman in this regard. See See Wayman 1977 (2005 Reprint), pps. 208-
214.
Mapping the buddhas onto regions of the body, although not as specific as onto
cakras or vital points allowed for a more location-oriented model than the Buddha-
skandha correlation. Though one might be able to imagine Vairocana to be in the nature
of the form aggregate in an abstract sense, it is difficult to locate the aggregates, in
particular the non-material aggregates, in specific bodily sites. It is necessary to look at
the individual sådhanas and commentaries to better understand how these two schema for
correlating the five Buddha families with bodily regions and skandhas respectively
developed in relation to one another. We may also look at later developments in
Buddhist tantra such as those of the Cakrasaµvara system in which preliminary practices
include the more abstract correlation of Buddhas with skandhas while the body mandala
practice proper locates deities more specifically upon the body.
!
! 75!
consecrated, the Buddha, the mind of enlightenment vajra, becomes visible to all the
tathågatas as the three-faced tathågata.’ So there is the teaching of atiyoga.”237
It seems likely the Mkhas grub is interpreting the quote through Någårjuna’s text.
Though Mkhas grub chooses not to quote any of Någårjuna’s commentarial notes from
the text, it is possible that he assumes his readership to be familiar with the Mdo bsres;
specifically, he may assume his readers are aware of the tradition of reading four stages
of generation stage practice, the four yogas (yoga, anuyoga, atiyoga and mahåyoga) onto
the root text of the tantra. Wayman describes how these terms are used by Någårjuna to
structure his Pin∂îk®ta-sådhana, a generation-stage text we will examine more closely
below. 238 Wayman observes:
“...in yoga there is the rite involving the recitation of the celebrated mantras, Oµ
!unyata...With verse 51, he (Någå®juna) mentions the anuyoga and this culminates in the
contemplation of the ‘primeval lord’ (ådinåtha). Then the atiyoga develops
vajrasattva...as the progressed self of the yogin with his body as a mandala...The
mahåyoga starts with verse 70 and involves the blessing of the empowerment of body,
speech and mind, using the mantras of Guhyasamåja, Chapter 6.”239
Wright 2010 specifies that the atiyoga “consists of transforming oneself, as the
Ådibuddha, into the form of Akßobhya and visualizing the deities and their attributes
placed on various parts of the body...”240 Understanding the context of the quote in
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
237 !mdo*bsres*[24.7M25.!2]:!!da*ni*sku'i*dkyil*'khor*gyi*lha*rnams*la*rjes*su*zhugs*pas*
zhal*gsum*lta*bus*gdul*bya*rnams*kyi*don*du*rdo*rje*'chang*gi*bdag*nyid*du*bsgom*par*
bya'o;*don*'di*nyid*bstan*pa*di*le'u*dang*po*las;*rig*pa'i*skye*bu*chen*po'i*gzugs*da*
(read!de)*bzhin*gshegs*pa*thams*cad*kyi*bdag*po*sngags*kyis*byin*gyis*brlab*so;*
byin*gyis*rlabs*ma*thag*tu*bcom*ldan*'das*byab*(read!byang)*chub*gyi*sems*rdo*rja*
(read!rje)*de*bzhin*gshegs*pa*zhal*gsum*pa*lta*bur*de*bzhin*gshegs*pa*thams*cad*kyis*
gzigs*par*'gyur*ro*zhes*gsungs*pa*ni*shin*tu*rnal*'byor*ro!
238
See Wayman 1977 (2005 Reprint), p.156-163 for a detailed description of these terms
and their deployment by Någårjuna and Tshongkhapa, among others.
239
Wayman 1977 (2005 Reprint), p.157-58. See also the description of the four yogas in
Wright’s 2010 translation and study of the Pin∂î-k®ta, pp. 31-33 and Appendix A.
Wayman also highlight an interesting passage from Tsong kha pa’s Sngags rim chen mo
(f.364b) in which the author correlates the four yogas with those presented in the
K®ßnayamåri-tantra. (Wayman, p157).
240
Wright 2010, p.32. For the atiyoga section of the Pin∂î-k®ta, see v.52-68. Wright
makes an interesting point about the preceding stage of anuyoga: “Later commentaries,
from those attributed to Candrakîrti onwards, explain this yoga as representing the
exercise of the body, speech and mind of the Saµbhogakåya, which is consistent with the
structure of the sådhana since the previous section manifested the Dharmakåya (v.44ff)
and the following section manifests the Nirmåñakåya.” (Ibid) Therefore, this might
suggest that the three Buddha body model that Tsong kha pa was so invested in reading
into the Guhyasamåja in terms of the stages of birth, death and the intermediate state does
! 76!
Någårjuna’s Mdo bsres, therefore allows us to situate it within the enactment of
generation stage ritual in the Årya tradition, the larger topic of Mkhas grub’s text.
Mkhas grub follows his citation from Chapter Eight of the root tantra via the Mdo
bsres with a citation from Candrakîrti’s Pradîpoddyotana. Here, the correlation of the
Buddhas with regions of the body is situated in the yoga and anuyoga stages of the ritual
drama:
“From the 'grel pa sgron gsal (it says):

‘In terms of the stages derived from the sådhana, in the yoga and subsequent yoga
(anuyoga), for the purpose of demonstrating oneself and one’s mudrå to make the five
families, it says “nu ma'i dbus,” and so forth.241
As for the statement, “nu ma'i dbus,” it is abode of the heart.
When it says “bar” (it means), that in the section in between there is the syllable ‘hum’
and Akßobhya’s family is arranged here.242
When it says “spyi gtsug mtha'” (the edge of the crown of the head), the words “spyi
gtsug mtha'” mean the edge of the hair [lit. starting with the hair].
From the edge of the hair up until the root [skye gnas], arrange Vairocana who arises
from the letter ‘aoum.’243
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
not appear in the interpretation of this tantric cycle before Candrakîrti. For more on
Tsong kha pa’s correlation of these triads, see Wayman 1977 (2005 Reprint).
241
Khenpo Choying Dorje explains rigs lngar bya ba as meaning ‘transforming into’ the
five familes, much like byin*gyis*brlabs*would be used to express blessing or
transformation. He adds that both Mkhas grub and Tsong kha pa say that one can’t
transform one’s actual body, but rather a future body. He also makes the important point
that the sgron gsal leaves room for disagreement on this issue. [Personal communication,
Spring 2011]
It is of note that the Pradîpoddyotana’s account refers to transforming the bodies of both
practitioner and consort into the bodies of the five Buddha families.
242
The Sanskit humkårajam makes explicit the fact the Akßobhya’s family is produced
from the syllable huµ.
243
Perhaps we might translate this as “as far as it is able to grow.” This section is difficult
to translate definitively. It seems to be saying that Vairocana is imagined from the root to
the tip of the hair, but might also mean from the crown up until the spot above the
forehead where hair begins to grow. The 2010 Sanskrit edition of the text reads:
!ikhåntaµ iti/ !ikhånta!abdena ke!å upalakßyante yåvat ke!åntaµ/ ke!abhumau
aomkårajam vairocana nyåset/
See Candrakîrti. 2010. Guhyasamåjatantrapradîpodyotana†ikå Ía†ko†ivyåkhyå of Åcårya
Candrakîrti. Ed., Ngawang Samten and S.S. Bahulkar. Dhîh: Journal of Rare Buddhist
Texts Research Department. Sarnath, Varanasi: Central University of Tibetan Studies.
ke!åntaµ corresponds with the Tibetan skra'i mtha' and is defined by Monier-
Williams as both “the border of the hair on the forehead” and “lock of hair, tuft,” the
latter presumably referring to the brahman’s topknot. Monier-Williams likewise defines
ke!abhümi (the Sanskrit translaion of skra'i skye gnas ) as the “the skull on which hair
grows. [MW definition accessed on 10-24-2013 via www.lexica.indica-et-
buddhica.org/dict/lexica]
! 77!
As for the “cho ga shes pas,” it is the one who knows the generation stage.
When it says “bar,” (it means) between the heart center and the crown.
If you ask why, (it is because) the region of the throat is the mandala of the mouth [lkog
ma'i phyogs kha'i dkyil 'khor].244 There, Amitabha is to be arranged arising from the
letter ‘å.’
When it says “rkang pa'i bar,” it means between the foot and the root of the thigh, (that
is) “between the foot/leg.”245
As for the word “dang”, the meaning condensed (therein) is the two feet.
There, Amoghasiddhi is to be arranged arising from the letter ‘hå.’
As for “lte ba rkad pa gsang ba,”246 [lit. navel-waist-secret]
Here, the seventh (case) [rnam dbye] is invisible.247
In these places, Ratnsambhava is to be arranged arising from the letter “sva” [Tib. soha ].
As for rgyal ba (“the victor”), that refers to Vairocana and so forth.
As for rgyal ba'i sras (“the bodhisattva”), there are the families of those (buddhas).
So “rgyal sras” refers to the Buddhas and bodhisattvas, Vairocana and so forth.
The ritual which explains the families of those from the sådhana says to arrange each of
the five families just like this.’
This teaching clearly explains generation in the manner of arising from the seed (syllable)
of each (deity).”248
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
In his contemporary writings on the Guhyasamåja practice, discussed in the first
part of this chapter, Geshe Lobsang Tsephel locates Vairocana “between the crown and
the hairline.” (Paths and Grounds, p.34). His description also correlates the buddhas with
the skandhas. He reads the buddhas onto five regions: between crown and hairline,
hairline and throat, throat and heart, heart and navel and navel and groin. This description
appears to co-exist alongside a more localized one focused on the crown, throat, navel
and groin. In that version, Akßobhya, the main deity is fused with the body of the
practitioner, though we might simultaneously interpret imagining him at the heart (though
this is not explicitly stated in this account.) See Geshe Lobsang Tsephel, Paths and
Grounds, p.26.
244
Khenpo Choying Dorje remarked upon the somewhat unusual use of kha’i dkyil ‘khor
(mandala of the mouth/throat) as we usually find mgrin pa’i dkyil ‘khor. I suggested a
possible translation as ‘speech mandala.’ [Personal communication, Spring 2011]
245
The fact that rkang pa may refer to either the foot or the leg may help to explain how
the bar is working here as referring to the entire span of the leg.
246
Read rked pa vs. rkad pa.
247
In Sanskrit grammar, the seventh case is the locative case.
248
de'i 'grel pa sgron gsal las;sgrub pa’i thabs [243.5]las byung ba'i rim gyis rnal 'byor
dang rjes su rnal 'byor byas la ; bdag nyid dang rang gi phyag rgya rigs lngar bya ba
bstan pa'i phyir; nu ma'i dbus zhes bya ba la sogs pa gsungs te; nu ma'i dbus zhes bya ba
ni snying kha'i gnas so ; bar zhes bya ba ni de srid kyi [243.6] bar 'dir yi ge hum ste
mi bskyod pa'i rigs dgod pa'o ; spyi gtsug mtha' zhes bya ba ni spyi gtsug gi mtha'i sgras
skra nye bar mtshon te; skra'i mtha' ste skra'i skye gnas ji srid pa'i (sde dge reads kyi vs.
pa’i) par du yi ge aom las byung ba'i rnam par snang mdzad dgod ro (sde dge reads do vs.
ro); cho ga shes pas [244.1] zhes bya ba ni bskyed pa'i rim pa shes pas so; bar zhes bya ba
! 78!
The most obvious addition provided by the Pradîpoddyotana is the explicit
correlation of sites on the bodies with specific buddhas and their association with
respective seed syllables from which they arise. Furthermore, there is enhancement and
revision of the bodily sites with which these buddhas are associated. For example, in
clarifying the location referred to as “between the feet,” Amoghasiddhi is located in the
region from the feet to the root of the thigh. The commentary also adds the “mandala of
the mouth” in the throat region, inhabited by Amitåbha, a site not explicitly named in the
root. On a general level, both the root text and Candrakîrti’s commentary appear to be
locating the buddhas of the five families in regions of rather than points on the body. The
root text only directly names three or four regions or spans of the body [nu ma'i dbus bar;
spyi gtsug mtha' yi bar ; rkang pa'i bar ; lte ba rked pa]. The Pradîpoddyotana, on the
other hand, specifically refers to a fourth site in the throat and explicitly parses the lower
body as the abodes of Ratnasambhava and Amoghasiddhi to total five. We might be
tempted to read other versions of mapping deities onto the body, such as those that
separate the navel and genitals as separate vital points, onto this model. However, the
Pradîpoddyotana reads them as a single region inhabited by Ratnasambhava.
The Pradîpoddyotana does, in a few instances, add further specificity to the
assignment of bodily locales; for example, it clearly identifies the area between the
breasts as the abode of the heart. It also appears to be emphasizing that bodhisattvas
associated with the Buddha families are likewise being mapped onto the body although it
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
ni snying kha dang sbyi gtsug gi bar te; gang yin zhe na; lkog ma'i phyogs kha'i dkyil
'khor te; der yi ge aa las byung ba'i od dpag tu med pa dgod par bya'o ; rkang pa'i bar zhes
bya ba ni rkang pa dang brla’i ( sde dge reads brla vs. brla’i) [244.2] rtsa’i ba'i bar du ni
rkang pa'i bar ro (sde dge reads bol vs. bar ro); dang zhes bya ba'i sgra ni rkang pa gnyis
su zhes bsdu ba'i don te; der yi ge haa las byung ba'i don yod par grub par dkod bar (sde
dge reads dgod par) bya'o ; lte ba rkad pa (sde dge reads rked pa) gsang ba zhes bya ba ni;
'dir bdun pa mi mngon par byas pa yin te; gnas 'di dag tu sba' [244.3] las byung ba'i rin
chen 'byung ldan dgod par bya'o ; rgyal ba ni rnam par snang mdzad la sogs pa'o ; rgyal
ba'i sras ni de'i rigs te; rgyal ba dang rgyal ba'i sras dag ni rgyal sras zhes bya ste; rnam
par snang mdzad la sogs pa dang; de'i rigs rnams sgrub pa'i thabs [244.4] las bshad pa'i
cho gas rigs lnga'i bdag nyid dgod pa ji lta ji lta (sde dge reads ji lta ji lta) bar bya'o zhes
pa'o ; zhes gsungs pa 'dis rang rang gi sa bon las byung ba'i ltar bskyed pa dngos su gsal
bar bshad cing
This quote is found on 118.4-119.3 of the sde dge bstan ‘gyur edition of the
Pradîpoddyotana-nåma-tîkå (sgron ma gsal bar byed pa zhes bya ba’i rgya cher bshad pa)
by Candrakîrti (zla ba grags pa) Vol.29 (ha) p.2-402 (ff.1v-201v). I have compared this
edition with Mkhas grub’s citation and annotated the minimal differences in parentheses.
The only variation of potential significance is 244.2 rtsa’i ba'i bar du ni rkang pa'i bar ro
for which sde dge (118.7) reads bol vs. bar ro. According to RY on thlib.org (Accessed
10/18/2013) bol can mean the upper part of the foot whereas for IW it may mean the
upper part of the leg.
For a Sanskrit edition of this passage see the 2010 publication by Ngabang Samten &
S.S. Bahulkar. pps. 114-115.!
! 79!
remains unclear where precisely they are being located. In other versions of the practice,
we commonly find the bodhisattvas mapped onto the sense spheres, vajra, and joints.
In order to reveal the evolution of notions of the subtle body inherent in changes
in modes of mapping deities onto the body, it is vital that we resist the tendency to
impose what may be a later formulation of the practice upon open-ended or idiosyncratic
passages we encounter. 249 Doing so allows us more freedom to generate potentially
fruitful questions with regard to this model of the body mandala. Such questions include
the following. Is it possible that the mode of mapping buddhas onto segments of the
body rather than cakras or vital points as found in the Pradîpoddyotana’s interpretation of
the verse from Chapter Eight of the root leaves open the question of whether the
manipulation of energies through the particular bodily centers is involved in the practice?
When did the practice of “piercing the pith” [gnad du bsnun], tapping into specific vital
energy centers of the body through mantra recitation and visualization practice, become
prominent within the Guhyasamåja system? 250 In the next chapter, we will begin to
closely examine this notion of “piercing the pith” and the way it functions as a defining
quality of body mandala for both Mkhas grub and Ngor chen. In terms of the evolution
of tantric ritualized knowledge of the body, however, the Pradîpoddyotana’s description
does not suggest a model of the body focused upon the “pith,” but rather a more general
localization of a bodily totality.
Mkhas grub concludes his discussion of these passages from the root tantra via
the Mdo bsres and Pradîpoddyotana as follows:
This is an explanation for arranging the five families and all the deities of those (families)
upon both one’s own body and that of the consort. Therefore, (in my mind) the one who
claims this arrangement of deities is not the explanation of 'phags pa yab sras [the Årya
fathers and sons] has not seen the Pradîpoddyotana.251
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
249
An excellent example of this tendency is found in Fremantle’s 1971 translation
of this verse as “knowing the ritual, he should place the Five Families, the sons of the
Jinas, on the forehead (fn5), the throat, the heart, the navel and the genitals.” Fremantle’s
translation reflects a degree of confidence in mapping Buddha families onto a set a five
cakras that is premature in the case of the root tantra. In this way, Fremantle obscures the
technology of parsing the body into regions suggested by the root text and preserved, to a
degree, by Candrakîrti’s commentary (despite the fact that the author relies heavily upon
this very text in the translation of the root tantra.) The compound lte ba rkad pa gsang ba
which in Fremantle’s Sanskrit edition of the root tantra corresponds with nåbhîka†îguhye
presents particular challenges. In the case of the root, Fremantle is parsing this as “navel
and genitals.” Her translation therefore creates a system of five sites that overlooks
“rkang pa'i bar” and divides lte ba and gsang ba without accounting for rkad pa.
250
The definition of body mandala practice as that which “pierces to the pith of the body”
[lus la gnad du bsnun] will be discussed in further depth in the next chapter of the
dissertation. I have yet to locate a Sanskrit equivalent for this term. This may suggest
that its origins and emphasis lie in the Tibetan rather than the Indian tradition.
251
Khenpo Yeshe confirmed that here Mkas grub is mocking the proponent who thinks
you just arrange syllables rather than the deities themselves. [Personal Communication,
Spring 2011]
! 80!
In other words, Mkhas grub is defending the validity of this mode of mapping
deities rather than just seed syllables onto the body by securing its position with the
authoritative texts of the Årya tradition.252 Finally, Mkhas grub proposes a theory of
accommodation of two different modes of mapping deities onto the body:
Furthermore, in terms of this quotation from the root tantra that was just cited, if one
follows the same model of (explaining) arranging Vairocana (in the region) from the edge
of the hair [skra mtshams ] to the crown and Amitabha from the throat [lkog ma ] to the
edge of the hair, and Akßobhya, from the heart center to up to edge of the throat, then it
seems permissible to arrange Ratnasaµbhava from the heart to the navel and
Amoghasiddhi from the lower edge of the navel up until the root of the thigh.
In accord with the literal understanding of the (section from the) Pradîpoddyotana that
was just cited,253 Ratnasambhava pervades the three, the navel, waist, and secret place,
and Amoghasiddhi appears to be arranged from the base of the thigh up until the toes.
However one chooses to arrange those two, it is fine.254
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
rang dang phyag rgya gnyis ka'i lus la rigs lnga dang de'i rigs [244.5] kyi lha thams cad
'god par bshad pa yin pas lha 'gong pa 'phags pa yab sras kyi bshed pa ma yin par smras
pas ni 'grel pa sgron gsal ma mthong bar go bar byas so
252
It seems a bit strange that Mkhas grub mentions the consort here yet does not quote
the almost immediately preceding verse from the root that explicitly mentions the
consort. For the moment we will bracket the significance of this omission, but it is
important to note that understanding how the relationship of the body mandalas of main
deity/practitioner and consort is a complex issue Mkhas grub himself grapples with
elsewhere within his text. See, for example, 239.4-241.5. Fremantle 1971 identifies this
chapter as dealing with the “secret consecration” (guhyåbhißekha) [Chap 8, fn1.]
If indeed the emerging Gelukpa tradition embraced and promoted the
Guhyasamåja system with an emphasis upon its power as a wisdom tantra as a part of a
larger monastic reform initiated by Tsong kha pa, the role and nature of the consort
within this equation would be significant. I am grateful to as suggested by Christian
Luczanits for bringing my attention to this aspect of the tantra in the Tibetan context.
[Private communication, 9/17/2013], One potential source for exploring how the
Guhyasamåja might be understood to fit within a paths and stages model of Buddhist
practice is Ati!a’s Lamp for the Path.
253
See Ocean of Attainment 243.4-244.4.
254 !deyang drangs ma thag pa'i rtsa rgyud kyi lung 'dis; rnam snang skra mtshams nas
spyi bo'i bar dang; 'od dpag med [245.2] lkog ma nas skra mtshams par dang; mi skyod
pa snying kha nas lkog ma'i ma mtha' bar du 'god par bshad ba'i rigs; 'gres254 sbyar na
snying kha nas lte ba'i bar la rin 'byung dang; lte ba'i ma mtha'i msthams nas brla'i rtsa
[245.3] ba'i bar la don grub bkod pas chog par snang la; drangs ma mthag pa'i sgron gsal
gyi tshig zin ltar na lte ba rked pa gsang gnas gsum la khyab par rin 'byung dang ; brla'i
rtsa ba nas rkang sor bar la don grub 'god par snang ste de gnyis [245.4] gang byas kyang
chog par bzhed do
! 81!
It is interesting that the two Buddhas whose positions are at issue are the buddhas
of the jewel and karma families; these two families are considered to be a later addition in
the evolution of mandala from a three to five family scheme typically associated with the
STTS.255 As such, earlier depictions of these two families within both visual and textual
representations were often unstandardized.256 Slippage in techniques of mapping
Ratnasambhava and Amoghasiddhi between different texts that serve as sources for body
mandala may be connected with the evolution from a three to five buddha family schema.
Therefore, divergences in methods of mapping buddhas onto the form of the mandala
found in visual and textual representations may also have occured in technologies of
mapping deities onto the body.257
What is most important for our purposes of understanding the body mandala
debate is attuning ourselves to the ways in which Mkhas grub copes with variant modes
of mapping the body. In our readings, Khenpo Choying Dorje highlighted the
significance of the phrase chog par bzhed do, as meaning something along the lines of
“it’s ok.” This phrase is far less definitive than other terminology Mkhas grub uses to
assert his position within the text, which typically takes the form of what is ‘logical’ or
‘illogical,’ ‘acceptable’ or ‘unacceptable.’ It signals Mkhas grub’s prowess in navigating
delicate issues of textual authority. If he were to exclusively assert his own interpretation
of the root tantra, then he would be contradicting Candrakîrti. Such a move would be
unacceptable, not simply because the latter is an Indian commentator (for Mkhas grub is
known to at times be surprisingly liberal in his critiques of some Indian commentators),
but because he is a hallmark writer of the Årya tradition as perpetuated by the emerging
Gelukpa tradition.258
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
255
On the transition from a three to five-family schema, see Snellgrove 1987 (2002
Reprint), pp.189-213.
256
I am grateful to have had the opportunity to study some examples of such
representations in a Fall 2009 seminar on the STTS with Alexander von Rospatt and Jake
Dalton as well as in a Spring 2011 seminar on Tibetan art with Christian Luczanits at UC
Berkeley.!!!
257
Note that one of these modes of mapping posits the thigh as the lowest point for
arranging the Buddhas; the other, derived from the Pradîpoddyotana, retains the feet as
the lowest point (as found in the root text.) A divergence of this kind raises further
questions, of the kind posed in mapping the proto-body mandala from Dunhuang in
Chapter Two. If the focus of mapping deities onto the body is to cover its full span,
rather than to locate specific vital points, is it a seated or a standing body that is the
prototype? Is it appropriate to locate a Buddha upon the feet? We will return to explore
different modes of mapping deities onto the body in dialogue with some non-Buddhist
examples in the conclusion of the dissertation.
!
258
I am deeply indebted to Khenpo Choying Dorje for bringing my attention to such
subtleties in Mkhas grub’s polemics. He also suggested a possible attempt on Mkhas
grub’s part to relate his own position to Tsong kha pa’s on this issue. More research is
needed on this point. [Personal communication, Spring 2011]
! 82!
IIB. Locating and Correlating the Four Goddesses
The next topic to be addressed from Mkhas grub’s text involves similar issues of how to
locate deities on the body, in this case, the four goddesses Locanå, Måmakî,
Påñ∂aravåsinî and Tårå. We have already noted the complex history of these goddesses
and their identifications with the buddha families, in particular the troubled relationship
of Måmakî and Tårå, in Chapter Two of this dissertation. Therefore, it is necessary that
we remain attuned to changes in the associations of these goddesses within the Indian
context. Such changes may lay the backdrop for Mkhas grub’s efforts to account for the
multiple strategies for identifying the goddesses and their connections to the buddha
families and bodily elements. Above, we referenced the comparison of frameworks
associating buddhas with bodily regions and with the skandhas. Here, in the case of the
goddesses, once again we encounter an initiative to more concretely ‘locate’ deities
within the space of the body.
In this portion of his argument, Mkhas grub is attempting to modify an existing
correlation of these goddesses with the elements to posit and solidify their relationship to
bodily winds. The passages to be discussed appear in the same section of Mkhas’grub’s
texts as those interpreted above. Therefore, both clusters of passages share a common
goal of articulating the mapping of deities vs. merely seed syllables onto the body and
thereby distinguishing body mandala practice from practices like nyåsa. This argument
regarding the goddesses is more complex than that of mapping buddhas onto bodily
regions. The incorporation of the body mandala of the consort, or “mother deity,” here in
relation to that of the male practitioner or “father deity” is one reason for the added
complexity of the argument surrounding the goddesses.259 Therefore, although it precedes
the latter in Mkhas grub’s text, it is presented after it here; this will allow us to build upon
the observations made in interpreting Mkhas grub’s discussion of mapping buddhas onto
bodily regions.
Mkhas grub begins the relevant section by referencing the arrangent of the
goddesses according to the Rnam gzhag rim pa [Samåja-sådhana-vyavasthole (sthåli)].
As mentioned above, this text, attributed to Någabodhi, is an Årya cycle text dealing with
the generation stage practice of the Guhysamåja.260 Mkhas grub asserts:
“The rnam gzhag rim pa intends for one to arrange the goddesses who are the five
mothers [yum] on the bodies of both the father and mother deity. It’s unreasonable [mi
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
259 !The use of the terms “father deity” and “mother deity” here reinforce the ritual notion
of the practitioners acting as the main deities of the Akßobhyavajra mandala of the Årya
Guhyasamåja tradition.!
260 !Toh 1809 Sde sde bstan ‘gyur Vol. 35 ff. 121a-131a by Någabodhi. For partial
editions, see Tanaka 2001-2, 2004 & 2009. For Tsong kha pa’s position on the
attribution of texts to Någabodhi, see Kilty 2013, p.65-66. See also Tsong kha pa’s
commentary on this text: Rnam gzhag rim pa’i rnam bshad dpal gsang ba ‘dus pa’i gnad
kyi don gsal ba. In gsung ‘bum/ Tsong kha pa blo bzang grag pa (Reproduced from
Lhasa zhol par khang blocks). New Delhi: Lama Guru Deva, 1978-9. Vol.6. pps. 5-166.
!
! 83!
rigs] to arrange the four, Locanå and so forth on the navel, heart center, throat, and
crown.”261
We have come to know the four goddesses Locanå, Måmakî, Påñ∂aravåsinî and
Tårå to be associated with the buddha families and positioned on the elements. However,
texts (such as the Piñ∂ik®ta discussed above) that map the goddesses onto elements
without specifying precise bodily regions or sites exhibit a somewhat imprecise variety of
mapping not unlike mapping buddhas onto skandhas rather than regions or points.
The reference to the five goddesses as the five mothers [yum] is also somewhat
perplexing. The set of five goddesses we are familiar with is the five rdo rje ma:
Rüpavajrå, ¸abdavajrå, Gandhavajrå, Rasavajrå and Spar!avajrå. The first four of these
are, in some cases, considered to be in union with four of the bodhisattvas in the body
mandala. The fifth goddess, Spar!avajra, is often united with Akßobhya, the main deity
of the mandala.262
The Rnam gzhag itself addresses the four goddesses as follows:
“The arrangement of the goddesses is taught.

Moharatî Locanå is the earth element.
Dveßaratî Måmakî is water
Rågaratî Påñ∂aravåsinî is fire
Vajraratî Tårå is rlung.
We look to the root tantra to clarify the meaning:
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
261
rnam gzhag rim par yum lnga'i lha mo rnams bkod pa ni yab yum gnyis ka'i lus kyi
[240.2] dbang du byas pa'i lha rnams la dgongs pa yin no
spyan ma la sogs pa bzhi lte ba snying kha mgrin pa spyi bo rnams su 'god pa yang mi
rigs te
262 !Furtherstudy is required to understand when and how Spar!avajra comes to be one of
the mothers of the five families. In one sense, she might be imagined at the heart in union
with Akßobhya, while in another, she may be the consort herself in union with the
practitioner as father deity. The question of how the consort was incorporated into body
mandala practice may be relevant to grappling with this ambiguity.
Von Rospatt 2010 grapples with a potentially related issue involving the
deployment of the goddesses in his study of a consecration ceremony based in
Kulada††a’s Kriyåsamgrahapa•jikå. He is dealing with a set of four Vajrî derived from
the Vajradhåtu mandala: Sattvavajrî, Ratnavajrî, Dharmavajrî and Karmavajrî. In the
adhivåsana mandala component of the consecration, these Vajrî exchange places with
Locanå, Måmakî, Pañ∂ara and Tårå. This exchanges defies the standard of the
Vajradhåtu mandala explicated in Abhayåkaragupta’s Nißpaññayogåvalî. Von Rospatt
observes: “This configuration is not a Newar innovation, but accords with the general
tendency in Indian Buddhism to substitute the Vajrî goddesses of the Vajradhåtu-
mandala...with Locanå and so on. The four goddesses of both sets are related each to one
of the five Buddhas in a largely but not completely congruent manner.” See von Rospatt
2010, especially p.221-2. In fn 45, he provides a reference to the Ma•juvajra mandala
(associated with the J•ånapåda lineage of the Guhyasamåja system) for comparison.
! 84!
‘As for the element of earth, it is explained as Locanå
As for the element of water, it is explained as Måmakî
[330] As for the element of fire, it is explained as Påñ∂aravåsinî
As for the element of air, it is known as Tårå’
So it is said.”263
The Rnam gzhag itself therefore seems to refer to this set of four as “goddesses” [lha
mo], not “mothers” [yum], and makes no mention of a fifth goddess or to the body of a
consort. It also provides alternate names for them. Furthermore, the Rnam gzhag
solidifies the relationship of these four goddesses to the elements through citation of
Chapter Seventeen verse 51 of the root tantra.264
When Mkhas grub critiques the practice of mapping these goddesses onto the
navel, heart, throat, and crown as unreasonable [mi rigs], he is taking a definitive stance
against this technique although he does not cite its source. Mkhas grub continues by
citing the Piñ∂îk®ta [Mdor byas], a text attributed to Någårjuna that also deals with the
generation stage practice of the Guhyasamåja. This text is believed to be a counterpart to
the more well-known Pa•cakrama by the same author which focuses upon the completion
stage practice.265 Roger Wright has dated the Piñ∂îk®ta to between 800 and 950 CE.266
Wright attests to the enduring significance of this text within the Geluk tradition in
revealing that it is the basis for Tsong kha pa’s Guhyasamåja sådhana, a practice widely
used today.267 It is perhaps of note that the alternate names for the goddesses provided in
the rnam gzhag citation above are found in the within the body mandala of the father
deity; in that text, the names Locanå... refer instead to the goddesses in the body mandala
of the consort. Mkhas grub cites the Piñ∂îk®ta as follows :
“In the Piñ∂îk®ta (it says):

‘As for Locanå and Måmakî, likewise Påñ∂aravåsinî and Tårå,
they are arranged by the mantrika on the earth (element) and so forth.’
And in terms of the explanation of arranging the four goddesses in the sites of the four
elements:
‘It is proper to arrange Locanå in (the area of) the genitalia, the abode of earth rlung, Tårå
at the navel, the abode of rlung rlung, Måmakî at the heart center, in the abode of water
rlung, and Påñ∂aravåsinî at the throat, in the abode of fire rlung.’ So it is said.”268
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
263
I have located and translated this passage using Kimiaki Tanaka’s partial critical
edition, which has been emerging over the course of a series of articles. See Tanaka
2001-2, 2004 & 2009.
264
This citation from the root text was identified by Kimiaki Tanaka.!!!
265
Wright 2010, p.8.
266
Wright 2010, p.16.
267
Wright notes that Tsong kha pa’s text elaborates upon the basis of the Piñ∂îk®ta but
“maintains the same sequence of visualizations and mantras.” Wright 2010, p.54. See
Tsong kha pa T5303: Dpal gsang ba ‘dus pa’i sgrub thabs rnal ‘byor dag pa’i rim pa. See
also Wright 2010, Appendix A in which he compares the structure of the two texts.
268
mdor byas su; spyan dang ma' ma ka'i dang ni ; de bzhin du ni gos dkar mo
! 85!
The initial quote from the Piñ∂îk®ta is derived from the mahåsadhana section of
that text, explaining the arrangement of the body mandala of the consort.269 [See
Fig.17] In the context of laying out the body mandala of the father deity, in the atiyoga
section of the text270, the Piñ∂îk®ta instructs [See Fig.16]: “With Moharatî, the mantrin
should place them on the earth (element), and so forth: that with solidity, that with
fluidity, that with warmth and that with airiness respectively.”271
Therefore, the Piñ∂îk®ta clearly connects these four goddesses with the elements
and their defining characteristics. The goddesses appear in the body mandalas of both
father deity and consort, albeit with different names [Locanå...vs. Moharatî...] However,
they are not located on sites in the body in any specific sense. In my own diagram of the
body mandala of the father deity based on the Piñ∂îk®ta and Wright 2010, I struggled to
locate these goddesses in their elemental abodes. [See Fig.22] Like mapping the
Buddhas onto skandhas vs. specific points or regions of the body, mapping goddesses
onto elements appears to be a more abstract correlation than is typically found in nyåsa or
body mandala practices. According to Wright’s study of the Piñ∂îk®ta-sådhana, the
buddhas of the five families are positioned upon the bodies of both the father deity and
the consort at the crown, throat, heart, navel and feet. [See Fig. 16 & 17] So if one
were to map the consorts onto those points by their associations with the buddhas of the
respective families we would have a similar layout to that critiqued by Mkhas grub, with
the addition of the feet.
As for Mkhas grub’s next citation, (“it is proper to arrange Locanå in the (area of
the) genitalia...”) it’s derivation is unclear, as it does not appear in Piñ∂ik®ta.272 The
closing phrase zhes zer ro suggests that it is a quotation, although it is possible that
Mkhas grub is simply paraphrasing, perhaps even from a Tibetan source.273 The
statement locates the goddesses [Locanå, Tårå, Måmakî, and Påñ∂aravåsinî] on four
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
sgrol ma yang ni sngags pa yis; sa [240.3] la sogs bar rnam par dgod; ces 'byung ba bzhi'i
gnas su lha mo bzhi dgod bar bshad pas; sa rlung gi gnas 'doms su spyan ma dang; rlung
gi rlung gi gnas lte bar sgrol ma dang; chu rlung gi gnas snying khar ma' ma ki dang; me
rlung gi gnas mgrin par gos [240.4] dkar mo dgod par rigs so zhes zer ro
!
269
See Wright’s 2010 translation v.96-97.
270 !See Wright’s 2010 v.52-69.!
271 !gti mug dga' sogs sngags kyis ni ; sa la sogs la rab tu bzhug; sra dang gsher dang dro
ba dang; rlung sogs der ni rim pa bzhin [Skt. moharatyådikair mantrî p®thivyådin
prave!ayet; kharatvaµ; dravatå außñyam îrñatvam ca te kramåt] See Wright 2010
translation and edition v61.
272
I was also unable to locate it in Tanaka’s partial editions of the Rnam gzhag.!
273
For example, Bentor 2006 has demonstrated Bu ston to be the target of some of
Mkhas grub’s critiques within the Ocean of Attainment. Among the texts she cites is
Bu ston’s commentary on the Piñ∂i-k®ta. See Bu ston Rin chen grub (1290-1364). Dpal
gsang ba ‘dus pa’i sgrub thabs mdor byas kyi rgya cher bshad pa bskyed rim gsal byed. In
The Collected Works of Bu-ston. New Delhi: International Academy of Indian Culture,
1967, Vol. 9, 683-877.
! 86!
specific sites on the body [genitalia, navel, heart and throat] which are, in turn, defined as
the abodes of elemental rlung [earth, rlung, water, and fire rlung]. These sites differ from
the set critiqued by Mkhas grub in that the genitalia [‘doms] replace the crown. Neither
the crown nor the feet, for that matter, are included here. Understanding how the sites on
the body are associated with different elemental rlung as well as with the elements
themselves is one point for further exploration.
Mkhas grub continues his critique a few pages later:
“There is the claim that it’s necessary to arrange (deities) on sites such as the secret place
based upon the explanation for arranging deities such as Locanå on (elements such as)
earth. For you who lack discerning minds when it comes to the meaning of the tantra, of
course you have doubts.”274
The polemical flavor of Mkhas grub’s writing is impossible to ignore here. He

then proceeds to set forth his own position:
Nevertheless, as far as I’m concerned, this is how it is:

Within the classification of five root rlung, the abode of the earth rlung, downward-
clearing rlung, is the secret place.
The abode of balancing, the rlung rlung, is the navel.
The abode of life-sustaining, water rlung, is the heart center.
The abode of the upward moving, fire rlung, is the throat.
The abode of the all-pervading, space rlung, is the whole body. 275
The list of five primary or root winds matches a common set [thur sel, mnyam
gnas, srog 'dzin, gyen rgyu and khyab byed].276 Here Mkhas grub is identifying each of
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
274 !spyan ma sogs bzhi sa la sogs pa la dgod par bshad pas gsang gnas sogs su dgod dgos
par 'dod pa de ni khyed cag rgyud don la zhib par 'byed pa'i blo dang mi ldan pa dag la de
lta bu'i dogs ba 'byung ba [245.5] bden no
Khenpo Choying Dorje and Khenpo Yeshe suggested that the use of dod pa may
contribute the derogatory tone of the passage (whereas dgong pa would have ben the
more neutral choice). [Personal communication, Spring 2011]
!
275 !'on kyang nga ni 'di yin te; rlung la rtsa ba'i rlung lngar phye ba'i sa rlung thur sel gyi
gnas gsang gnas dang; mnyam gnas rlung gi rlung gi gnas lte ba dang; srog 'dzin chu
rlung gi gnas snying kha dang; gyen rgyu me rlung gi gnas mgrin pa dang; khyab byed
[245.6] nam mkha'i rlung gi gnas lus thams cad la
276 !See
entries by OT, IW, & RY on thlib.org [Accessed 7/29/201] 5 [srog 'dzin, gyen
rgyu, khyab byed, me mnyam, thur sel] Garrett 2008 pp.65-66 describes how these five
root winds (together with five subsidiary winds) are common in tantric physiological
accounts, citing the twelfth-century Sakyapa patriarch Grags pa rgyal mtshan as one
example. She locates the winds at areas of the body: thur sel in the anus, mnyam gnas
(or me mnyam) in the navel, srog 'dzin in the heart, gyen rgyu in the throat, and khyab
byed throughout the body. They bear associations with the elements and with colors as
! 87!
the five root winds as an elemental wind and locating it within the body [at the secret
place, navel, heart, throat & bodily totality]. There is, however, no mention of the
goddesses, and we are working with a list of five rather than four. Mkhas grub continues:
“In light of this explanation, as for the one who explains the four (goddesses), Locanå
and so forth, as the four elements here (this may be said):
Generally speaking, there are many contexts for applying the four elemental
winds to the four such as Locanå. However, having construed the four goddesses as the
elements such as earth, here one generates the four goddesses as the aspects of bodily
solidity, moisture, heat, and motility. If one arranges them like that, having condensed all
five root rlung into just the element of rlung, it is necessary to make all of those as the
basis of accomplishing Tårå. So then it would not be fitting to apply the generation of
Locanå from the earth rlung and so on, on account of the absence of the characteristics of
solidity in the downward-clearing rlung.”277
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
well as bodily functions. On the medical conception of rlung, see pp62-63. The three
humors are rlung, bile and phlegm; each construed in terms of five types. Further
research into both tantric and medical systems to determine subtleties in their
understanding of rlung. Garett 2008 makes some important inroads in chapters four and
six. Khenpo Choying Dorje pointed out theat the medical definition of rlung is just one
aspect of what Vajrayåna describes as rlung. Translating rlung provisionally as “energy,”
he explained the Vajrayåna view of rlung as the horse and the mind as the rider.
Everything that is moving, everything in the physical world has rlung. Even the
bodhisattva still has rlung when reborn with karma. Within the thirteen-stage model of
the path, rlung is at play until you reach stage of Vajradhara. [Personal communication,
Spring 2011]
See also Kontrul 2005, Systems of Buddhist Tantra, Book Six Part Four, pp.176-
180, especially fn 47. The comparison of the relationship between the role of rlung in
cosmic creation and destruction (derived from the Abhidharma tradition) and its role in
tantric conceptions of bodily creation and dissolution is compelling. Kittay 2011 (p.133)
observes that the five winds presented in the Vajramålå accord with those found in the
Visuddhimagga 11:37.
!
277
...bshad pa yin la; 'dir spyan sogs bzhi sa la sogs pa'i khams bzhir bshad pa ni
spyir 'byung ba bzhi'i rlung dang spyan sogs bzhi sbyor ba'i skabs mang du yod kyang;
'dir ni lus kyi sra ba'i cha dang; [246.1] gsher ba'i cha dang; dro ba'i cha dang; g.yo ba'i
cha rnams la sa'i khams la sogs pa bzhir byas nas; de dag lha mo bzhir bskyed ba yin
zhing; de ltar bzhag pa na ni; rtsa ba'i rlung lnga ka yang rlung gi khams gcig bur byas
nas de thams cad sgrol ma'i bsgrub gzhir [246.2] byed dgos kyi sa rlung las spyan ma
bskyed pa sogs byar mi rung ste; thur sel gyi rlung la sra ba'i mtshan nyid ma tshang ba'i
phyir ro
! 88!
This is a difficult point to understand. We should begin by clarifying that the context to
which Mkhas grub refers with ‘here’ is indeed the Piñ∂ik®ta excerpt cited earlier by
Mkhas grub [240.2-.3]: “As for Locanå and Måmakî, likewise Påñ∂aravåsinî and Tårå,
they are arranged by the mantrika on the earth (and so forth).”
As observed above, the identification of the defining qualities of the elements to
which to which he refers is actually derived from the earlier atiyoga section of the Piñ∂i-
k®ta.278 This citation, not provided by Mkhas grub himself, has been discussed above:
“With Moharatî, the mantrin should place on them the earth (element), and so forth: that
with solidity, that with fluidity, that with warmth and that with airiness respectively.”279
To review, this statement is made in the context of the body mandala of the father deity
[i.e.the male practitioner], whereas that cited by Mkhas grub describes the body mandala
of the consort. The correlation of these goddesses with the elements and their defining
characteristics is the dominant mode of correlation for the Piñ∂ik®ta. There is no
reference to the locations of the elements or to elemental rlung. Likewise, the root tantra
[XVII.51]280 itself clearly correlates these goddesses with the elements, though there is no
mention there of their locations, elemental qualities (ex.solidity), or elemental rlung.
Mkhas grub is grappling with two alternative systems of correlation for the goddesses.
Unfortunately, mapping the goddesses onto the body through association with the
elemental rlung contradicts the already existing system of correlating them with the (un-
located) elements in both the root and the Piñ∂i-k®ta. In attempting to make sense of the
relationship between two networks of correlation, one based in the elements and the other
in winds, Mkhas grub copes with a clash:
“For that reason, conflating earth and earth rlung is mistaken.

If it is otherwise, do you claim the pervasive rlung is that of space?
If the downward-clearing wind has the characteristic of solidity, then you have to say that
the four, earth, water, fire, and rlung are non-contradictory.
[You are] like a child who’s never mastered the signs of related and contradictory
(phenomena). Go ahead and exert your minimal effort commenting on the meaning of
the sütras and tantras!”281
Here Mkhas grub invokes the terms of Buddhist logic and epistemology to deliver his
critique. In the next chapter of this dissertation, we will investigate various examples of
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
278
See Wright 2010, verses 52-69.!
279 ![gti mug dga' sogs sngags kyis ni ; sa la sogs la rab tu bzhug; sra dang gsher dang dro
ba dang; rlung sogs der ni rim pa bzhin] (Skt. moharatyådikair mantrî p®thivyådin
prave!ayet; kharatvaµ; dravatå außñyam îrñatvam ca te kramåt) Wright 2010, v61.
280
As identified by Tanaka in the context of its citation within the Rnam gzhag.
281
des na 'di 'dra ba ni sa dang sa rlung gi shan ma phyed pa ste; gzhan du na khyab byed
kyi rlung nam (m)kha'i yin par khas len nam; thur sel [246.3] gyi rlung la'ang sra ba'i
mtshan nyid tshang na sa chu me rlung bzhi mi 'gal bar khas len dgos mod kyi 'gal 'brel
gyi brda la gtan ma byang pa'i byis pa nyid nas mdo rgyud kyi don 'grel ba281 la snying
las chung ngur gyis shig
! 89!
Mkhas grub’s use of these discourses as a strategy for legitimizing his tantric
hermeneutics. In referring to “the signs of related and contradictory (phenomena)” ['gal
'brel gyi brda], Mkhas grub challenges the relationship of the elements to the elemental
rlung and the five root rlung through the authority of firmly established categories of
Budddhist logic and debate.282 He implies that the essential properties of the elements
are mutually exclusive; therefore, fusing them onto the elemental rlung conflates
contradictory phenomena. He attempts to show how earth rlung is not simply composed
of both rlung and earth; rather, it resembles earth in certain aspects. So, for example,
solidity is a property that can’t exist together with non-solidity; it is an exclusive property
essential to the definition of earth.283 Therefore, the simple conflation of correlation of
the goddesses to varieties of rlung and correlation of elements presents a potential clash
of systems of meaning.
As observed in previous chapter, system of correlation for mandala were often
modified and structures adapted to effectively incorporate or even conflate multiple
systems of “correlative correspondences.”284 The most familiar instances occurred in the
transition from a three to five buddha family system. For example, two poisons were
added to the standard set of three (desire, aversion and ignorance) in the transition to the
five family system. Other systems, like the skandhas, lent themselves to adoption
without modification. In that light, Mkhas grub’s polemics on correlating the goddesses
might be read as a form of iconography in the making in which elements, elemental
winds and root winds must all be accommodated in attempting to locate them more
concretely upon the human body. The polemical aspect of his approach is accentuated in
the final line of the passage. It bears the colloquial flavor of many of Mkhas grub’s
heated critiques, conveying something like “you’re not ready to write about those texts’’
or “go ahead and write your elementary commentaries!” In other words, “Go ahead and
embarrass yourself! Deplete your virtue in doing a bad job writing about sütra and
tantra!”285
Mkhas grub then tests out his theory, turning for support to yet another text of the
Årya Guhyasamåja tradition:
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
282
For more on the principal of contradiction in Buddhist philosophical debate, see
Dreyfus 2002, p.212-13.
283 !I am grateful to Khenpo Yeshe for clarifying the logic of this statement. [Personal
communication, Spring 2011]

284
This term was coined by Charlotte Furth in her 1999 study of womens’s medicine in
Imperial China to describe the multiple relationships of yin and yang as a correlative set
to one another as well as to other like sets such as male and female. I explored and
modified “correlative correspondence” as an interpretive tool for understanding the
modes of representations of the body in a set of seventeenth-century Tibetan medical
paintings in my unpublished 2007 Masters thesis.!
285
I am grateful to Khenpo Choying Dorje and Khenpo Yeshe for their suggestions in
better understanding the meaning and tone of this remark. [Personal communication,
Spring 2011]
! 90!
“If so, in this case, if you ask how do I know it’s related to setting forth the component of
solidity in earth, you can look at the Vajrasattva-sådhana in which Candrakîrti wrote286:
‘When arranging the four, Locanå and so forth, on the body, the so-called Moharatî is
Locanå and has the nature of the earth element.”
Up until: “The thus-called Vajraratî is Tårå and has the nature of the rlung element.
Having arranged (them thus), thoroughly envision solidity and moisture and heat and
motility.’ Thus, it is clearly taught.”287
Here the goddesses are located upon the body through reference to the elements
and their defining qualities; however, no specific bodily sites are named. In other words,
Mkhas grub seems to be asking how earth is related to solidity and not to the elemental
wind of earth (or to the downward-clearing root rlung). Luo Hong & Toru Tomabechi
have published a critical edition of both the Sanskrit and the Tibetan texts of the
Vajrasattva-sådhana.288 The editors suggest that the text is authored by the same
Candrakîrti who wrote the Pradîpoddyotana (although not the Madhyamika
Candrakîrti).289 Like the Mdor byas and the Rnam gzhag, it focuses upon generation vs.
completion stage practice. The editors present a useful overview of the sådhana in their
introduction that helps us to contextualize this quotation within the overall structure of
the text.290 It occurs within the atiyoga portion of the text in which the practitioner
manifests as the nirmåñakåya buddha body; it appears alongside the placement of
Buddhas on the skandhas, bodhisattvas on the sense faculties, and krodhas on the limbs.
The sådhana as a whole seems to espouse a three buddha body system. According to this
system, in producing the dharmakåya, the practitioner generates deities, absorbs them,
places them on the body, and dissolves them into emptiness. In producing the
sambhoghakåya, the practitioner manifests as white Vajradhara. Finally, in producing the
nirmåñakåya, the practitioner locates the deities on the body.291 The mapping of these
three bodies onto the sådhana structure becomes significant in the Tibetan interpretation
of the Guhyasamåja as a technology for manipulating the process of death and rebirth, a
theme to which we shall return below.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
286
See Vajrasattva-sådhana. Gser bris bstan ‘gyur Vol.35 436-61.
287
'o na skabs 'dir sra ba'i cha la sar [246.4] bzhag pa sogs kyi dbang du byas par gang las
shes zhe na; dpal ldan zla ba grags pas rdo rje sems dpa'i sgrub thabs las; lus la sbyan ma
sogs bzhi 'god pa'i skabs su ; mo ha ra ti zhas bya ba spyan ma sa'i khams kyi rang bzhin
zhes bya'o; zhes ba nas; [246.5] ba dzra ra ti zhes bya ba sgrol ma rlung gi khams kyi
rang bzhin gyis so; bkod nas sra ba nyid dang ; gsher ba nyid dang; dro ba nyid dang;
g.yo ba nyid du rab tu bsam par bya'o; zhes gsal bar gsungs la
288
For this quote, see 49.15-19 [III.3.2] in their edition. Luo Hong and Toru Tomabechi
(eds.), China Tibetology Publishing House & Austrian Academy of Sciences Press,
Beijing-Vienna, 2009.
289
Hong and Tomabechi 2009, Intro p.x!
290 !Hong and Tomabechi 2011, Intro p.x. See also pp.xii-xv!
291 !Later on in the sådhana, there is also a section devoted to consort yoga; this text
appears to identify Spar!avajra as Akßobhya’s consort.!
! 91!
Mkhas grub then proceeds to quote another text, one that extends the correlation
of goddesses and elements to include not only elemental qualities but also bodily
substances:
“In the Vajramålå Explanatory Tantra (it says):

‘Moreover, the Bhagavatî Locanå abides in the earth element, in the fat and so forth of
this one. The Bhagavatî Måmakî abides in the water element, the blood and so forth. The
Bhagavatî Påñ∂aravåsinî abides in the fire element, heat and so forth. The Bhagavatî Tårå
abides in the rlung element, trembling and so forth.’292
Thus it is clearly explained.”293
Based upon these correlations, Mkhas grub asks:
How could anyone in their right mind claim that the flesh of the body (is endowed
with/made up of) the earth rlung and the blood (with/of) fire rlung?294
The Vajramålå is an explanatory tantra [Tib.bshad rgyud/ Skt. vyåkhyåtantra] accepted as

buddha-vacana within the Årya Guhyasamåja tradition. It is, therefore, in a different
class than the commentaries attributed to Någårjuna and his disciples that have formed
the core of the body of citations discussed thusfar. This text is cited extensively by Mkhas
grub as well as in the writings of his teacher Tsong kha pa and in the body mandala text
of Ngor chen as well. It is perhaps most well-known for the forty verses which expound
upon the first-forty syllables of the Guhyasamåja Tantra. These verses are cited in both
Candrakîrti’s Pradîpoddyotana and in part within Åryadeva’s Caryåmelåpakapradîpa.295
No original Sanskrit text of theVajramålå survives (except portions cited in these texts);
however, Tibetan translations began to emerge in the eleventh century.296 Unlike the
other texts cited by Mkhas grub thusfar in the sections discussed, the Vajramåla focuses
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
292 !With regard to the translation of this passage, once again we are faced with the issue
of how best to translate the ladon particle. Kittay 2011 has opted for “in” (743) and I
have followed suit.
!
293 !bshad rgyud rdo rje 'phreng ba las kyang; yang 'di'i sha sogs [246.6] sa khams la ;
bcom ldan 'das yum spyan bzhugs so ; khrag sogs chu'i khams la ni ; bcom ldan 'das yum
ma ma bzhugs ; dro ba la sogs me khams la ; bcom ldan 'das yum gos dkar bzhugs;
bskyod pa la sogs rlung khams la ; bcom [247.1] ldan 'das yum grol ma bzhugs ; zhes
ches gsal bar gsungs te
This quotation can be found in the sde dge edition of theVajramålå 270a.3-.4 [539.3- .4]
where the only real difference in mi bskyod vs. bskyod pa. For Kittay’s translation se
Kittay 2011, p.743
294
lus kyi sha sa rlung dang; khrag me rlung du shes rig dang ldan pa su zhig khas len par
byed
I have emended ma to me and shas to shes.
295
Kittay 2011, 5.
296
Kittay 2011, 6.
! 92!
upon completion stage practices of the Guhyasamåja, although some generation stage
practices are included.
We are fortunate to be aided in our understanding of the Vajramålå by Kittay’s
2011 study and translation. Kittay regards the text as a compendium of different
practices inclusive of both Mahåyoga and yoginî-tantra based interpretations of the
Guhysamåja Tantra.297 This quote is extracted from Chapter Sixty-Four, entitled “The
Explanation of Mandala of Body, Speech and Mind.” Kittay summarizes this chapter as a
detailed description of the arraying the body mandala on the body of the guru298
according to the Årya tradition as follows:
“The beginning of the chapter explains that the enlightenment spirit in the thirty-two
channels in the crown cakra constitutes the body mandala, the division of the parts of the
letters forms the speech mandala, with twenty-seven members, and the distinguishing of
‘the instincts of desire and so forth’ is the mind mandala, with twenty-eight parts.” 299
The section cited here, the body mandala, (see v.4-17) describes the correlation of
the jinas (albeit in a more abstract sense of body, space, speech, action and mind) with
the skandhas, the four goddesses, the eight bodhisattvas on the sense faculties (plus joints
and sinews), and the ten krodhas on the limbs (and mouth). Mkhas grub explains further:
“In that case, there is the explanation for dissolving earth, water, fire and rlung and so
forth. At the time of the dissolution the twenty-five coarse (constituents),; the potential
for producing the consciousnesses [rnam shes kyi rten phyed pa'i nus pa] dissolves. The
bodily deities, the (set of) four, Locanå and so forth, are taught to be dissolved in accord
with the dissolution (of) those.300
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
297 !Kittay 2011, p. 188. As Kittay notes, some chapters of the text explicitly clarify the
Mahåyoga or Yoginî-tantric basis for interpretation.!

298
The issue of whether the mandala is arrayed on the body of the guru or upon the
practitioner’s own body is an important issue touched upon in Wayman’s translations of
Mkhas grub’s Fundamentals of Buddhist Tantra. This is a point for further research and
calls into question the relationship of the body mandala practice to guru yoga,
consecration rites, and deity yoga. Note that this description of arranging the body
mandala on the body of the vajråcårya from the Vajramålå does not include the five rdo
rje ma. The number of deities thus equals twenty-seven, as is the case for the body
mandala of the main deity in the mdor byas.
!
299 !Kittay 2011, p283!
300
des na rags pa nyi shu rtsa lnga thim pa'i skabs su; sa chu me rlung sogs thim [247.2]
pa bshad pa ni;lus kyi sra ba'i cha la sogs pa la sa la sogs par byas nas; de dag gis rnam
shes kyi rten phyed pa'i nus pa thim pa'i dbang du byas te; de dag thim pa dang mthun par
lus kyi lha spyan ma la sogs pa bzhi yang thim par gsungs pa yin [247.3] la; khams phra
pa thim pa'i skabs su; sa chu la thim pas smig rgyu lta bu'i nyams 'char ba sogs ni 'byung
ba bzhi'i rlung la phra rags du ma yod pa'i nang nas rags pa thim pa'i dbang du byas nas
gsungs pa yin te; 'di lta bu'i rnam dbyed dag ma shes na rags [247.4] pa thim pa dang
! 93!
That being the case, at the time of the dissolution of the subtle elements [khams
phra pa], a vision akin to a mirage manifests on account of the dissolution of earth into
water and so forth. So here is the teaching concerning the dissolution of the coarse among
the many subtle and coarse (components) possessed by the four elemental rlung. If you
don’t know how to make distinctions like this, not knowing how to distinguish the
dissolution of the coarse and the dissolution of the subtle, how would it be possible to
realize the essential point of the completion stage?”301
Mkhas grub is interpreting the correlation of goddesses with the elements, qualities, and
substances cited from the Vajramålå in terms of the order of dissolution of the body at
death enacted in completion stage practices.302 The dissolution of the bodily elements in
sådhana practice serves as preparation for the moment of death. Earth rlung is the most
coarse of the elemental rlung, while rlung rlung is the most subtle. The signs of death
can also be distinguished in terms of coarse and subtle. For example, when earth
dissolves into water, the appearance of mucus is a coarse sign that can be seen by anyone,
while the vision of the mirage is a subtle sign that only the dying person themself can see.
The sense consciousnesses reliant upon the presence of these elements in the body,
likewise, dissipate. When Mkhas grub refers to the dissolution of “the potential for
producing the consciousnesses” [rnam shes kyi rten phyed pa'i nus pa], he is referring to
the vital connections between the elements and the varieties of sensory consciouness; he
is solidifying the correlation of the goddesses with particular phases of this process.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
phra ba thim pa tsam gyi khyad par yang mi shes na rdzogs rim gyi gnad zab mo dag
rtogs par lta ga la 'gyur
301 !desna rags pa nyi shu rtsa lnga thim pa'i skabs su; sa chu me rlung sogs thim [247.2]
pa bshad pa ni; lus kyi sra ba'i cha la sogs pa la sa la sogs par byas nas; de dag gis rnam
shes kyi rten phyed pa'i nus pa thim pa'i dbang du byas te; de dag thim pa dang mthun par
lus kyi lha spyan ma la sogs pa bzhi yang thim par gsungs pa yin [247.3] la; khams phra
pa thim pa'i skabs su, sa chu la thim pas smig rgyu lta bu'i nyams 'char ba sogs ni 'byung
ba bzhi'i rlung la phra rags du ma yod pa'i nang nas rags pa thim pa'i dbang du byas nas
gsungs pa yin te; 'di lta bu'i rnam dbyed dag ma shes na rags [247.4] pa thim pa dang
phra ba thim pa tsam gyi khyad par yang mi shes na rdzogs rim gyi gnad zab mo dag
rtogs par lta ga la 'gyur
!
302 !Khenpo Choying Dorje clarifies that Mkhas grub is addressing two topics: the deities
of the body and the winds of which there are five different varieties. In addressing the
meaning of dbang du byas nas, he observes how while in some cases it indicates a clear
causal relationship, here it denotes a more general connection and can thus be translated
as ‘concerning’ or the like. When Mkhas grub says nus pa thim pa'i dbang du byas,
therefore, the meaning is concerning dissolving the potential of producing consciousness,
although this is not explicitly stated. Likewise, when the elements are dissolved, the four
dåkinîs are dissolved. [Personal communication, Spring 2011]
!
! 94!
Within the Guhyasamåja body mandala practice transmitted by Geshe Lobsang
Tsephel, this phase of dissolution occurs in generation stage practice after one has
generated oneself as the main deity, Akßobhyavajra, and arrayed all of the mandala
deities upon one’s body (five tathågatas, four consorts, eight bodhisattvas, five rdo rje ma
& and ten krodhas). The goal of such practice is to enable the practitioner to control their
progress through the death process by attaining the three Buddha bodies.303 Such
attainment requires a mastery of techniques designed to draw the bodily winds as
purveyors of consciousness into the central channel to enter the clear light. Many tantric
texts use such descriptions of the dissolution of the body characterized by both a
microcosmic-macrocosmic interplay as well as by tensions between emanation and
absorption and gross and subtle phenomena. The ritual drama of dissolving the body
reverses the order of composition involving the natural elements and particular bodily
winds. Tibetan interpreters of the Årya Guhyasamåja tradition such as Tsong kha pa and
Mkhas grub emphasized this aspect of the practice. When Mkhas grub specifically
identifies this practice of dissolution of course and subtle components as fundamental to
the “essential point of the completions stage,” he reveals the importance of ritual
technologies of dissolution to a two-stage sådhana structure for his interpretation of the
Guhyasamåja. While his own text is focused upon the generation stage, his consistent
references to theVajramåla and the logic of dissolution in mapping the goddesses onto
bodily constituents, elements, and winds emphasize this dimension of the practice.
It seems odd that Mkhas grub chose to cite Chapter Sixty-four of the Vajramålå
rather than Chapter Sixty-eight, “The Epitome of All Attainments” in which a more
detailed account of the body mandala appears. In the latter chapter, we find a complex
grouping together of buddhas, skandhas, elements, rdo rje ma as sense objects,
bodhisattvas as sense faculties, and krodhas (See v.27-37) in the context of “dissolving
and enjoying.” These correlations are followed by the grouping of buddhas with
elements and bodily constituents (v.37-41), a list of internal parts of the body to be
“known in succession” (v42-44) and the location of twenty-four external sites (familiar
from yoginî tantra) onto sites of the body (v. 45-50). Note, however, that the account
from Chapter Sixty-eight, does not include the four Goddesses (Locanå, Måmakî,
Påñ∂aravåsinî and Tårå) included in Chapter Sixty-four. Chapter Sixty-four, on the other
hand, does include the rdo rje ma. However, while Chapter Sixty-eight may not include
the goddesses at issue, it is there that the five root winds are itemized, located within the
body, and associated with the five Buddhas:
Now, further, I will explain

The supreme characteristics
Of the energy-winds.
The life-energy energy-wind
abides in the heart,
born from the particular clan
of Akßobhya. V52
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
303
The mapping of the bodies onto the Guhyasamåja sådhana may have begun with
works such as Candrakîrti’s Vajrasattva-Sådhana.
!
! 95!
The Evacuative energy-wind
That abides in the crotch
Is born from the aspect
Of Ratnasambhava.
The Ascending energy-wind
Abides at the end
Of the throat,
The nature of Amitåbha. V53
The Fire and Equalizing

Abide in the navel
The nature
Of Amoghasiddhi.
The pervading
Abides in all the limbs
Blessed by Vairocana. V54
The continuity
of life energy and effort
From the continuity
of the sense doors
moves at all times
explained as ‘life-energy.’v55304
The text then describes the bodily functions performed by each of these winds,
followed by their destruction; this description resonates with Abhidharmic accounts of
cosmic destruction like the excerpt from Chapter Eleven of the Visuddhimagga described
in Chapter Two of this dissertation. 305
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
304
Translation by Kittay 2011, p.773!
305
Then, the Vajramålå relays how the winds and consciousness arise again, together
with karma and kle!a, to reenact the birth process. This is a remarkable dramatic
enactment of the destruction of ignorance that is the goal of cultivating an understanding
of the processes of birth and death through Buddhist tantric sådhana practice:
And from that,

Again death and birth
Thus like a cakra turning
As before, vajra recitation
And so forth
Gradually divides that. V63
Then all the great beings,

With a spoonful of eye ointment,
Dispel the blindness of the mistakes of all.
! 96!
Mkhas grub continues his discussion of the dissolution of the body by parsing the
relationship of subtle and gross components:
“(With regard to) that which is referred to as the “dissolution of the subtle,”
the three, earth, water, and fire are coarse. Compared to these, the element of rlung is
subtle. There are many distinct degrees of coarse and subtle for the internal subdivision
of rlung itself.”306
Here Mkhas grub is breaking down the elements on a scale running from coarse to
subtle. Generally the order of emanation we witness in sådhana practice (typically
generation stage) and in material formation (of the cosmos or person) proceeds from
subtle to coarse, whereas that of dissolution in sådhana practice (typically but not
exclusively completion stage) and in material disintegration (of cosmos or person)
reverses that order, moving from coarse to subtle. Mkhas grub is showing how rlung is
the most subtle of the elements, but that rlung itself can be broken down further into
gross and subtle components. Until the most subtle of these components of rlung is
dissolved, the ultimate dissolution of the body is incomplete. Mkhas grub then elaborates
upon the breakdown of rlung to describe what we have termed the elemental rlung:
“Since the hue and function and so on of each of the five primary rlung individually
accord with the four elements in term of location and so forth, the earth rlung and so on
are set forth (systematically). There is rlung, moreover there is earth. So there is the earth
rlung. (This rationale) is not posited. Composed as yellow earth, white water, and green
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
They see perfect reality. V.64
Then, having prostrated again and again,

With eyes full of tears,
Having circled the lord
Three times
They praised vajradhara. V.65 [Translation Kittay 2011, p.775]
Chapter Sixty-Eight of the Vajramålå provides an extremely more detailed picture
of the human body together with the energies that move through it and impel it through
the cycle of saµsåra than found in Chapter Sixty-Four. As such it seems like the more
intuitive choice for citation if the ordering schema of dissolution is indeed the process is
attempting to foreground. However, the absence of Locanå, Måmakî, Påñ∂aravåsinî and
Tårå from the description in Chapter Sixty-Eight likely prevents Mkhas grub from
referencing that chapter in his argument on correlating the goddesses. Moreover,
although it establishes connections between five Buddha families and the five root winds
of the body, there is no explicit mention of “elemental winds.”
306
phra ba thim zhes pa yang; sa chu me gsum rags shing; di las rlung gi khams phra pa
yin gyi; rlung rang gi nang gses kyi dbye ba [247.5] la phra rags kyi khyad par rim pa du
ma zhig yod do
Khenpo Choying Dorje presente the possibility that Mkhas grub may be citing a root text
here. [Personal communication, Spring 2011]!
! 97!
rlung, though they are in accord with the color of the three mandala which are the base of
the world, (they) are not in accord with the colors of any of the three, earth, water, and
rlung which are on the earth [sa stengs].307 If the abode of the body’s earth element is in
the secret place, then it doesn’t contradict being in the navel.”308
From this passage we learn that each of the five root winds shares a common
color and location with the one of the four elements [earth, water, fire, rlung]. The next
point is more complicated. Mkhas grub appears to be distinguishing these varieties of
rlung from the larger category of the element rlung. He does so by producing an example
based in Abhidharmic theory and the pan-Indian conception of cosmos as composed of
three realms. Mkhas grub makes a parallel between macrocosm and microcosm,
comparing elements of the earth realm to those belonging to the region below it; this
comparison displays the manner in which earth may abide in both the navel and secret
place of the human body. Mkhas grub seems to be saying that although the other cosmic
elements share qualities with the elements as they appear in the realm of earth, they are
not identical. Likewise, although the internal subdivisions of elements of the body share
qualities amongst one another, they remain distinct. Mkhas grub’s concluding move is
more radical. Moving farther afield from the texts of the Årya Guhyasamåja cycle,
beyond the completion-stage focused explanatory tantra of that system, Mkhas grub
invokes an explanation from another explanatory tantra, the Saµputa:
Alternatively, then you must reflect on how to account for the explanation from the
Sampu†a Tantra of arranging Locanå in the navel, the abode of earth and Tårå in the
crown, the abode of rlung.309
The Vajramålå and Saµpu†a tantras will play an important part in the analysis of
the dynamics of tantric polemics and exegesis expressed within the texts of the body
mandala debate in this dissertation. The Sampu†a Tantra, regarded as common to the
interpretation of both the Hevajra and Cakrasaµvara systems, is a more radical source for
support than Mkhas grub’s previous choices. He builds upon his discussion of the
completion stage-focused interpretations of the Vajramåla and its incorporation of both
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
307
Khenpo Choying Dorje suggests that this may be a quotation of a tantric
commentary/root text. He referes to Abhidharma theory according to which wind holds
water, and the elements combine with other things on earth to make colors. In other
words, wind isn’t color but it has color.
308 !rtsa
ba'i rlung lnga'i rang dbang gi kha dog dang byed las sogs 'byung ba bzhi dang
rnam pa mthun pa dang gnas la sogs pa du ma zhig gi sgo nas sa rlung sogs su bzhag gi;
sa yang yin rlung yang yin pas [247.6] sa rlung sogs su bzhag pa ma yin la; sa ser po chu
dkar por rlung ljang khur byas pa yang; 'jig rten gyi 'og gzhi'i dkyil 'khor gsum gyi kha
dog dang bstun pa'i dbang du byas kyi; sa stengs kyi sa chu rlung gsum gang yin thams
cad kyi kha dog dang [248.1] bstun pa ma yin la; lus kyi sa khams kyi gnas gsang gnas na
yod pa dang lte ba na'ang yod pa 'gal ba ma yin te
!
309
gzhan du sam pu ta las; lte ba sa'i gnas su spyan ma dang spyi gtsug rlung gi gnas su
sgrol ma 'god par bshad pa ji ltar 'chad soms [248.2] shig
! 98!
Mahåyoga and Yoginî tantra-based approaches to the Guhyasamåja to extend the limits
of interpretation beyond the Guhyasmåja system. After working carefully and closely
through the interpretation of commentaries and an explanatory tantra associated with the
Årya tradition of the Guyasamåja, suddenly Mkhas grub has brought those approaches to
locating goddesses and elements within the human body into dialogue with a yoginî
tantra- affiliated explanatory tantra. The reference to the Sampu†a appears to be to a
passage cited by Ngor chen is his own response to Mkhas grub’s text:
“The karmamudrå, Locanå, she of great compassion, she of great method, she with the
domain of manifold nature, abides in the nirmåñacakra in the variegated lotus at the
navel. She should be known as baµ in/as the water element.
The dharmamudrå Måmakî, possessed of loving kindness and the manner of
prayer, preeminent goddess of the vajra family, abides in the dharmacakra within the
eight-petalled lotus in the heart. She should be called må in/as the fire element.
The mahåmudrå Påñ∂aravåsinî, with joy and vigor, goddess of the lotus family
abides in the sambhogacakra in the eight-petalled lotus at the throat. (With )yå, the
nature of rlung, she conquers all afflictions.
True samayamudrå, goddess of the karma family, with the yoga of equanimity
?and wisdom, the liberator from saµsåra, Tårå, abides in the mahåsukhacakra, the thirty-
two-petalled lotus...310
The four goddesses are densely encoded with correlations in this passage: they are
named, associated with a mudrå (karma-, dharma-, mahå-, samaya-), associated with
particular spiritual perfections, located within a cakra (envisioned as a lotus at a particular
site on the body), and finally, connected to a seed syllable and an element.
It is important to consider the significance of Mkhas grub’s decision to bring
Sampu†a Tantra, a text closely connected with the Cakrasaµvara and Hevajra tantras, into
his discussion of the Guhysamåja’s treatment of the goddesses, elements, and rlung. On
one level, this example makes a basic hermeneutic point, that one must develop
interpretations that allow for multiple readings and reworkings. Otherwise one runs the
risk of violating the fundamental rules of preserving the integrity of the Buddha’s word,
in this case, the tantras. On another level, it demonstrates that the human body is, to
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
310 This quote can be located in the sde dge version of the Sampu†a Tantra 81a.5-81b.1 as
cited in Ngor chen’s text:

!e nis sa ru shes par bya; las kyi phyag rgya spyan ma nyid; snying rje chen mo thabs
chen mo; rang bzhin sna tshogs sbyod yul ma; lte bar sna tshogs 'dam skye la ; sprul pa'i
'khor [562.3] lo rnam par gnas; wam ni chu ru shes par bya; chos kyi phyag rgya ma ma
ki; byams pa dang ni smon lam tshul; lha mo rdo rje'i rigs gtso ma; snying gar chu skyes
'dab brgyad la; chos kyi 'khor lo yang dag gnas; ma ni me ru brjod pa ste; phyag [562.4]
rgya chen mo gos dgar mo; dga' dang stobs kyi rnal 'byor gyis; lha mo pad ma'i rigs
'byung ma; lkod mar chu skyes 'dab brgyad gnyis; longs spyod 'khor lor yang dag gnas;
ya ni rlung gi rang bzhin te; nyon mongs thams cad rab 'joms byed; [562.5] dam tshig
phyag rgya chen mo nges; lha mo las kyi rigs gtso ma; btang snyoms ye shes rnal 'byor
gyi; sgrol ma 'khor ba sgrol ba mo; 'dam skyes 'dab ma sum cu gnyis; bde chen 'khor lo
chen por gnas
! 99!
some degree, a signifier that remains constant within the body mandala practice across a
variety of tantric cycles and their interpretation by Indian and Tibetan thinkers. It is the
deployment of symbols upon this body that changes. In the evolution of the practice
through tantric exegesis and polemics, there is a tendency to create links with other
systems of signification. The case of the goddesses is, therefore, an extreme instance of
phenomenon common to the evolution of mandala systems, a phenomenon we might
identify as “correlative correspondence.” Here, in the formulation and exegesis of body
mandala ritual, there is a repeated redeployment of a set of symbols in different
configurations in relation to the body and in different patterns of relationship to other
correlated sets such as the elements and buddha families.
Mkhas grub’s sources for citation reveal a great deal. Both of these “explanatory
tantras,” the Vajramålå and the Saµpu†a, will continue to be relevant to our discussion of
the body mandala debate; they will play an important part in our discussion of Ngor
chen’s reply later in the dissertation. The use of these explanatory tantras within the body
mandala debate highlights the delicate network of relations between root and explanatory
tantras as well as between different tantric textual cycles. Mkhas grub ends his body
mandala debate text with a comparison of the Guhyasamåja with the Cakrasaµvara; this
choice attests to the fact that articulating such relationships was a pressing concern for the
author. In tracing such choices in the texts of the body mandala debate, we witness the
reinvention of hermeneutic strategies; the body itself is encountered anew through such
acts of interpretation.
We noted the significance of the Vajramålå, as an “evolving text” focused heavily
upon completion stage practices and yogîni tantra-based readings of the
Guhyasamåja.”311 Likwise, we considered why Mkhas grub would conclude his
discussion of the goddesses with a quote from the Saµpu†a Tantra, an explanatory tantra
deeply connected with the Hevajra and Cakrasaµvara cycles.312 Both of these
explanatory tantras express more complex and sophisticated understandings of nature and
inner workings of the subtle body than the other texts Mkhas grub cites in this section.
Kittay 2010 notes that the correlation of the goddesses with elemental winds
rather than with the elements themselves was a form of the practice promoted by Tsong
kha pa.313 Therefore in the context of Mkhas grub’s writings on body mandala, it is
possible that Mkhas grub’s extensive exploration of this aspect of the mechanics of the
practice was motivated by a need to support the validity of a practice expounded by his
teacher.314 For example, as we observed, while the Vajramålå, which we have established
openly employs some of the techniques of yoginî tantra, incorporated winds into the
Guhyasamåja practice, the description of dissolution there does not precisely parallel
what we find in Mkhas grub’s text. Kittay remarks on the representation of the rlung
practices in the Vajramålå and its commentary by Alaµkakala!a as follows:
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
311
Kittay 2011, p.185 .
312
Note that Ngor chen’s text N1 cites this tantra several times.!!!
313
Kittay 2010, 258 note 923.
314
Of course, in light of Bentor 2006 work,we know that Mkhas grub’s approach does
not always match his teacher’s.
! 100!
“The Tantra does not explicitly identify which elemental mandalas and colors are
identified with which energy-wind. Alaµka identifies the Amitåbha fire element with the
Ascending energy-wind and the Amoghasiddhi wind element with the Equalizing energy-
wind, 168B, but makes no specific link to the others. In his detailed discussion of this
practice in the BIL (Rim pa lnga rab tu gsal ba’i sgron me), Tsong kha pa outlines the
correspondences with the other energy-winds indicated above, and gives greater detail on
the practice, which involves meditating the primary Buddha energy-wind with the other
goddess elemental-winds, so e.g. the red Amitåbha Ascending energy-wind emanating
from the left nostril is meditated with the four elemental energy winds of Påñ∂aravåsinî,
Tårå, Locanå and Måmakî in that order, which, influenced by the fire mandala, appear as
red, reddish green, reddish white, and reddish yellow respectively.”315
In his own discussion of the links between goddesses and elemental winds, Mkhas
grub exphasized the dissolution of the body and its attendant winds, constituents, and
elements at death as the primary organizing schema for establishing correspondences. As
referenced above, Gelukpa scholars consider the manipulation of the death process
through mastery of completion stage practices that mirror this very process to be a vital
aspect of the Guhyasamåja system as a tantric practice. Reinterpreting the role of the
goddesses in connection with the dissolution process emphasizes this dimension of the
practice. In her study of the varying uses of embryological accounts in Tibetan Buddhist
texts, Frances Garrett has explored tantric narratives of gestation and the formation of the
human body as models for spiritual transformation. In doing so, she has demonstrated
how these narratives were often produced in dialogue with narratives of the body’s
dissolution at death.316 In the tantric context, we might view the relationship of birth and
death processes as a highly evolved instance of the principle of “correlative
correspondence”; ritual reinforces this relationship through the logic of emanation and
absorption orchestrated between subtle and gross components.
In accounts of ordinary birth, wind and karma interact with the elements as well
as with the essences of father and mother to produce the human body. However, the goal
of tantric practice is to create not an ordinary human body but a buddha body. Referring
to the work of Brian Cuevas, Garrett notes the proliferation of ritual formulations of the
intermediate state in fourteenth and fifteenth-century Tibet based in the six doctrines of
Nåropå.317 Tsong kha pa, in particular, is credited with “synthesizing a comprehensive
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
315
Kittay 2010, 258 note 923. Kittay adds here a reference to a relevant passage in Tsong
kha pa’s writings: “See TK 2010 at 234.” For Alaµka’s commentary, see Alaµkakala!a
(Tshul khrims rin chen). Rnal 'byor chen po'i rgyud dpal rdo rje phreng ba'i rgya cher
'grel pa zab mo'i don gyi 'grel pa. In bstan 'gyur (sde dge). TBRC W23703. 34: 4 - 442.
delhi: delhi karmapae choedhey, gyalwae sungrab partun khang, 1982-1985.
Rnal 'byor chen po'i rgyud dpal rdo rje phreng b'i rgya cher 'grel pa zab mo'i don
gyi 'grel pa. In bstan 'gyur/?gser bris ma/. TBRC W23702. 34: 3 - 638. tibet: [snar thang],
[17-?]
!
316
Garrett 2008.
317
Garrett 2008, p.110. See Cuevas, Bryan J. 2003. The Hidden History of The Tibetan
Book of the Dead. New York: Oxford University Press and Tsong-kha-pa Blo-bzang-
! 101!
path of training based on Nåropå’s six doctrines” and integrating “intermediate stage
teachings with instructions for the purification of the three life stages.”318 In other words,
Tsong kha pa mapped the life trajectory of death, intermediate state [bar do], and rebirth
states onto sådhana practices based in highest yoga tantra.319 Employing sådhana as a
means to purify these states of existence is referred to as “bringing the three bodies to the
path.” [sku gsum lam ‘khyer]320 The trajectory of embodiement was thereby connected
with the production of three varieties of buddha bodies: dharmakåya, sambhogakåya, and
nirmåñakåya respectively.321 Garrett describes Tsong kha pa’s articulation of these
practices in his Sngag rim chen mo as part of the transition from generation stage to
completion stage practice: “Success in this type of meditation is said to result in the
winds of one’s subtle body entering, remaining, and then dissolving inside the central
channel, whereupon one may begin completion stage practices.”322 Manipulation of and
control over the channels, winds, and drops of the subtle body or perhaps, more
accurately, of the “vajra body” is therefore essential to this practice and ultimately to the
soteriological project.
In Garrett’s comparison of embryological narratives, she notes discrepancies over
the role of the elements in human conception and development and, in particular, over the
role of wind [rlung]. Garrett observes that: “the names and functions of the winds, as
taken from the Buddhist sutra, are the most prominent and consistent details these
medical commentators add to their accounts of the body’s weekly development.”323
These winds include the five root winds discussed above along with five subsidiary
winds, all drawn from tantric physiology. None of these, however, seem to be labeled
specifically as elemental winds. She notes that by the fifteenth century, the winds became
more important to Tibetan medical accounts of fetal gestation, suggesting that religious
texts actually influenced medical ones.324 Moreover, the role of the elements also
became more prominent over time. In this regard, Garrett observes: “the increasing
interest in the material nature of the human body, and in connecting the human individual
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
grags-pa, and Glenn H. Mullin. 1996. Tsongkhapa's Six Yogas of Naropa. Ithaca, N.Y.:
Snow Lion Publications.
318 !Garrett2008, p.110.!

319
Garrett 2008 also refers to Bentor 2006’s discussion of Tsong kha pa’s and Mkhas
grub’s participation in debates over this practice, in particular over the questions of
whether or not only birth is purified by the generation stage and whether thr rules only
apply for birth from a womb. Garrett 2008 p.114. See Bentor 2006 p.186 fn4 & p. 192
respectively. It is of interest to note that Tsong kha pa and Mkhas grub postulate the
purification of the three states occurs in the generation stage while the generation of the
Buddha bodies occurs in the completion stage.
320
Garrett 2008, p.112.
321
As discussed above, Candrakîrti’s Vajrasattva-sådhana organized generation stage
pracitice in terms of the production of these bodies.
322
323
324
! 102!
with the cosmos at the material level, may be an idea that comes to medical thinkers as a
result of the increasing influence of Buddhism on contemporary scholarly
communities.”325 Both elements and winds appear to have been given more weight and
attention than karma in this context.
Above, we have looked closely at Mkhas grub’s efforts to negotiate the
relationship of winds and elements embodied as buddhas and goddesses in aspects of
body mandala ritual. If we consider these efforts in light of Garrett’s discussions of
controversies within Tibetan embryologies, we find a shared discourse of themes of gross
and subtle, emanation and absorption, cosmic creation and destruction all located within
the body. Concerns with causality and especially with the causal efficacy of winds and
elements at the nexus of tantric and medical accounts provides us with a taste of the
intellectual climate expressed within the bodily discourses of late fourteenth and
fifteeenth-century Tibet. They also prepare to engage with comparable tensions between
tantric and sütric approaches to the human body to be discussed in the next chapter of this
dissertation. We will see how, like the tension between karma and human intervention
that Garrett describes in the context of medical rituals to alter the gender of a human
fetus, for example, Mkhas grub’s writings on body mandala reflect a tension between
“determinism and ritual agency.”
In the case of such medical rituals, Garrett suggests that Tibetan medical writers
referred to a wider domain of discourses such as Indian medical and Tibetan astrological
texts.326 In the case of body mandala ritual, Mkhas grub too, refers to a wider range of
discourses to navigate tensions in the interpretation of ritual practice focused upon the
human body; in his case, such tensions express concerns with the relationship between
that unenlightened body and a buddha body. In the next chapter we will observe Mkhas
grub’s mode of engagement with these discourses, while continuing to consider
exegetical details such as his sources for citation and polemical aspects such as tone. In
his account of what makes the body mandala practice special and distinct from other
mandala traditions, we will become more intimately acquainted with Mkhas grub, both in
terms of his polemics and his commentarial skill. In the process, conceptions of the body
and its role in tantric practice will begin to rise to the surface.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
325
326
Garrett 2008, p. 142. !
! 103!
Chapter Four: Imagined or Real?: The Use of the Category of
“Fabrication”[ bcos ma ] in Mkhas grub’s Body Mandala Chapter
This chapter focuses upon the notion of “fabrication” [bcos ma] introduced by Mkhas
grub in his chapter on body mandala in his Ocean of Attainmen. In particular, it
draws attention to how Mkhas grub uses “fabrication”as a standard in articulating
relationships, namely, the ontological relationship of the human body to the mandala and
the relationship of the body mandala to “outer mandala.” In this context, “outer
mandala” refers to mandala paintings or ritual altars made of painted powders, what we
might call ‘representations.’327 Through his deployment of this category of “fabrication,”
Mkhas grub probes the boundaries of the påramitånaya and mantranaya and brings to
light key aspects of the body’s role as a basis for tantric practice. He explores ritual and
philosophical implications of “fabrication” side by side to establish the superiority of the
body mandala to other mandala practices. In the process, he challenges a defining feature
of tantric practice, the act of imagining oneself as a deity. In examining these various
dimensions of “fabrication” in Mkhas grub’s argument, we will consider possible
motivations for his attempts to reconcile the methods and aims of tantric practice with the
påramitånaya at large through the language of Buddhist philosophical investigation. We
will also highlight particular sensitivities triggered by the topic of the human body.
I. The Positions of others
IA. Category confusion

Mkhas grub begins his chapter on body mandala with a discussion of a quote from
Ghantapa’s “Condensed Activities of the Cakrasaµvara initiation” [dpal ‘khor lo sdom
pa’i dbang gi bya ba mdor bsdus pa]:
“These sentient beings, are not separated from the naturally established mandala.”328
Mkhas grub uses his opponents’ inaccurate interpretation of this quote as the point of
entry for the portion of his text assessing the positions of others; he returns to the quote
later to articulate his own views on body mandala. We will first deal with Mkhas grub’s
critique of other’s interpretations as found in 234.1-238.2 of his text; the friction he
creates in juxtaposing tantric practice and the idea of visualizing the human body as
divine with påramitånaya practice at large is of especial relevance. In Part II of this
chapter, we will consider the original context in which this controversial statement
appears in Ghantapa’s text. However, for the present, we will focus upon the particular
points of critique that it enables Mkhas grub to raise.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
327
In the conclusion of the dissertation,we will explore the implications of this category
of fabrication for defining and evaluating representations from a Buddhist perspective.
328 !234.1!'gro ba 'di dag rang bzhin gyis; sgrub pa'i dkyil 'khor gnyis med pa'o
! 104!
Mkhas grub introduces the inappropriate usage of the category of “fabrication” by
his opponents immediately following the Ghantapa quote:
“They speak of the body mandala as an unfabricated mandala [ma bcos ba'i dkyil 'khor],
(and) they don’t understand in what way the bodies of sentient beings are primordially
mandalas. Therefore, they say that what already existed [sngar yod ] is cultivated
through visualization.”329
Mkhas grub seems to be highlighting tensions inherent in the classification of the body
mandala as ‘unfabricated.” If it’s already formed, what’s the point of performing the
body mandala practice? The tone of the critique resembles similar statements made in
the critique of subitism, the archetypical enemy within the realm of Tibetan philosophical
debate and reasoning. In other words, if we are already enlightened, why practice? Such
objections also characterized Buddha nature debates, a connection to be explored in
further depth below. Similar critiques are reiterated throughout this section of Mkhas
grub’s text, bringing attention to the importance of the path structure in legitimizing
tantric practice for the Gelukpas. In the context of Gelukpa interpretations of
Guhyasamåja practices, the structure of Någårjuna’s five stages was instrumental in such
acts of legitimation. As we will see below, in the section devoted to his own views on the
practice, Mkhas grub is not defying the classification of the body mandala as
“unfabricated’; rather, he is pointing out areas of confusion for those who lack his
training and clear perspective.
Mkhas grub then points to ways in which proponents who accept that the body is
already a mandala threaten the very structure of the Buddhist path, beginning with the
four noble truths:
“As for those who speak in this way, they are of improper understanding that is extremely
uncritical. If you see things in this way, isn’t it the case that the (noble) truth of suffering-
(the very fact) that the body of sentient beings is generated by karma and kle!a- is
eradicated?”330
In these terms, the body is virtually equated with suffering, making any view of it
as divine a category confusion of the worst kind. Mkhas grub then proceeds to make a
number of hyperbolic statements targeting the opponent who confuses human bodies with
mandalas. Such an opponent jeopardizes the very pillars of the Buddhist tradition,
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
329 !lus kyi dkyil 'khor ni ma bcos ba'i dkyil 'khor du gsungs pas; [234.2] 'gro ba thams cad
kyi lus gdod ma nas dkyil 'khor du yod pa la yod par ngo ma shes pa shes par bya ba'i
phyir du sngar yod gsal 'debs pa'i tshul gyis bsgom pa yin no zhes zer ro
!
330 !'di ltar smra ba ni ha chang yang ma brtags pa'i log bar rtog pa yin [234.3] te; de lta na
'gro ba rnams kyi lus las dang nyon mongs pas bskyed pa'i sdug bsngal bden ba ma yin
ba'am
! 105!
particularly the path model of gradual practice, ethical conduct, and the theory of karma.
For example:
“One would thus be lead to conclude that the mandala of Vajradhara (ie. the body
mandala) is established by the power of karma and kle!a. (Further, the state of) being
Vajradhara and (the state of) being a transmigrator would be inseparable [mthun yod pa
tsam du ma zad]. Not only that, but also, due to that perverse implication, the state of
being a sentient being who experiences the suffering of saµsåra would be pervaded by
(the state of) being Vajradhåra.”331
In other words, mixing inalienable categories like Buddha and sentient being
contaminates both and potentially obliterates the possibility of moving beyond saµsåra at
all. Even more dramatically, Mkhas grub elaborates upon his critique, asserting:
“From this perspective, one would have to claim that if there is defilement, it is
wisdom, or (for that matter) anything at all. If all sentient beings were actually buddhas,
the worldly container (of sentient beings) would have to be the celestial palace of self-
appearing wisdom. By virtue of this contradiction, the (inevitable) consequence would
be that saµsåra itself would be untenable as an object of knowledge. How could it be
that a Buddha who doesn’t recognize himself as being Buddha, not even knowing
himself, would be omniscient? The result would be a buddha who does not know any
object of knowledge at all. Such a foolish one is truly incredible!”332
In other words, one is leveling the very ground from which the path begins, the
recognition of suffering and of the nature of saµsåra. If saµsåra cannot be conceived of
as an object of knowledge [shes bya ], the entire imperative of liberation through
reasoning as well as the necessity of working from within the bonds of conventional
reality are called into question. These are methods embraced by the Gelukpas as essential
for liberation. Mkhas grub is therefore playing upon the resonance of the rhetorical power
of issues surrounding the relationship of the two truths (conventional and ultimate)
grounded in the Gelukpa Madhyamaka perspective. In the process, he is employing the
tools provided by Buddhist logic and epistemology for explaining human perception and
conception and their potential to access the true state of things.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
331 !rdo rje 'chang gi dkyil 'khor las nyon gyi dbang gis grub pa yod par thal ba'i phyir
dang ; rdo rje 'chang dngos yang yin 'khor ba ba yang yin pa'i gzhi [234.4] mthun yod pa
tsam du ma zad; 'khor ba'i sdug bsngal myong ba'i sems can yin na rdo rje 'chang yin pas
khyab par 'gyur te
!
332
de lta na ni sgrib pa yin na ye shes yin pa sogs su’ang 'dod dgos la; sems can thams
cad sangs rgyas dngos yin na snod kyi 'jig rten ye shes rang snang gi gzhal yas khang
dngos ma yin ba 'gal bas 'khor ba shes bya la mi srid par [234.6] thal zhing; rang nyid
sangs rgyas yin pa la yin par ngo ma shes pa'i sangs rgyas des ni rang yang ma shes na
chos thams cad mkhyen pa lta zhog ; de bas na shes bya gang yang mi mkhyen pa'i sangs
rgyas shin tu blun pa de lta bu ni ches ya mtshan no
! 106!
This is not to suggest that Mkhas grub is alone in invoking multiple frameworks
of Buddhist thought in the polemical context. Rather, it is our objective to highlight the
different ways these arguments and levels of discourse are employed by the author at
hand for different purposes. For example, Mkhas grub may be building upon or even
resurrecting charged controversies surrounding Buddha nature, for example, to give
momentum to an argument. Alternatively, the writer may garner rhetorical advantage by
accentuating certain aspects of the opponent’s argument to make it resemble the view of a
heretical tradition like the Såµkhya (as Mkhas grub does at one point in this text).333
Moreover, understanding Mkhas grub’s writings on Madhyamaka, for example, and his
encounters with other thinkers on the topic may inform our interpretation of Mkhas
grub’s polemical writings on tantra. Tsong kha pa’s positions and the responses of his
contemporaries to similar issues may also better contextualize Mkhas grub’s position. In
highlighting these connections, we will attempt to gain a more comprehensive view of
Mkhas grub’s literary persona, his voice and strategies of argumentation as well as of the
intellectual milieu of scholar monastics to which he contributed.
Next, Mkhas grub applies this critique of those who confuse sentient beings and
mandala to explicitly challenge some interpretations of the logic of tantric ritual:
“On the other hand, one who makes claims like that would have to (also) claim
that it is totally unnecessary for one who is a Buddha and recognizes oneself as such to
cultivate the path [phyin chad lam]. (Based on that) one would have to claim that
cultivating the path after encountering [brda 'phrod ] the body mandala one time is totally
unnecessary.”334
From the Sakyapa perspective, the body is considered, in a sense, to already be divine.
From that perspective, tantric practice is used to revolutionize the mind’s understanding
of the body, to expose it to the light per se. This is one potential target of Mkhas grub’s
critique here, although certainly not the only one.335 Initiation into the body mandala of
Hevajra for the Sakyapa entails the crucial phase of the guru showing that each part of the
disciple’s body is a part of the mandala inhabited by deities. Of course, such initiation
also demands a commitment to the continued practice of this embodied visualization.
Therefore the centrality of the tantric vow to the practice itself counters Mkhas grub’s
concerns with the abandonment of practice. Below we will return to the
phenomenological dimension of tantric ritual and initiation practice raised in this
example by the term “encounter” [brda 'phrod ]. We will see how such remarks such as
these within the body mandala debate alert us to the volatile status of sense perception
and cognition as instruments or catalysts for realizing enlightened awareness. For now,
we will take note of this passage as evidence of Mkhas grub’s strategy of alluding to the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
333
Personal communication, John Dunne Fall 2011.
334
gzhan yang de ltar [235.1] 'dod pas ni sangs rgyas yin pa la yin par ngo shes phyin
chad lam bsgom pa'i dgos pa ci'ang med par 'dod dgos pas lus dkyil lan gcig brda 'phrod
phyin chad lam gang yang bsgom mi dgos par 'gyur ro
335
Personal communication, Kurt Keutzer 12-16-2013.!
! 107!
opponent’s (/opponents’) understanding of the relationship of the human body to the
mandala as a threat to the very principle of the Buddhist path.
Mkhas grub proceeds to address the status of the tantric approach in relation to
that of the perfection vehicle [phar phyin theg pa'i lam.]. In doing so, he attempts to
demonstrate that those who reify the tantric path over the latter suffer from a fatal self-
contradiction:
“ [235.2] ..one who claims that it’s necessary to traverse the grounds and paths in stages
while cultivating the path is unstable. Having asserted the “Universal Illumination,” the
eleventh (bhümi), which is explained as the ultimate object of attainment of the path of
the perfection vehicle, to moreover be inferior to tantric Vajradhara, [235.3] (such a
proponent) establishes all sentient beings as primordially the mandala of Vajradhara. (So
matters really) become extraordinary.”336
The equality of the tantric path to that of the perfections is a standard formulation of what
we understand as the Gelukpa tradition. Other traditions, such as the Sakyapa, disagree,
holding the tantric path to be superior.337 This fundamental disagreement may shed some
light upon Mkhas grub’s motivation for measuring tantric practice according to the
standard of the “path of the perfection vehicle.” Mkhas grub’s use of hyperbole in his
elaboration upon such category confusions is charged and borders upon comical at times.
However, he touches upon significant points of tension for Tsong kha pa’s disciples in
their attempts to position their teacher’s tradition in relation to those of the others.
Before concluding our discussion of Mkhas grub’s concerns with category
confusion, I would like to draw attention to two important technical terms he introduces:
“In these positions, it is necessary to assert that the basis of purification [sbyang gzhi],
the world and its inhabitants, is impossible. Thus, any category of the purifier [sbyong
byed ] at all becomes (logically) untenable. [236.2] Since it would then be the case that
all sentient beings are simply Vajradhara, there is a resultant inseparability of those who
are endowed with the destiny of mantric action and those who are not.”
The term sbyang gzhi, translated here as “basis of purification,” is a vital

expression for evaluating the role of the body in tantric practice recurring throughout
Mkhas grub and Ngor chen’s texts. Bentor 2006 suggests that Geluk and non-Geluk
perspectives on the sbyang gzhi diverge. According to Bentor, several non-Gelukpa
schools of thought split the category of into a pure basis and a sbyang bya, what is to be
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
336 ![235.2] lam bsgom nas sa lam rim can du bdrod dgos par 'dod pa rig pa rnal du ma
bzhag pa yin la ; phar phyin theg pa'i lam gyi thob bya mthar thug tu bshad pa'i bcu gcig
kun tu 'od kyis sa thob kyang da dung sngags kyi rdo rje 'chang las ches dman par khas
blangs [235.3] nas; sems can thams cad rdo rje 'chang gi dkyil 'khor dngos su gdod ma
nas grub par khas len pa 'di las ngo mtshar du gyur ba ci zhig byar yod
!
337
Personal communication, Kurt Keutzer, 12-16-2013. For an example of this
perspective, see Chapter 5 in Verrill’s 2012 translation of the Sakyapa patriarch Sonam
Tsemo’s (1142-1182) The Yogini’s Eye.
! 108!
purified (i.e. the defilements).338In reference to the sbyang gzhi, Bentor states, “While in
the Dge lugs pa tradition it refers to the ordinary saµsåric state, according to others it is
the true nature of things.” In our interpretation of Mkhas grub and Ngor chen’s texts on
body mandala, we will devote attention to the use of such terms, terms describing the
basis or support for practice. In doing so, we may come closer to determining how
Gelukpa and Sakyapa attitudes toward the role of the human body in tantric practice
diverge and how this divergence informs their views on body mandala. 339 On a more
general level, attention to such terminology may reveal how these authors cope with the
problem of relating ordinary and enlightened realities. This is a problem that is central to
writings on tantra but features prominently in other genres of Buddhist discourse as well.
IB. The causal link

Tsong kha pa and his descendants express concerns with establishing a verifiable
causal link between this present human body and the Buddha bodies produced through
tantric practice. These concerns often took shape in discussions of the need to produce a
“similar type cause” [rigs ‘dra’i rgyu], as introduced in Chapter Three. Such a variety of
cause provided the necessary link between the ordinary and inherently flawed human
body and the form body of the Buddha. This cause was to be produced through mastery
of the completion stage of sådhana practice. Tsong kha pa describes the Buddha body as
a type of container:
“For those tormented by thirst their main focus will be the search for something
to drink, but for that they will need a container. Likewise, for those of the Great Vehicle
who are moved by a great compassion that is unable to bear living beings being
tormented by suffering and deprived of happiness, their main focus will be striving for
the welfare of others...actually appearing before sentient beings and then accomplishing
their needs is to be performed by the form body (rüpakåya) from the two types of
enlightened bodies, and not by the dharmakåya. Therefore the main focus of their
endeavor is the form body. Because of this, a special cause that is similar in type to the
form body, that is used as a method for achieving the form body, and that is a special and
peerless feature not found in other vehicles, other classes of tantra, and in the generation
stage has to be present in the completion stage.”340
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
338
Bentor 2006, pp.196-7!
339
The Sakyapa triadic framework of the cause, path, and fruit/result may also be helpful
in evaluating attitudes toward the body as a support for practice. In our spring 2011
meetings, Khenpo Yeshe explained the sbyang gzhi as the body and the sbyong byed as
the path. He added that while Ngor chen regards the body is the basis for purification,
Mkhas grub sees it as an impure thing needs to be left behind. Khenpo Choying Dorje
pointed out that the late Fifteenth-century Sakyapa interpreter Go rams pa explicitly
stated that the kle!as become wisdom.
340 !Kilty 2013, p.117.!
! 109!
In this quote, Tsong kha pa copes with a perceived challenge posed by tantric practice to
the Buddhist theories of causality that form the very basis for karmic law and human
embodiment. Resources for such theories of causality include the Abhidharma literature
and Madhyamaka reformulations of the ideas presented therein.341 If a Buddha body is
not the result of or equal to karma and kle!a, then how does it manifest? Among the five
standard categories of effect, the Buddha body seems to most closely resemble the
nißyanda-phala or the “natural outflow” effect; this is an effect that “corresponds to the
nature of its cause in activity and experience” and is produced by the sabhåga-hetu or
“homogenous cause.”342 Witihin Abhidharmic literature, the nißyanda is used to describe
how one’s present actions can positively influence future moral tendencies.343 In applying
such a formula within the tantric context, the requisite action is the completion stage of
sådhana practice. The cause appears to then be the series of simulations of self as
Buddha effected through such practice. Therefore, in primatizing the production of a
form body and substantiating the coherence of causal logic for producing such a body,
Tsong kha pa identifies tantric practice as central within a larger påramitånaya project of
compassionate activity.344
As evidenced by his repeated recourse to hyperbolic category confusions, Mkhas
grub expresses a comparable anxiety to negotiate the relationship of the human body to
the Buddha’s form body in terms of the larger påramitånaya: mantranaya framework.
However, in doing so, Mkhas grub challenges a fundamental aspect of tantric meditation,
visualizing oneself as a Buddha, by subjecting it to the principles of Buddhist logic and
epistemology:
Moreover, when cultivating oneself as a deity, some (say) that, ‘I am actually divine.’
In that case, that mind becomes a mistaken consciousness [log shes ] when cultivating the
self as divine even when it isn’t. Thus, it becomes inadmissible as the cause of
Buddhahood.345
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
341 !Frances Garrett 2008 has presented the complex theory of causes and conditions that
pervade embryological accounts from Tibetan medical works and their Indian sources.341
Garrett has explored the different uses of embryological accounts in Tibetan Buddhist
texts, as both a deterrent to craving rebirth in a human body and as positive model for
spiritual transformation.!
342
Hodge 2003, fn 10.5 & Hirakawa 1993, p.189
343 !See!Abhidharma-ko!a II.56-58 as referenced in Hodge 2003, fn 10.5. For an example
Madhyamaka understanding of Abhidharmic causality, see Kalupahana 1996 on

Någårjuna’s writings, pp. 110-111 & 284.
344 !I have encountered numerous references in secondary scholarship that suggest that the
ability to produce a form body by means of the “similar type cause” was regarded as
distinguishing feature of the tantric path, one that marked it as superior even, to the path
of perfections. For example, see Hopkins 2007, p.315.!
345
[236.3]yang kha cig rang lhar bsgom pa'i dus na lha dngos yin te; gal te lha ma yin
kyang lhar bsgom pa yin na blo de log shes su 'gyur pas sangs rgyas kyi rgyur mi 'thad
pa'i phyir ro;
! 110!
Here we find another manifestation of Buddhist notions of causality introduced within the
framework of Buddhist theories of perception and cognition. Namely, false cognitions
[log shes Skt. mithyå-j≤åna or viparyaya-j≤åna] are insufficient bases for liberation. The
passage suggests a irreconcilability of certain canonical Gelukpa philosophical notions
with the most essential principles of tantric practice, most notably, the practitioner’s
identification with a deity. Mkhas grub continues:
“...when performing imaginative activity [lhag mos] in cultivating clear realization

[mngon rtogs]:
All sentient beings are empowered. All obstacles and obscurations are purified. One
cultivates installation in the level of Vajradhara...
Is that the case or not? What is cultivated?
If it’s like the first (i.e. if these propositions are indeed true), all sentient beings
would become Vajrasattva. (These would be the consequences):
Beings who hadn’t overcome (the) obstacles (ie. non-Buddhas) would be
impossible. Beings who hadn’t attained empowerment would also be impossible.
Sentient beings who hadn’t completed the purification of obstacles and obscurations
would also become impossible.346
Because one becomes mistaken in cognition [log shes su 'gyur bas] if one
practices as though one were a Buddha even though one is not, (that consciousness) is
untenable as the cause of Buddhahood. If one speaks thus, what is to be said?”347
This passage probes the very basis of imagination itself and its potential as a tool
for liberation. Why is Mkhas grub pressing the issue of a seeming irreconcilability of
vital principles of the påramitånaya such as the two truths with visualization of oneself as
a Buddha, a vital aspect of tantric theory and practice? The author appears to question the
point of repeatedly imagining something that isn’t, logically speaking, true. Ultimately,
he begs the question as to how inculcating oneself in such delusions could possibly help
to defeat ignorance and to realize enlightened awareness. In invoking the category of log
shes or “false cognition” to describe such acts of imagination, Mkhas grub taps into yet
another level of Buddhist discourse, the realm of “valid cognition” [pramåna Tib. tshad
ma].
IC. Pramåña
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
347 !...mngonrtogs bsgom pa'i skabs su lhag mos kyi mdzad pa lta bu bsgom nas
sems can thams cad la dbang bskur; sdig sgrib thams cad sbyangs; rdo rje sems [236.5]
dpa'i go 'phang la bkod par bsgom pa sogs de ltar yin nas sam ; ma yin bzhin du bsgom
pa gang yin; dang po ltar na [yin]; sems can thams cad rdo rje sems dpar gyur zin pas
sangs ma rgyas pa'i sems can mi srid pa dang; dbang bskur ma thob pa'i [236.6] sems can
mi srid pa dang; sdig sgrib dag ma tshar ba'i sems can mi srid par 'gyur zhing; ma yin
bzhin du bsgom na log shes su 'gyur bas sangs rgyas kyi rgyur mi 'thad do zhes brjod na
ci smra
!
! 111!
The Sakya patriarch Sa skya pandita’s (Sa pan) thirteenth-century antirealist reading of
Dharmakîrti, the Tshad ma rigs gter and its autocommentary (itself a response to the
works of the Gsang phu tradition of Phya ba348) sparked centuries of debate. Its relevance
persists even today. Dreyfus regards Mkhas grub as part of the larger project of clarifying
Sa pan’s ideas in terms of the Gelukpa realist interpretation of Dharmakîrti’s philosophy.
A significant portion of this philosophy concerned theories of “valid cognition.”
[pramåña Tib. tshad ma]. Pramåña theory provides a forum for considering how the two
truths frame our processes of perceiving and conceiving objects. Do we see things
because they are really there? Can we rely upon our senses for an accurate portrait, view,
or understanding of the world? Is it ever possible to see things as they truly are? These
are among the many questions entertained by pramåña theorists. Building upon the
suggested equivalence of tantric acts of visualization with log shes or “false cognition”
discussed in the preceding section, we will examine Mkhas grub’s relationship to
pramåña theory. In the process, we will assess how this relationship may have factored
into his evaluation of tantric practice in the body mandala debate.
Dreyfus describes the climate that characterized writing on pramåña duing Mkhas
grub’s time as charged:“...at the end of the fourteenth and the beginning of the fifteenth
centuries, a conflict of interpretations opposed the partisans of Tibetan realism to those
who followed Sapan’s antirealism.”349 Dreyfus describes the Gelukpa perspective on
valid cognition shared by Tsong kha pa’s disciples, specifically Mkhas grub and Rgyal
tshab dar ma rin chen (1364-1432) (though they differ on various points), as distinct from
Sa pan’s in significant ways:
“the explanation of valid cognition as the elimination of a superimposition is an obvious

truth. This has important consequences for their understanding of perception...For Sa-
pan, perception does not eliminate superimposition, for its function is strictly limited to
the passive appearance of things.”350
These conflicting perspectives, inherited generally by the Gelukpa and Sakyapa

traditions, revolve around fundamental assumptions about the relationship between
conventional and ultimate realities. For example, if it is not possible to perceive reality
without projecting inaccurate frameworks of understanding onto it, then how are we to
practice within the bounds of conventional reality? What shape should such practices
take, and what are their limits?
Dreyfus has shown how Mkhas grub was embroiled in debates over identity that
dealt with how to properly understand the relationship of perception, cognition and
reality in Dharmakîrtî’s thought.351 Mkhas grub and his fellow “moderate realists”
distinguish identity in terms of substance and concept, perception and conception:
“As realist thinkers, Ge-luk philosophers tend to stress the importance of intelligibility as
a criterion of existence. Due to the moderation of their realism and their historical
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
348 !See Cabezon 2007, p.26-27.!
349 !Dreyfus 1997, p26.!
350 !Dreyfus 1997, pp.376-77.!
351
In particular, the question of the relationship of universals and particulars was at the
heart of the controversy.
! 112!
situation as commentators of a conceptualist system, they do not go, however, as far as
equating intelligibility and reality, as extreme realists do.”352
Whereas in the antirealist vein of interpretation set forth by Sa pan and many of
his Sakyapa followers, perception is inherently flawed, the moderate realists attribute a
vital role to perception in determining identity. For them, “Perception is understood to
have undistorted epistemic access to reality. Nevertheless, it does not establish identity
but only substantial identity.”353 As moderate realists, Gelukpa thinkers like Mkhas grub
emphasized the importance of the conventional tools of perception and intellect in
working towards a more accurate view of things as they truly are, the ultimate truth. This
driving imperative of the Gelukpa project may have inspired Mkhas grub to attempt to
reconcile the soteriological orientation of certain aspects of the philosophical and
practical or meditative systems.
In discussing Mkhas grub’s interpretation of Dharmakîrti’s chapter on
pramåñasiddhi, van der Kuijp demonstrates the link created between this chapter and the
Kadampa lam rim teachings.354 This link helps to “form the hermeneutic grid along
which the soteriology of Buddhist pramåñavåda came to be established.”355 Pace
Steinkellner, he suggests that Tsong kha pa, rather than his teacher Ren mda’ ba may
have started this trend356:
“If we recall that the last half of the fourteenth century witnessed an unprecedented
revival in the Bka’-gdams-pa stan/lam-rim and blo-byong cycles- among the prime
movers behind this revival were the highly influential Rgyal-sras Thogs med dpal-bzang-
po (1295-1362) and Tsong kha pa’is teacher Chos-skyabs dpal bzang po- which to some
degree culminated in Tsong kha pa’s Lam-rim chen-mo, it would not appear wholly
unreasonable to suggest that the specifically dga’ ldan pa linkage between Dharmakîrti
and the lam-rim teachings could very well have had its inception with Tsong kha pa.”357
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
352
Dreyfus 1997, p.176.
353
Dreyfus 1997, p.175.
354
See Jackson’s 1993 edition and translation of the pramåñasiddhi chapter. Jackson,
Roger R., and Rgyal-tshab Dar-ma-rin-chen. 1993. Is enlightenment possible?:
Dharmakîrti and rGyal tshab rje on knowledge, rebirth, no-self and liberation. Ithaca,
N.Y.: Snow Lion Publications.
355
van der Kuijp , Leonard. 1985b. "Studies in the Life and Thought of Mkhas grub rje I:
Mkhas grub rje's Epistemological Oeuvre and His Philological Remarks on Dignåga's
Pramåñasamuccaya," Berliner Indologische Studien 1 (1985b), 75-105: p.75.
356
See Steinkellner 1983, pp. 282-283. Steinkellner, Ernst. 1983. “Tshad ma’i skyes bu:
Meaning and Historical Significance of the Term,” In Csoma de Ko rös Memorial
Symposium, Ernst Steinkellner, and Helmut Tauscher. 1983. Contributions on Tibetan
language, history, and culture. Wien: Arbeitskreis für Tibetische und Buddhistische
Studien, Universität Wien: 275-283.
!
357 !van der Kuijp 1985b, p77.!
! 113!
As a lineal descendent of Tsong kha pa, Mkhas grub is therefore potentially involved in a
discourse on pramåña that incorporates the notion of a graded Buddhist path into the
realms of Buddhist epistemology and soteriology. Van der Kuijp describes how Mkhas
grub himself emphasized his reception of these teachings from Tsong kha pa rather than
from Ren mda’ ba.358 Determining the boundaries of perception and conception, how best
to access a more accurate view of reality, and, ultimately, how to liberate oneself from
the bonds of saµsåra are the underlying problems framing these discourses. Further
reinforced by the plenitude of polemical discourse on pramåna sparked by Sa pan, we can
imagine the ways in which Mkhas grub’s invocation of pramåna here within his
commentary on the Guhyasamåja Tantra resonated with the charged tenor of those
debates.
In his commentary, Mkhas grub continues to probe the boundaries between
different levels of Buddhist discourse in introducing the terms of pramåña into the the
realm of tantric ritual and imagination:
“Moreover, having consecrated a thangka, a statue, and so forth, it becomes necessary to

claim that it is really a Buddha. One imagines all the realms of space to be filled with
flowers at the time of making offerings. If this is as one imagines, there’s the problem of
it not being visually perceptible [snang du rung ba]. If it’s not (i.e. if the statues are not
really Buddhas, the sky is not really full of flowers...), it becomes a false cognition [log
shes]. (Therefore) it is rendered untenable as the cause of Buddhahood.”359
The object of perception is negated here based upon the fact that the flowers
offered imaginatively in sådhana practice are not readily observable by everyone [snang
du rung ba]. The categories of snang du ma rung ba and snang du rung ba are important
for structuring refutation within the Praj•åpåramitå literature; they refer to two basic
categories of phenomena, those that are visible and those that are invisible to the ordinary
person.360 In this context, the log shes is to say something like a statue is actually a
Buddha once it is consecrated or that an object of visualization is really there. This
passage requires further clarification and attunement with pramåña theory. However, on
a basic level we can see how Mkhas grub is playing upon and bringing attention to the
controversial status of acts of imagination.
He continues to target possible contradictions underlying the logic of tantric
ritual, bringing both pramåña and the force of the vinaya to bear upon the mechanics of
consecration:
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
358 !van der Kuijp 1985b, p76.!
359
gzhan yang bris sku blugs sku sogs kyang rab [237.1] gnas zin nas sangs rgyas dngos
su 'dod dgos par 'gyur zhing; nam mkha'i khams thams cad me tog sogs kyis gang bar
bsgom nas mchod pa 'bul ba'i tshe; bsgom pa ltar yod na de'i tshe snang du rung ba las mi
dmigs [237.2] pas khegs la; med na log shes su song bas sangs rgyas kyi rgyur mi 'thad
do
!
360
I am grateful to Khenpo Choying Dorje for clarifying this context for me in our Spring
2011 meetings.
! 114!
“At the time of consecration, the disciple has already been transformed [gyur zin] into
Akßobhya by the Akßobhya water consecration. So again, why would it be necessary to
purify the stains of pride and so on with the Ratnasaµbhava consecration and so forth?
When one cultivates oneself as Vajradhara, one’s own body is ornamented with the major
and minor marks (of Buddhahood). One is practicing without experiencing any direct
perception [mngon sum du rtogs pa ] of omniscience whatsoever. So then the pretense of
oneself as a Buddha as a direct perception is nuts. That being the case, assuming that
we’re not talking about the pride of conceit, then you know yourself to be totally devoid
of the qualities of that (Buddha) like longevity [tshe stobs] and so forth. Surely, it would
be a (major) transgression to claim that, at that time, one is a Buddha. Don’t be careless
by causing trouble [dbyen bcos pas] with faulty doctrine.”361
This passage refers to the logic of consecration ritual in which each empowerment
purifies one of the obscurations through association with one of the five Buddha families.
In the water empowerment, the Vajråcårya imagines himself and the disciple and the
water as Akßobhya.362 Mkhas grub evokes pramåña theory here by the use of the term
mngon sum du rtogs pa, “direct perception.” His use of this term alerts us to the tension
between theory and practice articulated through a kind of tantric epistemology in the
body mandala texts. How is direct perception employed in tantric ritual practice? How do
sense perceptions and mental cognitions interact in the process of realizing oneself as a
Buddha? Is their cooperation different than it would be in other non-tantric contexts?
Furthermore, in pursuing the full implications of such questions raised by Mkhas grub’s
comments, one would have to consider how such questions play out in the conflicting
positions of the Gelukpa and Sakyapa on direct perception [mngon sum Skt. pratyakßa].
For the Gelukpa, such perception is unmediated whereas for the Sakyapa, ultimately, it is
always mediated. Dharmakîrti and his Tibetan antirealist interpreters, the Sakyapas, differ
from his realist interpreters, the Gelukpas on the nature of this mediation:
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
361 !dbang bskur ba'i tshe bskyod pa chu'i dbang bskur ba nyid kyis slob ma mi bskyod bar
gyur zin pas; da de la rin 'byung gi [237.3] dbang la sogs pas nga rgyal la sogs pa'i dri ma
sbyang dgos pa 'gal zhing; rang nyid rdo rje 'chang du bsgom pa'i tshe rang gi lus mtshan
dpes brgyan pa dang; sems kyis chos thams cad mngon sum du rtogs pa gang yang med
par myong bas grub [237.4] bzhin du de'i tshe rang nyid sangs rgyas su khas 'che ba ni
mngon sum la smyon par byed pa yin la
mngon pa'i nga rgyal ma yin na ni rang nyid la de'i tshe stobs sogs kyi yon tan gang yang
med par shes bzhin du; rang nyid de'i tshe sangs rgyas su zhes bas [237.5] smras bas
pham pa bskyed par byed pa gdon mi za bas; grub mtha' ngan bas dbyen bcos pas bag
med par ma byed cig
!
362
This passage raises the question of what the transformative power of consecration
really achieves. In other words, if one were really already transformed into a Buddha like
Akßobhya, what need would there be to become Ratnasaµbhava? If one is truly
transformed, what is the relationship between these transformations? Moreover, how do
individual phases of consecration inculcate the practitioner in cultivating generation and
completion stages, respectively?
!
! 115!
“Dharmakîrti and his Sa-gya interpreters hold a position we will describe...as
representationalism, which postulates that awareness is directly in contact with only
representations, what Dharmakîrti call reflections or aspects. By contrast Kay-drup holds
a direct realist view, according to which mental episodes are in direct contact with
objects. Hence, the reflection of a thing in consciousness is not a representation but the
revelation of that thing itself.”363
While the Gelukpa don’t go so far as to say that we can (conventionally speaking)
see things as they truly are, there are certain aspects of our powers of perception that
Mkhas grub’s philosophical position requires him to salvage and even prioritize. These
notions of the link between conception and language bring our attention to the
controversial status of ‘representation’ for Tibetan thinkers, a theme to which we will
return in the conclusion of the dissertation.
Davidson 1999 addresses the use of pramåña terminology in the writings of
medieval Indian authors on meditative, and in particular, esoteric practice. Davidson
describes a “definite proclivity on the part of Buddhist meditative theoreticians to employ
philosophical and doctrinal terminology for their own purposes.” 364 Davidson associates
this tendency with the “institutionalization of esoteric Buddhism, a purpose well served
by this language.”365 Although such texts may be the province of a different time and
place, they are the inheritance of Tibetan scholar monastics like Mkhas grub. Therefore,
Davidson’s study of the Indian precedent for incorporating pramåña discourse into the
tantric context attunes us to the possibility that Mkhas grub may have followed the lead
of his Indian predecessors. In other words, he may have appropriated pramåña discourse
to confer prestige upon certain aspects and interpretations of tantric practice over others.
For the time being we will forego further investigation of the role of pramåña in
Mkhas grub’s account of tantric ritual. However, we will remain attentive to the ways in
which he treats perception and cognition in his discussions of visualization practice. In
particular, we will take note of occasions in which such references challenge or impart
validity upon the tantric project. Moreover, later in this chapter, we will consider how
Mkhas grub depicts perception as essential for substantiating the basis for body mandala
practice.
Mkhas grub’s use of another term in the passage cited above draws even more
explicitly upon what Davidson might call the “institutional” domain of Buddhism. The
tern mngon pa'i nga rgyal refers to the false conceit of oneself as an enlightened being.
Namely, if you are aware that you are not actually a Buddha, but still claim to be one,
you are lying. The focus shifts from faulty perception to flawed action. To claim false
accomplishments is to commit one the four major transgressions or “defeats” [pham]
outlined in the vinaya. Many thinkers consider the reconciliation of tantric practice with
the principles of the vinaya to be one of the major contributions of Tsong kha pa’s
monastic reform. Moreover, the clash of tantra and the vinaya characterizes much of the
early polemics of the Gsar ma era in Tibet. In this passage, Mkhas grub forces his
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
363
Dreyfus 1997, p.252. For more on the role of representations in Dharmakîrti’s
thought, see Dreyfus, p.220.
364 !Davidson 1999, p.26.!
365 !Davidson 1999, p.35!
! 116!
opponent into a position of inevitably violating the rules of either Buddhist epistemology
or of the vinaya, two charged levels of discourse, with the latter, in particular, invested
with inalienable authority.
ID. Buddha nature and Buddha bodies
Yet another level of Buddhist discourse generating polemical writings in fourteenth and
fifteenth-century Tibet concerns the questions raised by Buddha nature, loosely defined
as the potential possessed by beings to realize Buddhahood. Much of Mkhas grub’s
description of the positions of others and the attendant category confusions presented in
the juxtapsition of human body and mandala touches upon many of the same points and
utilizes many of the same formal arguments as found within contemporary writings on
Buddha nature. The evolution of the gzhan stong view at the hands of Dolpopa (1292-
1361), trained in the Sakya tradition but identified as a Jonangpa, in particular, is
regarded as having sparked controversies. Mather describes the Jonangpa gzhan stong
position as follows: “The Jonang tradition of zhentong Madhyamaka asserts a truly
existing ultimate that is endowed with all Buddha qualities and thus not “empty of an
own-being” (rang stong), but “empty of other” (gzhan stong) nonexisting stains.” 366 The
ontological implication of such true existence offended many, both in challenging classic
Madhyamaka fundamentals on emptiness and in resembling non-Buddhist heretical
views. Dolpopa explains his position on existence as follows: “The dharmakåya is free
from mental fabrications throughout beginningless time. Because of recognizing it as
being free from mental fabrications, it is truly established.”367 The term Mathes
translates as “mental fabrications” here is spros ma [Skt. prapa•ca], not bcos ma, as
found in Mkhas grub’s texts, but the connotations of the terms are similar; often taken as
“elaboration,” spros ma refers to the tendency of our minds to proliferate ideas and
concepts that obstruct our ability to experience reality accurately.
Salient issues within buddha nature discourse include the contested status of
Buddhahood as immanent or transcendent and the relationship of Buddha bodies to the
mindstream of sentient beings. The broader implications of Buddha nature theories for
understanding the relationship of the two truths and even the paths of påramitånaya and
mantranaya were also at stake. Buddha nature debates called into question the role of
perception and cognition in actualizing one’s potential to be liberated. These debates also
investigated the basis for Buddhist practice and it’s relation to the stains of karma and
kle!a.368 Within this particular aspect of the discourse, the notion of transforming the
basis of practice is introduced. These discussions of transformation pose an interesting
point for comparison with the varieties of transformation effected through tantric practice
in ritual acts of identitfying with the divine like body mandala practice.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
366 !Mathes 2008, p.45.!
367
Translation by Mathes 2008 p.45. Dol po pa Shes rab rgyal msthan. (1292-1361) The
Ocean of Definitive Meaning of Mountain Dharma. ( Jo nang ) Ri chos nges don rgya
mtsho. Beijing: Mi rigs dpe skrun khang, 1998.: 446.26-447.1
chos sku de ni gdod nas spros dang bral; spros dang bral ngo shes pas bden par grub
368
See Mathes 2008, pp.25-129!
! 117!
Concerns with providing a causal link between conventional and ultimate realities
in terms of the Buddha bodies factored into Buddha nature debates in a manner
comparable to Mkhas grub’s tantric context. Mkhas grub follows his vitriolic accusations
of the practitioner who declares himself to be a Buddha to be either an idiot or a liar with
the following direct reference to the Buddha nature problem: “Anyone who makes claims
like this, asserting that the continua of sentient beings possess a stable and permanent
svåbhåvikakåya together with all the qualities of results of separating (from obscurations)
is an intolerable proponent.”369 In debates surrounding Buddha nature, parsing the
qualities of different Buddha bodies provided a method for coping with tensions between
the categories of the natural and the fabricated, that which is inherently present and that
which must be actualized or acted upon. Dolpopa parses these qualities as follows:
“For example, in the same way as the inexhaustible treasure underground is naturally
present, not newly brought about by effort, and the tree with its fruits gradually grows in
the garden by having brought about [the necessary conditions] with effort, the Buddha
potential, which has the ability to bring forward the three kåyas, should be known to be
twofold as well. It is both the natural potential, [namely] the pure dharmadhåtu, which is
closely present as the nature of mind through beginningless time, and the fortified
potential, [which is] supreme in terms of virtues and conducive to liberation. [The
fortified potential] arises from [virtuous deeds] being newly acquired by effort, [namely
by] something being done, such as focusing on the naturally present potential] and
studying.
As to how the three kåyas are attained, it is [here] maintained that the fruit,
[namely] the three kåyas of the perfect Buddha, are attained owing to a cause, [namely]
these two naturally present and fortified potentials. First, the naturally present potential
is perfected through many accumulations of wisdom, and becomes free from all
adventitious stains, and.. the svåbhåvikakåya, the dharmatå endowed with both purities, is
thereby attained. Second, the accumulation of merit is perfected by increasing the
fortified potential, and the latter kåyas, namely the sambhogakåya and the nirmåñakåya,
which appear to disciples near and far, are thereby attained.”370
In this account of the development of the Buddha bodies by Dolpopa, concerns

with identifying a particular notion of causality attuned to the subtleties of differentiating
inherent and acquired elements and with defining the role of practice in bringing about
the result are prominent. Mkhas grub deals with these same issues through asserting the
distinction of the fabricated and unfabricated [bcos ma bcos kyi khyad par 'byed pa]:
If that is the case, many things could be said (such as):
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
369
'di ltar 'dod bzhin du gang dag sems can gyi rgyud la ngo bo nyid sku rtag brtan bral
'bras kyi yon tan thams cad dang ldan pa yod [237.6] par khas len pa la mi bzod bar smra
ba ni
!
370 !Translation by Mathes 2008, pp.83-84; Dolpopa; Nyi ma’i ‘od zer 987.3-.6. “Theg pa
chen po rgyud bla ma’i bstan bcos legs bshad nyi ma’I ‘od zer.” The ‘Dzam-thang
Edition of the Collected Works of Kun mkhyen Dol-po-pa Shes-rab rgyal-mtshan, vol.4
(ma), 883-1161. Delhi: Shedrup Books, 1992. !
! 118!
“For us, considering the suchness of the bodies of sentient beings and the suchness of the
deities of the mandala to be inseparable in nature, all sentient beings are asserted to be
mandalas primordially.’
If you say that the mandala of the Buddha is not actually accepted (to be primordially
existent in sentient beings), well in that case, (one might reply), by that reasoning, the
distinction of that which is fabricated and that which is unfabricated is illogical just like
the suchness of painted powders and cloth paintings being inseparable in nature from the
suchness of deities.”371
Bentor 2006 notes that although the issue of tathågatagarbha isn’t “central” for
Mkhas grub, it is one of the sectarian claims emerging in the present text. Through
reference to Ruegg ’68 and (Lessing &) Wayman ’68, Bentor clarifies Mkhas grub’s
position on the svåbhåvikakåya, namely that it is not found in the continua of all sentient
beings and cannot be equated with Buddha nature, counter the Jonangpa position.372 Bu
ston, on the other hand, equates the svåbhåvikakåya with Buddha nature but denies its
inclusion in the mindstream of all sentient beings.373
In his Rgyud sde spyi rnam, Mkhas grub gives attention to sütra teachings on
tathågatagarbha theory and explicitly differs both from the Jonangpa as well as from
certain aspects of Bu ston’s writings.374 Bentor also brings attention to the fact that in
that text Mkhas grub is explicit in identifying the objects of his critiques. The context in
which Mkhas grub introduces these positions in that text is a presentation of the three
turnings of the wheel of the Buddha’s teachings. The interpretation of the schema of the
three turnings of the wheel provided a forum for Tibetan thinkers to compare and stratify
multiple levels of Buddhist discourse. Where to fit the sütras dealing with Buddha nature
and how to rank them on the spectrum of definitive and provisional teachings were key
elements of such interpretations. Both Mkhas grub and Bu ston, counter the Jonangpas,
regarded the second turning of the wheel, classified by Mkhas grub as the teachings on
emptiness and the unity of all three vehicles, as the sole definitive teaching. The
Jonangpas, according to Mkhas grub, regard both the first and second turnings of the
wheel as provisional, taking only the final turning in which they include the
tathågatagarbha sutras, to be definitive.375
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
371 !gal te mang du smras mod; kho bo cag ni 'gro ba'i lus kyi de kho na nyid dang dkyil
'khor gyi lha'i de kho na nyid rang bzhin dbyer med pa la bsams nas 'gro ba thams cad
gdod ma nas dkyil 'khor du grub bar smra'o; sangs rgyas kyi dkyil 'khor dngos su khas mi
len [238.2] no zhes zer na;'o na de'i sgo nas bcos ma bcos kyi khyad par 'byed pa mi rigs
te; rdul tshon dang ras bris kyi de kho na nyid kyang lha'i de kho na nyid dang rang bzhin
dbyer med ba mtshungs pa'y phyir ro
!
372 Bentor 2006 !p195-7, Fn33; Ruegg 1968: 504-5 & Wayman 1968: 50-51.
Ruegg, David Seyfort. 1968/9. La théorie du tathågatagarbha et du gotra; études sur la

sotériologie et la gnoséologie du bouddhisme. Paris: École Française d'Extrême-Orient.!
373 See Bentor 2006, fn39 and Wayman 1968, p.48-49. !
374
See Bentor 2006 and Ruegg 1968, fn 501.
375
Wayman 1968, p49-51.
! 119!
While it is likely that the Jonangpa are indeed being targeted here in the chapter
on body mandala of Mkhas grub’s Ocean of Attainment , there may also be a critique
of the Sakyapa position implied as well. Such a critique may be employing a sarcastic
note, to the effect of: ‘you say that your theory is different from that of the Jonangpas, but
it isn’t.’376 Such a reading brings attention to the relation of the Jonangpas as a newly
evolving institution to the Sakyapas, from whom they descended formally but
distinguished themselves doctrinally. Cabezon 2007 speculates:“...both the Dga’ ldan pas
and Jo nang pas were attempting to create identities for themselves apophatically- by
distinguishing themselves from their rivals, and among these rivals were the Sa skya
pas.”377 Buddha nature was among the hotly debated topics in these encounters, and such
debates reflect a willingness to juxtapose multiple fields of Buddhist discourse in a
manner matching Mkhas grub’s discussion of the human body: mandala relationship.
Discussions of both Buddha nature and body mandala pivot upon a carefully
articulated distinction of the relationship of enlightened and ordinary bodies and
perceptions. One might even consider how polemical writings on body mandala transfer
many of the issues addressed in Buddha nature debates in terms of the mindstream and
the ålayavij•åna onto a new locus, the human body.378 One might further compare
descriptions of the ålayavij•åna and the human body to ask: what about the socio-
political and intellectual climate of fifteenth-century Tibet may have contributed to a
heightened interest in problems of embodiment?
In describing the eleventh and twelfth centuries, Frances Garrett encourages
connections between tantric, medical, and geographic traditions in forming
understandings of the body: “Geographic and geometric conceptualization-whether for
individuals, texts, or internal organs- was thus an important form of self-identification
and validation, and arguably it was often the most significant factor enabling a Tibetan
notion of history in which history itself serves as a validation of the present.”379
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
376
Personal communication, Khenpo Choying Dorje, Spring 2011
377
Cabezon 2007, p.43. Cabezon, likewise, highlights that these exchanges were
multidirectional, including Jonangpa critiques of Tsong kha pa and so forth (fn207).
Tsong kha pa’s Sakyapa teacher, Ren mda’ ba, refuted the Jonangs on many issues
beyond Buddha nature, including aspects of interpreting the Kålacakra system. (Personal
communication, Khenpo Choying Dorje, Spring 2011).
378
For a thorough exploration of the early Buddhist evolution of attitudes toward
embodiment in the Indian context, see Radich, Michael David. 2007. ‘The somatics of
liberation: Ideas about embodiment in Buddhism from its origins to the fifth century
C.E.’ PhD dissertation, Harvard University.
Radich makes some interesting observations using a rich array of source
materials. For example, he comments, “Early Mahåyåna texts are more central than has
been recognized to the rise of new, positive ideas of embodiments, which are connected
with a broader Mahåyåna shift towards greater emphasis on immanence.” On the relation
of the ålayavij•åna and the å!raya, see Radich, p. 1165. On å!raya-paråv®tti and å!raya-
pari!uddhi, see 1109-1188.
379
Garrett 2004, p.236. It should be noted that this statement, found in the “Concluding
Digressions” of Garrett’s dissertation does not reflect the scope of the work as a whole.
! 120!
Garrett’s approach suggests a willingness to combine the resources of multiple varieties
of Tibetan discourses on the body that is lacking in contemporary scholarship.380 The
conjunction of mantranaya and påramitånaya perspectives on the body within fifteenth-
century polemical discourses is one potential avenue for investigating such connections.
For the time being, it is sufficient to acknowledge the compulsion to account for the role
of the Buddhist path in actualizing the full extent of human potential common to
polemics on Buddha nature and on body mandala.
IE. ‘Grinding the axe’: Mkhas grub’s approach to sütra and tantra
van der Kuijp’s research on Tsong kha pa and Mkhas grub helps to relate Mkhas grub’s
presentation of the relationship of påramitånaya and mantranaya to the larger fifteenth-
century Tibetan intellectual climate. In his review of Thurman’s dissertation on Tsong
kha pa’s Legs bshad snying po, van der Kuijp brings attention to an unsubstantiated claim
by Thurman that Ren mda’ ba disapproved of Tsong kha pa’s enthusiasm for tantra. Van
der Kuijp traces this claim to Mkhas grub’s biographical writings on Tsong kha pa and
critiques its accuracy381:
“...Mkhas-grub-rje states that at that time, Tibetans were generally either involved in
mdo, or in sngags studies and exegesis, and that there existed a considerable rivalry
among them, with one faction belittling the other. While there is some, albeit meager,
tangible evidence for several fourteenth century Tibetan scholars who may have held this
view- such can perhaps be gleaned from a number of remarks found in Sgra-tshad-pa
Rin-chen rnam-rgyal’s (1318-1388) commentary on Bu-ston’s tathågatagarbha treatise in
which he explicitly states that categories should not be mixed- this state of affairs was by
no means as prevalent as Mkhas grub would want us to believe. A mere glance at the
oeuvre of the most famous masters of the fourteenth century would indeed strongly
testify that in fact the opposite was the case. Future research may very well establish my
hunch that, with this assessment, Mkhas grub was grinding his own axe.”382
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
The dissertation and resulting monograph are more specifically focused upon the
encounter of tantric and medical traditions.
380
Willa Blythe Miller’s 2013 dissertation (though not as broadly situated as the
approach suggested in this statement by Garrett) makes an important contribution to
better understanding fourteenth-century conceptions of the body in Tibet through the
juxtaposition of Buddhist discourses. I look forward to engaging more deeply with
Miller’s work in the future, and specifically, to exploring how her claims compare with
my own findings for the fifteenth-century context. See Miller, Willa Blythe. 2013.
‘Secrets of the Vajra Body: Dngos po'i gnas lugs and the Apotheosis of the Body in the
work of Rgyal ba Yang dgon pa.’ Ph.D. dissertation. Harvard University.
381
See van der Kuijp 1982, p.47.
382
van der Kuijp 1982, p.50; see Thurman 1972, p. 69.
! 121!
Van der Kuijp points to the importance of confirming a more specific date for
Mkhas grub’s biography of Tsong kha pa than the colophon provides. Based upon the
colophon’s statement that the text was written in Nyang stod in Gtsang, van der Kuijp
deduces that it was composed after Tsong kha pa’s death in 1419 and before Mkhas
grub’s tenure as throneholder at dga’ ‘ldan beginning in 1431.He speculates further that it
may have been written in the midst of the friction with prominent Sakyapas. 383 It is
important to note that the colophon of the text that is the focus of our study, the Ocean
of Attainment , also indicates Nyang stod as the site of composition. Therefore it is
possible that both texts were written during a period when Mkhas grub was “grinding his
own axe,” emphasizing the distinction of paths of sutra and tantra rather than their
complementarity. On the one hand, Mkhas grub may have been picking fights, in a
manner of speaking, to bring attention to the claims of his tradition over and against that
of others. He may have been creating an artificial conflict between these two genres of
Buddhist learning that he might skillfully resolve it for his audience.384
The gradual solidification of a “Gelukpa philosophical stance” based in
Madhyamaka values may have influenced Mkhas grub’s approach to the tantric path
here. For example, the relationship between the two truths, the importance of working
towards enlightenment from within the boundaries of conventional reality, the definition
of existence, and the status of the flawed tendencies of our cognitive impulses are issues
of the Madhyamaka that take on a different flavor in the tantric context. They are key
issues that must be accounted for if the two paths are to be reconciled.
II. Mkhas grub’s own perspective
We have been able to make some useful observations about Mkhas grub’s thought based
upon his descriptions of others’ views, i.e. those of his opponents who ‘misunderstand’
the relation of sentient beings to mandalas. This situation attests to the very nature of
Buddhist debate, in which one clarifies one’s own ideas and demonstrates their logical
verifiability in relation to the positions of others.
The second part of Mkhas grub’s text is specifically devoted to presenting and
validating his own views. It is divided according to the subjects of the body mandalas of
support and supported and of generating deities from seed syllables located on specific
bodily sites. The mandala of the support is the generation of the body as the celestial
palace; the mandala of the supported is the generation of deities therein. The discussion
of the generation of deities from seed syllables is here distinguished from nyåsa, the
placement of deities upon the body through mantra. Nyåsa is among the repertoire of
practices referred to in Chapter Two of the dissertation that resemble body mandala,
practices often classified as preparatory and/or purificatory in nature. We will discover
that the nature of the body as the basis of ritual practice is of vital importance to Mkhas
grub in distinguishing body mandala from such practices. However, the category of
“fabrication” may be the most important element for Mkhas grub in distinguishing body
mandala practice from other tantric ritual technologies. We have addressed the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
383
van der Kuijp 1982, fn8.
384
Catherine Bell 1992 has brought attention to the tendency of many Western scholars
to create such artificial conflicts in their studies of ritual.!
! 122!
multivalence of “fabrication” in the preceding section of this chapter, observing
connections with the discourses of Buddha nature and Buddhist logic and epistemology.
Here we will witness how Mkhas grub applies “fabrication” not only to describe the
ontological status of the body mandala, but also, to describe disputed aspects of the
mechanics of the practice. In the process, we will continue to develop a portrait of Mkhas
grub’s commentarial style, strategy in argument, and authorial voice.
IIA. The mandalas of support and supported

We will first deal with Mkhas grub’s views on the body mandalas of support and
supported as presented in an excerpt from Ocean of Attainment [250.5-253.4]. First,
Mkhas grub distinguishes body mandala from nyåsa:
“The point of what’s called cultivating the body mandala doesn’t mean only merely
cultivating a deity on each place on the body. Establishing each respective part of the
body as the foundation [bsgrub pa'i gzhir byas nas] for each deity [250.6] means
cultivating the deity. Otherwise, even from the lower sections of tantra, many body
mandala cultivations would be explained.”385
For Mkhas grub, unlike nyåsa, body mandala practice involves more than simply placing
deities on the body or imprinting them there through imagination. What does it mean to
take the body as a foundation for practice [bsgrub pa'i gzhir byas nas]? How precisely
does this act distinguish body mandala from ritual technologies like nyåsa? Moreover,
what kind of foundation is the body in this instance? These are important questions,
questions we will keep in mind as we work slowly through the details’s of Mkhas grub’s
account of body mandala as a unique and ultimately, superior form of tantric ritual
practice.
As for the reference to the lower tantric classes in this passage, biographical
sources as well as secondary scholarship suggest that Mkhas grub and Ngor chen clashed
on topics surrounding cultivating deities in practices associated with the lower tantric
classes; this ontroversy was addressed briefly in Chapter One of this dissertation.
However, it is important to recognize that classifying the varieties of deity cultivation in
relationship to other tantric practices was controversial for these fourteenth and fifteenth-
century scholars. Such controversies may indeed have impacted their views on body
mandala.
In this section of his chapter, Mkhas grub addresses the category of “fabrication”
as it relates to the basis for establishing a mandala and uses this category to differentiate
the body mandala from “outer mandala” [phyi'i dkyil 'khor] like paintings or altars made
of painted powders:
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
385
lus dkyil bsgom pa zhes bya ba'i don ni lus kyi gnas so sor lha bsgom pa tsam mi zer
gyi ; lus kyi cha de dang de lha de dang de [250.6] bsgrub pa'i gzhir byas nas lha bsgom
pa la zer ba yin te; gzhan du na rgyud sde 'og ma rnams las kyang lus dkyil bsgom pa du
ma zhig bshad par 'gyur ro
! 123!
“Moreover, painted powder and cloth and so forth are the basis of establishing [gang las
bsgrub pa'i gzhi] the celestial palace in the outer mandala [phyi'i dkyil 'khor]; the seed
syllable, symbol and so forth are the basis of establishing the deity. Since the mandala
has been newly fabricated by colors and by artist (i.e. the causes), it is called the
fabricated mandala. As for the body mandala, the particular parts of the body mandala
have not been newly fabricated. (Rather) being perfected since the time of generation
from the mother and father, (they) created the basis for establishing (the body mandala)
[bsgrub gzhir byas]. Since the (body) mandala is established from those, it is taught as
the unfabricated mandala.”386
In evaluating the nature of the basis of establishment [bsgrub pa'i gzhi] and
whether it is inherent or newly produced through causes and conditions, Mkas grub
defines body mandala as “unfabricated” in contradistinction to other mandala. The
rhetoric of the naturally present vs. the produced is familiar from the discussions of
Buddha nature described above; it also resonates with more general Buddhist theories of
cause and effect. As for the latter, on the most basic level, that which is conditioned is a
product of saµsåra, produced by and productive of karma and therefore, in a sense,
inferior. Deconstructing an entity in terms of its parts as well as the compounded causes
and conditions that have contributed to its formation was a core Buddhist technique. This
technique was most commonly used to break down the conventional illusion a self or of
the true ontological existence of an entity. It can be found throughout early Buddhist
literature and practice, as exemplified by some of the practices described in Chapter Two
of this dissertation, as well as in later developments in Buddhist philosophical pedagogy.
To show that an entity is conditioned is demonstrate its ties to defilement and its ultimate
impermanence.
In the context of Buddha nature debates, asserting the presence of a permanent
and unchanging Buddha nature or an enduring dharmakåya posed serious problems for
many thinkers. Such trouble appears to be at the heart of Mkhas grub’s critique of the
idea that sentient beings are mandalas in any primordial way. Perhaps it is a
consciousness of such ontological issues posed by the body mandala that inspires Mkhas
grub to distinguish it as not “newly fabricated.” He stills maintains a basic causal model:
the union of the fluids of father and mother provided the basis for personhood in
cooperation with karma and kle!a and, in turn, form the basis for establishing [bsgrub
gzhir] the body mandala. In qualifiying fabrication in term of its ‘newness,’ Mkhas grub
presents fabrication as a spectrum rather than a duality. The bodily basis must, of course,
be produced by causes. To say that it is primordially a mandala in any definitive sense
would be to deny this. However, Mkhas grub is able to maintains that the body is a
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
386
de yang rdul tshon dang ras bris la sogs pa'i phyi'i dkyil 'khor la ni gzhal yas khang
[251.1] gang las bsgrub pa'i gzhi ; dang lha gang las bsgrub pa'i gzhi sa bon phyag
mtshan la sogs pa tham cad tshon dang ri mo mkhan la sogs pas gsar du bcos nas
de dag las dkyil 'khor bsgrub pa yin pas bcos ma'i [251.2] dkyil 'khor zhes pa dang
lus dkyil la ni; lus dkyil cha de dang de gsar du ma bcos par pha ma las skyes pa'i dus nas
rdzogs par grub yod pa dag bsgrub gzhir byas nas; de dag las dkyil 'khor bsgrub pas ma
bcos pa'i dkyil 'khor zhes [251.3] gsungs pa yin no
!
! 124!
different kind of basis for practice by distinguishing it from “newly fabricated” mandala
such as a mandala painting.
Mkhas grub juxtaposes the categories of intrinsic and conditioned, fabricated and
unfabricated in a variety of ways. In the following example, he applies the tools of
Buddhist linguistic theory and logic to solidify his argument. Returning to further
expound upon the correct interpretation of the Ghantapa quotation introduced at the
beginning of the chapter, Mkhas grub asserts:
“Therefore, since those foundations for establishing [bsgrub pa'i gzhi ] the mandala are
intrinsic [rang chas su yod pa] as soon as one’s body is formed:
‘These beings are not separate from the naturally established mandala.’
So it is said. The basis of establishing the mandala is like using the verbal convention [tha
snyad] ‘mandala,’ before it (a mandala) is drawn with painted powders and so forth.”387
Mkhas grub appears to strike a balance, capitalizing upon the positive

connotations of rang chas su yod pa and its association with ultimate reality while also
tempering any ontological consequences of permanence and true, unchanging existence.
He does this with the help of the notion of “conventional designations” [tha snyad ].
Buddhist linguistic theory such as that based within the tradition of Dharmakîrti
emphasizes the conventional aspect of language. This view of language distinguishes
Buddhists from Hindu schools of thought like the Mîmåµså that posited a true existential
connection between words and referents.388 Buddhist thinkers like Dharmakîrti treat
language as an intermediary in the thought process: “Thought, then, conceives its object
through an object designated by words.”389
Mkhas grub connects the Ghantapa quote and this notion of “conventional
designation” to illustrate how the body is not a mandala from the beginning, but just the
basis. In other words, ‘we just call it mandala,’ for the sake of making communication
about it easier within the parameters of conventional discourse. Therefore, one should not
superimpose ideas about it’s ultimate primordial status based upon such statements.390 It
would be interesting to know more about how and why Mkhas grub employs linguistic
theory in the tantric context.391 For example, what does a comparison of the basis of
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
387
de bas na dkyil 'khor gang las bsgrub pa'i gzhi de dag rang gi lus grub tsam nas rang
chas su yod pas’ 'gro ba 'di dag rang bzhin ni gyis ; grub pa'i dkyil 'khor gnyis med pa'o
zhes gsungs pa yin te; dkyil 'khor [251.4] gang las bsgrub pa'i gzhi rdul tshon sogs kyis
bris pa dag la ma bsgrub pa'i gong nas dkyil 'khor zhes pa'i tha snyad byed pa dang 'dra'o
rang bzhi has been emended to rang bzhin as it appears in 234.1 where the same quote
by Ghantapa appears with the variation of sgrub vs. grub. Thanks to Khenpo Yeshe for
noting this discrepancy.
388
Dreyfus 1997, p.222
389
Dreyfus 1997, p.222
390
I am indebted to Khenpo Choying Dorje for his help in interpreting this passage.
Personal Communication, Spring 2011
391 !Mkhas grub refers to Buddhist linguistic theory again, later in the chapter, in a
discussion of the mandala of the support that seems to based on the Vajramålå. Mkhas
! 125!
naming things and the basis of generating deities reveal? However, for the purposes of
the current chapter, the connection between language, conceptualization, and
superimposition is sufficient.392
As explored above in the section on pramåña, conceptual acts of superimposition
or imputation are varieties of “fabrication” that hold a negative valence in the Buddhist
context. The fact that the imaginative acts of sådhana practice are fabrications of the
mind presents a problem for Mkhas grub:
Having established, when cultivating as a deity, regarding both, the outer and the body
mandala, there is no distinction in terms of what is fabricated by the mind [blos bcos pa]
and what is imputed by the mind [blos btags pa]. And therefore, in all the stages of
development from Vajra∂åkinî and the ‘drop of springtime’ [dpyid kyi thig le Skt.
vasantatilaka] and so forth, it is called the ‘yoga of fabrication’ [bcos ma'i rnal 'byor].393
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
grub anticipates doubts concerning the practice of generating the sense and mind
consciousnesses as part of the celestial palace/body correlation and responds accordingly:
“When there is the teaching of the manner of cultivating the celestial palace here, why do
you think there is the teaching of the sense and mind consciousnesses as individual parts
of the body? The meaning of the term ‘body’ is said to be a ‘conglomeration.’ The body,
a coarse vessel, is a conglomeration of the constituents of fat and so forth up until the
subtle particles; the consciousness is a conglomeration of various moments. Therefore,
there is no fault based on the definition of the term ‘body.’ Moreover, it is not pervaded
by being the basis of engaging ['jug pa'i gzhi] in being the foundation of defining [sgra
bshad pa'i gzhi] the term ’body.”
'dir lus gzhal yas khang du bsgom pa'i tshul ston pa na dbang shes dang yid shes rnams
gzhal yas khang gi cha rer bsgom [257.3] par gsungs ba ci snyam na; lus kyi sgra don
bsags ba zhas bya ba yin la; rags pa ‘bem391 bo'i lus 'di ni sha'i kham chad la sogs pa nas
rdul phrar rab kyi bar du bsags pa dang; shes pa rnams ni skad cig du ma bsags pas lus
kyi sgra [257.4] bshad pa'i gzhi yin pa'i phyir skyon medo; 'on kyang lus kyi sgra bshad
pa'i gzhi yin pa la 'jug pa'i gzhi yin pas ma khyabo
This is a complicated passage, and several aspects of it need further clarification. These
include the application and relation of the terms sgra bshad pa'i gzhi and 'jug pa'i gzhi .
In our discussion of this passage, Shabdrung Rinpoche summarized the meaning as
follows: while we can use the definition of the body for the mind (i.e. a
“conglomeration”), referring to the body as the mind doesn’t make sense.
392
See Dreyfus 1997 for more on Mkhas grub’s approach to Buddhist linguistic theory.
For a study of Buddhism and language, see Cabezón, José Ignacio. 1994. Buddhism and
language a study of Indo-Tibetan scholasticism. Albany, N.Y.: State University of New
York Press.
393
bsgrub zin nas lhar bsgom pa na phyid dang lus kyi dkyil 'khor gnyis ka la yang blos
bcos pa dang blos btags pa [251.5] tsam du khyad par med pas; rdo rje mkha' 'gro dang
dpyid kyi thig le sogs las bskyed pa'i rim pa thams cad la bcos ma'i rnal 'byor zhes gsungs
so . I have emended bsgrub bzhi to bsgrub gzhi on the basis of the numerous
appearances of that term in similar contexts within this text.
! 126!
Mkhas grub’s direct correlation between the categories of “what is fabricated by the
mind” [blos bcos pa] and what is “imputed by the mind” [blos btags pa] provides a prime
example of the relevance of accounting for Mkhas grub’s philosophical perspective in
better understanding his approach to tantric materials.
Further clarification of the ritual context referenced by the passage is required.
However, it is likely that bcos ma'i rnal 'byor refers to the generation stage of sådhana
practice.394 In our discussion of Bentor’s theories on the stages of sådhana practice in
Chapter Three, we observed the dichotomy of “similitude” vs. “true transformation” in
Gelukpa presentations of the generation vs. completion stages of the Guhysamåja.395 The
language of ‘similitudes,’ ‘fabrications’ and ‘imputations’ and the ‘contrived’ pervade
such descriptions of the generation stage. Such terms treat the generation stage as the
requisite ‘practice run’ for the actual realization we find in the completion stage.396 Tsong
kha pa uses this framework for distinguishing the two stages to cope with the notion of
fabrication in tantric practice. It is clear that Tsong kha pa, like Mkhas grub, regards
fabrication as an inferior quality. For example, Tsong kha pa extols the quality of non-
fabrication quoting the Sampu†a Tantra:
‘That with the nature of nonfabrication

Is known as wisdom...”397
Above we have discussed examples of Mkhas grub’s reservations about tantric

acts of imagination as “fabrications.” Tsong kha pa appears to neutralize the potentially
problematic status of tantric acts of imagination by relegating the qualities of fabrication
to the generation vs. completion stage and stressing their value as skillful means. In the
following passage, he associates the completion stage with the genuine, automatic
response produced by the repeated imaginings of the generation stage:
“Therefore, just as you can discard the boat when you arrive at the far bank of the river
but have to rely upon it to get you there, likewise the attainments of the natural and
uncontrived completion stage will mean the discarding of the generation stage, but to
attain them you will need the contrived generation stage. Thus, for the beginner, the
generation stage is worthy of great praise and very important. The Vajradåka Tantra
says:
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
394 !In his Fundamentals of Buddhist Tantra, Mkhas grub explains the “drop of
springtime” in the context of Cakrasaµvara body mandala practice as a drop in the heart
of the “ground Heruka” from which the main deities in union are generated. (Lessing and
Wayman 1968, p.305)
395
See Bentor 2006, p. 186 fn4.
396
Where the body mandala practice, included in body isolation [lus dben ], fits in
relation to these classificatory schema was an issue of concern for Tsong kha pa. This
concern may signal the problematic status of the body itself at the intersection of
fabricated and natural realities and ordinary and extraordinary bodies.
397
The context for the quote is a discussion of in the context of the bliss resulting from
the union of compassion and emptiness. Kilty 2013, p107.!
! 127!
‘To gain the insights of the natural yoga,
you undertake the meditations of the contrived
and perform the recitations of the contrived.
With the realization of the natural yoga,
The contrived yoga will be external,
So having realized the natural yoga,
You do not perform the contrived.
For example, you take a boat across a river,
And when you arrive, you leave the boat.
The contrived yoga is similar to this.
The activities of the mandala and so forth,
All undertaken with the contrived mind,
Are activities clarifying the external
And for the beginner are worthy of praise.
As all those siddhis are found there,
The reality of the conqueror is not known.”398
In other words, though not ultimately viable, acts of imagination are indeed necessary.
This is a familiar Buddhist mode for relating different phases of meditative practice; only
by forfeiting attachment to one level of experience may one progress to the next.
Without the make-believe buddhas of the generation stage, no actual Buddha bodies
could come about.
Both Tsong kha pa and Mkhas grub agree that the body mandala is unique in its
avoidance of fabrication. For example, in his own commentary upon the much-disputed
Ghantapa quote, Tsong kha pa remarks: “As the mandala which pierces to the pith in the
body [lus la gnad du bsnun], the body mandala which refrains from [mi bya] the two
fabrications is essential.”399 The two fabrications Tsong kha pa speaks of are mandala
paintings and altars of painted powders. What is most problematic about these
fabrications, their materiality or their role as mediators? In the conclusion of this
dissertation, we will have the opportunity to reflect upon such problem expressed in
Buddhist attitudes toward representation more specifically. In the next section of this
chapter, we will focus specifically upon the notion of “piercing to the pith” [gnad du
bsnun] in Mkhas grub’s body mandala writings and observe how it helps him to
distinguish the body mandala as superior and unfabricated. In the process, we will
consider what kind of basis that the body forms for tantric practice as well what ‘true’
bodily ‘transformation’ entails.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
398 ! Kilty 2013, p.85-6; see note 147!
399
lus la gnad du bsnun pa'i dkyil 'khor ni bcos ma gnyis la mi bya'i lus dkyil la bya dgos
te [61.3-4]
Tsong kha pa. The Jewel Treasury: The Rite of Empowerment of the Body Mandala of
Ghanatapa, the Lord of Yoga. Rnal ‘byor dbang phyug dril bu lugs bde mchog lus dkyil
gyi dbang chog rin po che’i bang mdzod. Tsong kha pa, Vol. 10 pps. 57-106 ; Toh 5327,
pp. 58.6 to 59.3. I tracked this source down based upon a reference in Bentor 2006.
! 128!
IIB. Piercing to the pith and the superior body mandala
“Piercing to the pith” has been described to me through the metaphor of an arrow hitting
its target, getting to the essential point or heart of the matter.400 It is a phrase used to
distinguish the påramitånaya, which “pierces to the pith of the mind” [sems la gnad du
bsnun], from the mantranaya, which pierces to the pith of the body [lus la gnad du
bsnun]. This distinction attests to the centrality of the body to the tantric project. In the
Sakyapa context, the profundity of practice is discerned by “...whether or not the path is
embellished with ecstasy is distinguished by [whether there is] piercing to the pith of the
mind or body.”401 Verrill elaborates: “This expression specifically refers to the
concentration meditation of Påramitå and the internal and external body yogas that are
practices in the completion process of Guhyamantra, respectively.”402 Tsong kha pa
further describes the latter as a defining factor of highest yoga tantra and its techniques
for manipulating the subtle body:
“In short, in the three lower tantras from yoga tantra downward and within the
Philosophical Vehicle, there are descriptions of many instances of uncontaminated and
non-worldly bliss achieved through meditating without error on the significance of
emptiness. Nevertheless, they do not describe the bliss of the melting bodhicitta brought
on by the blazing candali ignited by the force of the winds entering the dhüti from the
practice of penetrating the vital points of the channel cakras in the body, and therefore,
such bliss does not fulfill the criteria of being the bliss of bliss and emptiness united.”403
In the following excerpt, Mkhas grub proclaims the superiority of the body
mandala based upon the “unfabricated basis of establishment” discussed above. He then
uses the concept of “piercing to the pith of the body” to identify a crucial transition from
the repeated imaginings or mental ‘fabrications’ of the generation stage to the natural
appearances of the completion stage:
“So, if you ask, ‘why is the body mandala superior to the two fabricated external
mandala?’:
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
400
Khenpo Choying Dorje, Personal Communication, Spring 2011
I have been unsuccessful to date in locating the Sanskrit equivalent of gnad du bsnun or
in securing the Indian origins of this practice. “Piercing to the pith” may have different
meanings in the different cycles and transmissions of tantric practice. It is possible that it
is a practice that evolved in conjunction with acts of reading the yoginî tantras back into
the Guhysamåja system (as the Vajramålå does). The notion of the pith evolves in
representations of the subtle body in ritual practices like the body mandala. For example,
as discussed in Chapter Three, the Guhysamåja Tantra divides the body into regions
rather than cakras with clearly identifiable piths.
401
See Tsemo 2012 p.163, an excerpt from Notes on Ngor chen Kun dga’ bzang po by
dPal kyi rGyal mtshan [See bibliography, “NOTES”] where this statement is made in the
context of differentiating the practices of the different varieties of disciple.
402
Tsemo 2012, Fn 344.!
403 !Kilty 2013 p.105.!
! 129!
The distinction of the superior and inferior emerges based on the fabricated and
unfabricated basis of establishment. The completion stage, generated from meditation by
piercing to the pith of the body [lus la gnad du bsnun], is the main cause of establishing
the supreme siddhi. By cultivating the transformation [byin gyis brlabs pa]404 repeatedly
while generating all the current parts [da lta kyi cha thams cad] as a deities, the channels,
winds and drops of the body become workable [rung du gyur]. By piercing to the pith of
the body in meditation, the ripening of the effortless generation [bde blag tu skye ba] of
realizing the completion stage becomes supreme.”405
A key aspect of bodily transformation as described here is making the

components of the subtle body “workable” [rung du gyur]. Mkhas grub then presents
another comparison between the body mandala and outer mandala to emphasize the
limitations of the latter:
“One generates [bskyed] the painted powders and cloth and so forth as the deity (/deities)
and transforms (them). As in the case of the body mandala, the accumulation of merit and
initiation and so forth are possible to achieve. However, there is no way to generate
wisdom by meditatively piercing to the pith in colored powders and cloth.”406
Without an unfabricated basis for practice, it is not possible to pierce to the pith of the
body, to make the body “workable” as a basis for transformation. How do ritual acts of
imagination, acts of mental fabrication, transform the basis for ritual action? The final
portion of this chapter will explore the ramifications of the category of “fabrication” as
manifested in a few specific points of body mandala practice raised by Mkhas grub.
IIC. Transformation and the Ritual Context

As discussed in the early chapters of this dissertation, in examining the structure of body
mandala ritual, we encounter many practices that resemble it in some way. Such
practices include those that are Buddhist and non-Buddhist in origin. They are used to
purify the body, speech and mind as well as to transform and deify the body of the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
404
After much deliberation and consultation with learned teachers, I have chosen to
translate byin gyis brlabs pa here as “transformation” rather than the the more standard
definition of “blessing.” There have, of course been influential critiques of the language
of transformation in studies of Buddhist meditation practice by scholars such as Sharf.
See, for example, Sharf 1995, 1998 & 2005.
405 !lus la gnad du bsnun nas bsgom pa las byung ba'i rdzogs rim nyid yin la; da lta kyi cha
thams cad lhar bskyed cing byin gyis brlabs pa [252.2] yang yang goms pas lus kyi rtsa
rlung thig le rnams las rung du gyur te; lus la gnad du bsnun nas bsgom pa'i tshe rdzogs
rim gyi rtogs pa bde blag tu skye ba'i smin byed khyad par can du 'gyur la
!
406 !rdul tshon dang ras bris sogs lhar bskyed cing byin gyis [252.3] brlabs kyang
bsod nams kyi tshogs dang dbang bskur sogs kyi dgos pa bsgrub nus pa ni lus dkyil dang
'dra la; rdul tshon dang ras bris la gnad du bsnun nas bsgom pa'i stobs kyis ye shes
bskyed par byar med pa'i phyir ro
!
! 130!
practitioner. Understanding how these practices may relate to the evolution of body
mandala is a complex and intriguing avenue of inquiry for better understanding the
history of the body in tantric ritual. Mkhas grub’s discussions of the relationship of
different varieties of practices focused upon the body to body mandala demonstrate the
continued relevance of this evolution within the fifteenth-century Tibetan polemical
context. Bentor’s 2015 article introduces the possibility that they may also be the next
phase of evolution in the Guhyasamåja body mandala practice within the tradition of
Tsong kha pa’s descendants:
“Geluk scholars that followed Tsong kha pa were certainly willing to challenge, or
improve upon, the explanations of their founder on the working of the Guhyasamåja
sådhana. Their modifications seem to have resulted from the internal contradictions in
the notion of transforming the body by means of creative visualization. Their aim was to
find a way to bridge over between the limitations that general Buddhist theoretical
considerations put on the transformative power of the mind and the point of view of
meditators who were seeking a more substantial transformation than visualization alone
can provide.”407
Bentor’s observations are illuminating with regard to the themes of imagination,

fabrication, and transformation to be explored in the final portion of this chapter.
Multiple varieties of bodily practice are included within body mandala practice
and multiple phases of transformations occur; visualizations are created, dissolved and
recreated. In describing the method of generating deities from seed syllables located at
various sites on the body, Mkhas grub explains:
“...One makes each part of the aggregates, skandhas, elements and sense spheres and so
forth into the foundation for establishing the deity [lha bsgrub pa'i gzhir byas nas]. Once
they are transformed [yongs su gyur pa], they must be generated as each deity...For
example, one’s eye organ is changed into the syllable THLIM. Having been changed,
that then is visualized as changing into Kßitigarbha.[253.1] First, with your mind,
generate THLIM without a foundation. Having generated Kßitigarbha from that,
afterwards, it’s not enough to simply imagine [mos pa tsam] the inseparability of own’s
own eye organ and Kßitigarbha. Having finished arraying the deities on the body, there is
the basic meditation upon (those deities) as essentially inseparable from one’s skandhas
and so on. Even though imaginative activity [lhag mos]408 is the repeated dissolution (of
those deities) in the body [lus la bsdus pa], that itself is not like cultivating the body
mandala.”409
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
407
Bentor 2015, p.72.
408
Bentor 2006 translates lhag mos [Skt. adhimokßa] as “special visualization.”
409
The phrase lus la bsdus pa here is one of many instances of the term bsdus within the
body mandala texts. Although I have opted for “dissolution” as a translation here,
“gathering” is also appropriate.
... phung khams sky mched sogs kyi cha de dang de lha bsgrub pa'i gzhir byas nas
[252.6]; de dag yongs su gyur pa las lha de dang der skyed dgos so...
dper mtshan na rang gi mig gi dbang po yongs su gyur pa las yi ge thlim du gyur
! 131!
As for the ritual context being referenced here, there are several possibilities.
Bentor 2015 has interpreted this passage as evidence that Mkhas grub propounds a
different approach to the mechanics of the practice than Tsong kha pa. Bentor explains:
“Apparently Mkhas grub rje offers his divergent suggestion because for him a significant
transformation of all the psycho-physical elements of one’s body into the seed syllables
of the deities is crucial. For him merely visualizing that the essence of the psycho-
physical element abides in the appearance of the seed syllable, as instructed by Tsong kha
pa, in a insufficient initial step for the transformation of the impure body into the pure
divine mansion. Therefore he requires a complete transformation of each psychophysical
element into the respective seed syllable within the visualization, before the seed syllable
transforms into a deity. At the same time Mkhas grub rje never regards the creation stage
as capable of producing true transformations.”410
Whether the passage is interpreted in terms of the juxtaposition of varieties of

transformation realized through generation and completion stage practice, as a
modification of Mkhas grub’s teacher’s approach to the mechanics of body mandala, or
as an assertion of the distinction of body mandala from related preparatory practices
oriented around the body, Mkhas grub’s use of the language of imagination suggests
important subtleties. As observed above, not all imaginative activities are regarded as
equal. It is clear is that Mkhas grub wishes to distinguish ‘mere imagining’ [mos pa
tsam] from body mandala. Furthermore, he connects lhag mos (a terms that seems to be
associate with generation rather than completion stage practice), with the imaginative act
of dissolving deities into the body and distinguishes that particular activity from body
mandala proper.
Next, Mkhas grub makes a salient point about the connection between
imagination and “fabrication” in the context of practice with a consort. In doing so, he
produces a juxtaposition of the points of pramåña and tantric visualization suggested in
more general, theoretical terms by the excerpts discussed above in the first part of this
chapter:
“Therefore, afterwards, in the context of union with the mudrå (consort), if one arrays the
deities on the body of a karma mudrå, then it becomes the body mandala of the consort.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
de yongs su gyur pa las sa snying du [253.1]gyur par bsgom pa yin gyi; dang por thlim
rang gi blos gzhi med du bskyed cing; de las sa snying bskyed nas; de'i rjes su sa snying
rang gi mig dbang dang dbyer med yin par mos pa tsam gyis mi chog ste; lha lus la bkod
tshar nas rang gi phung sogs [253.2] dang ngo bor dbyer med du bsgom pa tsam ni lhag
mos lus la bsdus pa la yang yang mod kyang de nyid lus dkyil bsgom pa ma yin pa bzhin
no
410
Bentor 2015, p.70. Bentor also demonstrates how Tsong kha pa differed from later
Geluk thinkers on the mechnics of generating the body as the celestial palace; see Bentor,
pp.66-67.!
! 132!
There is the generation of seed (syllables) of the deity from that transformation of the
individual aggregates and so forth of the body. For the wisdom mudrå, (on the other
hand) one doesn’t establish (them) in actual sites that are not the mere fabrication by the
mind [blos bcos pa rtsam]. Therefore, the body that is the basis of establishment [bsgrub
bzhi'i lus] does not become the body mandala even though one arranges the deities in a
similar way.”411
In other words, actual transformation into the body mandala can only occur with an
actual human consort (i.e. the karma mudrå), not an imagined one. The wisdom mudrå,
the imagined consort, is regarded as a mere mental fabrication [blos bcos pa rtsam], and
therefore an insufficient basis for transformation. So not only is it necessary to do more
than merely ‘locate’ deities on the body, one must transform the bodily basis. And in
order for that transformation to be possible, that basis must be ‘real.’
In the final portion of his chapter, dedicated to his views on issues surrounding
the generation of deities from seed syllables, Mkhas grub solidifies the connection
between the practitioner’s own body and the generation of the body mandala. Again, the
emphasis is upon establishing an unfabricated basis for practice through reliance upon an
actual, empirically verifiable body:
“Moreover, though there are six eyes that are cultivated when cultivating oneself with
three faces and six arms, it’s not necessary to cultivate the Kßitigarbha from the right and
left faces. Cultivating (him) as the two eyes of the main face will suffice. Likewise, it’s
sufficient to arrange Vajrapåñi and Åkå!agarbha (respectively) in the two ears and two
nostrils of (just) the main face. The Vajramålå says, ‘In the two eyes is Tathågata
Kßitigarbha. In the two ears of that (one ie. the father deity) is Tathågata Vajrapåni.’ So
it is taught. Likewise, it’s fine to arrange Yamåntaka and Aparåjita on the two main
hands. Why? It is the basis of purification or the basis for establishing the deity. (As for)
those limbs of the body, if there are many, it’s necessary to generate many deities (on
them). It’s like cultivating two Kßitigarbhas on account of there being two eyes. For that
reason, it’s necessary to generate Samantabhadra within the (bodily) sites in accord with
however many joints there are in the body. In the Vajramåla (it says):’In all the joints of
the body, Samantabhadra.’412 It is taught like that.
Like generating Kßitigarbha from the right and left faces without there being an
actual entity (there), the basis for practice becomes fabricated. As a result, the manner of
generating deities of the unfabricated mandala does not happen..”413
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
411 !des na 'og tu phyag rgya dang snyom par 'jug pa'i skabs su; phyag rgya'i lus la lha 'god
pa yang; phyag [253.3] rgya de las rgya yin na de'i lus kyi phung sogs de dang de yongs
su gyur pa las sa bon dang de las lha bskyed pa yin na yum gyi lus dkyil du 'gyur ba yin
gyi; ye rgya la ni bsgrub bzhi'i lus blos bcos pa rtsam ma yin pa'i dngos gnas la grub pa
med pas; [253.4] de'i lus la lha 'god tshul ji ltar byas kyang lus dkyil du mi 'gyur ro
!
412
This quote appears to be derived from Chapter 68 of the Vajramålå. See Kittay 68.12.
413 !de yang rang zhal gsum phyag drug par bsgom pa'i tshe bsgom pa'i spyan drug yod
kyang; g.yas g.yon gyi zhal gyi spyan la sa snying [260.5] bsgom mi dgos kyi; rtsa zhal
gyi spyan gnyis la bsgom pas chog la ; de bzhin du phyag rdor dang nam snying yang rtsa
! 133!
Despite the very technical nature of this excerpt, the general point is clear. It is
not possible to achieve the body mandala just by imagining an abstract correlation
between imagined deities and imagined body parts. Even though one might imagines
oneself (or one’s consort) as a deity with multiple pairs of eyes, only the eyes that are
really there can be the basis or point of departure for imagining deities. Even mental
fabrications must be founded upon an actual physical reality to ensure efficacy. The
human body is that reality. The ground for practice must be stable and the rules of
conventional reality followed for true transformation to occur.
Mkhas grub concludes his chapter on body mandala by critiquing yet another
aspect of fabrication, the invention of spurious Tibetan body mandala practices. He
identifies such practices as “baseless mental imputations [rgyu med pa'i blos btags]
masquerading as the oral instructions and the profound dharma.”414 This climactic
concluding statement confirms our suspicions about the negative valences of the category
of fabrication, valences we have explored within multiple levels of Buddhist discourse. In
the context of questions of textual authority, fabrication is decidedly negative. Though
Tibetan authors were great innovators, innovation in and of itself was regarded as a
transgression rather than a virtue. The authorial ideal was to seamlessly transmit the
teachings of the Tibetan and in particular, the Indian masters of the tradition, without
superimposing one’s own ideas. In order to be appreciated, innovations needed to be
presented as elucidations of the teachings of the past.
Whether one is discussing fabrication in the context of Buddha nature, pramåña,
the Madhyamaka or on the more general level of determining textual authority, it is a
volatile category. It is often used to identify flawed human tendencies to superimpose
false concepts upon reality, obstructing our ability to access a clear vision of things as
they really are. Yet we have also observed the manner in which tantric acts of
imagination, in particular, the act of identifying oneself with a deity, shed new light on
fabrication, revealing tensions that lay at the very root of Buddhist practice. Such
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
zhal gyi rna ba'i nang gnyis dang sna bug gnyis bkod bas chog ste; bshad rgyud rdo rje
'phreng ba las; de mig dbang po gnyis la ni; de [260.6] bzhin gshegs pa sa snying po; de
yi rna ba'i dbang gnyis la; bde gshegs phyag na rdo rje'o; zhes gsungs pa ltar ro; de bzhin
di khro bo gshin rje gshed dang gzhan gyis mi thub pa yang rtsa phyag gnyis la bkod pas
chog go; 'di dag gi rgyu [261.1] mtshan ni; yang na sbyang gzhi'am lha'i bsgrub gzhi lus
kyi yan lag de dngos po la du mar yang na lha yang du ma bskyed dgos te; mig gnyis yod
pas sa snying gnyis bskyed pa bzhin no; de'i phyir lus la tshigs ji snyed yod pa'i grangs
[261.2] dang mthun pa'i kun bzang gnas du dang der bskyed dgos te; rgyud rdor 'phreng
las; lus kyi tshigs ni thams cad la ; de zhin gshegs pa kun tu bzang; zhes gsungs pa ltar ro;
dngos po la med pa g.yas zhal dang g.yon zhal gyi [261.3] mig las sa snying lta bu
bskyed kyang; bsgrub gzhi bcos mar song bas ma bcos pa'i dkyil 'khor gyi lha bskyed
tshul du mi 'gyur ro;
!
414
The passage, found at 263.5, reads: bod la grags pa'i lus dkyil mang po zhig la de 'dra
ba'i rnam gzhag gang yang sbyar rgyu med pa'i blos btags 'ba' zhig la man ngag dang zab
chos su byed par snang bas
! 134!
tensions resonate across multiple levels of Buddhist discourse. In this light, we begin to
see how “fabrication” assumes the status of a kind of ‘necessary evil,’ in the tantric
context. In this chapter, we have had the opportunity to explore the ways in which
Tibetan writers like Mkhas grub developed strategies for coping with the contradictions
“fabrication” presents; these means of coping are productive, telling us more about why
ritual, and more precisely, tantric ritual acts of imagination like body mandala, are
deemed necessary or efficacious.
! 135!
Chapter Five: The Body Mandala Debate: Body as Explanatory Tantra
In Chapter Three we observed the complex ways in which Mkhas grub’s manner
of engaging with his many “unknown opponents” reflects both a tendency to play to his
strengths as well as a desire to tap into the main issues characteristic of the intellectual
climate of the times. In terms of the former, he embraced the methods of Buddhist logic
and epistemology and the tone of philosophical debate for which the emerging Gelukpa
sect gained such renown. As for the latter, he grappled with the contradictions inherent
in human embodiment in the terms of the Buddha nature debates, with a particular
interest in reconciling the paths of the mantranaya with the påramîtånaya at large. In this
chapter, we turn to the body mandala debate proper, in other words the explicit encounter
of Mkhas grub and Ngor chen on the body mandala practice. As in discussed in Chapter
One of this dissertation, this is just the beginning of a debate that extends far beyond the
limits of Mkhas grub’s chapter and Ngor chen’s response, into at least two further texts
by Mkhas grub as well as into the writings of Ngor chen’s students and successors.415 In
another sense, it is not the beginning but rather a particular crystallization and
redeployment of the views of Mkhas grub and Ngor chen’s predecessors. As such, it
indicates how these two authors formulated and distinguished their traditions and
contributed to the fifteenth-century moment. Attending to the tension between such
vastness and particularity, a tension that might be said to characterize the practice of
Tibetan exegesis, we turn to the ‘body mandala debate.’
Ngor chen begins his text by citing a passage from Mkhas grub’s body mandala
chapter and then proceeds to refute the critiques presented therein. Basically, he is
responding according to the structure laid out by Mkhas grub in part two of his chapter on
body mandala; this triadic structure contains Mkhas grub’s own views on the body
mandala of ‘support’ [rten] and ‘supported’ [brten pa] as well as on generating seed
syllables on the body. Ngorchen’s response to these three topics makes up just under half
of his text. The remainder is devoted to issues of textual authority, a defense of the
particular body mandala practice critiqued by Mkhas grub through recourse to a variety
of tantric texts, commentaries, and oral instructions. Ngor chen’s defense is a testament to
his prowess as a tantric commentator extraordinaire as well as to the Hevajra
abhisamaya’s centrality to the Sakya tradition.
In this chapter we will explore the subtle dynamics of Mkhas grub’s accusations
and Ngor chen’s responses. In doing so, we will also situate these accusations within the
larger context of Mkhas grub’s text, adding to the observations on his method and style of
argumentation made in the previous chapter. Furthermore, we will begin to build a
portrait of Ngor chen based upon his response. The next chapter of the dissertation will
add further dimension to this portrait by examining the heart of Ngor chen’s defense of
the Hevajra body mandala on issues of textual authority and comparing the two versions
of Ngor chen’s body mandala text. While exploring the details of this debate, we may
reflect upon how the body mandala practice allows these two thinkers to articulate and
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
415 !For!one!example,!see!(Go rams pa) bsod nams seng ge (1429-1489). Illuminating the
Pith: Dispelling Objections to the Moonrays of the Pith. Gnad gyi zla zer la rtsod pa
spong ba gnad kyi gsal byed,!
! 136!
distinguish their views and the views of their traditions on larger issues. In the process,
we will begin to see why the body mandala debate matters. For it is through technical
details of the mechanics of visualization practice and commentarial method that identities
are concretized, authenticated, and even reinvented.
I. Ngor chen’s Introduction

As is typical of response texts within the polemical genres, Ngor chen introduces his
argument with a statement of intent, to set right the accusations of the sectarians [phyogs
‘dzin] and repudiators [skur pa ‘debs pa].416 He then paraphrases the opponent’s most
offensive claim that the oral instructions of Virüpa in the Hevajra body mandala contain
lies and fabrications [mun sprul dang rtog brtag]. What is somewhat odd is that nowhere
in Mkhas grub’s text does he explicitly refer to the Hevajra system by name. It is
completely standard practice to elide the names of one’s opponents or of their traditions,
and to cite rather liberally from texts without providing precise citations. However,
initially it seems strange that Ngor chen would construct his entire project as a recovery
of the Sakyapa Hevajra practice if Mkhas grub never explicitly referred to that tantric
cycle. In proceeding, we must therefore attend to the clues provided by Mkhas grub’s
text that the Hevajra practice is indeed at issue. In doing so, we must also consider both
the centrality of the Hevajra-based Lam ‘bras tradition to the Sakyapa tantric system and
the history of critiques waged against this system in the earlier history of Tibetan
Buddhism.
II. The Opponent’s View

Ngor chen follows standard protocol in first laying out the opponent’s objections [rtsod
pa’i ngag dgod pa] before refuting them. This entails citing from Mkhas grub’s body
mandala chapter which he identifies as the “hoarse screaming’ of the logician [rtog ge pa]
who hasn’t even engaged in Vajrayåna practice for a long time.” [Destroyer of the
Proponents of Evil (N1) 548.1] Though not necessarily indicative of the dominant
tone of Ngor chen’s writing style, comments such as these demonstrate that Ngor chen
also participates in the spirit of the polemical drama so adamantly embraced by Mkhas
grub. Ngor chen present the opponent’s objections by citing Mkhas grub’s text [see
Ocean of Attainment , 254.6-255.5]; the citation, provided below, may be broken
down into three parts. The first part describes what ‘some Tibetans’ claim as body
mandala practice in the mother tantras. The second describes the position of Bu ston on
the version of the practice described in Nag po pa’s Samvara-vyåkhå (a Cakrasaµvara-
affiliated text). The final part critiques various Tibetan versions of the body mandala
practice as inauthentic with a particular focus upon problems in relating internal and
external mandala.
This excerpt appears in Mkhas grub’s chapter on body mandala in Ocean of
Attainment early on in the section of the chapter asserting his own tradition with regard
to the mandalas of support and supported and the placement of seed syllables on the body
[Ocean of Attainment, 250.4-263.6]. The first couple folia of this portion of the text
contain the discussion of the Ghantapa quote from his “Condensed Activities of the
Cakrasaµvara Initiation” together with Mkhas grub’s distinction of fabricated and
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
416 !In!the!conventions!of!Tibetan!polemics,!see!Lopez!1996!and!Cabezon!2007.!
! 137!
unfabricated mandalas [250.4-252.2]. This is followed by a more detailed discussion
[252.5-254.5] of how this notion of fabrication plays out in the actual mechanics of the
body mandala practice. In particular, Mkhas grub describes the transformation of the
bases of aggregates and other components of the body and their subsequent generation as
deities through the use of seed syllables and consort. Mkhas grub’s treatment of these
issues is familiar from the two preceding chapters of this dissertation. Mkhas grub then
grapples with the relationship of the body mandalas of support and supported and with
the relationship of the multiple varieties of celestial palace generated and dissolved over
the course of the body mandala practice.417 In the process, he refers to two of
Guhyasamåja texts discussed in Chapter Three, the Rnam gzhag rim pa and the
Vajramålå explanatory tantra, as evidence for the necessity of generating the body as the
celestial palace (the body mandala of the support).
Understanding the different ways in which Mkhas grub and Ngor chen use the
Vajramålå will prove to be an important avenue for exploring their methods of tantric
exegesis. We first encountered this text in Chapter Three of the dissertation, where we
noted the significance of the Vajramålå as an explanatory tantra of the Årya tradition with
a strong tendency to interpret the Guhyasamåja Tantra and its sådhana practice from the
perspective of yoginî tantra. Kittay regards the text as an anthology composed of both
mahåyoga and yoginî tantra-based practices. This very quality of compiling and
combining different kinds of practices in new ways is as an important theme for
understanding the significance of the Vajramålå to the body mandala debate. This quality
makes it potentially useful to our authors but also makes it controversial. One way of
interpreting such controversy is in terms of a tension between innovative syncretism and
a conservative anxiety, a tension that plays out in terms of a variety of category
confusion.418
By observing the ways in which Ngor chen and Mkhas grub exhibit tendencies
both of syncretism and of conservatism, we obtain a glimpse of some of the larger
dynamics fueling the art of Tibetan commentary. We will explore these authors’
particular ways of using the Vajramålå in light of these dynamics in further detail later in
this chapter. However, in embarking upon a discussion of their body mandala debate, we
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
417
See folia 253.5-254.6 of Mkhas grub’s text for his interpretation. This is a complicated
and very technical section of Mkhas grub’s argument. The gist is that it is necessary to
generate the body as the celestial palace for body mandala practice. Mkhas grub’s
rationale seems to rest in the continuity between the phases of visualization. As indicated
by her 2013 IATS and 2014 Berkeley Tantra conference papers, as well as her recently
published 2015 article, Yael Bentor is dealing more extensively with this particular
aspect of Mkhas grub’s argument and how it relates to the position of his teacher, Tsong
kha pa. See Bentor 2015, pp. 66-67. I will engage briefly with this interpetation below.
However, I refer the reader to Bentor’s article as well as to her forthcoming translation of
the complete text of Mkhas grub’s Ocean of Attainments for a more in-depth treatment of
this issue.
418
I am grateful to Christian Luczanits for bringing my attention to the larger significance
of the artful navigation of issues of syncretism in the Tibetan tradition. In particular, we
discussed the centrality of such issues to reform movements like the one initiated by
Tsong kha pa and perpetuated by Mkhas grub. Personal communication 7/17/2014
! 138!
can begin to see how this tension between syncretism and conservatism pervades the
interpretation of both bodies and texts. We have already introduced the problem of
category confusion relating to the tension between ordinary and samsåric bodies in the
previous chapter. The category confusion we encounter here, on the other hand, pertains
to the classification of texts, broadly conceived as “doxography.” Establishing which
texts may be brought to bear upon others, to support or confound the authenticity of a
given interpretation, is a vital part of both polemical and exegetical processes. Authors
typically draw upon established systems for classifying texts. Texts may, for example, be
grouped according the deity they focus upon, the philosophical view they express, or who
authored them: whether they are regarded as the word of the Buddha, the work of an
Indian scholar or accomplished practitioner, or the work of a Tibetan master. Tantric
texts, in particular, often defy clear classification and inhabit grey areas. Davidson has
coined the term “grey text” to refer to tantric texts regarded as the work of Indian masters
in collaboration with their Tibetan disciples and translators.419 Some of these “grey texts”
were transmitted orally from master to disciple for generations before being written
down. Tantric teachings focused on a particular deity like Cakrasaµvara or Hevajra,
deities whose qualities and worship may take a very particular form (potentially sexual or
violent in nature), may be further stratified on a scale of esotericism and profundity.
Only more advanced practitioners may be deemed capable to engage with the more
‘profound’ texts and their associated ritual practices.
Mkhas grub’s final remark before the contested section of his text cited and
disputed by Ngor chen reflects a concern with preserving a category distinction between
tantric cycles:
“Therefore, this is how it is for the manner of establishing the body mandala in accord
with the tantra pi†aka and the texts of the mahåsiddhas and the texts of the Indian
panditas. This is not the case for particular extraordinary situations such as those
pertaining to Cakrasaµvara. Terrifying with power, (that tradition) is not explained here,
but it should be understood from the great exegesis of Luipa, the discourse(s) of Rje rin
po che himself (i.e. Tsong kha pa) and so forth.”420
We have already observed the importance of the Cakrasaµvara Tantra to Mkhas

grub’s argument through the detailed exploration of the Ghantapa quote in Chapter Four.
We will continue to explore the role of that tantric textual cycle in both Ngor chen and
Mkhas grub’s arguments on body mandala over the course of the present chapter.
However, for now, we should note that, for some reason, Mkhas grub feels compelled to
introduce this system into his discussion of the Guhyasamåja body mandala while also
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
419
Davidson, 2005.!!!
420
des na rgyud sde dang grub chen gyi gzhung dang rgya gar mkhas pa gzhung dang
mthun pa'i lus dkyil bsgrub tshul ni de ltar yin la; bde mchog la sogs pa'i thun mong ma
yin pa'i khyad par dag ni skabs ma yin pa dang mngas pas [254.6] 'jigs te 'dir ma bris
gyi; rje rin po che nyid kyi gsung rab lu i pa'i rnam bshad chen mo sogs las shes bar bya
zhing
!
! 139!
attempting to distinguish the two systems.421 Moreover, unlike the Hevajra cycle, he
refers to Cakrasaµvara explicitly by name. The need to articulate these complex relations
and the nature of continuity and conflict between different textual cycles and their related
practices sets the stage for the contested section of Mkhas grub’s text, the excerpt that
sparks the body mandala debate.
As texts and bodies are repeatedly collapsed and disambiguated in new ways, the
nature of their relationships is potentially recreated. For example, in certain contexts but
not others, an author may choose to bring Cakrasµvara-related materials to bear upon the
interpretation of Guhyasamåja-related materials. Likewise, in certain instances the nature
of enlightened bodies may be brought to bear upon the experience of saµsåric
embodiment, while in others, the distinction must be preserved. Tantric ritual practice,
and particularly deity yoga, plays upon this tension between merging and separation. As
this chapter progresses, we will begin to see a correlation between the impulses to
classify bodies and texts. Furthermore, we will observe how tensions between
conservatism and syncretism, distinction and merging, exhibited in both polemical and
ritual practice, bear the potential to transform the limits of textuality and corporeality.422
IIA. The Protective Circle and the Body Mandala of the Support
Frameworks for classification are also used to articulate the relationships between
different varieties of mandala and different phases of ritual practice. In Chapter Four, we
discussed Mkhas grub’s views on the relationship of external to internal (i.e. body)
mandalas in terms of their status as fabricated and unfabricated bases for tantric practice.
We recall how Mkhas grub praised the body mandala as a superior basis on account of
this unfabricated quality. In this chapter, we will look more closely at an even more
subtle level of distinction within the body mandala practice itself, between the mandalas
of support and supported. Although the definitions of the mandala of support and
supported vary from one tantric cycle to the next, generally speaking, the mandala of the
support is the body as the celestial palace, in some cases also including the cremation
grounds and protective circle. The mandala of the supported is the deities inhabiting that
bodily palace, and in some instances, the channels and chakras together with other
psycho-physical elements like the winds and drops. The first part of the citation from
Mkhas grub’s critique describes a version of the protective circle and mandala of the
support:
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
421
In her March 2013 paper at the Berkeley Tantra conference, Yael Bentor also
questioned the relevance of Mkhas grub’s remarks on the Cakrasaµvara to the purported
focus of his text on the generation practice of the Guhyasamåja tantra. I look forward to
seeing more of Bentor’s conclusions in her forthcoming work and would add that this is
not the only aspect of Mkhas grub’s work that seems to transcend his focus. For
example, his repeated reference to the Vajramålå, a text more focused on completion
stage practice for support seems to exceed his own text’s focus the generation stage.
422 !Of course, there are moments in which Tibetan authors experiment with the
metaphorical relation between frameworks governing the classification of texts and of

bodies. A particularly illuminating example to be discussed below presents the body as
the explanatory tantra upon the root tantra that is the mind. See Stearns 2006, fn19.!
! 140!
Some Tibetans say for the many body mandalas of the mother tantras:
“The crown is the vi!va-vajra;
The soles of the feet are the vajra foundation.
The ribs are the vajra fence.
The skin is the vajra tent and canopy
The body hairs are a net of arrows.
The fingernails are the utterly blazing fire mountain.”
They also say:
“The four channels of the heart are the four gates.
The eyes are the tiered walls of the palace.
The nose is the jeweled beam423.
The teeth are the lace curtains.424
The tongue and lips are the sense pleasures.”425
This quote [254.6-255.1] forms the bulk of a section of Mkhas grub’s text cited by Ngor
chen [254.6-255.5]. As the point of direct exchange of Ngor chen and Mkhas grub and
Ngor chen’s explicit object of refutation, it is vital to our understanding of the body
mandala debate. This quote of ‘some Tibetans’ can be found in contemporary Hevajra
sådhanas used by the Sakya school.426 However, these sådhanas which are a core part of
daily practices for initiates in the Hevajra practice and central to the Sakya Lam 'bras
tradition, were formulated and compiled after Ngor chen’s time, primarily by the tenth
abbot of Ngor, Dkon mchog lhun grub (1495-1557). In looking to the works of the five
great Sakyapa patriarchs for an earlier precedent for this version of the practice, we find
that it appears both in a work attributed to Grags pa rgyal mtshan, Lus kyi dkyil ‘khor,427
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
423
pha gu can also be translated as “cornice.”
424
DM on defines dra ba dra phyed as: “Fabric temple-hanging forming a kind of
latticework of jewels and auspicious symbols which seem to be spit out of the mouth of a
kîrtimukha.” [thlib.org - Accessed 1/9/2013]
425
bod dag gis ma rgyud kyi lus dkyil mang por; sbyi ba sna tshogs rdo rje ; rkang mthil
rdo rje'i sa gzhi; rtsib ma rdo rje'i rwa pa; pags pa rdo rje'i [255.1] gur dang bla re; ba spu
mda'i dra ba; sen mo me ri rab tu 'bar ba; zhes pa dang; snying kha'i rtsa bzhi po sgo
bzhi; mig pa ‘tsa re ga'i rtsig pa; sna rin po che'i pha gu; so dra ba dra phyed; lce dang
mchu 'dod pa'i yon tan
Ngorchen N1 reads: lce dang mchu 'dod yon gyi snam bu, ‘ the corridors of the desire
goddesses.’ snam bu may also be translated as “terrace.”
426
I am grateful to Drapa Gyatso of the IBA for first bringing my attention to this fact.
[Personal communication, Spring 2011]
427
See the digital Sakya Lam ‘bras collection, Vol.10, p.140-143. I am grateful to Rory
Lindsay for his help in locating this text. This appears to be the same text referred to in
Davidson 1992 fn 26 within the collection of the Pod gser ma, or the “Yellow Book”
compilation of esoteric instructions [(?) Lus kyi dkyil 'khor, in Pod-ser-ma (Bhir:'Jam
dbyangs lung-rtogs dpal-bzang, 1970), pp. 169.3-173.4] and in Sobisch 2008 Title list
#290 [ (Lam la sogs pa’’chos nyi shu la) lus kyi dyil ‘khor, Sa skya Lam ‘bras series Vol.
11, 68r-69v].The Yellow Book was intended to transmit the esoteric teachings passed
down orally until the time of Sa chen. This text is one of twenty-three in that collection
! 141!
as well as in ‘Phags pa’s Hevajra body mandala sådhana.428 Davidson 1992 points out
that Grags pa rgyal mtshan, one of the two sons of the Sakyapa patriarch Sa chen kun
dga’ snying po, instituted a tradition of composing addendum texts to the Hevajra
sådhanas to describe the body mandala practice; this tradition was continued by 'Phags
pa, the fifth of the great five Sakyapa patriarchs.429 Davidson adds : “ 'Phags pa, however,
went one step further and, following instructions from the Pod-ser-ma, combined the
structures of the internal mandala and the reception of the consecration (abhißekha) into
one unified work.”430 In order to better understand how this quote fits into the tradition
of Sakyapa body mandala practice, we will briefly summarize and compare these two
addenda body mandala texts.
Beginning with Grags pa rgyal mtshan’s text, we find the first part of our quote
identified with the protective circle and the second with the support [rten]. (140.3) The
author then informs us that while in (fetal) gestation [lus chags tshul] the deities of the
mandala of the supported reside in the navel, in the context of meditation, they are
visualized from the heart center.431 The remainder of the short text describes the mandala
of the supported, the five cakras or mansions [pho brang] at particular sites on the body
inhabited by the respective pairs of Buddhas and consorts and their retinues of
goddesses.432 Details of their source seed syllables, appearance, attributes, purification of
particular skandhas and poisons and resulting realization of the five wisdoms enrich the
account. Then, the practitioner is instructed to recall the ten krodhas as the syllable
hum/hung and the six (deities) of senses and sense objects. The text concludes with the
spontaneous generation of the four Buddha bodies and summarizes that this is the
visualization of the support [rten], protective circle, celestial palace, and the one hundred
and fifty-seven inhabitant deities. Finally, it declares that the practitioner should
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
intended to clarify Virüpa’s teachings and Sa chen’s explanations of them. While
Davidson attributes this particular text to Grags pa rgyal mtshan, Sobisch does not
specify its authorship. Many of the twenty-three clarifying texts are indeed by Grags pa
and the inclusion of this text as one of the twenty-three within his own title list of the
work might support this attribution. On the Yellow Book, see Sobisch 2008 Chapter One
(pp85-101).
428
(‘Phags pa) Blo gros rgyl mtshan. (1235-1280). Hevajra Body Mandala Sådhana.
"Kyai rdo rje lus dkyil gyi sgrub thabs/." In Sa skya bka' 'bum. Collected writings of the
first five great patriarchs of the Sakya order: Includes the three supplementary volumes
recently published by Khenpo Tsultrim Gyaltsen, listed separately [W20751]. Reprinted
from a set of Dege Parkhang prints. TBRC W22271. 13: 538 - 542. dehra dun: sakya
center, 1992-1993.
429
Davidson 1992, p112 and fn 26.!
430 !Davidson 1992, p112 and see fn 27.!
431
For a detailed description of the formation of the body by Grags pa rgyal mtshan, see
his Rin po che’i ljon shing, pp. 117-119.
The Wish-Fulfilling Tree. Mngon par rtogs pa rin po che'i ljon shing /." In gsung 'bum
?dpe bsdur ma?/_grags pa rgyal mtshan/. TBRC W2DB4569. 1: 19 - 293. pe cin: krung
go'i bod rig pa dpe skrun khang, 2007
432
The number of retinue goddesses for each palace equals the number of channels or
channel petals of which each of the five cakras is composed.!
! 142!
maintain the four consecrations of the path through one of the three practices of extended,
middle, and condensed (sådhana) as appropriate. These are the three versions of the
Hevajra sådhana that initiates employ in daily practice based upon their level of
attainment as well as upon the time available for meditation sessions (of which optimally,
there are four).
'Phags pa’s text begins with a statement similar to the much-discussed Ghantapa
quote, that primordially the bodies of sentient beings are spontaneously produced as
undifferentiated from the mandala. He specifies that the body mandala is taught for the
purpose of actualizing this primordial status. He locates the practice within the ritual
chronology after the dissolution of the wisdom beings into the pledge beings, and then
proceeds with the order of visualization. ‘Phags pa identifies the first part of our
controversial quotation (beginning with “The crown is the vi!va-vajra...”), asscociated by
Mkhas grub as the view of “some Tibetans” on mother tantra body mandalas, with the
protective circle [Hevajra Body Mandala Sådhana 526.6-527.1]. ‘Phags pa identifies he
second part of the quotation (beginning “The four channels of the heart are the four
gates...) as the celestial palace [Hevajra Body Mandala Sådhana 527.1-.2]. Both texts
contain an additional section between these two passages; this section, not cited by
Mkhas grub, describes four portions of the body as four cosmic elements in ascending
order, and the spine and top of the head as portions of Mt. Meru. This aspect of the
practice resonates with the abhidharmic imagery of cosmic creation and destruction
discussed in the Chapter Two of the dissertation. This portion of the sådhana, while
elided by Mkhas grub, is, in fact, cited in Ngor chen’s response; therefore, it will be
discussed further below as the “elided section.” This three-fold framework of correlations
of parts of the body with parts of the protective circle, with the elements of the universe,
and with the parts of the celestial palace is common to both Grags pa rgyal mtshan and
'Phags pa’s texts. Below, we will have the opportunity consider why the component of
cosmic correlation is absent from Mkhas grub’s account.
'Phags pa then describes five cakras named dharma, emanation, great bliss,
enjoyment, and (again) great bliss (though he presents them in a different order from
Grags pa). The descriptions are similar in the two texts; here it is summarized as the
channel mandala of the one hundred and fifty-seven deities gathered by the five
mansions. 'Phags pa provides more detail on the sense goddesses and krodhas, naming
and locating them individually. He concludes the discussion of deities with a meditation
on the empty aspects of the divine forms and the cultivation of the channel body as the
nirmåñakåya (i.e.emanation) the channel letter as the sambhogakåya (i.e.enjoyment), the
elemental nectar as the dharmakåya and the heart wisdom wind as the svåbhåvikakåya.
As suggested by Davidson, the next portion of the text adds the element of consecration,
prescribing that the practitioner first imagine the deities of the body mandala together
with the outer mandala. 'Phags pa adds an explanation of how this fits in with other
versions of the Hevajra practice: “Concerning the connection with the extended outer
abhisamaya, there is no contradiction between cultivating oneself as just a single lord and
cultivating the expanded body mandala.”433 'Phags pa specifies that this method of
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
433
[Hevajra Body Mandala Sådhana 529.5-.6] 'di ni phyi'i mngon par rtogs pa rgyas ba
dang 'brel pa'i dbang du byas pa yin la; bdag bskyed dpa bo gcig btsam zhig bsgoms nas
lus kyi dkyil 'khor rgyas pa bsgoms kyang 'gal ba med do
! 143!
practice is found in the oral instructions of the lineage [brgyud pa’i man ngag]. 434 In
terms of the systems of classification discussed above, therefore, this text inhabits what
some interpreters regard as a grey area, with its authorship hovering somewhere between
Indian and Tibetan. As a result, the practices described within the text might be deemed,
by some, to be of questionable authenticity.
In comparing these two works, we have established a ritual context for the version
of body mandala problematized by Mkhas grub and located it firmly within the Sakya
Hevajra and Lam ‘bras tradition transmitted by the early Sakya patriarchs. In the process,
we have produced a rough outline of what the Hevajra body mandala practice looks like.
It is possible this description of the body mandala might be found in even earlier texts as
well, such as in the works of Sa chen or even in an Indian source. The Lam ‘bras
tradition was passed down for two hundred years before Sa chen put it in writing; this
aspect of the textual history of the core teachings of the Sakya tantric tradition provoked
critiques from skeptics who doubted its authentic basis in Indian Buddhist tradition at
several key instances in Tibetan history.
IIB. Bu ston on Nagpopa

In the second part of the Mkhas grub quote, he contests the validity the manner of
generating the protective circle portion of the body mandala described above. He sets
forth the claim that Bu ston himself, in writings that appear to deal with the
Cakrasaµvara body mandala, objected to the authenticity of such a practice:
With regard to that writing, the omniscient Bu ston says:

“In the sdom pa bshad pa [Saµvara-vyåkhyå ] (it says):
‘Generating the body in stages, Vajrakîla is spontaneously generated. The binding with a
string is firmly fastened without a gap while the tent is the garland of bones itself.’
Except for this, there is no explanation like that one anywhere in any Indian system.”
That is correct.435
Bu ston is citing Nag po pa’s text, the Saµvara-vyåkhyå [13.4].436 The context in which
the source passage appears in Nag po pa’s text is one in which there is a description of
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
This is followed by zhes gsung go, indicating that is a quote or a summary of transmitted
instructions.
434 !His conclusion also contains an interesting verse reflecting upon the nature of the
internal [nang] body mandala. The verse can be roughly translated as follows: “In order
to supremely practice inside, in agreement with the inner way, the manner of cultivating
the inner mandala is taught through the perception of internality itself.” [530.1] nang la
mchog tu gzhol ba’i phyir; nang gi tshul dang rjes mthun par; nang gi dkyil ‘khor bsgom
pa’I tshul; nang*gi*nyid*gzigs*pas*gsungs*
!
435
zhes sogs bris pa la; thams cad [255.2] mkhyen pa bu ston rin po ches; sdom pa bshad
par ; go rim bzhin du lus skyes nas435 ; rdo rje phur bu lhan cig skyes ; bcings pa thig
skud rab tu ldan ;shin tu mtshams med brtan par bcings ; gur ni rus pa'i phreng ba nyid;
ces ba tsam ma gtogs rgyud gzhung [255.3] gang na'ang de 'dra ba bshad pa mi 'dug go ;
zhes gsung ba ni shin tu med do
! 144!
Cakrasaµvara (technically Heruka in the Cakrasaµvara mandala) in his spontaneously
generated form [lhan cig skyes pa] continuously abiding in the heart, equal inside and
out.437 [Saµvara-vyåkhyå 13.2-.3] Then, we are told that, “The array having been
explained in terms of the mind of the mandala [dkyil ’khor gyi yang sems], the sites of
the body are to be explained.” [13.3] Our passage is part of the instructions for what the
yogin should cultivate after spontaneously generating Hevajra. As such, it can likely be
understood as the body mandala produced after generating oneself as the deity and then
imagining the deity positioned on one’s heart.438
The passage in question is introduced in Nag po pa’s text as “the generation of the
body in stages” [go rims bzhin du lus bskyes ]; this is likely intended to distinguish it
from the spontaneous generation. It is is followed there by further embellishment of the
appearance and qualities of Heruka [13.4-.5] and then by a passage [13.6-.14.1] that
resembles the description of the mandala of the support as the elements of the cosmos
elided from Mkhas grub’s citations. To review, that “elided passage” emerged in some
part from Hevajra sådhana and their addenda within the work of the Sakya patriarchs; in
those texts, it formed part of a three-fold method of correlating the body with the
protective circle, cosmic elements, and celestial palace of the mandala. This passage
from Nag po pa’s text describes elemental mandalas located in the soles of the feet,
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
436
Nag po pa. Saµvara-vyåkhyå. Sdom pa bshad pa. In bstan 'gyur (sde dge). TBRC
W23703. 22: 13 - 22. Delhi: delhi karmapae choedhey, gyalwae sungrab partun khang,
1982-1985.Toh:1460; bstan 'gyur gser bris 179 vol 14 ff. 8r-14v (pp.15-28); Otani
Beijing 2177.
437
For an interesting inquiry into the meaning of lhan cig skyes pa [Skt. sahaja], see
Kvaerne, Per. 1975-6. “On the Concept of Sahaja in Indian Buddhist Tantric Literature,”
Temenos: Studies in Comparative Religion, vol.11-12: 88-135.
!
438
Jake Dalton has pointed out how this citation sounds like a body-based or ultimate-
truth oriented interpretation of the preparatory rites for constructing a mandala. [Personal
communication, June 2015] Drapa Gyatso suggested thatVajrakîla’s lus dkyil is the
object of this spontaneous generation. [Personal communication, Spring 2012].
Preparatory tantric rites often involve the spontaneous generation of the form to which
the main rite as a whole is devoted to building in stages. For example, rites for creating a
sand mandala are typically prefaced by the mental production (or imagination) of that
mandala. Likewise, initiation rites typically require the vajråcårya to first generate self as
deity before ritually guiding the initiate in cultivation those same divine qualities through
as series of consecrations. In the discussion of the “proto-body mandala rite” from
Dunhuang in Chapter Two of this dissertation, we discussed its possible relation to such a
preparatory rite. One might even hazard to say that tantric preparatory rites are, by nature,
prototypical in the sense that they lay out a form or structure upon which later versions
are modeled.
See Beyer 1973 [Reprint 1978, p. 73-74] for a helpful description of the Cakrasaµvara
body mandala practice based upon Pad ma dkar po’s Snyan rgyud yid bzhin nor bu
bskyed rim, fols.12b ff.
! 145!
stomach, chest, throat and crown (as well as perhaps an elusive description of the secret
place) [14.1].439
Our goal is to understand Mkhas grub’s use of Bu ston’s interpretation of Nag po
pa’s text. Locating the source for this comment on Nag po pa’s text within Bu ston’s
writings proves challenging since Bu ston wrote extensively on the Cakrasaµvara
practice. For example, while his Nag po pa’s Cakrasaµvara Sådhana, Free from Errors or
Impurities seems a likely candidate, no such comment appears there.440 The description
of the mandala of the support in that text does not seem to include the protective circle;
however, it does, once again, resemble the description of the body as cosmos common to
the Sakyapa Hevajra sådhanas but elided by Mkhas grub in his citation of them.441 In
continuing to search for the source of Bu ston’s statement, clues may be derived from
Ngor chen’s rebuttal [549.2-5]. There Ngor chen asserts that his opponent’s statement
contains a misreading of Bu ston. In doing so he refers to two texts, the Yoginî-saµcåra
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
439 !rkang pa'i mthil du rtsom pa'i rlung; yang dag gnas shing lter son par; me yi dkyil
'khor gzi brjid 'bar; lto bar ches ni dkar ba yi; chu ni mdzes pa nyid kyi gzugs; brang du
yang ni dbang chen ni; kha dog ser pos rnam par spyas; mgrin pa'i steng du brten ba yi;
rtse brgyad ldan pa'i ri rab bo; bdun ldan nyi ma a li zla; dri med zla ba'i gzi brjid nyid;
nang du 'grod pa 'ong ba dang; chu skyes 'dab ma sum cu drug; ldan bar rdo rje'i dbus son
par; aa li ka' li las byung grub; mdzes pa'i spyi bor hum byas te; der gnas gser gyi gnas su
ni; sna tshogs dbus su dang po'i ming; zhes bya ba lhan cig skyes pa yang
Shortly after, Nagpopa writes, “As for the mandala that is produced in terms of the three,
it is the mandala of that one’s body.”[14.3] [gsum*gyis*bskyed*pa'i*dkyil*'khor*ni;*de*yi*
lus*su*dkyil*'khor*nyid].!! “The three” may refer to the circles of deities of body, speech
and mind that characterize many descriptions of the Cakrasaµvara body mandala. It is
also possible, though perhaps less likely, that it refers to the rungs of the protective circle
(or even more tentatively, to the body as protective circle, cosmic elements, and celestial
palace).!
440 Nag po pa’s Cakrasaµvara Sådhana, Free from Errors or Impurities. Bde mchog nag
po pa’i sgrub thabs ‘khrul ba’i dri bral. Toh 5049. In gsung 'bum/_rin chen grub (zhol
par khang). TBRC W1934. 7: 151 - 186. [lha sa]: [zhol par khang], [2000].
I did find compelling descriptions of body mandala practice of the Nag po pa tradition
there. Cumulatively, these practices reflect a dense layering of elements, deities, cakras,
channels and in particular, mantras and syllables upon the human body as well as subtle
body practices involving the manipulation of winds and drops in close cooperation with
mantra and syllable recitation and visualization.
441
tshur 'dus nas rang yab yum la thim pas ro gcig tu gyur p'i sku la rten and brten par
bcas pa'i dkyil 'khor rang bzhin gyis gnas pa gsal gdab par bya ste; rkang pa gnyis bgrad
pa rlung gi dkyil 'khor ; sum mdo grug sum me; lto ba zlam po chu; brang gru bzhi sa;
skal tshigs ri rab; mgo bo ri rab kyi stod kyi cha; lus 'dom gang gru bzhi ni gzhal yas
khang gru bzhi; sgo dgu ni sgo mtshams; yan lag gan rkang brgyad ni ka ba brgyad ; lus
ngag yid gsum ni 'khor lo gsum
The detail that distinguishes this description of the mandala of the support in Bu ston’s
text from the one we quoted above from Nag po pa’s text is the correlation of the shape
and measurement of the body with the celestial palace and of the limbs of the body with
the pillars.
! 146!
and the Commentary on Nag po pa’s Sådhana.442 Additional possibilities include Bu
ston’s Eliminating Errors in the Commentary on the Cakrasaµvara Tantra443, his Garland
of Perfection Yoga444, or even his Ghañ†apa Cakrasaµvara Mandala Abhisamaya.445
Another possibility is that the reference is to a text Bu ston composed on another tantric
cycle.446
We will see why the difficult task of locating precisely where this statement by
Bu ston is made is important when we evaluate Ngor chen’s critique. The main point
here, however, is to determine how Mkhas grub is using it. Bu ston, the great fourteenth-
century cataloguer and compiler of the Tibetan canon is invoked as an authority on what
is authentic, in other words, what texts and practices can be definitively identified as
Indian in origin. Mkhas grub is therefore saying that this body mandala description
derived from Nag po pa’s text is the only one of it’s kind in an Indian source. Therefore,
he implies the preceding quote of “some Tibetans,” which we have traced to the Sakyapa
Hevajra body mandala texts, is an account of an illegitimate version of the practice.
Mkhas grub’s claims direct our attention to the relationship among various phases of the
body mandala practice, among the various components of the mandala of the support (the
protective circle, the ‘container’ built of cosmological elements, and the detailed
formation of the celestial palace), and of the mandalas of support to the supported.447 The
relationship between different structural components of body mandala and the transitions
between the phases of the ritual practice are an important aspect of Mkhas grub and Ngor
chen’s methods for defining body mandala.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
442
As discussed below, the former, the Rnal 'byor ma kun spyod, probably refers to Bu
ston’s Rnal 'byor ma kun tu spyod pai rgyud gyi bshad pa bde mchog gi don rab tu gsal
ba [Bu ston’s Collected Works (rin chen grub (zhol par khang), Vol. 6, pp. 725-876
(76ff) Toh 5045]. The latter, “Nag po pa'i sgrub thabs kyi 'grel pa,” is likely the text
briefly discussed above, Bu ston’s Bde mchog nag po pa’i sgrub thabs ‘khrul ba’dri bral
[Toh 5049].
443
Dpal ‘khor lo sdom pa’i sgrub thabs kyi ‘grel pa ‘khrul bas pong par byed pa [Toh
5050]. I have reviewed about one third of this text to date, and have yet to find a
connection.
444
Rnal 'byor rdzogs pa'i 'phreng ba . This text on the Cakrasaµvara abhisamaya appears
to be referred to by Mkhas grub at 260.1. See 6 ff scanned dbu med manuscript found as
TBRC W1CZ1191.
445
Bde mchog dril bu lugs kyi dkyil ‘khor gyi mgon rtogs. There is an dbu med
manuscript of this text by Bu ston on the Ghantapa Cakrasaµvara abhisamaya. See
TBRC W1CZ1248.
446
In her 2006 essay, Bentor refers to several ways in which Mkhas grub and Bu ston
differ in their understanding of the Årya Guhysamåja practice. For example, Bentor
refers specifically to Bu ston’s commentary on the Mdor byas [Piñ∂ik®ta ], his Dpal
gsang ba ‘dus pa’i sgrub thabs mdor byas gyi rgya cher bshad pa bskyed rim gsal byed.
Collected Works (New Delhi International Academy of Indian Culture, 1967, vol.9, 683-
877).!
447
No mention is made of the cremation grounds here, but as a key feature of both the
Cakrasaµvara and Hevajra body mandala and outer mandala, of of yoginî tantra more
largely, one might be lead to consider how they fit in as well.
! 147!
IIC. What Really Qualifies as a Body Mandala?
The final section of the quote from Mkhas grub’s text identifies and denounces various
fraudulent forms of body mandala practice:
“There is that method of generating the body as a celestial mansion like that, of arranging
the one hundred and fifty-seven deities of the body mandala and so forth in the outer
mandala of only nine gods, of arranging the deities of the outer mandala in the inner
mandala, and without deities common to the two (inner and outer). (Such methods) are
not explained in the tantra or any authentic Indian system nor even in the implicit
meaning [don thob].
In investigating with a mind that doesn’t even distinguish between a body
mandala and establishing deities on the body, many lies masquerading as the superior
instructions appear in the Tibetan methods of establishing the body mandala. The
affirmation and negation are not elaborated at length here.” 448
In this section, Mkhas grub appears to refer to version of the Hevajra (outer)
mandala popular in artistic representations; this mandala contains nine deities, Hevajra
united with Nairatmya and surrounded by the eight retinue goddesses. The Hevajra body
mandala or inner mandala on the other hand, contains one hundred and fifty-seven
deities. I have not encountered any other version of body mandala practice from another
tantric cycle that shares this enumeration; therefore, it seems likely that a scholar monk,
the most likely taudience for the body mandala debate texts, would have easily identified
Mkhas grub’s comments here with the Hevajra system. Therefore, we can understand
why Ngor chen construed his remarks as an attack on the Hevajra practice of the Sakyapa
despite the fact that Mkhas grub never identifies that system by name.
Mkhas grub is troubled by the question of how precisely to relate the deities of
these versions of inner and outer mandala, presenting us with another variety of category
confusion (and also an iconophilic challenge). He, likewise, repeats his charge of
fraudulence with regard to the basis of such practices in the Indian tradition.
Furthermore, he clarifies that his critique includes the full range of traditional exegesis
ranging from explicitly to implicitly-rendered interpretations of the texts. The division of
meaning into explicit or clearly apparent meaning [nîtårtha Tib. nges don] as well as or
interpretable meaning [neyartha Tib. drang don], meaning that must be sought, is a
standard Buddhist exegetical principle. This two-fold principle of meaning- making is
often placed in dialogue with two truths of conventional [saµv®ti] and ultimate
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
448
lus gzhal yas khang du bskyed tshul de 'dra ba dang; phyi'i dkyil 'khor la lha dgu las
med pa la; lus dkyil gyi lha brgya rtsa lnga bcu rtsa bdun la sogs pa bkod cing; lus dkyil
[255.4] gyi lha'i nang na phyi dkyil la 'god pa dang thun mong ba lha gcig kyang med pa
sogs; rgyud dang rgya gzhung tshad ldan gang nas kyang ma bshad cing don thob la'ang
med kyang; lus dkyil dang lus la lha 'god pa ci ma gyi khyad par yang ma phyed pa'i blos
btags [255.5] la; man ngag mchog tu ming btags pa'i mun sprul du ma zhig bod kyi lus
dkyil bsgrub tshul rnams la snang ste; 'dir dgag sgrub rgyas par ma spros so
! 148!
[paramårtha]. 449 The category of implicit meaning introduced in this passage, don thob,
is literally meaning “obtained” or “made manifest.” While it is juxtaposed with what’s
been “explained” [bshad], Mkhas grub does not provide any clues as to how don thob can
be accessed. Finally, Mkhas grub critiques those who fail to distinguish body mandala
from practices like nyåsa. We discussed this distinction in Chapter Four; Mkhas grub
understands nyåsa as simply placing syllable upon the body’s exterior; body mandala, on
the other hand, involves a holistic transformation of the parts of the body to prepare them
to support the generation of deities.
Once again, in this passage from Mkhas grub’s text, the relationship of inner and
outer mandala comes to the forefront. However, it is not their ontological status or their
aptness to serve as a basis for tantric practice that is at issue. Rather, in this instance, the
problem is the specific manner in which they are related through the mechanics of body
mandala practice. A few additional points are clarified as well. First, we now know that
Mkhas grub is troubled specifically by the manner of generating the celestial palace in his
citation of the practice of ‘some Tibetans’ [pod dag]. According to Dongsung Shabdrung
Rinpoche, some of the Mkhas grub’s problems with the relationship of inner and outer
mandala can be attributed to the ritual context, the dissolution of inner into outer
mandala.450 The ritual order of such acts of dissolution does indeed seem relevant, as the
preceding portion of Mkhas grub’s text negotiated the relationship of earlier and later
visualizations of the celestial palace.
“Gathering/Dissolution” [bsdus pa] is a vital ritual act that blurs the boundaries
between emptiness and form. This very act call the relationship of different stages of the
practice to one another into question. The first verse of the quote from the practice of
“some Tibetans” suggests another source of confusion in the relation of inner and outer
mandala. Shabdrung Rinpoche identified this particular kind of issue as “go rims
‘khrugs,” a confusion of the order or stages of practice.451 This confusion pivots upon the
placement of the vi!vavajra or crossed vajra. The vi!vavajra typically appears at the base
of the protective circle in outer mandala representations, whereas this body mandala
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
449 !Lamotte 1988 p.17 and Broido 1988, pp. 80-81. Lamotte, Etienne, 1988. “Assesment
of Textual Interpretation in Buddhism,” in Buddhist Hermeneutics.Ed, Lopez, Donald S.

Honolulu: University of Hawaii Press.
Broido, Michael M. 1988. “Killing, Lying, Stealing and Adultery: A Problem of
Interpretation in the Tantras,” in Buddhist Hermeneutics. Ed. Lopez, Donald S.
!
450
Dongsung Shabdrung Rinpoche, Personal Communication, Fall 2013. Bentor’s
findings reported in her 2015 article would appear to support this interpretation in the
case of the visualizing the body as the selectial palace. The evidence presented by Bentor
suggests that Tsong kha pa maintained that the earlier visualization of the celestial palace
plays a parts in the practitioner’s later generation of the parts of the body as the parts of
the palace. Mkhas grub, on the other hand, intends to first dissolve that earlier
visualization.
451
Personal Communication, Fall 2013.
! 149!
practice locates it on the crown of the head.452 Several sources have suggested that the
location of the vi!vavajra in this version of body mandala practice provoked Mkhas grub
by turning the practice on it’s head, so to speak. In his writings on the Lam 'bras
teachings of the “Outer Creation Stage” of Hevajra, sixteenth-century Sakyapa master
Jamyang Khyentse Wangchuk (1524-68) distinguished the vi!vavajra at the base of the
celestial palace from that at the base of the protective circle. His comments suggest that
there was some controvery over the vi!vavajra at the base of the celestial palace: “Some
claim that this is the crossed vajra of the protection cakra, but we maintain that that is
invisible, obscured by the tiny vajras spread out like barley. This is the seat of the
celestial mansion.”453 In this case, the exegete compensates for the absence of clear
relationships between these elements by suggesting that there is more than one
vi!vavajra. Mkhas grub’s concerns with the connections between the components of
inner and outer mandala, as exemplified in the problem of the vi!vavajra, highlight the
significance of both establishing a proper basis for generation and for maintaining order
in the relationship of different types of mandala and of their parts.
III. Ngor chen’s Reply Part One: on Bu ston and the Mandala of the
Support
Ngor chen begins his defense by challenging the validity of Mkhas grub’s citation of Bu
ston. He claims that Bu ston never denied the existence of such teachings on the mandala
of the support in the Indian system. In doing so, he refers to what appear to be two
commentaries by Bu ston, one on the Yoginî-saµcåra-tantra, the other on Nag po pa’s
sådhana. The latter is likely the text briefly discussed above, Bu ston’s Nag po pa’s
Cakrasaµvara sådhana, Free from Errors or Impurities454. The former probably refers to
Bu ston’s The Commentary on the Yoginî-saµcåra-tantra, the Utterly Clear Meaning of
Cakrasaµvara.455 The Yoginî-saµcåra-tantra was classified by Bu ston as an uncommon
explanatory tantra of the Cakrasaµvara cycle.456 Both texts, therefore, are
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
452
I am grateful to Dongsung Shabdrung Rinpoche for his guidance on this point and to
Dan McNamara who, in his readings of Chogay Trichen Rinpoche’s body mandala
commentary, first brought my attention to this important detail.
453 !Translated in Stearns 2006, 503. Summarizing Notes of the Outer Creation Stage
From the Expansion of the Great Secret Doctrine: Summarizing Notes of Guidance for
the Precious Teaching of the “Path with the Result” by Jamyang Khyentse Wangchuk. !
454 !Bde mchog nag po pa’i sgrub thabs ‘khrul ba’dri bral [Toh 5049]!
455
Rnal 'byor ma kun tu spyod pai rgyud gyi bshad pa bde mchog gi don rab tu gsal ba.
Bu ston’s Collected Works (rin chen grub (zhol par khang), Vol. 6, pp. 725-876 (76ff).
456
Gray 2007, p16. See Pandey 1998 for a Sanskrit and Tibetan edition of this tantra.
(Gray 2007, fn52). According to English 2002, this tantra was significant for the study of
body mandala practice because it “emphasizes the importance of the practice in the
Cakrasaµvara tradition by ascribing it to the mythical Låkßåbhidhånatantra (althougth the
practice was in fact ¸aiva in origin).” (p.197) For more on the relevant passage from the
Yoginî Saµcåra and on Sanderson’s work in tracing body mandala prototpes to ¸aiva
sources such as the Tantrasadbhåva see also fn 470 and Sanderson 2001.
! 150!
Cakrasaµvara-related. Ngor chen argues that it was only in the case of Cakrasaµvara that
Bu ston denied an Indian textual basis for this version of the mandala of the support.
He then cites the “elided section” from the Hevajra body mandala quote
referenced above (although, he too does not yet identify it explicitly with the Hevajra
cycle): “The soles of the feet, the rlung; the abdomen, fire; the stomach, water.; the heart
center, earth; the vertebra, Mount Meru, the highest part of the summit of (/which is) the
head; The armspan, the 4 cornered celestial palace, The eight bones, the pillars...”457 We
will recall that this teaching was included both in the body mandala text attributed to
Grags pa rgyal mtshan and in that by ‘Phags pa. Similar versions of the practice were
also found in Nag po pa’s text, the Saµvara-vyåkhyå (albeit in a portion also not quoted
by Mkhas grub) as well as in Bu ston’s Nag po pa’s Cakrasaµvara Sådhana, Free from
Errors or Impurities.
IIIA. A Diagram of the Mandala of the Support?: Cultivating

“Iconophilia” in Approaching Sådhana as Bodily Discourse
If we are having difficulty imagining what such a vision of the human body would
actually look like, we can refer to a cakra diagram dated by Amy Heller to eleventh-
century Western Tibet.458 [See Fig. 23] We will look more closely at this diagram in
our discussion of representations of the subtle body in the conclusion of this dissertation.
At present, we are merely using it as a point of reference. The front side of the drawing
depicts a body containing a series of cakras while the reverse, of primary interest to us
here, reduces the body to just a series of stacked forms, with no bodily outline or cakras
present. One noteworthy detail of the recto drawing in the boars’ head emerging from the
head of the main figure. It seems likely that Heller’s interpretation of this detail as
indicating a connection to the practices of Vajravårahî, and thereby connected at least
peripherally with the Cakrasaµvara system, is correct. Using English’s 2002 monograph
as a guide, Heller has identified the stacked shapes and associated seed syllables that
make up the verso drawing with the fundamental elements of the universe upon which the
celestial palace is built.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
457
rkang mthil rlung; sum mdo me; lto ba chu; snying ga sa; sgal [549.5] tshigs ri rab;
mgo bo'i rab kyi stod kyi cha; lus 'dom gang gru bzhi gzhal yas khang; rkang brgyad ka
ba rnams 'don pa yin na ni
458
Heller 2009. I am grateful to Nancy Lin for bringing this drawing to my attention.
More research is required on the dating of the drawing. Heller based her dating on four
factors: a radio-carbon analysis of the paper, mention of a Zhi ba‘Od in the inscription
(which she takes to connect it to the royal family of the Guge-Puhrang empire of Western
Tibet), stylistic aspects of the drawing and “archaic orthography.” In consulting with
other scholars such as Christian Luczanits, I am lead to doubt the validity of the latter
points of evidence. This diagram was recently displayed in the Rubin Museum’s “Bodies
in Balance” exhibition. I am grateful to Christian Luczanits for discussing some of the
details of the drawing with me at that time. [Personal communication, 7/17/2014].!
! 151!
While we have not had the opportunity to explore the Cakrasaµvara body
mandala practice in depth, this diagram and Heller’s interpretation of its parts helps us to
think about how the “elided section” of the citation from the body mandala practice of
‘some Tibetans’ in Mkhas grub’s text relates to the cited portions. Theoretically, the
element of wind resides in the soles of the feet, and the universe/body is built up from
there, gradually assuming more coarse levels of embodiment up until the earth element in
the chest. However, the location of the vi!vavajra, the foundation for the celestial palace
and the base for the protective circle, in unclear. In this drawing, likely associated with
the Cakrasaµvara system, the vi!vavajra appears above all the other elements rather than
at the base.
Heller envisions the vi!vavajra in this diagram as the base of the palace and the
vajra as its tip. Therefore, in her view, the palace is represented only elliptically, and no
direct correlation is specified between the palace and the parts of the body. Instead, the
palace appears to be founded upon the basis of the body as a container of cosmic
elements. English’s 2002 work on the practices of Vajrayoginî (upon which Heller relied
in part), presents a relevant passage on the generation of the deity’s abode upon a
foundation of cosmic elements:
“The commonest method of visualizing the deity’s dwelling place in mainstream

sådhanas...is not a cremation ground but as the traditional Abhidharmic universe. This
begins with the visualization of the elements that underpin the earth’s surface; the yogin
sees the axial Sumeru (or Meru) rising up into the heavens. Above this (or encompassing
it all), he installs the circle of protection and the dharmodayå, or “origin of existents.”
Finally, upon the mountain’s peak, he visualizes an elaborate and decorative temple
palace (kütågåra˙) as the future abode of the deity.”459
English compares the vision of the Abhidharmic universe with the Vajravåråhî
sådhana that is the focus of her study. In the process, she demonstrates how certain
details were modified in the sådhanas of higher yoga tantra, such as the correlation of
space with the principle of emptiness as well as the order of the elements. Fire comes to
follow wind and therefore to agree with the description of the body in Chapter Three of
the Abhidharmako!a (v.44b).460 In the case of the Vajravåråhî sådhana, English observes
how this correlation of macrocosm with microcosm plays out with particular salience in
the body mandala practice.461 Finally, she interprets the absence of the palace from that
sådhana as a significant indication of the evolution of the practice to a “more integrated
higher tantric practice” placing more emphasis upon the cremation grounds as the abode
of the deity and thereby a key site in the ritual drama. Based on this observation, one
might speculate that the absence of visual or textual representations of the body as
celestial palace in a given account of body mandala practice reflects a similar emphasis
upon the higher yoga tantra agenda.
As observed in Chapter Two, the mandala, as a three-dimensional form that can
be collapsed along its axis to a pith or central point, contains a tension between center
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
459
English 2002, 144.
460 !English!2002,!146.!!!
461 !Ibid.!
! 152!
and periphery, top and bottom in its very structure. The mapping of these forms onto the
body in body mandala practice further complicates this system of relations, by
introducing the form of the body with its own structures and hierarchies. It is therefore
not surprising that questions of the relation of its parts would arise in the creation and
exegesis of ritual practice. English, for example, encounters comparable difficuties in her
interpretation of the relationship between the protective circle and the cosmos in
theVajravåråhî sådhana:
“Sådhanas that directly follow the emptiness meditation with the visualization of the
cosmos must postpone installing the circle of protection until after the cosmos has been
set in place. This differs from the Vajravåråhî sådhana, in which the emptiness
meditations lead on directly to the circle of protection, and in which the cosmos...is
visualized inside the circle of protection...The difference is more apparent than real,
since, in this case, the circle of protection presumably encompasses the visualized
cosmos, or on “top”, perhaps “superimposed”? The dharmodaya and temple palace are
then visualized within the circle of protection, on top of Mount Meru.”462
English’s comments highlight the connection between structure and chronology in

visualization practice. They also suggest that sådhana themselves do not always provide
clear transitions between parts of a visualized structure or phases of the visualization
process. English is grappling with a similar problem in interpreting her sådhana as we do
in viewing the diagram. As we recall, Mkhas grub appears to have also shared similar
concerns with the versions of body mandala practice from the mother tantras (presumably
Hevajra). We described this problem above as go rims ‘khrugs, a confusion of the order
of the stages of practice. In the case of thevi!vavajra introduced in the context of relating
inner and outer mandalas, Mkhas grub expressed similar concerns for establishing a
proper basis for generation and for maintaining order in the relationship of different types
of mandala and of their parts.
In order to fully understand the role of the citation from the Hevajra body
mandala practice in the body mandala debate, one would need to engage with multiple
aspects of the order of ritual practice of the mandalas of support and supported.
Furthermore, comparison of their roles in the Cakrasaµvara and Hevajra practices would
be necessary. Even in just beginning with the basic details of the citation, we are lead to
question how the body can function simultaneously as a circle of protection, the universe,
and a palace. In the excerpts from the Hevajra practice quoted by Mkhas grub, all of the
body parts correlated with parts of the palace seem to belong to portions in and above the
heart. However, there is also an emphasis upon the all-encompassing aspect of the body
as palace, both in the span of the body and the limbs as foundational elements that also
demarcate the borders of the palace. Therefore, despite the localization in the upper part
of the body that might be interpreted as an expression of vertical hierarchy, there is also a
decidedly holistic or totalizing aspect to the body: palace correlation.463 Moreover, when
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
462 !English!2002,!fn321.!!
463
In the early chapters of this dissertation, we have discussed the different modes of
representing hierarchies in the mandala. We have also considered how the hierarchies of
! 153!
we address Ngor chen’s discussion of the mandala of the supported below, we will find
that deities of the Hevajra body mandala inhabit cakras along the central channel of the
body from top to bottom. We can see, therefore, how the relation of top to bottom, inside
to outside, in this version of Hevajra practice is far from straightforward.
How then are we to interpret the absence of transitions between representations of
the body in sådhana and diagrams like this one? If we were to interpret the diagram as a
representation of body mandala, the front side with the the drawing of the cakras would
be the mandala of the supported; the reverse would be the mandala of the support
emphasizing the correlation of body with cosmos through reference to the elements, Mt.
Meru, etc... But we are still left with the problem of navigating the transition between the
two. Are we perhaps dealing with multiple bodies or versions of the body? Are these
different phases in a ritual process rather than a collection of forms simultaneously
mapped onto a single body? The mode of representing the body in this case, plays upon
the implied relation of the elemental composite on the reverse side to the drawing of the
body with the cakras on the front side, a implied relation of support and supported.464
However the phases of constructing a protective circle from the body or of generating
parts of the body as parts of the palace are absent.465
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
body and mandala interact in body mandala ritual. We will revisit this topic in the
discussion of representations of the subtle body in the conclusion of the dissertation.
464
A similar logic might be said to be at play in relating a painting to the consecratory
inscriptions and forms that appear on its reverse. See the introduction for a brief
engagement with such representations in terms of the principle of container and
contained.
!
465
Bentor 2015 has suggestion a distinction between Tsong kha pa’s interpretation of the
relationship between the different versions the visualization of the body as the celestial
palace and those of later Geluk thinkers. Bentor cites Tsong kha pa’s ‘Dod pa ‘jo ba to
clarify this distinction: “From now on, the continuum of your earlier visualization of the
stacked up physical elements, Mt. Meru and the celestial palace proceeds without being
dissolved. Therefore when you begin your meditation on your body as the celestial
mansion, on the basis of each former similar moment and each part of the body, a
subsequent similar moment arises.” [‘Dod pa ‘jo ba folio 122b, p442.5-.6 as translated
by Bentor 2015, p.66.] The Wish-Granting Extensive Explanation of the Cakrasaµvara
Abhisamaya. Bde mchog mngon rtogs rgya cher bshad pa 'dod pa 'jo ba/." In gsung
'bum/_tsong kha pa/?bla brang par ma/?. TBRC W22273. 9: 195 - 592. [bla brang]: bla
brang bkra shis 'khyil, [199?]
Bentor also provides a reference (p.67 fn23) to Ngor chen’s commentary on the
Ghantapa transmission fo the Cakrasaµvara body mandala practice to demonstrate that
Ngor chen disagreed with this interpretation. [Bentor cites folio 375b, p.402.2.1-2. in the
following version: Ngor chen Commentary on the Ghantapa Body Mandala Practice. Sa
skya pa’i bka’ ‘bum. Tokyo: The Toyo Bunko, 1969, vol.10, folios 117b-140a, pp.
398.1.1-405.4.1]. These differences in interpretation nicely exemplify the very challenges
of connecting different versions of a visualization or phases of a ritual practice faced by
ritual exegetes. I am using Latour’s concept of “iconophilia” to suggest that the gaps in
many sådhana between these visualizations or ritual phases are productive.
! 154!
LaTour’s notion of “iconophilia” may provide some clues for interpreting the
ambiguity of relations between internal and external mandala and as well as between the
different phases of body (inner) mandala ritual. We have observed the problem of
locating the vi!vavajra, a foundational element of the protective circle of the mandala in
both practices, reinforced this ambiguity of connecting different phases or varieties of the
ritual. In the examples we have encountered, diagram and sådhana, image and text, both
challenge their audiences to be iconophilic, to focus upon “the movement between
images” and, in doing so, to bring them to life, to animate or even embody them. We will
continue to examine the ways in which ambiguities of relationship between
representations of the body as mandala power the body mandala debate.
IIIB. Ngor chen and the Sampu†a Tantra

Bearing these problems of the relation of the mandalas of inner and outer and
support and supported in mind, we will return to our discussion of Ngor chen’s reply. To
review, first, in the section devoted to correcting “literal flaws” [tshig la skyon] Ngor
chen points out that Bu ston only claimed that such versions of body mandala are absent
from the Indian sources on Cakrasaµvara in particular, with the exception of the Nag po
pa verse. Then, he cites the “elided section” of the quote correlating the body with the
cosmic elements that we know to be part of the Hevajra body mandala sådhana and
prepares us to engage with counter-evidence from the Sampu†a Tantra.466 Next, in the
section devoted to correcting “flaws in meaning” [don la skyon], Ngor chen interprets the
quote from Nag po pa’s Sdom pa bshad pa [Saµvara-vyåkhyå]. He clarifies “tent” in
“as for the tent, the very garland of bones” as a reference to the protective circle. He
supports his interpretation with a description of the protective circle by the Indian author
Abhayåkaragupta:
“Abhayåkara clearly explained:

‘In the limit of nadir, in whatever size, it is all firm and indestructible. The vajra
foundation (which has) the nature of the flaming vajra is bounded at the border by a
garland of lightrays like the fire at the end of a kalpa. From the nadir to the top is very
high. The flaming vajra fence is thick, solid, and firm. Above the vajra fence, there, the
blocks become all solid wall without a break. Above there is a lattice of arrows.
Below, there is the flaming vajra tent which is ornamented by vajra canopies.”467
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
466
[549.5] De rnams sam pu tar bshad pas rgyud ‘gog par ‘gyur la
467
slob dpon a bha ya ka ras; sa 'og gi mthar thug par ji ltar mngon par 'dod pa'i rgyar
gcig tu sra zhing mkhregs la; [550.2] 'bar ba'i rdo rje'i rang bzhin gyi sa gzhi bskal pa'i
mtha'i me ltar 'bar ba'i 'od zer gyi phreng bas mtshams bcings pa dang ; sa 'og nas steng gi
bar du shin tu mtho pa; mthug cing mkhregs pa sra zhing 'bar ba'i rdo rje'i ra ba dang; rdo
rje ra ba'i steng du phar [550.3] mtshams med par dum bu gcig tu gyur cing; steng rdo
rje'i mda'i dra ba dang; 'og tu rdo rje'i bla res brgyan pa'i rdo rje'i gur 'bar ba'o; zhes gsal
bar gsungs pa'i phyir ro
This quote is derived from Chapter One of Abhayåkara’s Nißpannayogåvalî [rdzogs pa'i
rnal 'byor gyi phreng ba] (Toh 312 1) sde dge bstan ‘gyur Vol.75 ff. 94v-151r (pp.188-
301) 189.4-.6.
! 155!
Here Ngor chen moves beyond the purview of Bu ston to show that such
descriptions of the body mandala of the support as presented by Mkhas grub for scrutiny
(familiar form Hevajra body mandala sådhana) can indeed be found in Indian sources.
This move is tricky and presents some difficulties for interpretation. He seems to claim
that indeed there is a practice of generating the body as the protective circle in the
Cakrasaµvara body mandala practice.468 We will not be able to explore all of the
subtleties it suggests here.469 Instead, we will focus Ngor chen’s next move: the
presentation of evidence from the Sampu†a Tantra.
The Sampu†a Tantra plays a prominent role in Ngor chen’s rejection of Mkhas
grub’s charges that a particular version of body mandala connected with Hevajra is
absent from the Indian sources and therefore, not authoritative.470 The Sakyapas classify
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
I am grateful to have had the opportunity to read portions of this text in a seminar with
Professor Musashi Tachikawa during his time at UC Berkeley in Fall 2012. The passage
appears in that text as a description of the protective circle; I do not know of any explicit
mention of body mandala within the Vajråvalî cycle. In his 2003 dissertation, Lee claims
that although Abhayå does not use the term “body mandala,” he refers to the practice in
some of his ritual descriptions in theVajråvalî . [Lee 2003, p.130 fn4]
I explored the possibility that this quote might be found in Abhayå’s extensive
commentary on the Sampu†a Tantra, the Åmnåyama•jari. Although the precise quote
does not appear there, there are references to aspects of the protective circle at 100b,127b
& 231b of the sde dge edition. Mori 1997 has located eight references to the Sampu†a
Tantra within the Vajråvalî itself. He also points out that Abhayå composed this
commentary simultaneously with the Nißpannyogåvalî or the Vajråvalî. See Mori 1997
p.26.
468
des na; lus kyi spyi bo'i rus pa sna tshogs rdo rje; rkang mthil gyi rus [550.4] pa rnams
rdo rje'i sa gzhi; rtsib rus rdo rje'i gur la sogs par lung 'dis bsnyon med par grub pas bde
mchog gi skabs su srung 'khor lus dkyil gtan nas ma bshad pa ma yin no
Ngor chen’s N2 reads bsnyon med par ‘grub pa yin no without the final phrase.
469
Among them are further possibilities for interpreting the relation of descriptions of the
protective circle in body mandala to those in other mandala rituals and in Cakrasaµvara
to Hevajra systems. More specifically, the final portion of the same chapter from
Abhayå’s text, not cited by Ngor chen, includes a detailed discussion of the role of the the
protective circle; this section compares its role in the generation and perfection processes
as well as its application in imagined [bsgom bya’ dkyil ‘khor] vs. drawn [bri bya’s dkyil
‘khor] mandalas. [See Sde dge 194.1-195.1]
470
For partial translations of the tantra, see Elder 1978 and Skorupski 1983. For a short
philological assessment of the tantra, see Szanto 2013. For an overview of the Indian and
Tibetan literature on the Sampu†a embraced by the Sakya tradition, in particular, see
Sobisch 2008. I have strong intuitions that pursuing the Sampu†a is of value to the
project, motivated in part by a comment relayed by Drapa Gyatso of the IBA. He recalled
! 156!
the Sampu†a as one of the “three tantras” of Hevajra; the text provides a wealth of
information on the completion stage of sådhana practice.471 The three tantras are: the
Hevajra Root Tantra; the Vajrapa•jara Tantra, often labeled an “uncommon” explanatory
tantra (meaning in pertains only to the interpretation of the Hevajra Tantra); the Sampu†a
Tantra, a “common” explanatory tantra applied to both the Hevajra and Cakrasaµvara
systems.
The classifications of root and explanatory tantra and, further, of common and
uncommon explanatory tantra are important for understanding the underlying structure of
Ngor chen’s commentarial method. These systems of classification provide another
example of the concerns with articulating relationships between different varieties of
texts (and bodies) expressed by the authors of the body mandala debate. We first
encountered the genre of explanatory tantras [Tib. bshad brgyud Skt. vyåkhyå] in our
discussion of Vajramålå in Chapter Three. How explicit is the relationship of an
explanatory tantra to the ‘root’ with which it is connected? In the case of the Sampu†a,
Elder claims that while the Hevajra Tantra is cited multiple times, the Guhyasamåja is the
only tantra explicitly referred to by name.472 Szanto 2013 identifies the Sampu†a as “a
compilation from most major tantras, such as the Hevajra, the Herukåbhidhåna, the
Catußpîtha, etc..”473 Lee 2003, likewise, suggests that “the principal purpose of the SPT is
to synthesize several Yoga tantras and Yoginî tantras in terms of theory and practice.474
However, the notion of the “common” explanatory tantra, one that can be shared by
multiple tantric systems, poses some questions about Tibetan hermeneutic and exegetical
practices.
Why are these systems for classifying the Sampu†a relevant for the body mandala
debate? They may be used to set the boundaries for textual interpretation. Sufficient
ambiguity in the relationship of texts allows for the potential modification of those
boundaries. The rules of exegesis may therefore be restructured through polemical
encounters like the body mandala debate. We will continue to explore these issues in our
examination of how Mkhas grub and Ngor chen use the Vajramåla and Sampu†a
explanatory tantras.
How have authors understand the scope of application for a tantra like the
Sampu†a? Verrill’s translation of an excerpt from a text he refers to as “Notes on Ngor
chen kun dga’ bzang po by Dpal gyi brgyal mtshan in A mes zhabs”475 suggests the scope
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
how the great 20th century Sakya master Khenpo Appey Rinpoche suggested that Mkhas
grub’s issue with the Hevajra body mandala was based on a misunderstanding of the
Sampu†a Tantra.
471
Sobisch 2008, p.6.
472
Elder 1978, 15.
473
Szanto 2013, p.5. Szanto also provides clues for dating the text such as the
“conspicuous absence” of the Kålacakra (c.1030) from its contents as well as the
likelihood it was quoted by Durjayacandra. [c. 1000c.e.]
474
Lee 2003 dissertation, p. 35.!
475
This is the same text Sobisch 2008 is working with as “Notes” but a different edition.
(“Notes for the Correct Explication of How to Enter into the Writings of the Venerable
! 157!
was vast. According to that text, Bu ston regarded the Sampu†a as applicable to the
interpretation of thirty-two tantras and for Bsod nams rtse mo, to sixteen or seventeen
tantras. The excerpt continues: “[in addition to the Two-Part Hevajra Tantra], it is most
importantly an explanatory tantra for the Guhyasamåja..., Vajra Catu®pitha...,
Cakrasaµvara..., and [four others] ending in Guhya, because it clarifies their uncommon
philosophical systems.]” This list includes both yoga and higher yoga tantras of both the
father and mother classes. 476 The variety in this list raises the question of whether or not
all of the parts of the Sampu†a are equally applicable to the understanding and practice of
all of these tantras. More specifically, for the purposes of the body mandala debate, can
the Sampu†a be applied to both the Cakrasaµvara and Hevajra tantras (higher yoga
“mother” tantras) in part or in their totalities or even to the Guhyasamåja (a higher yoga
“father” tantra)?477 Tsuda 1994 briefly assessed the state of the field of study of the
Sampu†a, comparing it with other tantras of the Cakrasaµvara cyle as follows: “The
relations between these tantras and the principles of classification which have been
adopted by Bu ston and other scholars should be discussed after studying each of the
tantras carefully. Nearly everything is left to be done in this regard.”478 Although Tsuda
made that comment two decades ago, many questions remain unresolved.
Ngor chen introduces this first quote from the Sampu†a Tantra explicitly within
the context of Hevajra body mandala practice [kye rdor gyi skabs su] rather than the
Cakrasaµvara. It describes what has become familiar to us as the body mandala of the
support:
“As for the body, the mandala, pleasant, the four gates; so it is taught.
(It) abides surrounded by the pillars which arise (from) one’s own eight limbs.
Because of being equal in all dimensions, it is known as a square.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Sa skya pas: Opening Wide in a Hundred Directions the Dharma Gates to which All
Beings of Tibet are Guided”) Rje btsun sa skya pa’i gsung rab la ‘jug tshul legs par bshad
pa’i yi ge bod yul ‘gro kun bsgrod pa’i chos sgo phyogs brgyar ring du phye ba. Edited
by A-med-zhabs and originally composed by Chos dpal bzang po. Collected works, vol.
kha, fols. 384r-393v. Verrill cites another version of A-med-zhabs’s collected works,
Vol.21, p.76.2. This text will be explored in greater depth later in this chapter.
476
Verrill 2012, p.328.
477
Szanto 2013 has narrowed the scope, based upon a comment from Bu ston’s ‘Extended
Categorization of Tantric Classes” suggesting the Saµvara as the primary object of
interpretation for the Sampu†a. The quote from p. 429 of Bu ston’s Rgyud sde rnam
bzhag rgyas is cited in Szanto 2013 fn25. [See The Jewel Ornament of Tantric Classes:
the Classification of the General Tantric Classes. Rgud sde spyi’i rnam par gzhag pa
Rgyud sde rin po che’i mdzes rgyan. Lokesh Chandra (ed.) The Collected Works of Bu
ston (vol.Ba). New Delhi: International Academy of Indian Culture 1966.] I have
translated it as follows: “Being the explanatory tantra of many tantras generally, the
Saµvara is primary. This is on account of the distinguishing Vajrasattva Saµvara as the
main deity of the mandala of this one. (Another reason is) to be explained in the
introduction of the Saµvara. (And also since) the commentaries count it as an
explanatory tantra of Saµvara.”
478 !Tsuda!1994,!p.40!fn1.!
! 158!
As for the nature of body, speech and mind, the three cakras are taught as one.
As for the mountain, in the center of the head.
Likewise, the stage of rtsom chen and so forth.
Through the stage of the method of the time of the guru,
As for that, the mandala abides completely.479
The passage itself, taken together with its context within the Sampu†a Tantra, refers to
generating the body as the celestial palace and likely also as the cosmic elements as in the
Hevajra sådhana.480 The way Ngor chen uses the Sampu†a reveals some key aspects of
his strategy as a tantric commentator. For example, Ngor chen is using the Sampu†a in its
capacity as an undisputedly authoritative tantric text (of the explanatory tantra category)
to establish the centrality of the guru’s oral instructions at “the time of the guru” to the
successful completion of the Hevajra body mandala practice. He identifies the quote as an
explicit explanation [dngos su bshad pa] and then adds that even without such a direct
reference, there are implicit meanings [don la thob pa] that can be sought out for
clarification. He then introduces another authoritative source, the oral instructions of the
lineage gurus [bla ma brgyud pa’i man ngag]. He refers to the process of seeking support
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
479
lus ni dkyil 'khor nyams dga' bar; sgo bzhir ji skad gsungs pa'o; rang gi yan lag brgyad
byung ba'i; ka ba de yis bskor nas gnas; dngos po kun gyis mnyam pa'i phyir; gru [550.6]
bzhi par ni rab tu grags; sku gsung thugs kyi ngo bos ni; 'khor lo gsum ni gcig tu gsung; ri
ni mgo po'i ze'u 'bru la; ji ltar rtsom chen la sogs rim; bla ma'i dus thabs rim nyid kyis; de
ni rdzogs par dkyil 'khor gnas
This quote is found in 113b.3-.5 (i.e. p226) in the sde dge edition of the Sampu†a Tantra
[Toh 381].
480
Resolving all of the questions raised by this citation may not be possible at present. A
few suggestions are included here. The three circles likely refer to the three rungs of the
protective circle, here taught as one, a detail not mentioned in the sådhana quote. I am
indebted to Drapa Gyatso for suggesting this interpretation. [Personal communication
Spring 2012]. However, we can also see how the text may allow for interpretation of the
“three” as the cakras of body, speech and mind in the case of the Cakrasaµvara cycle.
Other details, like the four gates, also suggest reference to the Cakrasaµvara system in
which there are four gates at the heart for body mandala.
The reference to the stage of rtsom chen requires further investigation in the
context of the tantra itself; there, the verses that follow it describe the location and shape
of the cosmic elements in parts of the body in a similar way to that description quoted
from the Hevajra body mandala practice above.
The reference to the ‘method of the time of the guru’ [bla ma'i dus thabs rim] is
also somewhat difficult to interpret. My Tibetan mentors had some suggestions. Drapa
Gyatso of the IBA suggested the reference is to understanding the formation of body in
accord with the fifth instruction of guru [Personal communication, Spring 2012];
Shabdrung Rinpoche proposed the reference is to the completion stage practice and the
Vajråcårya initiation [rje slob dpon kyis dbang]. [Personal communication, Fall 2013]
Further clarification is needed on this point.
!
! 159!
from this particular type of source as ‘supplemental’ [‘kha bskangs]. Moreover, he
claims that while there are many methods of supplementing or filling in the blanks left by
the teachings, the oral instructions of Jetari prove most reliable for the Sakyapa masters.
As discussed above, Mkhas grub’s critique of the (Hevajra) body mandala
practice targeted both explicit and implicit varieties of interpretation: “(Such methods)
are not explained in the tantra or any authentic Indian system nor even in the implicit
meaning [don thob].”[KJ 255.4] While we referred to the more general Buddhist
exegetical framework of apparent meaning [nîtårtha Tib. nges don] vs. interpretable
meaning [neyartha Tib. drang don] in that discussion, we observed that Mkhas grub
provided no clear reference for don thob. Ngor chen’s response fills that gap by asserting
that the oral instructions of the lineage gurus may be used as a valid resource in the
exegetical process. Moreover, he uses a tantra, an undisputedly authoritative text, to
validate this creative process of supplementing from these instructions in determining the
ritual application of somewhat obscure tantric literature. He also begins to forge a bond
between the teaching of one particular realized Indian tantric master [mahåsiddha], Jetari,
and the lineage instructions of the Sakyapa gurus.
To determine the significance of the conflict over don thob within the body
mandala debate and the implications it holds for understanding the broader controversies
over textual exegesis within fifteenth-century Tibet, we will recall a few points from
scholarship introduced earlier in the dissertation. First, historically the Sakyapas have
been critiqued for the ambiguous origins of the Lam ‘bras teachings as “grey texts”
produced as a collaboration between Indian masters and Tibetan disciples and translators,
passed down orally for centuries before being written down.481 While the Sakya tradition
makes a distinction between the transmission of the Lam ‘bras according to the
“explanatory system” ['grel lugs] of scriptural exegesis based in the Hevajra Tantra and
the “oral instructions system” [man ngag lugs] whereby the practices themselves are
explained, both are regarded as valid.482 However, Cabezón has observed that Mkhas
grub expresses a characteristic skepticism on the part of the Gelupkas toward the oral
instructions. According to this view, “the way to a true understanding of Buddhism is not
through mystical oral tradition, passed down in secret from master to disciple, but
through long and arduous study and analysis of scriptures...”483 In synthesizing these two
forms of knowledge, Ngor chen demonstrates his prowess as a tantric commentator and
an iconophile, creatively linking different classes of representation in his exegetical
process.
Ngor chen proceeds with a description of the mandala of the support [551.3-
552.1] extracted from Jetari’s Sådhana of the Four Seals [phyag rgya bzhi yi sgrub thabs
].484 Each seal refers to one aspect of the mandala of the support, namely, of the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
481
Davidson 2005.
482!Davidson!1991,!12!
483
Cabezón 1992, p.418 fn53.
484
Jetari (Slob dpon dgra las rnam rgyal). Caturmudrå-sådhana. Phyag rgya bzhi yi sgrub
thabs gser bris bstan 'gyur 2695 Vol 75 pp.387-393; snar thang bstan 'gyur vol.76,
ff.142v-144v (pp284-288); Otani Beijing edition: 4690.
! 160!
generation of the body as the protective circle, as the cremation grounds, as the cosmic
elements and as the celestial palace. Ngor chen uses this quote to make his concluding
point on the mandala of the support and to issue his defense of the Sakyapa interpretation
of the practice. Ngor chen claims that Bu ston wasn’t denying the validity of the
Sakayapa method of generating the body as the mandala of the support in Hevajra
sådhana practice. First of all, Bu ston didn’t actually deny the generation of the mandala
of the support in the context of Cakrasaµvara (though he may indeed say that the
quotation from Nag po pa’s text is the only evidence for such a practice in the Indian
sources). Moreover, since the Sakyapa masters only accept the form of body mandala
cited and critiqued by Mkhas grub for Hevajra practice and not for Cakrasaµvara
practice, the oral instructions on Hevajra practice (of whose view Jetari’s writings are
representative) are not in error. By this point in Ngor chen’s text, it is clear that, as the
title of the work has indicated, his driving imperative is to defend the Sakyapa Hevajra
practice despite the fact that Mkhas grub never named the Hevajra system explicitly. The
centrality of the oral instructions to that system requires a display of legitimacy to support
the method of filling in the blanks from sources that were not strictly textual. The quote
from the Sampu†a helps Ngor chen to identify this practice of supplementing as a
legitimate means of attaining the implicit meaning of a tantric text.
IV. Ngor chen’s Reply Part Two: On the mandala of the supported
The second portion of Ngor chen’s argument [552.3-560.5] concerns issues of the
relationship of external and internal mandalas raised by Mkhas grub as well as the
distinction of body mandalas from practices such as nyåsa (the placement of seed
syllables and/or deities on the body). The majority of the sources cited by Ngor chen
here are clearly Cakrasaµvara-related, including references to the Samvarodaya and
Abhidhånottara ‘explanatory tantras’ as well as to sådhanas by Indian masters such as
Darikapa and Tilopa. Ngor chen’s basic strategy in the discussion of the relationship of
inner and outer mandalas is to show how these various accepted authoritative texts
contradict Mkhas grub’s critiques regarding the disjunction of the number of deities of
the outer mandala [phyi dkyil] and of body mandala and the lack of gods common to
both. Once again, the explanatory tantras (and their commentaries) play a crucial role in
Ngor chen’s reformulation of the issues.
Buddhaguhya and Bu ston both classify the Abhidhånottara as an explanatory
tantra exclusive to the Cakrasaµvara cycle.485 Tsuda 1994 problematized the unqualified
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
I have located this quote in p388.4-389.2 of the gser bris bstan 'gyur 2695 Vol 75 pp.387-
393. It also appears on p286.5-287.2 of the snar thang bstan 'gyur vol.76, ff.142v-144v
(pp284-288), however the print is difficult to read; there is also an Otani Beijing edition:
4690.
485
See English 2002, p.7 and Gray 2007, p16 respectively. Gray cites both the
Abhidhånottara and the Yoginîsaµcåra (to which we referred above in our discussion of
Bu ston) as sources providing the “correlations to body parts and consitutents” not
included in the Cakrasaµvara tantra itself (Gray 2007, p.18 fn 60). Gray dates the
Abhidhånottara together with the Cakrasaµvara itself among the oldest strata of texts of
this tantric cycle; he bases this claim upon references to these texts within other
! 161!
application of the label “explanatory tantra,” adding depth to our understanding of
different perspectives on the status of the various types of tantric literature of the
Cakrasaµvara cycle. On the vast amount of “internal evidence” still to be collected from
these tantras, Tsuda remarks: “we must be content with the bare fact that some mutual
relation exists between the Laghusaµvara, the Samvarodaya and the Abhidhånottara
which, apart from the Yøginîsamcåra, can also be taken as a müla-tantra.”486 Looking
beyond the formally doxographical activities of scholars like Bu ston to tantric exegesis
and polemics as exemplified by the body mandala debate will contribute to the
understanding of how such classificatory schema are applied in new ways and,
potentially, thereby transformed.
Typical representations of the Cakrasaµvara “outer mandala” include the most
basic, with five deities, the main father and mother deity in union surrounded by four
dåkinîs, or one with thirteen deities, adding the eight goddess guarding the gates and
corners of the mandala. The body mandala typically enumerates sixty-two deities made
up of the main father and mother, the four dåkinîs and eight goddesses and adds twenty-
four dåkinîs with their hero consorts located at various bodily sites. These sites
correspond to twenty-four sacred sites of the Indian landscape. The Wish-Fulfilling Jewel
Sådhana [sgrub thabs yid bzhin nor bu is] is an example of a text cited by Ngor chen that
provides a complementary description of the body mandala practice.487 It first describes
what we might call the “outer mandala,” though it is to be cultivated in the womb (of the
consort presumably). This mandala is composed of five Heruka with consorts in the
cardinal directions and center, with skullcups in the intermediary directions. Then the
practitioner transitions to the body mandala of the main deity, described by Nåropå as
follows: “When cultivating the body mandala in the body of that main deity, there are the
stages of self-consecration [rang byin brlabs pa'i rim ] and the transformation [gyur pa ]
of the superior body mandala.”488 First, one imagines the main deity and consort at the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
explanatory tantras such as the Saµvarodaya (while acknowledging that Tsuda has
problematized the use of the term “explanatory tantra” to refer to the latter). (Gray 2007,
p20) As with many of the tantras, the issues surrounding the dating of the texts of the
Cakrasaµvara cycle are highly complicated and remain unresolved. For a partial
translation of the Abhidhånottara, see Kalff 1979. Gray is currently working on a new
translation of the Abhidhånottara.
The Abhidhånottara is one of many Buddhist tantras within which Sanderson has
located sections derived from ¸aiva tantras.( Sanderson 1995 as cited in English 2002 p7)
English identifies chapters nine and fourteen in particular as sources for descriptions of
the Cakrasaµvara mandala. (English 2002, p.109)
486
Tsuda 1994, 45. See Tsuda pp.40-45, in particular, for discussion of these issues in
dialogue with the tantric classifications of both Bu ston and Tsong kha pa.!
487
This text is likely Nåropå’s dpal 'khor lo sdom pa'i sgrub thabs yid bzhin nor bu,
translated by Marpa chos kyi blo gros) in mkha' 'gro snyan brgyud kyi yig rnying pp.
299-320; Drapa Gyatso had also suggested that the text referred to here is by Nåropå.
488
Ngor chen 556.4: de'i*gtso*bo'i*lus*la*lus*dkyil*bsgom*pa*na; rang byin brlabs pa'i rim
pa ste; lus kyi dkyil 'khor mchog gyur pa
!
! 162!
center of the lotus at one’s heart with the four dåkinîs at the petals in the cardinal
directions, with skull-cups of nectar on the intermediary petals. These deities are
common to the descriptions of both inner and outer mandala. Ngor chen refers to
elements of the twenty-four sites, of the twenty-four heroes (the male consorts of the
yoginîs linked with those sites) as well as of the eight guardian goddesses that reside on
the bodily gates or orifices in the inner mandala.489
The transition between ‘outer mandala’ to inner or body mandala, however,
appears to vary from one sådhana to the next. The variety of descriptions even among the
texts selected by Ngor chen texts suggests that the Indian sådhana literature (and perhaps
even the explanatory tantras) varied widely in their descriptions of this ritual transition
within Cakrasaµvara body mandala.490 The variety of practices was troubling enough to
prompt Ngor chen to deal extensively with the relationships and transitions between inner
and outer mandala in another body mandala text not engaged within this dissertation491
In the quote aboce, from the Wish-Fulfilling Jewel Sådhana, the transition from external
to internal mandala is is effected through rang byin brlabs pa [svådhiß†håna].492
Determining the meaning of rang byin brlabs pa [svådhiß†håna ] in this particular context
is important for understanding this form of the body mandala practice. byin brlabs itself
has undeniable meaning as “blessings.” These blessings infuse the body of the
practitioner or even other objects like precious pills or blessing cords, momentos of ritual
encounters with a sacred person or place. One possible translation of rang byin brlabs pa
is therefore, “self-blessing.” “Self-consecration” or “self-empowerment” are also
possible translations. The Sanskrit term adhißthana denotes the act of installing a
presence within an abode and is commonly used to describe the infusion of sacred
presence into an image in image consecration rituals. Abhißekha, on the other hand, is
consecration proper; the root abhi-ßic conveys the act of anointing in a royal consecration
and has been extended to the consecration of practitioners in tantric initiation rites. In
Tibetan, the inconsistent use of dbang to refer to both Sanskrit terms has generated
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
489
These citations are are elliptical, in that they use a limited number of members of a set
to refer to the set as a whole. Both names and locations are provided for some of these
members.
490 !Tsunehiko Sugiki’s work illuminates this diversity brilliantly. See his “Five
Types of Internal Mandala Described in the Cakrasaµvara Buddhist Literature: Somatic

Representations of One’s Innate Sacredness.” See also Suguki 2009. “The Structure and
Traditions of the Systems of Holy Sites in the Buddhist Saµvara Cycle and its Related
Scriptural Cycles in Early Medieval South Asia: The Geography of Esoteric Buddhism in
the Eyes of the Compilers of the Scriptures,” Genesis and Development of Tantrism,
edited by Shingo Einoo. Tokyo: Institute of Oriental Culture, University of Tokyo,
Institute of Oriental Culture Special Series, 23: 515-562.!
!
491
This is Ngor chen’s Explication of Difficult Points of the Ghantapa Cakrasaµvara
Body Mandala [dpal ‘khor lo bde mchog lus kyi dkyil ‘khor gyi sgrub pa’i thabs kyi dka’
bas gnas kyi mtha’ rnam par dpyad pa ].
492
Another quote Ngor chen provides from Darikapa’s sgrub thabs de kho na nyid bsdus
pa, describes this transition after generating the ‘outer circle’ of thirteen deities, as
proceeding from the cultivation of the nirmåñakåya to that to the sambhogakåya.
! 163!
confusion. Generally, my Tibetan mentors used the language of “transformation” vs.
“blessing” for translating rang byin brlabs pa [svådhiß†håna], emphasizing its role in
transforming ordinary nature to Buddha nature. Drapa Gyatso explained how in body
mandala practice, all ordinary forms must be transformed into the forms of the deity and
celestial palace; dissolving one form into another is often a key moment in this
process.493
The use of the term “transformation” of course has serious doctrinal implications,
suggesting a change from one state to another.494 Some of these implications were
explored in Chapter Four in the context of Mkhas grub’s critique of interpretations of the
quotation from Ghantapa’s “Condensed Activities of the Cakrasaµvara initiation” [dpal
‘khor lo sdom pa’i dbang gi bya ba mdor bsdus pa]: “These sentient beings, are not
separated from the naturally established mandala.”495 Mkhas grub raised ontological
tensions between the claim that the body is inherently pure and the view of ritual as a
transformative and efficacious event:
“They speak of the body mandala as an unfabricated mandala [ma bcos ba'i dkyil 'khor],
(and) they don’t understand in what way the bodies of sentient beings are primordially [
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
493
Personal communication, IBA, Spring 2012.
494
See Sharf 2003 for a critique of the use of this term in the context of Japanese Shingon
ritual.
Davidson 1992 describes the ritual phase of svådhiß†håna in the Hevajra
abhisamaya practice. He shows how the body mandala practice is initiated during the
generation stage after the merging of the jnånasattva and samayasattva. Previous to this,
the meditator has visualized the union of Hevajra and Nairatmya as the causal Vajradhara
(Hetu-Vajradhara ). According to Davidson, it is the “cause (hetu), in the sense of the
ground of purification (sbyong gzhi), (and) operates as the fruit and to utilize the five
types of gnosis of the Buddha as the path.” This causal Vajradhara is dissolved, and the
fruitional Vajradhara (Phalavajradhara) produced. Once the body mandala is created, the
meditator requests consecration from the deities, and the result is the “sealing” of “four
internal centers.” All of this is performed regularly to maintain the “stream of
consecration.” [See Davidson 117-119]. Davidson refers to Ngor chen’s gNad kyi zla zer
235.3.1-258.2.4 on rjes chags (the first part of the process described here )and 258.2.3-
261.2.1 on dbang ; the fact the a different term is used here, dbang vs. rang byin brlabs pa
suggests further inquiry is required.
See also Tsong kha pa’s writings on the stage of svådhiß†håna in the context of the
ritual system of the Guhyasamåja [for example, see Kilty 2013]. See Wayman 1977
(2005 Reprint) pps. 170-173 on the place of svådiß†håna within Candrakîrti’ six-
branched yoga and Någårjuna’s five stages.
!
495 !234.1!'gro ba 'di dag rang bzhin gyis; sgrub pa'i dkyil 'khor gnyis med pa'o
! 164!
gdod ma nas ] mandalas. Therefore, they say that what already existed [sngar yod ] is
cultivated through visualization [gsal 'debs pa'i tshul gyis bsgom pa].”496
Mkhas grub’s argument played upon the fundamental tension between subitism and
gradualism in the Tibetan Buddhist philosophical approach together with doctrinal
skepticism toward the mind’s tendency to impute false constructs upon reality. Building
upon this foundation, he challenged his opponents’ views on the role of ritual and the
status of the human body by juxtaposing the categories of fabricated and unfabricated and
what naturally exists with what is perfected through practice in interpreting body
mandala. The challenges posed for translating and interpreting the language of
transformation in the body mandala debate texts invite us to engage body mandala in
understanding Tibetan Buddhist ritual theory. How do tantric ritual exegetes like Ngor
chen and Mkhas grub explain the function of rituals like body mandala? Do rituals
actually change one’s unenlightended form into an enlightened one? Do they effect a
change in perspective or a change in matter, and on what basis does this purported
transformation occur?
Transforming the Basis
The last page or so of this portion of Ngor chen’s text [559.3-560.5], distinguishes body
mandala from nyåsa, referred to here as ‘placing deities on the body’ [lus la lha dgod pa],
and presents us with some additional points to reflect upon. First, Ngor chen mentions all
three tantric systems here by name: the Guhyasamåja, Cakrasaµvara and Hevajra. In
addition, the sources he cites are less predictable and may contain clues for better
understanding how he envisions the relationship of body mandala practices between these
systems.497 Finally, these passages provide a glimpse of Ngor chen’s views on the body
as the basis for or support of tantric practice and the role of body mandala practice in
effecting embodied liberation. Likewise, they give us the opportunity to briefly revisit
issues discussed in Chapter Four, issues of fabrication and imputation, the nature of the
body as a basis for generating deities, and the role of imagination in realizing tantric
ritual goals.
Earlier in the dissertation, we discussed how there are a variety of such techniques
of imagining deities or seed syllables on the body such as nyåsa and hasta-püja-vidhi.
Mkhas grub directly challenged the confusion of such techniques with the actual
transformation of the bodily basis effected through body mandala practice; he argued that
rather than merely locating deities on the body in a superficial way, body mandala entails
a more holistic transformation of this basis into enhlightened form:
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
496 !lus kyi dkyil 'khor ni ma bcos ba'i dkyil 'khor du gsungs pas; [234.2] 'gro ba thams cad
kyi lus gdod ma nas dkyil 'khor du yod pa la yod par ngo ma shes pa shes par bya ba'i
phyir du sngar yod gsal 'debs pa'i tshul gyis bsgom pa yin no zhes zer ro
!
497
Specifically, Ngor chen refers to the “Tantra of Six faces,” which appears to be a
reference to the Krßñayamåri tantra, as well as to Ghantapa, Åcårya Praj•årakßita,. The
latter two references are Cakrasaµvara-related. However, ngor chen refers to the school
of a slob dpon sgron ma can in the case of Hevajra as well.
! 165!
The point of what’s called cultivating the body mandala doesn’t mean only merely
cultivating a deity on each place on the body. Rather, one establishes such and such a
part of the body as the foundation [bsgrub pa'i gzhir byas nas] for such and such a deity
[250.6] and thus cultivates the deity.498
Likewise, Mkhas grub states:
(In piercing to the pith through body mandala practice) moreover, having taken that
particular portion of the aggregates, skandhas, elements and sense spheres and so forth as
the foundation for establishing the deity [lha bsgrub pa'i gzhir byas nas], they are
completely transformed [yongs su gyur pa ] and must be generated as that particular
deity.499
In Chapter Four, we discussed how rendering the body as the proper basis for practice
[bsgrub pa'i gzhi] was a defining aspect of body mandala for Mkhas grub, one that at
least partially reveals his understanding of the role of the body and imagination in tantric
practice. Moreover, Mkhas grub distinguished body mandala from “outer mandala” by its
unfabricated basis and its unique capacity of “piercing to the pith” of the body [lus la
gnad du bsnun]. The latter refers to a process of making the elements of the subtle body
malleable [las rung du gyur]. In this section, we will see how Ngor chen regards the
body as the basis for generating deities [bskyed gzhi] and how he understands the nature
of this basis to also be inextricably tied to the psycho-physical constituents of the subtle
body.
Ngor chen begins by describing the purification of the sense spheres, a
preparatory practice resembling body mandala in some ways. In terms of the ritual
context, he appears to refer to the stage of generating emptiness, in which the practitioner
visualizes deities on the sense spheres as a means for merging with the body, speech and
mind of the j•ånasattvas.500 Ngor chen provides a quotation from the “Tantra of Six
Faces,” [gdong drug gi rgyud] that describes applying deities such as Moha-vajra to the
sense spheres to empty the sense orifices; then, one uses the the seed syllable or name
letter to generate the form (of the deities presumably) together with the attribute of their
respective buddha family. [559.5] Ngor chen articulates the distinction of such
preparatory practices from the body mandala proper by showing how: “any part of the
body, in particular the three (main subtle body consituents) of channels, winds, and
drops, etc... must be the cause [rgyur byas pa ] of the body mandala.”501 So what is the
relationship between Mkhas grub’s notion of turning the body into a basis for generating
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
498
lus dkyil bsgom pa zhes bya ba'i don ni lus kyi gnas so sor lha bsgom pa tsam mi zer
gyi ; lus kyi cha de dang de lha de dang de [250.6] bsgrub pa'i gzhir byas nas lha bsgom
pa la zer ba yin te
499
'dir yang phung khams sky mched sogs kyi cha de dang de lha bsgrub pa'i gzhir byas
nas [252.6]; de dag yongs su gyur pa las lha de dang der skyed dgos so
!
500
Drapa Gyatso, Personal communication, Spring 2012.
501
lus [559.6] dkyil ni; rnal 'byor pa rang gi lus la gnas pa'i rtsa thig rlung gsum sogs lus
kyi cha shas gang yang rung pa cig rgyur byas pa dgos te
! 166!
divine forms as indicated by bsgrub pa'i gzhir byas nas in order to “pierce to the pith of
the body” [lus la gnad du bsnun] and Ngor chen’s focus upon subtle bodily consituents as
the “cause” of body mandala? In order to apprehend the impact of Ngor chen’s
understanding of the body as a cause, we must first consider some fundamentals of the
Sakyapa perspective on tantric practice.
The standard Sakyapa view distinguishes reality in terms of three continua [rgyud
gsum], the cause, the path, and the result, corresponding to the “universal ground,” the
body and the mahåmudrå respectively. However, the distinction of the three continua is
complex, as attested by the description of tantric practice by the Sakyapa patriarch Grags
pa rgyal mtshan as “taking the result as the path” [‘bras bu lam du byad pa].502 Stearns
2006 provides an insightful synopsis of this view:
“The result-the essence or innate true nature of a living being-is actually present at all
times. Were this otherwise, the practices of the spiritual path would be futile. This
essence is not transformed by the practices of the path because it is beyond conceptual
elaboration. The qualities of a Buddha, or enlightened being, are obtained instead
through removing obscurations and transforming one’s body, speech and mind. This
process involves concentrated focusing on the essence itself. The result that is already
present at the beginning is thereby taken as, or made into, the spiritual path by means of
tantric techniques.”503
The archetypical clash of sudden vs. gradual approaches embodied in the figures of
Kamala!ila and Mohoyen structures the very narrative foundation of Tibetan Buddhism.
Therefore, it is not difficult to anticipate that this perspective of “taking the result as the
path” would trouble some Tibetan thinkers. In Chapter Four, we explored tensions
between theories of primordial enlightenment and graded spiritual practice in Mkhas
grub’s writings. We contextualized these tensions as attempts to resolve the methods and
aims of Mantranaya practice with those of the Påramitånaya. In the case of Mkhas grub’s
text we saw how the role of the body in tantric practice often produced anxiety about
category confusions of the worst kind, between saµsåric and enlightened bodies. These
anxieties often prompted detailed explorations of the chain of causality linking these two
types of bodies; one example of such an explanation was Tsong kha pa’s description of
the “similar type cause.” Employing the rhetoric garnered from charged issues of the era
such the Buddha nature debates, Mkhas grub exploited the full scope of the tensions
between sudden and gradual practice and enlightened and saµsåric bodies. Perhaps the
most radical aspect of Mkhas grub’s text is the manner in which he brings all of these
tensions to bear on his analysis of a fundamental aspect of tantric practice: imagining
oneself as a buddha.
Mkhas grub challenged the very act of imagination itself, using pramåña theory to
show how reliance on false imputations resulted in a mistake consciousness [log shes].
According to Mkhas grub, such a consciousness could never serve as the cause for
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
502
Stearns 2006 fn1 traces this reference to Grags pa rgyal mtshan’s commentary on Sa
chen’s Explication of the Treatise for Nyak (11b) translated by Stearns in this same
volume.
503
Stearns 2007, p1.!
! 167!
enlightenment. [236.3-.4] We also touched upon connections between the “moderate
realism” of the Gelukpa philosophical tradition and their approach to tantra. In doing so,
we considered some basic philosophical differences from the Sakyapas. In introducing
the Sakyapa tantric perspective and how it structures the terms of the body mandala
debate, it will be useful to keep these basic philosophical concepts and related tensions
with the Gelukpa in mind. Therefore, we will briefly revisit Dreyfus’s distinction of
Mkhas grub’s (and rGyal tshab’s) view on pramåña from that of the Sakya patriarch Sa
skya pan di ta:
“the explanation of valid cognition as the elimination of a superimposition is an obvious

truth. This has important consequences for their understanding of perception...For Sa-
pan, perception does not eliminate superimposition, for its function is strictly limited to
the passive appearance of things.”504
The Gelukpas envision the goal of philosophical practice as the removal of

‘superimpositions’ or ‘imputations’ in order to actualize a form of “direct perception”
[mngon sum Skt. pratyåkßa] of the nature of reality. The Sakyapas, on the other hand.
see the mind itself to be bound by dualistic perceptions; for them, such direct or
unmediated perception is impossible. For the Sakyapa, all phenomena are just the
appearance of the mind. The manner in which this particular conviction translated into
their tantric perspective may have inspired accusations that they embraced a Cittamåtrin
view, fueling a clash with the Pransangika Madhyamaka view embraced by the Gelukpa.
One of Ngor chen’s most respected polemical texts, the Overcoming Objections to the
Three Tantras [Rgyud gsum ngan ‘joms], is a defense against such accusations (and the
particular accusation that Virüpa himself was a Cittamåtrin.). Now that we have outlined
the basic parameters of these philosophical tensions and planted the seeds for interpreting
their relevance to the tantric perspectives of Mkhas grub and Ngor chen respectively, we
will consider the Sakypa view of tantric practice and the role of the body as a means of
liberation in more depth.
For Ngor chen, the psycho-physical constituents of the subtle body, the channels,
winds and drops are the cause of the body mandala. By properly ‘aligning’ those
constituents through imaginative activity in body mandala practice, the mind can attain
realization. This kind of imaginative activity focused on the body’s vital points and
energies is not a mental imputation [blos brtag pa ] because those points and energies are
“naturally established.” Thus Ngor chen states:
The five channel cakras are naturally established [rang grub du yod pa] in the vajra body.
Having generated the goddesses and the drops in the center of the five cakras together
with outer husk channels as the five yab-yum deities, one visualizes them within
individual sites within the body. Therefore, this is not a mental imputation [blos brtag
pa].
As for the twelve deities of sense objects and faculties and the ten wrathfuls,
both the explanation in the context of the body mandala of the Årya cycle of the
Guhyasamåja and the explanation here in the oral instructions are undifferentiated in
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
504 !Dreyfus 1997, 376-77!
! 168!
terms of the arrangement and basis of generation [bskyed gzhi]. So these lack any error
as far as not becoming the body mandala.505
Here Ngor chen establishes the common ground of the Hevajra and Guhyasmåja
systems: the ground or basis of generation [bskyed gzhi]. He also describes a particular
kind of body, the vajra body [rdo rje’i lus], the body of the tantric practitioner. In this
passage, Ngor chen describes the cakras as “naturally established” [rang grub du yod pa].
506
This term is quite similar to rang chas su yod pa, the very term Mkhas grub used to
emphasize the superiority of the body mandala as an “unfabricated” mandala. Likewise,
the term blos brtag pa or “mental imputation” is also familiar from Mkhas grub’s writings
and from our extensive discussions in Chapter Four. There we observed a parallel
between his use of the categories mental fabrication [blos bcos pa] and mental imputation
[blos btags pa].507 While Ngor chen does not elaborate upon the problem of “fabrication”
[bcos ma] in the manner that Mkhas grub does, the authors express a common concern
with navigating the delicate balance between the naturalness and artifice in interpreting
body mandala. This concern is not particular to body mandala alone but rather underlies
all tantric ritual acts of imagination, in particular, acts of identifying oneself as a buddha
in deity yoga.
An account of the implications of imagining oneself as Hevajra by Jamyang
Khyentse Wangchuk (1524-68) serves as a counterpoint to Mkhas grub’s diatribe on the
problems of generating oneself as a Buddha discussed above:
“Briefly, this ordinary body arises as just a confusing appearance, in which the very
essence of the united lucidity and emptiness of the mind is not recognized. No body of
saµsåra exists as the ground of purification established outside this ordinary mind. When
precisely this momentary pure awareness arises in the form of the Hevajra of the time of
the path, those grounds of purification transform into the essence of the purifying path.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
505 rtsa 'khor lo [560.4] lnga rdo rje'i lus la rang grub du yod pa de; lha mo rnams dang
'khor lo lnga'i dbus kyi thig le phyi shun gyi rtsa dang bcas pa gtso bo yab yum lngar
bskyed nas; lus kyi nang so so'i gnas su bsgom pas blos brtag pa ma yin no;!yul yul can
gyi lha bcu [560.5] gnyis dang khro bo bcu ni; dpal gsang ba 'dus pa 'phags skor gyi lus
dkyil gyi skabs su bshad pa dang man ngag 'dir bshad pa gnyis; dgod pa dang bskyed
gzhi spyi tsam505 la khyad par med pas; 'di dag lus dkyil du ma song [560.6] ba'i skyon
med do505
I am grateful to Lama Kunga Thartse Rinpoche for directing my attention to the fact that
the Hevajra system is indeed at issue here. Personal communication, Fall 2010.!
506
This term is discussed in depth by Willa Blythe Miller in her 2013 dissertation on
Yang dgon pa’s thirteenth-century text, Secrets of the Vajra Body. Miller investigates the
author’s use of the term dngos po’i gnas lugs translated as “the nature of things” or “the
nature of material substance” to describe ordinary vs. enlightened embodiment.
Davidson 1992 fn 47 provides a helpful reference to Grags pa rgyal mtshan’s rgyud kyi
mngon rtogs rin po che’i ljon shing [Complete works Vol. 3, pp.29.1.3ff] for a detailed
account of the Sakyapa view of the “vajra body.” On the wider use of the term in early
Buddhist literature, see Radich 2007 p1485.
507
See especially Ocean of Attainment, 251.4.
! 169!
Because the result of purification, or what is to be obtained, is also not established
outside one’s mind, precisely this is Hevajra residing in the thirteenth spiritual level of a
vajra holder. Thinking that, apply the pure appearance and take it as the path. At that
point, the ground, path, and result have become indivisible in the perception of the yogin.
Precisely that is the indivisibility of saµsåra and nirvåña.”508
As we have mentioned, the Sakyapa philosophical view of perception defines all

thoughts and forms as “appearances” or representations in the mind. As
“perspectivalists,” they define enlightenment as a radical shift in perspective, a re-
ordering of the boundaries that produce the illusion of an apparent separation between
ordinary and enlightened being.509 Tantric ritual provides many opportunities to enact
this breaking down of boundaries, through acts of dissolution that blur the distinction
between one variety of form and the other. In the Sakya tantric context, the ordinary
body is just another representation appearing to the mind; the goal of tantric practice,
embodied in the form of the deity, also exists only in the mind, but, as a form of
awareness itself, it may be used to structure one’s practice. The Sakyas identify the body
with the path or method of practice, and thus it is not a cause per se, but rather a kind of
means to an end. However, as shown in this quotation, there is often an intentional
blurring of the three categories of cause, path and result in describing the process of
overcoming the ultimate duality of saµsåra and nirvåña.
Jamyang Khyentse Wangchuk used the term “ground of purification” [sbyang
gzhi] to discuss the illusionistic quality of the ordinary saµsåric body; by generating the
divine form in the mind, the ground of purification becomes the path. The term the
sbyang gzhi is familiar from our discussion of Mkhas grub’s writings in Chapter Four and
his concerns with the aspects of deity yoga construed as acts of mental fabrication [bcos
ma] and imputation [blos rtag pa]; one example of Mkhas grub’s use of the term occurred
in addressing the question of whether on generates deities on all the hands of the main
deity or just on the two principal hands. He concluded that it is only necessary for the
practitioner to generate deities on the two main hands of the body (as the other imagined
hands are a fabricated basis for practice). There, Mkhas grub referred to the two main
hands as the “ground of purification or the ground of generating deities” [sbyang gzhi'am
lha'i bsgrub gzhi]. [261.1] As mentioned above, Bentor 2006 suggests that Geluk and
non-Geluk perspectives on the sbyang gzhi diverge, with several non-Gelukpa schools of
thought splitting the category of sbyang gzhi into the pure basis and the sbyang bya, what
is to be purified (i.e. the defilements).510 In this regard, Bentor states, “While in the Dge
lugs pa tradition it refers to the ordinary saµsåric state, according to others it is the true
nature of things.”511
The category of the ground or basis provided a way for thinkers to deal with the
ambiguous role of the body in tantric practice. In other words, referring to the body as a
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
508
Stearns 2006, 484. Translation of Summarizing Notes in the Outer Creation Stage by
Jamyang Khyentse Wangchuk (1524-68).
509
I am grateful to John Dunne for his comments on the realist vs. perspectivalist
distinction. Personal communication, Fall 2011.
510
Bentor 2006, pp.196-7
511
Bentor 2006, p.197.
! 170!
ground or basis allowed thinkers to account for the causal dimension of the body’s
instrumental role in achieving liberation without blatantly designating it as a cause.512 As
expressed in the renowned quote from the Hevajra Tantra: “Great bliss resides in the
body. Casting off all conceptualizations (and) pervading all entities, what resides in the
body is not produced by the body.”513 In identifying the body as the cause of body
mandala practice, Ngor chen plays with the ambiguity in the frameworks of cause, path
and result and ground of purification, purifier and purified. While he carefully resists
explicitly defining the body as a cause of enlightenment, he maintains its centrality as the
ground for ritual action.
VI. Explanatory Continuum/ Explanatory Tantra [ bshad rgyud ]:

Corporeality and Textuality in the Body Mandala Debate
In commenting upon a verse from Virüpa’s Vajra Lines, “For the method continuum of
the body and so forth, there is the causal initiation with four triads, the seats and so
forth,” the Sakyapa patriach Sa chen remarks:
First, this is called the “method continuum of the body” because the alignment of the
dependently arisen connections in the body is the method that brings about the realization
of the mind, the universal ground, which exists in the manner of a seed or cause. The
“and so forth” includes the meaning “the body is the explanatory continuum” because
that mind that is the root of saµsåra and nirvåña can be realized by means of aligning the
dependently arisen connections in the body.514
This well-elaborated framework explains the relation of body and mind in tantric practice
and their respective roles in the liberation process. Sa chen identifies the role of the body,
commonly referred to as as the “method continuum” [thabs rgyud], here as the
“explanatory continuum” [bshad rgyud], the very same term used for “explanatory
tantra.” The body provides the necessary context for realizing the mind. The final section
of this chapter will consider the relation of body and mind from the Sakyapa tantric
perspective in conversation with the metaphor of explanatory and root tantras. Building
upon our discussion of the nature and function of the body as the basis for tantric
practice, we will begin by thinking more deeply about what it means to “align the
dependently arisen connections in the body.”
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
512
A similar dynamic may be at play in the use of the term “support” to designate the
body or even an image. We will pursue this possibility in the conclusion of the
dissertation.
513
Hevajra Tantra, Snellgrove 1.12. lus la ye shes chen po gnas; rtog pa thams cad yang
dag spangs; dngos po kun la khyab pa po; [2b.1] lus gnas lus la ma skyes pa’o [see sde
dge 2a.6-2b.1]
514
Stearns 2006, p28. Translation of Sa chen kun dga’ snying po’s Explication of the
Treatise for Nyak. In this text, Sa chen further identifies the body mandala as the inner
dependently arisen connection necessary for enlightenment, while the outer connections
are receiving the initiations from a nirmåñakåya emanation. See Stearns 2006, p.96.
Davidson 2005 translates bshad brgyud as “articulate continuity.” See his reference to
variations among the commentaries on this point. (Davidson 2005, Appendix 2 Fn3)
! 171!
I had the privilege of an audience with His Holiness Sakya Trizin in which I
introduced my research on the body mandala debate. When I asked His Holiness about
the role of the body in tantric practice, he remarked that it is difficult to purify the mind
while focusing on the mind itself. In other words, by focusing on the body, one can more
effectively purify the mind.515 In clarifying this perspective, His Holiness cited a
metaphor of the mind as the scent inhering in the flower of the body. This metaphor is
introduced in the Hevajra Tantra [II.2.36] in reply to Vajragarbha’s inquiry as to the
necessity of generation if the completion process [utpanna-krama] is so blissful:
“Just as the perfume of a flower depends upon the flower, and without the flower
becomes impossible, likewise without form and so on, bliss would not be perceived.”516
This metaphor is discussed in the Sras don ma, yet another commentary by Sa chen on
Virüpa’sVajra Lines.517 It also shares a context with the quote from Sa chen’s
Explication of the Treatise for Nyak discussed above. That context is the elaboration of
the three continua: the causal continua of the foundation [kun bzhi], the method continua
of the body [lus] and resultant continua, the mahåmudrå:
“The two, support [rten] and supported [brten pa], are undifferentiated like a mixture of
water and milk or the mixture of earth and water called mud. What is separate becomes
uniform. The container is like a flower, and the contained is like a scent. Likewise, the
scent residing in the flower would be imperceptible without the flower...518
Further along in the same section, Sa chen revisits the metaphor:

“Residing in the cause’ (means) the coarse and subtle wheels of the channel body. The
emanation body is said to spontaneously manifest in that; it is like a flower. The
supported [rten pa], the mind, the foundation [kun bzhi Skt. ålaya], is like the flower’s
scent, and the seed is flawless.”519
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
515
Personal Communication, Lumbini, November 2011
516
ji ltar me tog la gnas dri; me tog dngos med shes mi 'gyur; de bzhin gzugs sogs dngos
med pas; bde ba nyid kyang ngos med 'gyur//
Translation by Snellgrove, p 92. I am grateful to Kurt Keutzer for identifying the source
of this passage.
517 !Sa chen kun dga’ snying po. Sras don ma. Lam ‘bras literature Series 12, foilios 1r-
222v (pp.1-446); NGMPP L 170/4, Lam ‘bras gzhung sras don ma (xylograph from
Tyangpoche), 22 fols. [As cited in Sobisch 2008, Title list # 283.] (See also contemporary
two volume series)
!
518 !rten dang brten pa gnyis chu dang 'o ma 'dres pa'am; sa dang chi 'dres pa la 'jim pa zer
ba bzhin tha mi dad par gnas; grub sde gcig par gnas te; rten ni me tog lta bu la; brten pa
ni dri lta bu ste; ji ltar me tog la gnas dri; me tog dngos med shes mi 'gyur
Sa chen kun dga’ snying po. Sras don ma (Contemporary series, Lam ‘bras stod cha)
p.33.!
519
rgyus la gnas pa rtsa lus kyi 'khor lo rags pa dang phra ba; de la sprul sku rang bzhin
gyia lhun grub ces bya ste me tog dang 'dra la; rten pa sems kun gzhi me tog gi dri dang
'dra ste sa bon gas chag med lta bu'o Sa chen kun dga' snying po, Sras don ma (Lam ‘bras
stod cha) p.46.
! 172!
In these excerpts, Sa chen presents the mind as subtle and elusive, an aroma perfuming
the body. Through the method or explanatory continuum of the body, the practitioner
realizes the mind:
“[The method continuum] is also called the ‘explanatory continuum’ because the root to
be realized, which is the mind, is realized and mastered through the alignment of the
dependently arisen connections in the body, which is the agent of realization, and thus the
explanatory continuum.”520
Thus the body is not merely the ground for ritual action, but also the “agent of
realization.”
Ritual and exegetical action, body and text are poetically intermingled in these
statements. Just as the explanatory tantras provide access to the elusive meaning of the
‘root’ tantra, so the body provides access to the true nature of the mind. Stearns points to
the double entendre of the root and explanatory continuums [rtsa rgyud and bshad
brgyud] as root and explanatory tantras; he suggests that some Tibetan commentaries
deliberately exploited the semantic overlap of these terms.521 In light of the important
role of explanatory tantras in the arguments that make up the body mandala debate, this
double meaning might add depth to our understanding of the relation of root and
explanatory tantras and of mind and body.
In approaching the final chapter of the dissertation, we transition from exploring
the mechanics of ritual practice, their ontological implications for the status of the human
body, and the soteriological efficacy of tantric acts of imagination. However, we should
maintain this framework of interplay of body and text in progressing to investigate the
centrality of Ngor chen’s commentarial prowess to his polemical strategy. In the process,
we will observe how issues of textual authority are mapped onto the body, making it what
Foucault would call a “cultural text”522 or Sa chen would call an “explanatory
continuum,” or even an “explanatory tantra.” 523
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
The description appears in the context of discussing the vase consecration within a larger
section on the three continua.!
520 !Stearns 2006, p46. Translation of Sa chen kun dga’ snying po’s Explication of the
Treatise for Nyak.!

521
Stearns 2006, fn19.!
522
edited by Donn Welton (Malden, MA & Oxford: Blackwell Publishers, Ltd., 1998).
523
For one approach to body as text in the Kålacakra Tantra, see Wallace 2009.!
! 173!
Chapter Six: In Defense of the Hevajra Body Mandala
I. Ngor chen’s reply Part Three: Textual Authority
Ngor chen devotes over half of his text to responding to Mkhas grub’s claim that the
body mandala as rendered by “some Tibetans” is absent from the tantras or Indian
sources. [560.6- 580.6] We have already established the association of this version of
body mandala with the Hevajra practice despite Mkhas grub’s choice not to identify it as
such. However, we have also observed how Ngor chen’s sources for the first half of his
text are predominantly connected with the Cakrasaµvara cycle. The role of the Sampu†a
Tantra as an explanatory tantra common to both Cakrasaµvara and Hevajra systems
proved to be of particular interest. In this second half of his text, Ngor chen shifts to
focus predominantly on the Hevajra system. It is here that we begin to see how Ngor
chen has built his momentum in the first part of the text, in which he directly responded
to the form and content of Mkhas grub’s argument. In doing so, he has lead us to a space
in which he will reformulate the debate on his own terms. In this space, the text
transforms and Ngor chen actualizes the intent suggested by his title. Through a series of
maneuvers, Ngor chen asserts his defense of the Hevajra body mandala.
Ngor chen’s defense is divided into two parts. The first is based in the Hevajra
commentarial tradition, as made up of tantras, commentaries, and oral instructions. Ngor
chen includes the three tantras of Hevajra [rgyud gsum]: the Hevajra Root Tantra; the
Sampu†a Tantra, the common explanatory tantra; and the Vajapa•jara, the uncommon
explanatory tantra. For Ngor chen, the oral instructions as well as the commentaries of
other mahåsiddhas (namely Jetari, Darikapa, Indrabhüti and Vajragarbha) are also
essential components of the tradition. The second part of his argument is based in “other
tantric commentaries,” but, in actuality, it refers exclusively to the Vajramålå, the
explanatory tantra attributed to the Årya tradition of the Guhysamåja.
Ngor chen introduces his defense with a critique of the paucity of the opponent’s
(whom he addresses directly here as “you” [khyed]) scriptural knowledge. He offers a
citation that appears to be from Sakya Pandita to enhance to potency of his critique: “Just
because you don’t see it, doesn’t mean it isn’t there.”524 [561.1] In our discussion of
some of the basic differences between the Sakya and Geluk philosophical perspectives,
we encountered the problem of direct perception [mngon sum]. While the Gelukpa
accept the possibility of such unmediated contact with things as they really are, the
Sakyapa insist that all perception is mediated. For the Sakyapa, empirical observation
cannot account for all phenomena, and therefore “direct perception” is not an accepted
means of valid cognition [tshad ma Skt. pramåña]. In citing Sa pan, Ngor chen seems to
be playing upon this philosophical tension and responding in his own way to Mkhas
grub’s use of pramåña in his tantric polemics. This move is part of Ngor chen’s style;
while his own argument adheres far more closely to the standards of tantric exegesis than
of philosophical debate, he peppers his text with choice comments that engage with his
opponent’s approach. The tone of these comments varies from playful and ironic to more
impassioned. For example, in refuting Mkhas grub’s interpretation of Bu ston’s
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
524 !ma*mtong*phyir*na*med*pa*min!
! 174!
statement on the body mandala of the support, Ngor chen remarks: “ ...what’s the point of
responding to this lie about direct perception [mngon sum]?”525 Also, in the conclusion
of his text, Ngor chen refers to direct perception once again to critique Mkhas grub even
more explicitly: “Therefore, the proponent who denies the existence of the explicit
explanation from the tantra(s) and the authentic Indian system (does so) in sheer
disavowal of direct perception.”526 While Ngor chen himself does not rely upon pramåña
discourse in formulating the core of his defense, these comments not only play upon a
difference in philosophical perspectives, but also suggest a critque of Mkhas grub as the
master of pramåña who can’t see what’s sitting right in front of him.
We will now turn to closely examine how Ngor chen navigates the complexities
of tantric literature to reveal what is not immediately apparent about his sources and how
they support the validity of Hevajra body mandala practice. In the final portion of this
chapter, we will compare the two ‘versions’ of Ngor chen’s text. In focusing upon a few
key distinctions, we will enrich our portrait of Ngor chen to better understand the
significance of the body mandala debate for informing the connection of exegesis and
polemics in Tibetan scholasticism.
IA. On the Hevajra Tantras and Oral Instructions
Ngor chen begins, appropriately, at the root, with a quote from the Hevajra Tantra that
correlates the four goddesses, Locanå...with the seed syllables E VAM MA YA.527
[561.3] In Chapter Three, we explored the role of these goddesses in the evolution of the
Guhyasamåja body mandala practice. Ngor chen’s goal here is to show how what is not
readily apparent in the text of the Hevajra Root Tantra can be accessed, a process he
refers to as “attaining the meaning” [don thob]. Much of his supporting evidence is
derived from the Sampu†a Explanatory Tantra. First, he quotes the description of the
mandala of the support of cosmic elements discussed in detail above and insists that this
prototype is shared by both the Hevajra and Cakrasaµvara systems.528 Ngor chen
therefore provides an example for understanding the scope of exegesis of the Sampu†a as
an explanatory tantra common to both systems; we have already suggested the possibility
that this scope invited innovative acts of parsing and exegesis reinvented through
polemical and hermeneutic contexts like the body mandala debate. Then, Ngor chen
proceeds to the mandala of the supported with a few verses correlating each of the four
goddesses with a seed syllable (E WAM MA YA), a mudrå, a buddha family, qualities
identified with the perfections and immeasurables, and a cosmic element (earth, water,
fire, rlung).529 [562.2-.6] The verses also locate each of the goddesses in a cakra
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
525 ![N1!549.4]!mngon sum la rdzun du smra bas dgos pa ci zhig bsgrub
!
526 !des na rgyud dang tshad ldan gyi rgya gzhung rnams las dngos su bshad cing don thob
la yod bzhin du de las [588.1] bzlog ste smra ba ni; mngon sum la bsnyon pa kho nar zad
la
!
527
See the Hevajra Tantra, sde dge 2b.7.
528
Sampu†a Tantra, sde dge 113b.5 &.5-.6.
529
Sampu†a Tantra, sde dge 81a.5-81b.1.
! 175!
(nirmåña, dharma, sambhoga, mahåsukha) at a specific bodily site (navel, heart, throat,
lotus). 530
Ngor chen further expands upon the correlation of the goddesses with “cakras and
so forth” by referring to the fourth “cluster” [snye ma]. [562.6- 563.2] This is one of
several references, in fact, to the snye ma, a somewhat elusive term referring to a
commentary on the Sampu†a Tantra, Abhayåkåragupta’s Sam pu †a 'i 'grel pa man ngag gi
snye ma [Skt. Åmnåyama•jarî].531 Sobisch 2008, in reviewing the Hevajra literature
outlined by A mes zhabs (1597-1659), discusses the ambiguous status of
Abhayåkåragupta for the Sakyapa commentators, an ambiguity observed by A mes zhabs
himself.532 The problematic aspect of the Indian commentator’s approach is traced to a
critique by Sa pan of his inclusion of the four initiations in the lower tantras and both
generation and completion stages in the Amoghapå!a practice.533 Most importantly,
Sobisch points out that the earliest Sakyapa patriarchs, Sa chen, Sonam Tsemo and
‘Grags pa rgyal mtshan disregarded Abhayåkåragupta’s perspective in their interpretation
of the Lam ‘bras teachings. However, apparently Bla ma dam pa, upon whom Ngor chen
is regarded to have relied heavily for his writing s on the Lam ‘bras, referred to
Abhayåkåragupta often, and Bu ston took different positions with regard to the author on
different issues. Therefore, these citations in Ngor chen’s texts may also provide clues to
the evolving attitudes of Sakyapa authors toward Abhayå’s texts in the particular case of
Hevajra-affiliated teachings. We have seen Ngor chen refer to Abhayåkaragupta by
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
530
This passage from the Sampu†a was discussed in Chapter Three of the dissertation in
the context of Mkhas grub’s argument surrounding the location of the goddesses in the
Guhyasamåja body mandala practice. [KJ 248.2] Mkhas grub referred to the Sampu†a’s
arrangement of the goddess Locanå in the navel, the abode of earth, and of the goddess
Tårå in the crown, the abode of rlung , in the process of his attempt to establish the
goddesses’ association with elemental rlung rather than merely with the elements.
531
Abhayåkaragupta. Dpal yang dag par sbor ba'i rgyud kyi rgyal po'i rgya cher 'grel pa
man ngag gi snye ma. Åmnåyama•jarî. Toh 1198 cha 1v-316r. In bstan 'gyur (snar
thang). TBRC W22704. 21: 5 - 708. [narthang]: [s.n.], [1800?].
The fourth snye ma appears to be roughly equivalent to the fourth chapter, 44b-59a of the
sde dge edition; this citation is found at 56a. Is seems that rab byed is the term used to
demarcate a chapter within the Sampu†a Tantra; within the Åmnåyama•jarî, snye ma is
the term used for the “cluster” of commentary upon the equivalent chapter [rab byed] in
the tantra.
I am grateful to the ACIP and Kurt Keutzer for their efforts in making a digital version of
this extensive text accessible. Without this resource, it would have been very difficult to
definitively identify the Åmnåyama•jarî as the source. For another Indian commentary on
the Sampu†a, See Sobisch 2008 Title list #18: the Sam pu ta’i ‘grel pa chen po by
Indrabodhi [Toh 1197].
532
Sobisch 2008, p.76. Sobisch interprets A mes zhabs’s remarks that contemporary
scholars are more inclusive of Abhayåkaragupta contributions to be ecumenical in nature.
533
Sobisch has identified this critique within Go rams pa’s commentary on Sa pan’s text,
the sDom gsum rab dbye’i rnam bshad (94r). See Sobisch p.76 fn 217 and reference to
Rhoton 2002: 105, 186n. 20!
! 176!
name in a more general context in his citation describing the protective circle discussed
above [550.1-550.3].534
Szanto remarks on the Åmnåyama•jarî: “The influence of this text on Tibetan
authors, most significantly Tsong kha pa, is a well-known fact to Tibetanists and should
not be insisted on further.”535 Support for such a claim can be found in the frequent
citation of the Åmnåyama•jarî in texts like Tsong kha pa’s Lamp to Illuminate the Five
Stages, a Guhyasamåja based commentary.536 Therefore, another line of inquiry might be
to ask how the Sakyapas use Abhayå’s work differently than Tsong kha pa and his
disciples do.
Here in Ngor chen’s text, the citation from the Åmnåyama•jarî reinforces the
identification of the goddesses with the cakras and particular mudrås, dhåtus, and so forth
and describes their nature in terms of emptiness and compassion, method and wisdom.
The quote also explains that while, on some occasions, the letters are the focus of the
practice, on others, the mudrås may be the focus. [562.6-563.2] The commentary
therefore allows for variety amongst sådhanas. Ngor chen then interjects an explanation
to account for the enumeration of four cakras, motivated, we would assume by the fact
that the Lam ‘bras system teaches five cakras or palaces. By his reckoning, the great bliss
cakra refers to both the crown and secret place on account of the movement of bodhicitta
between them; he suggests that the commentaries themselves have produced this
explanation. In the final section of this chapter, we will see how Ngor chen’s other
version of his text [N2] further elaborates on conflicts between different systems in
enumerating the cakras. Such problems are similar in kind to those addressed early in the
dissertation, both in our exploration of the proto-body mandala text from Dunhuang as
well as in the case of Mkhas grub’s discussion of the goddesses and their larger place in
the exegetical tradition of the Guhysamåja system. Therefore, Ngor chen’s text
manipulates systems of correlation and reconciles conflicting accounts in a manner
typical of Indian and Tibetan mandala iconography and exegesis.
Next, Ngor chen refers back to the Sampu†a Tantra to describe the “generation of
five Buddha families in the center of the cakras from a drop.” He quotes the tantra,
naming the five wisdoms and five jinas and identifying the pure nature of the drop with
the goddesses. He then supplements the description with yet another quote from the snye
ma, naming the five consorts: Locanå, Måmakî, Påñ∂aravåsinî, Tårå, and rdo rje snyems
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
534
Recall also Ngor chen’s citation of Abhayåkaragupta on the Cakrasamvara mandala of
the support above. (549.6-550.3)That citation directly preceded Ngor chen’s explicit
introduction of the Hevajra body mandala context with a quote from the Sampu†a itself.
Lee 2003 remarks upon the importance of the Sampu†a Tantra for gaining insight in
Abhayå’s tantric writings, especially for the Nißpannayogåvalî and Vajråvalî mandala
manuals. See Lee 2003, p.29.
535
Szanto 2013, p6.
536
See Kilty Translation, 2013 and Tsong kha pa & Thurman 2010. A : 1 1)
2F1 1 A 1 - 2 B( 1 1 1A 01 1
' , , , 0 :
( 1 AB B B ) AB .B A!
! 177!
ma [564.1-.2]537 Ngor chen identifies this last goddess with Vajradhåtu-î!varî, who we
often find as the consort of Vajrasattva, and explains how all five families are sealed by
Vajrasattva. This inclusion of Vajrasattva is a standard technique in mandala logic for
elaborating a structure from four to five-fold. Therefore, cakras and mandalas are
expanded, classified, and reformulated in similar ways.
Moving beyond the central deities of the mandala to the limb or ancillary deities
[yan lag gi lha], Ngor chen includes another quote from the Sampu†a. [564.3-.4] This
citation describes how beings of the three realms are generated from the bodhisattvas and
krodhas. These deities are included as part of the body mandala but do not seem to be
counted in the enumeration of the one hundred and fifty-seven main deities of the
Hevajra body mandala. The tantra specifically refers the reader to the guru’s instructions
for the details of the practice. This reference to the guru’s instructions is an important
part of Ngor chen’s citations from the Sampu†a Tantra. Ngor chen uses this aspect of the
tantra as an exegetical tool for linking the Hevajra tantras with the oral instructions.
Then, once again, Ngor chen directs us to the Snye me for more explicit detail. [564.4-.6]
538
The passage describes the pure nature of the inner and outer sense spheres in
association with the bodhisattvas (ex. Kßitigarbha), sense spheres (ex. Mohavajra), and
sense objects (ex. Rüpavajrå) and the pure nature of the elements in association with the
goddesses (ex. Locanå). There is also mention of the krodhas.
Ngor chen then responds to an anticipated critique that the evidence he has
provided only authenticates the practice of generating these ancillary deities, but not the
deities of the body mandala proper. Ngor chen’s response is a quote from “further along
in the text” [lung de’i rjes kho na la] (meaning in the snye ma ) :“The skandhas [phung
po] of sentient beings are nothing more than the very nature of the causal continuum
Vajradhåra.”539 As addressed above in our discussion of the Sakyapa tantric view, the
causal continuum is based in the storehouse consciousness [kun gzhi] or what might be
called the universal basis. It has the nature of mind and accords, in some ways, with
Buddha nature.540 Therefore, the quote, on the most basic level, communicates that all
beings are Buddhas or possess Buddha-potential. 541
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
537
Åmnåyama•jarî, sde dge 227a. On his blog, Tibeto-logic [07/13/2011], Dan Martin
speaks to the problem of determining the Sanskrit equivalent for Rdo rje snyems ma.
Citing Vajrāvalī ch. 28 [ed. Mori 2009, vol. 2, pp. 428-429] and Abhayākaragupta's
Abhayapaddhati [Chog Dorje 2009, skt. p. 3, tib. p. 93], he suggests Vajragarvå.
538
Åmnåyama•jarî, sde dge 52b.
539
[N1 565.1] sems can gyi phung po yang de bzhin kho nar rgyu'i rgyud gyi rdo rje 'dzin
pa'i rang bzhin kho na'o
540
I am grateful to Drapa Gyatso for his patience in explaining the continua for me.
Spring 2012.
541
More specifically, in translating phung po as “skandhas” rather than merely as
“masses,” the quotation might be more precisely interpreted as a reference to the ritual
correlation of the body’s psycho-physical constituents with their inherent purity through
envisioning them as deities. As discussed in previous chapters, this ritual correlation is a
technology shared by body mandala and preliminary practices.
Davidson’s account of the Hevajra abhisamaya helps to clarify the specific ritual
context of the causal Vajradhara referred to here. The body mandala practice is initiated
! 178!
Ngor chen then summarizes the import of these exegetical maneuvers through the
root (Hevajra) tantra, Sampu†a Explanatory Tantra, and its commentary. He state that
while only a little bit of the body mandala practice is described explicitly in the root
tantra, the Sampu†a or “King of Explanatory Tantras” provides the implicit meaning.
[565.2] Most importantly, he declares that the main point [gnad gyi don] to be derived
from all of this is that the lama’s oral instructions are key to realizing the body mandala, a
point the Sampu†a itself reinforces. His final citation for this portion of the argument is
derived from the ‘uncommon’ explanatory tantra of the Hevajra cycle, the
Vajrapa•jara.542 As a way of completing the extent of the body mandala practice within
all three of the Hevajra tantras, he provides two brief references [mdor bstan] to the
introduction [gleng bzhi] to Chapters Seven and Eight of that text. The quote for Chapter
Seven is an abridged version of that chapter’s correlation of the constituents with the five
Buddha families and of the elements with the goddesses. 543 Ngor chen describes the
material from Chapter Eight as providing a bit of information on arranging deities on the
sense organs as well as descriptions of the deities of the five families in the outer
mandala. He suggests the latter are useful for envisioning the deities of the inner
mandala.544 [565.4-.5] Ngor chen therefore solidifies the basis of the practice in all three
tantras of the Hevajra cycle, provides justification from within the tantras themselves for
incorporating oral instructions, and forges connections between inner and outer mandala.
Ngor chen concludes this portion of the argument by foregrounding the
consolidated explanation of body mandala from the teachings of Virüpa and from another
text, the Lam bsdus pa. The latter likely refers to Sa chen’s Lam bsdus pa’i bshad pa,545 a
short exposition of Virüpa’s teachings included in the Red Book compilation of esoteric
instructions.546 The Red Book (pod dmar/pusti dmar chung) was compiled by Ngor chen
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
during the generation stage during the “Passion” [rjes chags Skt. anuråga] after the
merging of the jnånasattva and samayasattva. Previous to this, the meditator has
visualized the union of Hevajra and Nairatmya as the causal Vajradhara
(Hetuvajradhara). According to Davidson, it is the “cause (hetu), in the sense of the
ground of purification (sbyong gzhi), (and) operates as the fruit and to utilize the five
types of gnosis of the Buddha as the path.” This causal Vajradhara is dissolved, and the
fruitional Vajradhara (Phalavajradhara) produced. Once the body mandala is created, the
meditator requests consecration from the deities, and the result is the ‘sealing’ of ‘four
internal centers. Using this ritual context to interpret the quote, we might say that
through the production and dissolution of Buddha bodies, the practitioner enacts a
ritualized transition from ordinary to enlightened being although the ‘results’ of that
process are present from the start. Davidson 1992, pp. 117-18.
542
Yang dag par sbyor ba zhes bya ba’i rgyud chen po [Toh 381]!
543
See Vajrapa•jara Tantra, sde dge 45b.4 -45b.6.
544
See Chapter Eight of the Vajrapa•jara Tantra, sde dge 45b.6-51a.5, especially 46a.5-
46a.7. .
545
Sa skya Lam ‘bras Series 13, 119r-119v. See Sobisch 2008 Title list #393. See also
references in Sobisch pp.107 &120.
546
Alternatively, it is possible that the reference is to one of the three versions of Hevajra
practice: the extensive, middle and condensed.
! 179!
himself and includes writings by sixty previous Sakya masters as well as ten works by
Ngor chen.547 His nephew, fourth abbot of Ngor, Rgyal tshab dbang phyug (1424-78),
organized the title list.548 This particular work by Sa chen is classified in a section
exhibiting the “authenticity of the treatise,” one of the “four authenticities” [tshad ma
bzhi ], “of the guru, of experience, of the treatise, of the basic scriptures”; this
framework, first established by Grags pa rgyal mtshan, was often employed to organize
esoteric literature. 549 The nine works by Sa chen grouped in this section of the Red Book
were transmitted from his own teacher, Rje Dgon pa, Zhang-ston Cho-bar.550 Ngor chen
solidifies the final link connecting the Indian and Tibetan masters in the Sakya lineage of
the Hevajra tradition in stating: “So it is taught from the condensed meaning and
ascertained by the mahasiddhas as the whispered lineage in the inviolate oral
instructions.”551 Utilizing the full range of exegetical potential of the resources at his
disposal, Ngor chen has artfully navigated the reader through the complexities of
relationships between root and explanatory tantras and their commentaries, of explicit
and implicit meanings, of extensive and pithy and of scriptural and oral instructions to
reify the “inviolate oral instructions” that are the unique inheritance of the Sakya
tradition. As direct inheritors of this tradition, the Sakyapa hierarchs are further
validated, supporting Davidson’s claim that Ngor chen attempted to place the work of the
first three Sakya patriarchs, Sa chen and his sons, on par with the Indian masters as fully
authenticated authorities on esoteric practice.552 According to Davidson, Ngor chen
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
547
On the Red Book, See Sobisch 2008 Chapter Two: pp. 103-112 and Title List #344-
435.
548
Sobisch 2008, p.103.!
549
See reference to the ‘tshad ma bzhi in the conclusion of Ngor chen’s text [578.3]:
“Strive to practice in accord with these oral instructions which are established through the
four authenticities through the yoga of the four sessions.” [tshad ma bzhis grub pa'i
gdams ngag 'di lta bu la thun bzhi'i rnal 'byor gyis nyams len la brtson par gyis shig dang]
On the ‘tshad ma bzhi, see Sobisch 2008, 99-100.
550
On this identification of Rje dgon pa, See Sobisch 2008 p.106 and references there to
Stearns 2001: 63,242, n.162; 2006:220 ff and to Davidson 2005 301 &303.
551
don bsdus nas bstan zhing; man ngag ma nyams par grub chen rnams kyis snyan
brgyud du gtan la phab pa yin no
It is, moreover, possible that the don bsdus refers to a specific text, such as the Kyai rdor
don bsdus referred to in the following section by its other name, the Nå ro ‘grel chen
[567.5]. This text is cited below by Ngor chen at 567.5. See below for further detail.
552
Davidson 1992, p.112 provides a reference to Ngor chen ‘s gnad kyi zla-zer
[p.179.3.6] composed in 1419, therefore seven years before the present text. As
discussed in the introduction to the dissertation, Ngor chen’s writings on the Hevajra
practice in this text prompted controversies that required his successors to continue to
defend his perspective. See for example, Go rams pa’s dPal kyai rdo rje’i sgrub pa’i
thabs kyi rgya cher bshad pa bskyed rim gnad kyi zla zer la rtsod pa spong ba gnad kyi
gsal byed. See Sobisch 2008 Title List #167.
! 180!
envisioned the Hevajra and Lam 'bras instructions as incomplete until the compilations of
theYellow Book [Pod ser ma] based on the work of these three patriarchs.553
While we have focused thusfar upon the manner in which Ngor chen is defending
the Hevajra body mandala system against the charges issued by Mkhas grub, it is likely
that Ngor chen, too, has other unnamed opponents. Sobisch’s 2008 study of A mes
zhab’s (1597-1659), Notes on How to Enter Into the Writings of the Sakapas [NOTES], a
text primarily composed by Chos dpal bzang po (fifteenth century) based on Ngor chen’s
own teachings, offers some clues to who these opponents might be.554 The two systems
of pith instructions on Hevajra practice are those passed from Nåropå to Marpa and from
Virüpa to “Nagpo of the East” [Kanha]. 555 While the former system was favored by the
Kagyupa, the latter was the province of the Sakyapa. In the “Notes” text, Ngor chen
quotes a couple of polemical comments by Kagyupa authors that elevate Marpa’s
transmission and even critique the efficacy of the Sakyapas’ received transmission.556
There, in the “Notes,” Ngor chen (as rendered by A mes zhabs) proceeds to expose the
weaknesses of the Marpa transmission, in particular the absence of the Sampu†a
transmission within it as well as of the commentaries on the Hevajra and Vajrapa•jara.557
Verrill’s translates one such excerpt from the “Notes” as follows: 558
“In the path tradition of the great siddhas Padmavajra, Shantipa, and others, the Sambhuti
(Sampu†a) was considered unnecessary as an explanatory tantra; but in the special
Lamdre (Path and Result) instructions in the tradition of the great siddha Virüpa, the
Samputi is a necessary explanatory tantra following the revelation of the Two-Part
[Hevajra Tantra].”
Sobisch very tentatively sets forth a compelling theory distinguishing the

Kagyupa and Sakapa exegetical styles on Hevajra:
“One possible interpretation of a ‘Sakyapa’ and ‘Kagyupa’ approach to tantra may

therefore be that the Sakyapas consider the Indian commentaries (and the Sampu†a) to be
of major importance, while the Kagyupas stress the importance of the pith instructions
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
553
Davidson 1992, p.111. See also Davidson’s reference to Ngor chen’s gnad kyi zla zer
179.3.5-.6 in fn 16.
554
(“Notes for the Correct Explication of How to Enter into the Writings of the
Venerable Sa skya pas: Opening Wide in a Hundred Directions the Dharma Gates to
which All Beings of Tibet are Guided”) Rje btsun sa skya pa’i gsung rab la ‘jug tshul legs
par bshad pa’i yi ge bod yul ‘gro kun bsgrod pa’i chos sgo phyogs brgyar ring du phye
ba. Edited by A-med-zhabs and originally composed by Chos dpal bzang po; collected
works, vol. kha, fols. 384r-393v. [as cited in Sobisch 2008]
This text was cited above in our discussion of the classification of the Sampu†a Tantra.
555
On Kanha, see Stearns 2006 pp.9-13 and Part One, fn12.
556
Sobisch 2008, 5-6!
557
Sobisch 2008, 6.
558
The text is referenced by Verrill 2012 on p.331 as “Notes on Ngor chen by dPal kyi
rgyal mtshan in A mes zhabs,” Vol.21, p.76.4.
! 181!
alone, and in particular advise refraining from mixing these with commentaries or mixing
different systems together.”559
What we have encountered thusfar of Ngor chen’s exegetical style in the body
mandala debate suggests that there is some truth to Sobisch’s theory. Indeed Ngor chen’s
inclusion of all three Hevajra-related tantras, as well as commentaries and oral
instructions in his claim for textual authority displays a totalizing strategy that likewise
garners the associated prestige of his sources. Having gleaned the significance of the
identities of particular mahåsiddhas to the nature and authority of a tantric transmission,
we will proceed to examine how Ngor chen employs citations from the systems of four
other mahåsiddhas. Sobisch’s 2008 study of the Notes on How to Enter Into the Writings
of the Sakapas will prove particularly useful in this enterprise.
IB. On the commentaries of the other mahåsiddhas
Ngor chen divides the next section of his argument into four parts: on Nag po pa, Jetari
(whom we have already encountered above), Indrabhüti, and Vajragarbha. In order to
better understand how these masters fit into the larger schema of the transmission of the
Hevajra teachings from India to Tibet, we turn to Sobisch’s study of the “Notes.” The
“Notes” outlines eight categories of transmission of the Hevajra instruction; these include
the two systems of pith instructions of Marpa(-Nåropå) and of Virüpa(-Kanha) discussed
above as well as “six great chariot systems” (shing rta’i srol chen po drug) of teachings
connected with: Îombî[heruka]; Mtsho skyes rdo rje (Saroruhavajra/Padmavajra); Nag
po Dam tshig rdo rje (K®ßña Samayavajra); Shåntipa (Ratnåkara!ånti); Snyan grags bzang
po (Ya!obhadra?); Gnyis med rdo rje (Advayavajra/Avadhütipa/Maitrîpa).560 The
“Notes” further specifies that of these six chariots, the first three are perfectly transmitted
by the Sakyapas at that time.561 The centrality of the first two “chariots” to the Sakyapa
Hevajra practice is attested by the fact that they serve as the basis for the middle and
extended Hevajra sådhana practice respectively.562 The significance of the others is more
nuanced. We will refer to these eight categories of transmission to determine the import
of Ngor chen’s strategy of citation from the works of the Mahåsiddhas.
IBi. Nag po pa’s Position
The “Notes” associates the teachings of Nag po pa’s transmission with Nag po Dam tshig
rdo rje (K®ßña Samayavajra) and his student Nag po Zhi ba bzang po (K®ßña ¸ånti
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
559
Sobisch 2008, 6.!
560
Sobisch 2008, p.30.
561
Sobisch 2008, 39.
562
As explained by Davidson 1992 p.111, Bsod nams rtse mo developed the four-limbed
sådhana of the extensive practice based on Padmavajra’s system, while Grags pa rgyal
msthan developed the six-limb sådhana of the middle-length practice based on
Dombiheruka’s system. Davidson fn17 refers to this as Grags pa rgyal msthan’s own
understanding of the systems as conveyed in Go rams pa’s gnad kyi zla zer la rtsod pa
spong ba gnad kyi gsal byed, Collected Works, Vol. 12, 598.1-.3.
! 182!
bhadra).563 ‘Gos Khug pa lha bstas, transmitter of the Guhyasamåja, studied Hevajra with
Nag po Dam tshig rdo rje in India and proceeded to transmit the teachings in Tibet as
well as to translate the Hevajra commentaries.564
Davidson 1992 identifies four main systems transmitted by the Sakyapas as those
of Kanha, Dombiheruka, Padmavajra, and K®ßña Pandita (the nag po’i lugs). With regard
to the latter, Davidson observes that while Grags pa rgyal mtshan considered it important,
“it...did not seem to receive the intense interest that the other three meditative cycles did
and, while maintained down to the present, it appears to be more of an archaic appendage
than a vital part of the Sakya heritage.”565 He accounts for the exclusion of Kanha from
Ngor chen’s classification as based on the dearth of materials from that system.
The possibility of translating Nag po pa as either Kanha, “Virüpa’s disciple” or
K®ßña, “the scholar” presents an obstacle for definitive interpretation of his use by Ngor
chen.566 If the Nag po pa referred to in this portion of the body mandala debate is indeed
Nag po Dam tshig rdo rje (K®ßña Samayavajra) and/or his student Nag po Zhi ba bzang
po (K®ßña ¸ånti bhadra), Ngor chen’s choice of citations may enrich our understanding of
the weight given to this system of transmission at Ngor chen’s time. If the reference is to
Kanha instead, then Ngor chen’s use of the quotation may reveal what part Kanha’s
system played in the Hevajra exegetical tradition despite his choice not to include it in the
“six chariots” framework of the “Notes.”
The first citation, in this section, however, is incomplete; Ngor chen merely
indicates that it was composed by Nag po pa.567 The citation describes the body mandala
of the support, associating regions of the body with the elements in a manner familiar
from the discussion above; however, it adds the bliss cakra and the correlation of the
womb with space. [566.2-.5] Furthermore, it locates the cremation grounds/sacred sites
on the orifices, and appears to include a reference to the dåkinîs, the 72,000 channels and
the cakras. Ngor chen clarifies the meaning as a description of the five palaces or cakras
of the body mandala, made up of the one hundred and fifty-two retinue goddesses of the
channel petals plus the five mothers of the Buddha families (thus totaling one hundred
and fifty-seven). Next, Ngor chen refers to the “Commentary on the Vajra Song”; this is
likely to be a reference to the portion of the Hevajra Tantra in which the goddesses sing
to Hevajra to manifest himself.568 The quote describes the six Cakravartin males and
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
563
Varying explanations appear in the works of different authors attempting to parse the
authorship of the works of Nag po Dam tshig rdo rje (K®ßña Samayavajra) and Nag po
Zhi ba bzang po (K®ßña ¸åntibhadra). Davidson 1992 fn9 points to a reference to K®ßña
¸ånti bhadra as K®ßña Pandita in Ngor chen’s gnad kyi zla zer 175.3.3.
564
Sobisch 2008, p.36.
565
566
See Stearns 2001, p.242 fn 160. Stearns provides as example here of an instance of
dealing with this conundrum. Sobisch 2008 fn also refers to an instance on Nag po pa as
Kanha in Stearns 2002, 171 n.112.!
567 !If it refers to the Rnal ‘byor rin po che sbyor ba’i phreng ba (Toh 1183, for which
Snellgrove 1959 provides an edition), then it is Kanha vs. K®ßña that is the author. See
Sobisch 2008 Title List #6. !
568
However, it is also possible that the reference is to either Saroruha’s Rdo rje’i glu’i
grel pa (Toh 1207) or to Sgrol ma can gyi sde’s Rtsa rgyud phyi mar (Toh 1208).
! 183!
females and compares them with subtle particles [rdul phra rab]. Ngor chen explains that
the cakravartins are enumerated as six because Vajrasattva is added and clarifies the
analogy to subtle particles through reference to the later part of the (two-part) Hevajra
root tantra (i.e. “yoginîs equal to the subtle particles of Mt. Meru”). Furthermore, he
clarifies the ritual context as one in which the practitioner has generated the channels as
goddesses and arranged them on the body of Hevajra (referred to here Pi tsu ba jra).569
The final citation in this section is from the Nå ro ‘grel chen.570 It specifies five
seed syllables together with the five associated buddhas which are to be arranged on the
bodily sites (crown, navel, throat, secret place and heart) of pi tsu ba jra, in other words,
on the body of the practitioner which has been generated in the form of Hevajra; it also
refers to the subsequent realization of the vajra dåkinis.571 If we look to the context of
this quote within the Nå ro ‘grel chen itself, we find that it is preceded by the perfection
of wisdom through purifying the skandhas and followed by the mantra for actualizing the
sixteen arms (of oneself as Hevajra). The ritual described seems more like a preparatory
ritual for purifying the body and/or the initial generation as the deity and less like body
mandala proper.
In general, these citations share a vision of the body mandala composed of five
Buddha families, with the exception of the cakravartin system, which adds Vajrasattva.
While Nag po pa’s description emphasizes the role of the goddesses as channels, focuses
upon the subtle body, and incorporates cosmic factors like the elements and cremation
grounds, the Nå ro ‘grel chen’s description of the buddha families parses the body in a
more basic way. Ngor chen’s motivations for incorporating the Nå ro ‘grel chen as part
of his description of Nag po pa’s system remain unclear.
IBii. Jetari’s position
Jetari’s system is one of the two sets of pith instructions include in the framework of the
“Notes” and is regarded as unique in its basis in all three Hevajra tantras.572 Ngor chen
has already cited the mahåsiddha Jetari’s Sådhana of the Four Mudrå [Phyag rgya bzhi yi
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
569
I am grateful to Kurt Keutzer in helping me to clarify this.
570
Sobisch 2008 describes this text as a commentary on the Hevajra tantra, the Kyai rdor
don bsdus, with an incomplete sub-commentary, the Kyai rdor don bsdus kyi 'grel pa. See
Sobisch 2008 Title List # 12 & 134 (Toh 1186) respectively. According to Sobisch, Ngor
chen’s disputed this text’s attribution to Nåropå; however, he maintained its value and
compatibility with the generation stage of mahåsiddha Saroruha and with the completion
stage of Kålacakra.( Sobisch 2008, p.43. The “Notes” classifies this text within sNyan
grags bzang po’s cycle of teachings.
571
Thanks to Kurt Keutzer, this quote has been located at p.926 of OCR’d version of Toh
1186. Rdo rje tshig gi snying po bsdud pa’i dka’ ‘grel (Vajrapada-såra-saµgraha-
païjikå) also Nå ro ‘grel chen, Kyai rdor don bsdus, and Rdo rje tshig gi snying po bsdud
pa; Nåropåda. P 54/2316, 69r-169v, A. Naro-zhabs, Toh 1186, ka 58v-146v. A. Snyan-
grags-bzang-po. [cited in Sobisch 2008, title list #134].!
572
Sobisch 2008, p48
! 184!
sgrub thabs], earlier in his argument on body mandala, in the section devoted to the
proper interpretation of Bu ston’s comments on the body mandala of the support. [551.3-
552.1] In that quote, the four mudrå schema was used to describe four aspects of the
mandala of the support, correlating the parts of the body with the protective circle,
cremation grounds/sacred sites, cosmic elements, and the celestial palace, respectively.
In that context, Ngor chen endorsed Jetari’s views as the quintessence of the Sakyapa
interpretation of Virüpa’s teachings.
Here, in the evaluation of Jetari’s system among other mahåsiddha traditions,
Ngor chen once again cites the Sådhana of the Four Mudrå. He establishes the ritual
context for the quote, explaining that the practitioner first establishes the nine deity outer
mandala deities and Heruka and then proceeds to array the body mandala onto the main
deity’s body. [568.2-570.1] In this quote Jetari redeploys the four mudrå schema to
describe the four goddesses (Locanå, Måmakî, Påñ∂aravåsinî and Tårå) inhering in their
respective cakras (composed of particular numbers of petals); he also identifies these
goddesses the with particular elements and colors.573 The buddhas dwelling in the center
of those cakras are mentioned as well. Next, the fourfold schema is used to describe the
six bodhisattvas and six goddesses of the sense objects and the ten wrathful males in
union with ten wrathful females; the text locates them on the sense spheres, sense objects,
and limbs respectively. The text concludes: “The skillful ones diligently visualize all
things that are produced from the channels, the supreme mandala of body, speech and
mind within one’s own body as the cause of accomplishment (Tib. dngos grub Skt.
siddhi).”574
Ngor chen again endorses Jetari’s view as on par with Virüpa’s oral instructions
and then proceeds to explore potential contradictions or “doubts” pertaining to Jetari’s
description of the body mandala practice.575 The first point deals with the order of
generating the mandalas of the support and supported in the Luipa transmission of the
Cakrasaµvara practice. As mentioned above, Ngor chen wrote a whole text dealing with
the different versions of Cakrasaµvara body mandala practice; such questions of ritual
order and of the relation of different aspects or phases of the mandala assume a
prominent position in that text.576 In the present text, we find Ngor chen anticipating
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
573
This account resembles the one critiqued by Mkhas grub and described in Chapter
Three of this dissertation in the context of mapping and correlating the goddesses in the
Guhyasamåja body mandala. The resemblance is worthy of further attention.
574
thams cad rtsa las skyes pa'i dngos; [570.1] sku gsung thugs kyi dkyil 'khor mchog;
mkhas bas rang gi lus nyid la; dngos grub rgyur ni 'bad pas bsgom
575
It is of note that in Ngor chen’s other version of the text, N2, he uses the exploration
of doubts [dogs pa dbyed pa] as a larger organizational schema for the second half of the
text (on scriptural authority). There he divides the text according three topics:
explanation from the tantra, by the siddhas, and investigating related doubts. [N2 607.5]!
576
Commentary on the Ghantapa Body Mandala Practice. Dril bu pa'i lus dkyil gyi bshad
pa. TBRC W11577. 4: 735 - 766. [dehra dun]: [sakya centre]Vol.4 .[see also 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, pp. 398.1.1-405.4.1 (cited in Bentor 2015)].
I hope to present further research on this work in future.
! 185!
issues of what we referred to above as go rims ‘khrugs or “confusion of the order of
things” to be of concern for his interlocutors. Such discussions express both practical
concerns with ritual efficacy as well as their more theoretical implications of ontological
hierarchies.
For Ngor chen, however, the focus of the present text, and in particular of this
latter section on textual authority, is to articulate the relationship between different kinds
of sources that compose the Hevajra body mandala tradition. One way in which he
manifests this goal is by comparing apparently contradictory accounts and neutralizing
the contradiction. In the process, he reveals an intricate mastery of textual exegesis based
in, and in some ways particular to, the Sakya tradition. For example, he validates the
process of generating the cremation grounds on the bodily orifices from Jetari’s text by
showing how even though it has no basis in the gdams ngag [Skt.upade!a], it appears in
many “textual commentaries of the precious oral instructions” [gsung ngag rin po che'i
gzhung bshad du ma] [570.4-.5]. The latter most likely indicates commentaries on
Virüpa’s Rdo rje tshig rkang [Toh 2284] composed by the Sa chen and his sons. This
line of reasoning would further support Davidson’s theory that one of Ngor chen’s
driving motivations was to elevate the authority of these three early Sakyapa patriarchs to
match that of the Indian masters.577
In other examples, the potential contradiction is neutralized by using references to
different versions of the practice as alternatives or options [gdam ga] in the gdams ngag.
Finally, Ngor chen skillfully negotiates the relationship of orally and textually transmitted
practices, a relationship deeply layered with subtle dynamics of authority. In one case,
Ngor chen applies the common strategy of explaining different versions of a practice in
terms of the differing capacities of practitioners. [571.6-572.3] In this particular example,
the version in Jetari’s text agrees with the Vajrapa•jara Explanatory Tantra but disagrees
with the oral instructions. Ngor chen carefully manipulates the versions derived from the
work of one Indian master, from an explanatory tantra and from the mouth of the Indian
progenitor of the Sakya Lam ‘bras tradition to preserve a variety of authority for each.
IBiii. Indrabhüti’s position
Indrabhüti composed another (Indian) commentary on the Sampu†a Tantra, the dPal kha
sbyor thig le zhes bya ba rnal ‘byor ma’i rgud kyi rgyal po’i rgya cher ‘grel pa yang dag
par lta ba’i dran pa’i snang bar. [Toh 1197] Sobisch 2008 makes a point of the fact that,
according to the “Notes,” this particular commentary is “very unrefined” [shin tu gyong
ba].578 If this is indeed Ngor chen’s stance on the text, then his choice to incorporate
Indrabhüti’s system in his defense of the textual authority of a heterogeneous Hevajra
system is significant. In fact, the longer version of Ngor chen’s text [N2], to be discussed
in the final portion of this chapter, appears to eliminate Indrabhüti’s position from the
argument entirely.
The citation includes descriptions of both a six cakra body mandala and a five-
cakra body mandala. The first model includes cakras at the navel, heart, throat, crown,
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
577
578
Sobisch 2008, p.76!
! 186!
the center of the crown [spyi gtsug], and the secret place. Indrabhüti names the cakras
and designates their respective colors, syllables, and number of petals which total one
hundred and fifty-seven (the standard number of deities in the Hevajra body mandala). In
addition, the citation provides the succession of associated Buddhas and goddesses (as
elements). Finally, the three main channels are located in the center of the structure. The
second model is a less elaborate depiction of the pentadic structure that links the
syllables, Buddha families, channels, and cakras. In citing these two different models
for body mandala from Indrabhüti’s text side by side, Ngor chen allows for multiple
iterations of the subtle body structure within a single Indian text.579
IBiv. Vajragarbha’s position
The final system invoked here by Ngor chen is Vajragarbha’s; the “Notes” classifies
Vajragarbha’s two commentaries on the Hevajra Tantra among the minor works related
to the “six chariot system.”580 The two commentaries interpret the first and second part
of the two-part Hevajra Root Tantra respectively. The citation here appears to be from the
former, the Kyai rdo rje bsdus pa’i don gyi rgya cher ‘grel pa [Piñ∂årtha-†îkå ].581 The
description of the cakras are complex and initially appear contradictory; both incorporate
a cakra at the forehead, but only one includes the crown cakra. However, the varying and
intricate descriptions crescendo in a neutralizing statement by Ngor chen. His overall
explanation for variation in the number of cakras is in accordance with the number of
Buddha families presented in a particular system, further reinforcing the connections we
have drawn between the evoloution and exegesis of mandala iconography and
descriptions of the cakras.
Ngor chen concludes the section on the commentarial tradition of the
mahåsiddhas with a pramåña-inflected admonition of his opponent: “Taking merely not
seeing as your reasoning, you criticize the profound oral instructions of the
mahåsiddhas.”[575.6]582 As noted in the beginning of this chapter, Ngor chen plays off of
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
579
In our discussion of this portion of the text, Drapa Gyatso of the IBA in Kathmandu,
observed that much of Indrabhüti’s description differs from the system of oral
instructions that is the basis for the Hevajra sådhanas commonly in use by the Sakyapas
today. Noting the multiplicity of systems for cultivating Hevajra, Drapa Gyatso recalled
how the twentieth-century master Khenpo Appey Rinpoche enumerated some twenty-
eight different methods. In terms of the structure of the subtle body described by texts
such as Indrabhüti’s commentary, Drapa Gyatso explains how tantric texts account for
the existence of an infinite number of channels in the body, all of which originate from
the three main channels. The ‘palaces’ [pho brang] or cakras are composed of a main
channel surrounded by the branch channels. [Personal communication, Spring 2012]
580
Sobisch 2008, p.43.
581
Piñ∂årtha-†îkå [Toh 1180]. See Sobisch 2008 Title list #7. See also Shengde’s 2004
edition and the forthcoming one by Francesco Sferra (as indicated by Sobisch 2008
fn103).!
582 !des na rang gis ma mthong ba kho na rgyu mtshan du byas nas; grub chen gyi gdams
ngag zab mo la skur pa 'debs pa ni

!
! 187!
his opponent’s integration of valid cognition into his tantric polemics. His comments, on
the one hand, remind the reader that for the Sakyapas, the possibility of unmediated,
direct perception is impossible. Recall his remark, derived likely from Sakya Pandita:
“Just because you don’t see it, doesn’t mean it isn’t there.” [561.1] Such comments also
suggest that perhaps the opponent is missing something right before his eyes or even that
his command of the texts is lacking. Building upon this bit of a polemical flair, Ngor
chen concludes the section on the mahåsiddhas with a dramatic citation of a prophecy by
the bodhisattva Vajrapåni damning those who misconstrue tantric materials to a sentence
in hell. [576.1] Ngor chen shows that he too can fight fire with fire. However, even in this
rare glimpse of the polemical tone in Ngor chen’s text, it is clear that he relies primarily
upon the discourse and methods of tantric commentary to articulate and defend his
position. Of course he is familiar with the strategies of philosophical debate relished by
Mkhas grub as any monk trained in the Sakya scholastic tradition would be. He even
pauses at choice moments in the text to playfully exhibit that familiarity with a few
loaded comments. But ultimately, Ngor chen defends his tradition through mastery of the
tantric corpus and skillful display of that mastery.
IC. The Vajramålå
The final part of Ngor chen’s argument investigates methods for explaining from other
tantric commentaries but is actually devoted entirely to the Vajramålå. As discussed
above, Mkhas grub relied heavily upon the Vajramålå, particularly in articulating his
argument on the body mandala of the support. The majority of his citations were derived
from two chapters, Chapter Sixty-Four, “The Explanation of the Mandalas of Body,
Speech and Mind,” and Chapter Sixty-Eight, “The Collection of All Siddhis.” Ngor chen
begins his discussion of the Vajramålå by referring to three different varieties of body
mandala described in those very same chapters and identifying them with three different
tantric cycles: the Guhyasamåja version in Chapter Sixty-Three (by which he actually
means Sixty-Four); the Hevajra version in Chapter Sixty-Seven (by which he actually
means Sixty-Eight); the correlation of twenty-four places with twenty-four internal sites
in the same chapter. [586.6-587.1] He then turns his attention to an entirely different
chapter of the Vajramålå, Chapter Seventeen, “The Sites of the Successive rlung.” The
first verses he quotes describe five cakras (rlung, fire, enjoyment, dharma, and bliss) with
six, three, sixteen, eight and thirty-two channel petals respectively as well as an
additional version of one-hundred thirty one channels.583 Ngor chen skips the next
twenty verses of the Vajramålå, which name all of the channels that make up the cakras;
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
583
The verses referred to here are verses one through five in Kittay’s 2011 translation.
Different accounts enumerate the petals as 129 or 131. Kittay 2010, fn 981: ”The Tantra
says that these total ‘more than 131’ but all of the sDe dge, Peking and Snar thang
versions of the Commentary have 129 here, and the various numbers do in fact add up to
129, Alamka 125A.”
See Alaµkakala!a (Tshul khrims rin chen). Rnal 'byor chen po'i rgyud dpal rdo rje
phreng ba'i rgya cher 'grel pa zab mo'i don gyi 'grel pa. In bstan 'gyur (sde dge). TBRC
W23703. 34: 4 - 442. delhi: delhi karmapae choedhey, gyalwae sungrab partun khang,
1982-1985
! 188!
he briefly summarizes them and then resumes his citation with an series of verses
translated by Kitty 2011 as follows:
The body with the nature of the five aggregates

Is well known as the five dåkinîs.
That very thing,
Through the five elements,
Abides as the five wisdoms.
Therefore, with all effort,
You should know the channel wheels .v26
Just as a tree
In the middle of water
Grows quickly,
From it there is fruit
And so forth,
It will give.
In this there is no doubt. v27
Similarly, the aggregates

Are like a tree,
Having grown by the water
Of the channels.
Increasing, they bestow
The perfect fruit
Of omniscience.
You should meditate on
The nature of the channels
Through the actualities
Of the forms
Of deities v28584
These verses suggest that knowledge of the elements of the subtle body and their
relationship to the ordinary body of psychophysical aggregates is essential. They also
describe the relationship between the subtle body and the ordinary body: the channels
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
584
The verses appear in Ngor chen’s text as follows:
[577.2] de nyid 'byung ba lnga yi ni; ye shes lngar ni rnam par gnas; de phyir 'bad pa
thams cad kyis; rtsa yi 'khor lor shes par bya; ji ltar chu yi dbus kyi shing; myur du yang
ni 'phel nas 'gro; de las 'bras bu sogs ldan pa; ster [577.3] bar 'di la the tshom med; de
bzhin phung po'i shing zhes bya; rtsa yi chu yis 'phel nas ni; 'phel bas yang dag 'bras bu
ni; thams cad mkhyen pa nyid ster byed; lha yi rnam pa'i ngo bo yis; rtsa rnams rang
bzhin sgom par byed
! 189!
support the aggregates and catalyze their development.585 According to this model, the
coarse elements of embodiment actually rely upon their subtle invisible counterparts.
Body mandala is the ritual practice that re-enacts this underlying reality of human
embodiment through cultivating the channels in the form of deities. Ngor chen chooses
not to belabor the ontological implications of the verses; in other words, he could have
used the verses to engage with Mkhas grub’s argument regarding the interpretation of
Ghantapa’s position on the body’s primordial equivalence with the mandala. Moreover,
there are other verses in the section of the Vajramålå framing the passage Ngor chen cites
that he might have used to reinforce the soteriological importance of bodily knowledge
and even the metaphorical equivalence of embodiment with omniscience.586 Instead,
Ngor chen applies these verses from the Vajramålå to bare bones ritual mechanics:
generating the five dåkinîs in the center of the five cakras and arranging the one hundred
and thirty-one goddesses as the channel petals.587 This final quote describes the ancillary
deities (tathågatas, goddesses, sense object deities, bodhisattvas, ten wrathfuls), thereby
completing the depiction of the body mandala.588
Ngor chen shows the Vajramålå to contain versions of the Guhyasamåja, Hevajra,
and Cakrasaµvara body mandalas and yet selects passages for interpretation that are not
explicitly linked with any of these three cycles. Ngor chen’s choice to cite from this
particular section of this extensive and diverse text indicates a subtle mastery of the
materials as well as a nuanced understanding of the relationship of coarse and subtle
bodies. Through careful citation practices he is able to suggest this relationship without
diverting his focus from the topic of textual authority. Through interpreting the ritual
mechanics of body mandala, Ngor chen makes an implicit, rather than explicit, argument
for bodily knowledge, skillfully infused into a highly technical orchestration of source
materials in defense of the Hevajra body mandala practice.
II. Ngor chen’s Conclusion
As in his introduction to the text, Ngor chen uses the conclusion to rally sentiment in
defense of the Hevajra body mandala practice and to establish its authenticity and
superiority. However, he also presents some final insights into why the body mandala is
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
585
See the description of Kontrul’s Treasury of Knowledge, Book Six, Part Four:
Systems of Buddhist Tantra in which he explains the relation of gross and subtle bodies
as follows: “that which is supported, the body of habitual tendencies; and the support, the
innate body.” p.169!
586 !For!example,!Kilty!2011!translates!verse!25!as!follows:!“Thus, you should know
From the guru’s speech, The true stages Of the channels. The aggregates and so forth
Will not arise Without ascertaining The stages of the channels. Without the aggregates,
The yogi cannot achieve Great wisdom.” Verse 36 reads: “At that time, A seed arises.
From the seed Arises a living being. Therefore, you arise perfectly. From the channels,
The wonderful fruit is born.”
587 !See!note!above!on!the!problem!of!enumerating!!131!vs.!129.!
588 !Although Ngor chen identifies this final quotation as being from from “that chapter,”
the quote is actually derived from Chapter Eighteen, “The Gathering of the Channels of
the Body of the Yogin.” !
! 190!
so important as well as into how his exegetical style compares with that of his more
polemical opponents. As for the centrality of body mandala practice to liberation, Ngor
chen states: “It is necessary for the Buddha to consecrate the channel winds in the
generation of the wisdom of bliss of the completion stage. Other than the body mandala,
there is no profound shortcut for achieving that.”589 [578.6] After an invective targeting
frauds and proclaiming the benefit of the text for “other impartial scholars” [gzur gnas
mkhas gzhan], Ngor chen eloquently justifies his own method of argument in opposition
to the style of harsh polemics so familiar from Mkhas grub’s text as follows:
Even without wielding sharp weapons
or wearing strong armor,
How could the armies of flawless scripture and reasoning
fail to defeat the opponent?590
The final portion of this chapter will compare the two versions of Ngor chen’s text body
mandala text. Through this comparison, we will gain more privileged access to Ngor
chen’s authorial voice. In the process, we will enhance our understanding of the authorial
choices he makes in striking a subtle balance between polemical and commentarial
methods.
III. “Dispelling Evil Views” vs. “Destroyer of the Proponents of Evil”: a

Comparison
A mystery surrounds the existence of two ‘versions’ of Ngor chen’s body mandala text.
The longer “version,” Dispelling Evil Views by Eliminating Objections to the Hevajra
Body Mandala (Kye rdo rje'i lus kyi dkyil 'khor la rtsod spong lta ba ngan sel) has a
virtually identical title to the text that has been the focus of our study thus far.591 Mkhas
grub’s citations from Ngor chen’s argument in his Thunderbolt Wheel of Reply to
Ngor suggest that Dispelling Evil Views is the version of Ngor chen’s text he was
reading and responding to. So what makes Dispelling Evil View(s) different from
Destroyer of the Proponents of Evil ? The texts appear side by side in Ngor chen’s
collected writings, with Dispelling Evil View(s) ” [N2] immediately following
Destroyer of the Proponents of Evil . [N1] N2 is markedly longer, approximately
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
589
rdzogs rim bde ba'i ye shes bskyed pa la; rtsa rlung rgyal bas byin gyis brlabs pa dgos;
de nyid sgrub la lus kyi dkyil 'khor las; gzhan ba'i myur lam zab mo yod ma yin
590
[579.4 ] gang gis rno ba'i mtshon cha ma bzung zhing; sra ba'i go cha lus la btags min
kyang; skyon med lung dang rigs pa'i dpung tshogs kyis; phyir rgol pham par byas pa
min nam ci
591
Ngor chen kun dga’ bzang po (1382-1456). The Collected Teachings of Ngor chen
kun dga' bzang po (Ngor chen kun dga' bzang po'i bka' 'bum). Compiled by Bsod nams
rgya mtsho and reproduced from the Sde dge block prints. Dehra dun: photomechanical
print from a set of prints from the Sde dge kdgon chen blocks.Vol.1. N1 appears on
pp.545-580, and N2 follows on pp. 580-625.
! 191!
forty-five folia sides relative to N1’s thirty. The texts share identical colophons (with the
exception off their titles ) suggesting that they were regarded as interchangeable:
“Having negated the imposters who assert conceptual imputations and lies as the
profound practice of the Hevajra body mandala, this treatise establishes the oral
instructions as superior through pure scripture and reasoning. This Destroyer of the
Proponents of Evil, by arriving at the far shore of the ocean of one’s own and others’
tenets, becomes the crown jewel of the Buddha’s followers.
Through raising the dust on the feet of Rje btsun mkhas pa'i dbang po Ye shes
rgyal mtshan dpal bzang po to the crown, Sha kya'i dge slong kun dga' bzang po has
increased the appearance of understanding of the infinite tantric scriptures a little bit. He
excellently composed (this) in the year of zil gnon, on the third day of the month of
Buddha’s consecration in the school of wise ones at Sakya, the origin of many precious
jewels. May demons, opponents, and dissenters from all directions be conquered by
this!”592
The only apparent difference in the colophons is the title provided for the
respective texts. Why would the compiler choose to include both texts in the collection if
one was truly a version or draft of the other? Did they perhaps address the needs of
different audiences? Is one a revision or condensation of or elaboration upon the other?
The scholar monks of the Sakya tradition with whom I consulted could not explain this
anomaly. Van der Kuijp 1985a observes: “Written in the first half of 1426, these are two
prints of the same text, with some interesting variant readings.”593 Ngor chen’s biography
by Sangs rgyas phun tshogs (1649-1705), discussed in the introduction to the dissertation,
mentions both texts but does not provide any additional insights into their relationship:
“..there was a terrible misconception of imagining the Hevajra body mandala not to be
explained anywhere in the Indian tantric system. Through scripture and reasoning and
the oral instructions, he (Ngor chen) thoroughly refuted that circumstance of the Hevajra
initiate admitting wrong.594 He composed the great treatise that establishes the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
592
N1 580.2-.6: dpal kye rdo rje lus kyi dkyil 'khor gyi sgrub thabs zab mo la rtog btags
dang; [580.3] mun sprul du 'dod pa'i shes byed ltar snang rnams bkag nas yang dag pa'i
lung dang rig pas man ngag mchog tu sgrub pa'i bstan bcos smra ba ngan 'joms zhes bya
ba 'di ni; bdag dang gzhan gyi grub pa'i mtha' rgya mtsho'i pha rol du son pas bde bar
gshegs pa'i [580.4] ring lugs ba rnams kyi gtsug gi nor bur rgyur pa rje btsun mkhas pa'i
dbang po ye shes rgyal mtshan dpal bzang po'i zhabs rdul spyi bos blangs pas; sngags
gzhung rab 'byams la blo gros kyi snang ba cung zad rgyas pa sha kya'i dge slong kun
dga' bzang [580.5] pos zil gnon gyi lo sangs rgyas dbang bskur ba'i zla ba'i yar tshes
gsum la mkhas pa'i chos grwa dpal sa skya yon tan rin po che du ma'i 'byung gnas su legs
par sbyar ba'o!;'dis bdud dang phyir rgol ba dang mi mthun pa'i phyogs thams cad las
rnam par [580.6] rgyal bar gyur cig
See N2 624.5-625.3.
593
Van der Kuijp 1985, p.88.
594 !As!discussed!in!the!introduction!to!the!dissertation,!the!phrase!lam dus blangs pa
mthol bshags byed pa'i skabs byung ba is compelling but difficult to translate. Khenpo
! 192!
insurpassable intention of the tantric system called Destroyer of the Proponents of
Evil through Eliminating Objections to the Body Mandala and Dispelling
Evil View(s ).”595
Although their titles are abridged, the two texts are clearly mentioned side by side
as part of the same polemical project. Moreover, that project is identified as chiefly
concerned with the textual authority of the Hevajra body mandala system. N1 is
mentioned before N2, and it is possible that the phrase “treatise that establishes the
insurpassable intention of the tantric system” refers only to N1 or that they are regarded
as the same treatise. However, nothing conclusive can be determined based on the
conjunction dang. In fact, large portions of the texts are identical, but the differences, or
as van der Kuijp refers to them “variant readings,” are significant and worthy of scholarly
attention. While there are thirteen major variations between the texts, we will focus upon
a few key examples representative of the types of variation they exhibit.
The most general standard of variation between the texts concerns strategies of
citation. As we have already observed, N2 appears to eliminate the tradition of the
Mahåsiddha Indrabhüti from discussion. Abhayåkåragupta’s Åmnåyama•jarî, a text
whose significance has been discussed above, appears to be cited an additional time
[589.5-590.5]. Ngor chen concludes that citation with the following remark: “Although
some monks perform detailed analysis of the meaning of this text, flaws in your
intelligence replace them.”596 This more pointed critique of the opponent may come as a
surprise. While the title “Dispelling Evil Views” [N2] suggests a less personal attack
than “Destroyer of the Proponents of Evil” [N1], there are instances of more polemically
charged and directly targeted attacks within N2 than might be expected.
We will begin by examining the details of one of the longer divergences of N2

from N1, an approximately three and a half page section [N2 596.6-602.3] that departs
from 559.3 of N1; the text resumes at the same point at which it diverged. The context is
Ngor chen’s refutation of the charges based on an absences of deities common to inner
and outer mandala. Before diverging, both texts cite Chapter Twenty-Six of the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Tashi Dorje of the IBA suggested that it refers to Mkhas grub as someone who received
the lam ‘bras initiation but later critiqued the legitimacy of the transmission. Personal
communication, Fall 2011.!
595
Sangs rgyas phun tshogs 1688, p.546.4-.5
yang phyis kyis kye [546.4] rdo rje'i lus dkyil rgyud rgya gzhung gang nas kyang ma
bshad pa'i rtog brtags yin no zhes pa'i log rtog 'jigs su rung ba lam dus blangs pa mthol
bshags byed pa'i skabs byung ba de lung rigs man ngag gi sgo nas legs par sun phyung
nas; rgyud gzhung gi dgongs pa bla na med pa sgrub par byed [546.5] pa'i bstan bcos
chen po'i lus dkyil rtsod spong smra ba mngan 'jom zhes bya ba dang; lta ba ngan sel zhes
bya ba gnyis mdzdad do
!
596 !kye
dge sbyong dag; lung 'di'i don la zhib tu dpyad cig kyang; khyed kyi blo gros
[590.6] kyi mtshang dag dod rtog par 'gyur ro!
! 193!
Abhidhånottara Tantra [N1 557.6-559.3 & N2 595.3-596.6]. Ngor chen sums up the
import of the citation as follows:
“So, there is the explanation for arranging the eight blue deities of the mind cakra at the
heart center of the principal deity, the eight red deities of the speech cakra at the throat,
(and) the eight white deities of the body cakra at the crown. ‘All inner and outer
phenomena are purified by the union of body, speech & mind as buddhahood.’597 So, it
is correct to say that, combining the thirteen deities of the outer mandala and the twenty-
four deities of the inner mandala,598 there are thirty-seven factors conducive to
enlightenment in all.”599
However, N2 then proceeds to add an additional quote from Chapter Twenty-Seven of

that same tantra. [N2 596.6-598.1] The quote in N2 first describes the nine deities of the
outer mandala (Vajrasattva, four consorts and four offerings). It then arranges the body
mandala on the main deity of that array with the three cakras of body, speech, and mind
from the syllables hung, Å, and Aoum respectively. Ngor chen adds that here the
arrangement of the twenty-four heroes and goddesses of the three cakras is taught without
arranging the nine deities of the outer mandala. [598.1-.2] Therefore, here N2 specifies
additional information about the relationship of the two types of mandala in ritual
practice.
Ngor chen then cites a prophecy by Vajrapåñî regarding the root downfalls,
reminding monks that misunderstanding tantra can land you in hell. [N2 598.2-.3] This
same quote appears at a later point in N1 [575.6-576.1] as the dramatic conclusion to the
section on the support for Indian sources in the works of the mahåsiddhas. There it
follows right on the heels of the phrase mentioned above “Just because you don’t see it,
you criticize the profound pith instructions of the mahåsiddhas,” likely as a play upon the
verse from Sa pan: “Just because you don’t see it, doesn’t mean it isn’t there.” In N2,
Vajrapåñi’s ominous prophecy is followed by a warning that Mahåsiddha La ba pa’s
Cakrasaµvara commentary would be invalidated along with all the versions of the
practice that don’t arrange the deities of the outer mandala in the inner mandala; there
was no mention of La ba pa in N1. Ngor chen then proceeds to assert that if his
opponent’s position were valid, then Jetari’s Sådhana of the Four Seals would be
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
597 !This verse is found in mngon par brjod pa’p rgud bla ma [Abhidhånottara Tantra]
searchable sde dge 311b.1-2.!

598 !Note here that Ngor chen uses the term ‘outer mandala’ to refer to the eight goddesses
of the gates and corners and the main deity (no consort mentioned here) with retinue of
four goddesses. !
599 !zhes gtso bo'i lus kyi snying gar thugs 'khor gyi lha brgyad sngon po; mgrin par gsung
'khor gyi lha brgyad dmar po; sbyi bo'i 'khor lor sku 'khor gyi lha brgyad dkar [559.2] po
rnams dgod par bshad pa dang; sku gsung thugs kyi sbyor ba yis; phyi dang nang gi chos
rnams kun; byang chub phyogs kyis rnam par dag599; ces phyi dkyil lha bcu gsum dang
nang dkyil nyer bzhi po gnyis char bsdoms nas; byang phyogs so [559.3] bdun dang
sbyar ba'i dag pa gsungs pa'i phyir ro
!
! 194!
untenable on account of not explaining arraying nine deities of the outer mandala on the
body. 600
To review Ngor chen’s strategy in this section of N2, first he elaborates upon the
quotations from the Abhidhånottara Tantra of the Cakrasaµvara cycle shared by both N1
and N2, providing an additional citation from that tantra. He then cites Vajrapåñi’s
ominous prophecy, refers to Mahåsiddha La ba pa’s Cakrasaµvara commentary, and
interjects a somewhat snide comment about silencing his opponent. [596.1-598.6] Next,
he invokes Jetari’s phyag rgya bzhi'i sgrub thabs, a text that, according to Ngor chen, is
essential to the Sakya tantric perspective. [598.6-599.4] That text was presented in N1,
first in interpreting Bu ston’s contested statement on the absence of protective circle body
mandala in the Cakrasaµvara cycle [N1 551.3-552.1] and, later on, in the section on the
commentaries of the mahåsiddhas. Ngor chen’s next move is to call upon the writings of
the Guhyasamåja cycle to break down his opponent’s argument regarding the lack of
gods common to inner and outer mandalas. [N2 599.5-602.3].
In this next phase of the argument, Ngor chen’s polemics become increasingly
targeted. His use of the Guhyasamåja system narrows the focus of his critique, and he
stops just short of naming Mkhas grub as his opponent. Ngor chen first disputes the claim
(we know to have been made by Mkhas grub), that the Hevajra body mandala practice
(not named explicitly in Mkhas grubs’ text, but rather identified as the inner mandala of
one hundered and fifty-seven deities and the outer mandala of nine deities) is invalidated
because it lacks deities shared by both inner and outer mandala. Ngor chen argues that if
possessing deities common to both varieties of mandala were to be taken as the criteria
for assessing the efficacy of of a body mandala system, then even Årya Någårjuna’s
commentarial lineage contradicts the Guhyasamåja system. In support, he provides an
extensive quotation from Någårjuna’s student Åryadeva’s Caryåmelå-paka pradîpa
[599.6-601.4].601 Ngor chen summarizes the import of his citation of Åryadeva as
follows:
“In short, not arranging twenty goddesses taught in outer mandala within the inner
mandala and not arranging the thirty-two deities of the inner mandala in the outer
mandala and without any deities common to both, this supreme tradition is negated. If
you think one can’t establish an absence of deities common to inner and outer (mandala)
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
600 !The same section appeared in N1 [551.3-552.3]; however the citation there was
somewhat different and noticeably less fragmented. In N2, it appears much later in the
text; see N2 617.2-6 where it is combined with the preceding part of the Jetari quotation
[N2 615.2-617.2]. We have already established that Ngor chen regarded Jetari’s view as
the quintessence of the Sakyapa interpretation of Virüpa’s teachings. Tracing the
deployment of quotes from this incredibly short text by Jetari across Ngor chen’s two
texts provides a tangled web of repetition, modification, and fragmentation that we will
have to bracket for the time being. What is important to note is that the system of five
palaces or cakras of the body mandala described in this quote is shared by the Sakya
Hevajra Lam ‘bras system.!
601 !Note that Mkhas grub’s chapter on body mandala did not quote Åryadeva’s text,
relying more heavily upon Candrakî®ti’s rescension of Någårjuna’s Årya transmission.

Wedemeyer 2008 provides annotated translation of Åryadeva’s text.!
! 195!
in this tradition on account of the fact that the deities are arranged on the body of the
main deity, you’re wrong. The reason is that in the outer mandala the main deity is in
union with a consort. For the inner mandala, the main deity is at the heart, and the
consort is placed at the vajra door. Monks, in claiming to understand the texts of the
Årya cycle, (there is) this insincere speech which neglects the texts of Aa cu de ba
[Åryadeva]. What is it but aversion towards phenomena and persons?”602
In turning the attention back to the Årya Guhyasamåja system, the main topic of
Mkhas grub’s text as a whole, Ngor chen hones in on the identity of his opponent. The
passage also employs a common polemical strategy of augmenting a defense by showing
the opponent is not only wrong about your views but contradicting their own.
Ngor chen employs this very strategy in an even earlier instance of divergence in
N2. N1 552.4 references the opponent’s writings (which we can of course trace to Mkhas
grub) and criticizes a particular point within them: “You (rtog ge pa) say (that we)
“arrange one hundred and fifty-seven deities of the body mandala for only the outer
mandala which has only nine deities.” Although you write that, you’re wrong.”603
However, rather than just negating the opponent’s view with ‘you’re wrong,’ N2 instead
reads: “This is mere impudent speech without regard for the vast Vajrayåna scriptural
system. This is shown to be erroneous in three regards: obstructing highest yoga tantra,
obstructing the Hevajra system; obstructing your own claims.” Ngor chen’s critique hits
even closer to home within a longer divergence [N2 591.5-594.6]. This section is devoted
to defending against the claim of a discordance in the enumeration of deities in inner and
outer mandala. Ngor chen connects the opponent so explicitly with the Guhyasamåja here
that he seems to barely stop short of naming Mkhas grub:
“You babble senselessly without recalling your very own claims. Likewise, your very
own Årya Guhyasamåja commentary becomes invalidated on account of arranging the
forty-nine deities in the inner mandala within the mere thirty-two of the outer mandala.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
602 !mdor na gzhung 'dir phyi dkyil la lha mo nyi shu bshad pa de nang dkyil la ma bkod
nang dkyil gyi lha so gnyis po de phyi dkyil la ma bkod pas; phyi nang thun mong ba'i
lha gcig kyang med pas; mchog tu gyur pa'i gzhung 'di yang [602.2] 'gog par byad dam
gal te lugs 'di la phyi nang thun mong ba'i lha gcig kyang med pa ma grub ste; gtso bo lus
la bkod pa'i phyir ro snyam na ma yin te; phyi dkyil gyi gtso bi ni yab yum zhal sbyor yin
la; nang dkyil la gtso bo snying ga dang; yum rdo [602.3] sgor dgod pa'i phyir ro; kye dge
sbyong dag; 'phags skor gyi gzhung shes par khas len bzhin du; grub chen aa cu de ba'i
gzhung ma dran par bab col du smra ba 'di chos dang gang zag la zhe sdang bar gyur pa
las gzhan ci zhig
!
603
khyed kyis phyi'i dkyil 'khor la lha dgu las med pa la lus dkyil kyi lha brgya lnga bcu
rtsa bdun la sogs pa bkod cing ; zhes bris pa'ang mi 'thad de
instead of pa'ang mi 'thad de N2 587.4 reads just pa yang then adds:
[...587.4]rdo rje theg pa'i gzhung lugs rgya chen po ma mthong bzhin du; spyi brtol gyi
s?mra ba tsam zad de; 'di 'khrul par bstan pa la gsum; rnal sbyor bla med spyi dang 'gal
ba; kye rdo rje'i gzhung dang 'gal ba; rang gi khas blangs dang 'gal ba'o
! 196!
The reason is your very own claim: ‘arranging Amoghasiddhi on the two channels of
vital basis, Kßitigarbha on the two eyes, Vajrapåñi on the two ears, Rüpavajra on the two
eye orifices, ¸abdavajra on the two nostrils, Samantabhadra on the twelve joints,
Sumbha, on the two feet.’ You yourself wrote it. Remember?”604
Although this paraphrase of Mkhas grub’s text is not immediately recognizable

from his body mandala chapter, it corresponds roughly with Mkhas grub’s citation from
the Pin∂ik®ta. [KJ 257.5-259.1] However, it is also possible that it refers to another
portion of Mkhas grub’s text outside of the body mandala chapter or even to another one
of his texts. Ngor chen may even have undertaken a deeper investigation of Mkhas
grub’s writings between the time of composing N1 and N2. This longer “version” may
therefore represent Ngor chen’s findings.605 In the instances of divergence between N1
and N2 discussed thusfar, themes of variation in citation and increased polemical tone
have been established. The final standard of variation to be discussed here concerns
Ngor chen’s sophisticated approach to tantric commentary.
A particular passage from N2 [585.4-587.3] reveals an impressive attempt to
synthesize the versions of body mandala from different tantric texts and systems as well
as to evaluate their relationship. Here, Ngor chen synthezes the Vajramålå, Sampu†a ,and
Cakrasaµvara together with the writings of Darikapa and Ghantapa. The context is the
interpretation of Bu ston’s contested statement on the absence of protective circle body
mandala in the Cakrasaµvara cycle discussed in Chapter Three. Ngor chen quoted
Abhayåkåragupta’s Nißpannayogåvaî and the Sampu†a Tantra and referred the reader to
the explicit and implicit layers of textual meaning. He emphasized the importance of
“supplementing” [kha bskangs] from the oral instructions of the Sakya lineage gurus and
expressed the equivalence of those instructions with the teachings of Jetari. The
complexity of tracing citations of Jetari in Ngor chen’s texts has been referenced above.
In N2, Ngor chen postpones the citation of Jetari’s Sådhana of the Four Seals, for the
time being, and instead diverges to expound upon the precise methodology of
“supplementing” or “amending” [kha bskangs]:
“What’s the contradiction in supplementing in the context of Cakrasaµvara? If you think

it’s inappropriate to supplement what’s not explained in the Cakrasaµvara texts by
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
604
rang gi khas ci blangs ma dran par ci rung rung du smra ba'i bab col te; de ltar na;
khyed rang gi gsang 'dus 'phags lugs [594.5] kyi sgrub thabs kyi ti ka de'ang mi 'thad par
'gyur te; phyi dkyil la lha so gnyis las med pa las; nang dkyil la lha bzhi bcu zhe dgu
bkod ba'i phyir ro; rtags khyed rang gis khas blangs te; don yod grub pa brla rtsa(emend
to bla rtsa) gnyis dang; sa snying mig gnyis dang; [594.6] phyag rdor rna ba gnyis dang;
gzugs rdo rje ma mig gi sgo gnyis dang; sgra rdo rje ma rna ba'i sgo gnyis; kun bzang
tshigs bcu gnyis dang; gnod mdzes rkang mthil gnyis la bkod pa rgyu mtshan dang bcas
te bris pa de; nga [?da] dran par gyis shig
605
I am grateful to Professor Jacob Dalton for proposing this possibility. Ngor chen’s
own compositions on the Guhyasamåja system between 1423 and 1425 which just
preceded his 1426 body mandala texts may also have involved investigation of Mkhas
grub’s and Tsong kha pa’s writings on the Guhyasamåja system.
! 197!
drawing upon the Hevajra (corpus), then likewise, your own use of theVajramålå to
explicate all cases in which the Cakrasaµvara body mandala (appears) ...[long quote
from Vajramålå describing the body mandala of the support in equating the parts of the
body with parts of the celestial palace] ...would be invalid. This is because you’re
supplementing what’s not actually explained in the Cakrasaµvara text from a
Guhyasamåja explanatory tantra. To supplement the Cakrasaµvara body mandala with
the Vajramålå Explanatory Tantra would be wrong”.606
The quotation from the Vajramålå, elided here in my translation, is the very same
quote cited by Mkhas grub [Ocean of Attainment 255.6-256.4] to describe how to
generate the body mandala of the support. We have observed above the somewhat
puzzling significance of Mkahs grub’s invocation of the Cakrasaµvara system in a text
intended to focus upon the Guhyasamåja. For example, Mkhas grub’s choice to begin his
critique of other approaches to the body mandala practice as well his pronouncement of
his own interpretation with a quote from Ghantapa’s Cakrasaµvara text reinforces the
connection between the two systems in his text. [See Ocean of Attainment 234.1 and
251.3 respectively] Mkhas grub did make some effort to distinguish his remarks on
generating the body as the celestial palace from the Cakrasaµvara system, in stating:
“Therefore, as for the manner of establishing the body mandala in accord with the tantra
pi†aka and the texts of the mahåsiddhas and the texts of the Indian panditas, it is like this.
As for particular extraordinary cases such as Cakrasaµvara, it is not the situation.
Terrifying with power, it is not explained here, but it should be understood from the great
exegesis of Luipa, the discourse(s) of rje rin po che [Tsong kha pa] himself and so
forth.”607
Yet, Mkhas grub’s critique of the version of the body mandala practice
propounded by “some Tibetans” for “body mandala practice in the mother tantras” on the
basis of Bu ston’s interpretation of Nag po pa’s Cakrasaµvara commentary marks the
persistent relevance of the Cakrasaµvara system to Mkhas grub’s argument. We have
discerned how Ngor chen uses the Sampu†a Tantra to invalidate Mkhas grub’s claims and
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
606
[585] bde mchog gi skabs su'ang kha bskangs pa yin bas 'di la 'gal ba ci yod; gal te bde
mchog gi gzhung du ma bshad pa; kye rdor rje nas kha bskang du mi rung ngo snyam na;
de ltar na khyed rang gis bde mchog gi lus dkyil kyi skabs thams cad du; bshad rgyud rdo
rje 'phreng ba nas bshad pa ltar ...[586.3] zhes pa ci rigs pas kha bskang la gsal gdab; ces
bris pa yang mi 'thad par 'gyur te; bde mchog gi gzhung du dngos su ma bshad pa; gsang
'dus kyi bshad rgyud nas kha bskangs pa'i phyir ro; bde mchog lus dkyil gyi skabs su
bshad rgyud rdo rje phreng bas kha bskang ba yang mi thad de
607 !des na rgyud sde dang grub chen gyi gzhung dang rgya gar mkhas pa gzhung dang
mthun pa'i lus dkyil bsgrub tshul ni de ltar yin la; bde mchog la sogs pa'i thun mong ma
yin pa'i khyad par dag ni skabs ma yin pa dang mngas pas [254.6] 'jigs te 'dir ma bris gyi
; rje rin po che nyid kyi gsung rab lu i pa'i rnam bshad chen mo sogs las shes bar bya
zhing
!
! 198!
observed how the application of this explanatory tantra to both the Hevajra and
Cakrsamavara tantras is a source of potential confusion.
In our discussions of Ngor chen’s use of the Sampu†a Tantra above, we
considered the particular ambiguities presented by explanatory tantras for textual
interpretation. We also explored the possibility that polemical and exegetical contexts
like the body mandala debate are the very ground upon which the boundaries of tantric
hermeneutics are established. Ngor chen’s statement about supplementing the
Cakrasaµvara body mandala with the Vajramålå Explanatory Tantra further substantiates
that hypothesis. The Sampu†a is an explanatory tantra common to both Cakrasaµvara
and Hevajra systems and a text that has been labelled a “compilation” or “synthesis” of
other tantric texts.608 The Vajramåla, an explanatory tantra of the Guhyasamåjå, has been
described as an “anthology” of mahåyoga and yoginî tantra approaches and itself refers to
those different orientations in its respective chapters.609 Both texts present gray areas
regarding the scope and limits of their application. While Mkhas grub and Ngor chen
exploit this interpretive range, Ngor chen takes a further step in this particular divergence
in N2. Rather than merely synthesizing traditions, he is drawing a line to demarcate the
limit for the Vajramålå’s range of application.
Ngor chen provides a host of short excerpts and references in the following order:
Cakrasaµvara, the previously quoted Sampu†a Tantra, (Abhayåkåragupta’s)
Åmnåyama•jarî, Mahåsiddha Darikapa, the Vajramålå, Cakrasaµvara again, Mahåsiddha
Ghantapa and finally, the Vajramålå once again. These citations are organized as
evidence that it’s wrong to supplement the Cakrasaµvara from the Vajramålå and are all
familiar from Ngor chen’s larger argument. The essential differences Ngor chen suggests
concern the architecture of palace/body correlation, specifically the dimensions of the
four sides of the palace and the four gates of the mandala compound. He is attempting to
show that the explanations from the Vajramålå are incompatible with those associated
with the Cakrasaµvara; he includes the Sampu†a Tantra and its commentary, the
Åmnåyama•jarî, among the latter. While at first glance, this may appear to be a mere
summary of the ground covered in Ngor chen’s larger argument, N2 actually adds an
important element:
“For these reasons, it’s wrong to supplement the Cakrasaµvara from the Vajråmå¬a since
it (the Cakrasaµvara) agrees with the mother tantras. It’s utterly appropriate to
supplement it from the Hevajra cycle. This manner of scriptural explanation will be
expounded upon below.”610
In this display of commentarial mastery, Ngor chen thereby not only limits the
scope of application of the Vajramålå, but also, solidifies the hermeneutic link between
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
608
See Szanto 2013, p.5 and Lee 2003 , p.35
609
Kittay, p.189.
610
[587.1] rgyu mtshan de rnams kyi phyir; bde mchog gi skabs su rgyud rdor phreng gis
kha bskang ba mi 'thad par grub la; ma rgyud nang mthun pa'i phyir ro; kye rdor gyi chos
skor nas kha bskang ba shin tu 'thad par grub pa yin no; lung las ji ltar gsung tshul yang
'og nas rgyas par 'chad par 'gyur ro
! 199!
the Hevajra and Cakrasaµvara cycles. Ngor chen’s biography itself explains his tantric
polemics, as embodied in particular in his gnad kyi zla ser (and various letters and
replies) as a refutation of those who misunderstand the forefathers’ approaches to all
three systems: Hevajra, Cakrasaµvara, and Guhyasamåja.611
Our comparisons of the two ‘versions’ of Ngor chen’s text have highlighted three
important themes of variation: in citation strategies, polemical tone, and enhanced
doxographical or syncretic emphasis. Questions still remain as to the precise relationship
of these two texts. However, examining instances of these variations has enriched our
portrait of Ngor chen and the ways in which his identity as tantric commentator is
harmonized with his polemics. More largely, the comparison has reinforced the
importance of the body mandala debate as a site for experimenting with the boundaries of
tantric exegesis.
IV. Reflections on the Body Mandala Debate as Bodily Discourse
We will conclude by reflecting upon the significance of Mkhas grub and Ngor chen’s
engagement with the ritual mechanics and textual bases of body mandala practice. To
what degree does the body mandala debate have anything to do with the body? Catherine
Bell’s theories of the “ritual body,” discussed in the introduction, suggest that through
ritual the body both internalizes and even subverts external forces of discipline. The
issues raised in this debate speak to a similar variety of corporeal potential for the body as
both instrument and agent. The body mandala practice by definition internalizes the
cosmic structure of the mandala, transforming the practitioner’s view of his/her own body
as well as its relationship to the environment and community. Controversies over
techniques of mapping this structure onto the body explore its limits and strive to
articulate vital relationships between different ways of viewing the body. Perhaps like
simultaneously contemplating a vivisection and a skeletal drawing, mandalas ask us to
simultaneously view the world in two ways at once. To see the mandala is to see both a
bird’s eye view and an embedded view, to engage with the world with an apprehension of
the big picture.
To construct a mandala is to construct a world; to dissolve it is to abandon form
itself to the boundless expanse of emptiness. Through the embodied experience of the
practitioner, the mandala is spontaneously redeployed in new ways. The body mandala
debate presents a nuanced concept of the body as a “basis” or “support” [rten] for
generating deities and ultimately, for soteriological practice. The status of the body as
support raises questions about the relationship between ordinary and enlightened bodies
as well as between the ordinary body and the subtle body, or specifically in this context,
the vajra body. We have discussed the complex formulations of causality produced by
Tibetan scholars to describe the body’s role in facilitating liberation. To construct a body
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
611
Sangs rgyas phun tshogs, 564.5-.6. [546.5] gzhan yang bla ma gong ma'i lugs kyi bde
kye gsang gsum gyi don ma rtogs pa dang; log par rtogs pa mtha' dag sun 'phyin pa'i
bstan bcos chen po legs [564.6] bshad gnad kyi zla zer; spring yig dbang po'i rdo rje; de
bzhin du dris lan sna tshogs kyi sgo nas log rtog zlog par byed pa'i bstan bcos du ma
mdzad do
! 200!
mandala is to ‘pierce to the pith’ of the body, to access its subtle energies and to harness
their potential.612 Body mandala is therefore a ritual technology for interacting with the
body in a deeper way; the body mandala debate suggests competiting systems of
knowledge and interpretation of the body’s points of power and vulnerability and how
best to control them.
In the body mandala debate, the body also functions as what Foucault would call
a “cultural text.”613 As summarized by Judith Butler, Foucault describes the body as “a
site where regimes of discourse and power inscribe themselves, a nodal point or nexus for
relations of juridicial and productive power.”614 Mkhas grub and Ngor chen use the body
as a site for contesting issues of textual authority. Through exegesis and polemics, they
reinvent the relationships between texts in interpreting the body mandala practice. The
body provides a site for establishing the legitimacy of a tradition through
complementarity and conflict, and therefore as the ground for deploying “discipline” as
defined by Foucault. The body mandala debate therefore suggests complex dynamics of
institutional identity that help us to discern the contours of the “social body” of fifteenth-
century Tibet in a new light.
Butler highlights important paradoxes in Foucault’s treatment of the body,
especially problems of its ontology and materiality:
“In a sense, for Foucault, as for Nietzsche, cultural values emerge as the result of an
inscription on the body, where the body is understood as a medium, indeed, a blank page,
an unusual one, to be sure, for it appears to bleed and suffer under the pressure of a
writing instrument. ”615
Butler’s investgation of Foucault’s perspective on body as “surface,” “resistance,” and

“cultural or discursive practice” might be translated into our interpretation of the body
mandala debate:
“That history is ‘inscribed’ or ‘imprinted’ onto a body that is not history suggests not
only that the body constitutes the material surface preconditional to history, but that the
deregulation and subversion of given regimes of power are effected by the body's
resistance against the workings of history itself...Yet his statements on ‘history’ appear to
undermine precisely the insight into the constructed status of the body which his studies
on sexuality and criminality were supposed to establish.616
In order to avoid limiting our search for ways of knowing the body through the
body mandala texts, it is necessary to combine Bell’s ideas about the “ritual body” with
Foucault’s “cultural text.” Butler’s critique of Foucault is also valuable in prompting us
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
612
On the etymology of mandala as ‘taking the pith,’ see Lee 2003 p. 130 fn4.
613
edited by Donn Welton (Malden, MA & Oxford: Blackwell Publishers, Ltd., 1998).
614
Butler 1989, p.601. Butler, Judith (1989). “Foucault and the paradox of bodily
inscriptions.” Journal of Philosophy 86 (11): 601-607.
615 !Butler!1989, p.604.!
616 !Butler!1989, p.607.!
! 201!
to question our own motivations for seeking the body in the body mandala texts. In
combining these different approaches to the body together with those of Mkhas grub and
Ngor chen, we can see that the body is not just the ground for resistance or a blank slate,
the remnant of karmic defilement or the basis for transformation, illusive or real.
Interpreting the body mandala debate presents the possibility that body is,
perhaps, best known as what Sa chen would call an “explanatory continuum,” or even, an
“explanatory tantra”:
“[The method continuum (i.e.the body)] is also called the ‘explanatory continuum’
because the root to be realized, which is the mind, is realized and mastered through the
alignment of the dependently arisen connections in the body, which is the agent of
realization, and thus the explanatory continuum.”617
Just as the body helps us to better apprehend the subtle perfume of the nature of
the mind, it is always pointing beyond itself.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
617 !Stearns 2006, p46. Translation of Sa chen kun dga’ snying po’s Explication of the
Treatise for Nyak.!
! 202!
Conclusion: Iconoclasm and the Subtle Body: Art, Ritual, and the Body in
a Single Painting of the Cakrasaµvara Body Mandala
This conclusion examines an unusual painting of the Cakrasaµvara body mandala [Tib.
lus dkyil Skt. deha-mañ∂ala] practice created in Nepal in the nineteenth or twentieth
century.618 This anomalous painting provides an opportunity to revisit the broader issues
of embodiment, ritual, and representation framing this dissertation and discussed in its
introduction. [Fig.24] In Chapter Five, we initiated a comparison of visual and textual
representations of embodied ritual practice. In the process, we experimented with
LaTour’s concept of “iconophilia” as a tool for interpreting sådhana as bodily discourse.
In this conclusion, we will re-introduce a dialogue between visual and textual
representations of the body and also more deeply explore the relationship of images and
bodies.
How do we interpret the absence of comparable examples of paintings of body
mandala alongside the array of other varieties of mandala paintings created by Newar and
Tibetan artists? Does this absence qualify as a form of iconoclasm? Within the visual
culture of tantra, there are no apparent reservations about creating images of divine or
human bodies, although there are certainly iconographic convention and guidelines for
distinguishing the two. However, images of the subtle body are more rare. By “subtle
body,” we mean a body defined by invisible structures and processes realized exclusively
by the advanced tantric practitioner through sustained ritual practice. Does the subtle
body perhaps pose a problem of potential category confusion of divine and human
bodies, a confusion on par with the kind discussed by Mkhas grub in Chapter Four in the
context of body mandala?
The first part of this conclusion deals with particular modes of representing the
human body in the visual culture of India and the Himalayas. 619 We will explore the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
618
I presented my research on this painting at a workshop on Buddhist art and text in
September 2013. I am grateful for the feeback I received from the workshop participants
at that time. See my forthcoming publication: Dachille, Rae Erin. “Iconoclasm and the
Subtle Body: a Study of Art, Ritual and the Body in a Single Nineteenth- Century
Nepalese Painting.” Reading Outside the Lines: A Workshop on the Intersection of
Buddhist Art and Texts September 13-15, 2013, Ludwig-Maximilians-Universitœt,
Munich (Contemporary Issues in Buddhist Studies) (Publication forthcoming 2016).
!
619
Swiss art critic Heinrich Wölfflin (1864-1945) introduced the term “mode of
representation” within art historical parlance in his Principles of Art History, a
comparison of ways of seeing in the sixteenth and seventeenth centuries, as embodied in
the classical and baroque movements in Europe. Melville 2009 provides an overview of
Wölfflin’s contributions to the study of the role of representation in the “history of
vision.” Melville observes: “two levels of linguistic analogy are run constantly together
in this text, thus tangling together problems of translation and representation. Such terms
as ‘one’s own language’ or mode of representation as such’ introduce a deep
complication to notions of medium, genre, and relation within art history…all of them, I
! 203!
solutions the creator of this nineteenth/twentieth-century Nepalese body mandala painting
produced in response to the challenges of representing the body mandala practice
alongside other representations of the subtle body. We will then move on to reconsider
the concept of “fabrication” [Tib. bcos ma] familiar from our discussion of Mkhas grub’s
writings on body mandala in Chapter Four and explore its implications for understanding
material representations of this ritual practice. In doing so, we bring to light a significant
aspect of Tibetan Buddhist ritual and philosophical understandings of material creation
that may enrich art historical approaches. In addition, we will call into question
assumptions about the relationship of embodiment and materiality, suggesting that the
categories of body, image, and matter be re-evaluated in light of Buddhist sources.
Finally, we conclude by reflecting upon the relationship of representation and reality
expressed through body mandala to make sense of how images and bodies function as
“supports” for liberation.
I. Modes of Representing the Subtle Body

A. The microcosm-macrocosm correlation
The body mandala painting from a Private Collection has been identified by Jeff
Watt of the Rubin Foundation as a nineteenth or twentieth- century creation from Nepal
based in the Newari Cakrasaµvara tradition.620 [Fig.24] However, the painting appears
to be an anomaly, the only one of its kind. Why would Buddhists choose not to depict
the body mandala tradition in material form? The rich legacy of paubha paintings of
mandala created by Newar artists, often for Tibetan patrons, and specifically of
Cakrasaµvara mandala attests to the fact that in Nepal itself esotericism didn’t
necessarily limit artistic production. Despite the prohibition on revealing tantric practices
to the uninitiated, there is a prolific tradition of representing material (or what will be
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
suggest, work to maintain a constant reference to reading within the field of the history of
vision, and to maintain it as at once fleeting and natural, something like a metaphor- but a
metaphor without which one cannot manage, a catechresis then.” Stephen Melville,
“Deconstruction and the Limits of Interpretation,” The Art of Art History: a Critical
Anthology, edited by Donald Preziosi (ed) (Oxford & New York: Oxford University
Press, 1998 & 2009): 271-316. See pp. 279-280.
I have explored modes of representation as an interpretive tool in my studies of a
set of seventeenth century Tibetan medical paintings: Rae Erin Dachille-Hey, “The Case
of the Disappearing Blue Woman: Understanding How Meaning is Made in Desi Sangye
Gyatso’s Blue Beryl Paintings,” Asian Medicine Vol. 6, Issue 2 (2011): 293-320. See
also Rae Erin Dachille, “Modes of Representation and Meanings in the Blue Beryl
Paintings of Desi Sangye Gyatso” (unpublished Masters thesis, University of Wisconsin,
Madison, 2007).
This approach to representations through emphasizing mode over content has also
been articulated by Vidya Dehejia in the context of early Indian Buddhist art. See Vidya
Dehejia, “On Modes of Visual Narration in Early Buddhist Art,” The Art Bulletin Vol.72
No.3 (Sept. 1990): 374-392.
!
620
See www. himalayanart.org HAR Item no. 59648 for image and identification.
! 204!
termed ‘outer’) mandala, the mandala as a celestial palace inhabited by deities depicted in
two-dimensional form; such representations take the form of murals paintings, paintings
on cloth or more ephemerally, altars of painted powder that are ritually destroyed. The
absence of examples for comparison with our body mandala painting suggests that
perhaps there is something special about the body itself that resists such representation.
Let’s look for a moment at a few examples of representations of the body in South
Asia and the Himalayas that may suggest comparable modes of representing aspects of
the body invisible to the untrained eye. The first is a Hindu depiction also from Nepal
and likewise dated fairly late, the eighteenth century. [Fig.25] Here we find the
mapping of the universe onto the human form in a union of microcosm and macrocosm.
This image type may be classified as one of the Vißñu vi!varüpa forms in which the body
of the god contains the universe in its entirety. Howard has detailed aspects of the
evolution of the vi!varüpa type from Vedic through to Upanißadic and Puråñic sources.621
The lord’s body is encircled by a large black and white serpent, presumably the serpent
upon whom he rests during periods of cosmic gestation. A small red serpent also appears
at his feet and a rope-like form extends from a small white house beside his waist along
the lower contours of the body and up to an identical structure on the opposite side.
Deities inhabit subtle points of the body like the set of seven cakras as well as more
external points on the arms and legs and the gates or orifices of the body. Demons,
beasts and humans all appear within the serpentine enclosure, while the lower registers of
the painting seem to depict an epic struggle with demonic forces. On the upper portion of
the torso a crowned red deity approaches a mythical beast and what appear to be four
human figures. The main figure is positioned in three-quarter view circle with hands in a
a•jali mudrå facing a red circle containing a seated figure facing him. While the precise
identification of all the elements of this painting lies beyond the scope of this article, on a
general level we may compare the manner of mapping deities onto the body, the
correlation of macrocosm and microcosm, and the dialogue between internal and external
produced therein with the representation of the body mandala.
Another image of the Vißñu vi!varüpa from the collection of the Victoria and
Albert Museum appeared in the recent Smithsonian exhibition, Yoga: the Art of
Transformation, an exploration of the visual culture of yoga from the first millennium to
the present day.622 [Fig. 26] In this representation from Jaipur, dated to between 1800
and 1820, the entire cosmological drama is located within the boundaries of the blue
body of the god rather than spilling outward across the different domains of the painting.
The correlation of lord and cosmos is complete, reinforced by the many-headed serpent
upon which he stands and the cosmic elements of sun and moon appearing as the eyes.
Clouds, vapor and rain are mediating elements between the body of the god and the
otherwise empty background. Unlike the previously discussed example, the body of the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
621 !Angela Falco Howard, The Imagery of the Cosmological Buddha (Studies in South
Asian Culture). Leiden: Brill, 1986. See especially pp.58-64.!

622
The exhibition, curated by Debra Diamond, first appeared at the Arthur M. Sackler
Gallery in Washington, D.C. (10/19/2013-01/26/2014) before proceeding to the San
Francisco Asian Art Museum ( 02/21/2014-05/25/2014) where I had the opportunity to
view it. The final leg of the tour was the Cleveland Museum of Art (06/22/2014-
09/07/2014).
! 205!
god is forward-facing, adorned with jewelry, and possessed of his signature attributes like
the conch and discus. Both deities and architectural forms make up the structure of the
body. Deities are shown residing on key points of the body such as the forehead, throat,
heart and genitals, but no circles forming cakras or lotuses contain them. Other deities
appear on the arms and in the nostrils, reflecting some concern with occupying both
peripheries and orifices of the body. Only the shoulder region and central abdomen are
vacant spaces. Gold lines create registers in the lower part of the body and order the
interaction of various human, divine and animal figures; the effect is an apparent
hierarchy of cosmological domains.
The palatial architectural structures, in particular those at the crown, between the
eyebrows, and at the throat, suggest a correlation of human and divine through the
metaphor of the space of the royal court. David Gordon White has located some early
imaginings of the subtle body within the Kå†haka Upanißad (c. third century BCE) that
depicts the body as a “fort with eleven gates” inhabited by the soul in the form of thumb-
sized human who is the locus of divine adoration.623 One way of interpreting
architectural spaces within representations of the body is according to a socio-political
model assuming a correlation of human and divine orders.624 A painting from the Buryiat
Museum exemplifies the application of architectural and courtly metaphors for describing
the body within the textual and visual culture of Tibetan medicine. [Fig.27] For
example, registers three and four of the painting depict the heart as the king and the
members of his court as the supporting organs.625 Registers one, two and five show the
other parts of the anatomy as architectural forms.
One way of defining mandala is as an assembly of buddhas and attendant deities,
in other words, as a royal court.626 The mandala palace invokes multiple modes of
expressing hierarchy, through vertical as well as through radial stratification, negotiating
both the interplay of top and bottom as well as of center and periphery. Both of these
modes co-exist in this Victoria and Albert Vißñu vi!varüpa. Key deities are aligned on a
central axis, although the identity of this axis with the central channel of the subtle body
is not explicit; deities are also clustered around the figure at the heart and stratified in the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
623
David Gordon White, “Yoga in Transformation,” in Yoga: The Art of Transformation,
ed. Debra Diamond (Washington, D.C.: Freer Gallery of Art and Arthur M. Sackler
Gallery, 2013), 35-45. See p. 36 and Kå†haka Upanißad 4.12; 5.1,3.
624
Indeed the correlation of king and god is a common trope articulated in Indian myth
and ritual.
625
Such correlations can be found within the Chinese medical system as well. See Livia
Kohn’s discussion of the courtly model in Daoist representations of the body-cosmos
relation though reference to texts such as the Huainanzi, Huangdi neijing suwen and
Laozi zhongjing. Livia Kohn, “The Daoist Body of qi,” in Religion and the Subtle Body
in Asia and the West: Between Mind and Bod, edited by Geoffrey Samuel and Jay
Johnston (Routledge Studies in Asian Religion and Philosophy) (London & New York:
Routledge, 2013), 16-32. See pp. 29-30.
626
This trope of the “imperial metaphor” is appropriated by tantric literature as well,
especially through the template of the mandala. See Ronald M. Davidson, Indian
Esoteric Buddhism: A Social History of the Tantric Movement (New York: Columbia
University Press, 2003).!
! 206!
registers in the lower body. Relationships between figures also manifest in their particular
postures and gestures of exchange; multiple dramas are enacted within the space of the
body.627
Angela Falco Howard has studied the impact of Vißñu vi!varüpa iconography
upon Buddhist representations of the cosmological Buddha found in the caves of China
and Central Asia.628 Howard shows how early Hindu representations of the cosmic Vißñu
interpreted his manifold and all-encompassing nature as a many-headed form expanding
within the fixed limits of an aureole. However, the creators of Buddhist representations
were inspired by textual accounts of the cosmic aspect of Vißñu to create a cosmic
Buddha whose manifold nature was expressed not merely through a multiplication of
heads and arms but rather as a body populated by an infinite variety of human, animal
and divine forms.629 This infinity of forms indicated the totality of the realm of rebirth
within the form of this universal Buddha as exemplified by an early representation from
Northern Wei- (386-534) Dunhuang, Mogao Cave 428. 630 [Fig 28] A Sui-dynasty (581-
618) example, a gray marble statue in the collection of the Freer Gallery, exploits the full
potential of its three-dimensionality; all sides of the Buddha’s garment are covered with
scenes of the realms of rebirth as well as with narratives from the life of ¸åkyamuni.631
[Fig.29] The latter example, in particular, employs a multiplicity of architectural forms
in combination with mountain formations to frame and demarcate the various narrative
vignettes. Both examples employ a vertical hierarchy in depicting the levels of rebirth on
the garment draping the Buddha’s form while marking a central point on the body. In the
Dunhuang example, this central point is demarcated by a demi-god holding the sun and
moon while in the Freer example, the heart center is marked by Mt. Meru and a pair of
någas. Stephen Teiser has described these images in his study of representations of the
Buddhist wheel of life, considering similar issues of vertical vs. radial modes of
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
627 !Within the textual traditions of Buddhist yoginî tantra, it is a common trope to locate
the ritual drama within the space of the womb of the ∂åkinî; within certain varieties of
body mandala ritual, the practitioner projects this drama onto the womb of the consort.
However, I am not aware of any visual representations depicting the ∂åkinîs in this way.
Therefore, the Vißñu vi!varupa imagery may be an unusual occasion in which the rich
detail of mythological drama connected with a totalizing bodily form actually appears in
a visual representation. !
628
Howard 1986, especially pp.58-64. I am grateful to Monika Zinn for encouraging me
to explore these materials during Reading Outside the Lines: A Workshop on the
Intersection of Buddhist Art and Texts September 13-15, 2013, Ludwig-Maximilians-
Universitœt, Munich.
629 !Ibid.!
630
See figure 13 in Angela Falco Howard, “A New Chronology for the Kizil Mural
Paintings,” Archives of Asian Art, Vol. 44 (1981): 68-83. Here Howard explains that
there is some discrepancy in the dating of cave ?428 as Northern Zhou ca. 575 by the
Dunhuang Institute vs Northern Wei (ca. 525) by Paul Pelliot. See p.74.
631
For a more detailed discussion of the Freer example, see Angela Falco Howard, “The
monumental ‘Cosmological Buddha’in the Freer Gallery of Art: Chronology and Style,”
Ars Orientalis, Vol. 14 (1984): 53-73. The Dunhuang example also appears in this article,
as figure 2, but is labeled as cave 428 vs. 429 as in Howard 1981.
! 207!
representing the complete span of the wheel of rebirth.632 Teiser reflects upon the
embodiment of narrative and cosmos as Buddha as follows:
“The cosmos is part of his body-a point made equally well, in different form, by the
common mythological motif in which entire world systems emanate from various parts
(pores, tongue, ürnåke!a [wisp of hair between the eyebrows]) of the Buddha’s body. In
one mode, the whole world is projected outward from a single part of the Buddha’s body.
In the other mode, the world is inscribed upon the entire body of the Buddha. In both
modes-unlike pictures of the wheel of rebirth-the Six Paths are tied directly to the
physical body of the Buddha.”633
What is achieved by representing a multiplicity of forms within the body of a

deity? Is the body itself a narrative frame with the capacity to simultaneously contain
multiple temporal and spatial dimensions of reality? One key distinction between the
Vißñu-vi!varüpa forms and many of the cosmological buddhas discussed by Howard is
that while the former display cosmic bodes inhabited and pervaded by or even composed
of manifold forms, in the case of the cosmological buddhas, the variety of forms might be
described as ornamental. In other words, the narrative detail is translated into visual form
as a tapestry, a garment worn by the buddha that appears more exterior than interior (if
indeed those categories are deemed useful for examining this issue). From a
philosophical perspective, this garment is more like the screen upon which the
proliferation of forms of the phenomenal world manifest in an ephemeral manner than a
representation of a fixed ontological state.
In the case of the Vißñu vi!varüpa examples we have examined, locating the
narrative and formal details within or surrounding the deity’s body conveys a more vivid
equivalence of the body of the deity with the cosmos. Ornament does, however, play a
role in these examples as well. In the Nepalese example, the body of the deity itself is, in
a sense, ornamented by cakras and various figures. The Indian example exhibits an even
stronger principle of ornamentation, with the golden architectural structures at the throat,
between the eyebrows and at the crown bearing a strong resemblance to jewelry. The
deity’s body is marked and adorned with the forms of various deities and their respective
abodes. As with the thirty-two lakßañas or “marks” that identify the Buddha’s body as
extraordinary (like his elongated earlobes and the wheels that appear on the soles of his
feet), these ornaments or markers indicate the divine nature of this body.
We will compare two illustrations from a Rajasthani manuscript dated to 1824 of
Gorakßånatha’s hatha yoga text, the Siddha Siddhanta Paddhati. This comparison
provides the opportunity to juxtapose the two different modes of representing the
structure of the subtle body of the yogic practitioner: in terms of architectural forms and
of cakras. [Fig. 30 & Fig. 31] The text itself, which Mallinson 2011 has dated to the
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
632
Stephen F. Teiser, Reinventing the Wheel: Paintings of Rebirth in Medieval Buddhist
Temples (Seattle: University of Washington Press, 2007), 130-135. I have found Teiser’s
work helpful in identifying the central points of these figures.
633
Teiser 2006, 131.!
! 208!
early eighteenth century, propounds a nine-cakra system.634 These folia from the
Mehrangarh Museum Trust (also featured in Yoga: the Art of Transformation) use a
common template of the body as flesh-colored, facing front against a solid blue
background, and adorned with jewelry. Only a decorative mountain or cloud element
around the shoulders in Figure 31 bleeds beyond the boundaries of the body. Both
cosmic elements and deities inhabit the body; the sun and moon are located on the
cheeks, and ¸iva resides in the genital region, with Brahma in the area between the throat
and heart. Hosts of beings appear on the upper arms and inside the elbows, and a blue
figure, perhaps K®ßña, appears on the upper left shoulder. A male and female retinue
congregates near the navel, listening to the teachings of a divine being standing upon a
cosmic tortoise. Debra Diamond has identified further cosmological elements within this
painting, equating the palaces, for example, with the fourteen worlds.635 Both
architectural and environmental/geographic forms structure the space in a similar manner
to the cosmological Buddha images discussed by Howard but with less density of form.
The artist is, in a sense, punctuating the space of the body, making it legible to the
viewer. In the case of the two folia from the Siddha Siddhanta Paddhati, while Figure
31 imagines the body as built environment composed of architectural forms, Figure 30
maps the cakras onto the body. There, only two deities are shown residing within key
points along the central axis of the body, at the navel and genital region respectively. The
figure is more akin to a human body than the other examples we have discussed, neither
resting upon a mythically charged animal nor containing cosmic elements like the sun
and moon.
Thriving upon multiple correlations of human, divine, and cosmic bodies, the
body mandala practice plays with conceptions of inner and outer in comparable ways to
the cosmic Vißñu and Buddha representations. The ritual destabilizes any fixed boundary
between self and other, inside and outside, transforming the perception of the practitioner
by simultaneously revealing the nature of one’s own body as a divine container inhabited
by deities and of the whole universe as part of the mandala. The question of what it
means for deities to reside within the body has larger philosophical implications that may
lie beyond the purview of this paper, but a few basic aspects of the problem will be
introduced. First, one must consider the question of what kind of body is being inhabited.
In the case of the Vißñu vi!varüpa, it is a divine body that encompasses or permeates the
realms of existence and articulates a strong microcosmic-macrocosmic relationship. In
the case of representations of the bodies of yogic practitioners, these bodies reveal
previously invisible and inherent forces manifested through practice.
Methods for Interpreting Representations of the Subtle Body
A Ka!miri scroll also featured in Yoga: The Art of Transformation instantiates the
template of the body in the format of the scroll itself.636 The scroll combines the
representation of the cakras and deities residing therein with architectural forms, all
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
634
Debra Diamond, Yoga: The Art of Enlightenment (Washington, D.C.: Smithsonian
Insitution, 2013), 166 fn9.
635 !Diamond 2013, 164.!
636
See Diamond 2013, fig 11c.
! 209!
aligned along a vertical hierarchy. Therefore the “corpus” becomes the body upon which
the cakras and cosmological systems are diagrammed. This fusion of corporeal spaces
evokes Foucault’s interpretation of the body as a “cultural text” upon which society maps
its norms and desires; for Foucault, this body/text is the ground for regulation through
“discipline.”637 638 Catherine Bell’s theories of the “ritual body,” discussed in the
introduction, suggest that through ritual the body both internalizes and even subverts
external forces of discipline. If we regard visual representations of the cosmic man or the
yogic practitioner in light of Foucault and Bell, what can we see about the systems of
authority they internalized and externalized, and transformed or were transformed by?
Debra Diamond, Associate Curator of South and Southeast Asian Art at the Freer
and Sackler Galleries and curator of Yoga: The Art of Transformation, observes that “a
multiplicity of subtle body systems flourished in medieval and early modern India.”639
This diversity of systems should be taken into account in comparing different visual and
textual representations of the subtle or yogic body and different versions of ritual
practices like body mandala. The hierarchical placement of deities and cosmological
sites should be evaluated in relation to ritual, medical and even socio-political
understandings of the sites of power and vulnerability of the human body. For example,
the cakra system was gradually absorbed by Hindu yoga and Ayurveda from Buddhist
and Hindu tantric practices.640
Samuel has identified the principle of pråña as a key concept in subtle
formulations linking human physiology and cosmos, a concept intertwined with theories
of selfhood or åtman.641 Building upon the work of Larson, he indicates Pata•jalis’s
fourth-century Yoga Sütras as one source for tracing the “localization of yogic processes”
in particular sites within the body.642
In comparing different versions of ritual practices oriented around the subtle
body, practices such as body mandala, it is necessary to take into account variant
understandings of the movement of psycho-physical energies along pathways within the
body’s boundaries and beyond. Such understandings of the body’s hidden structures and
processes underlie visual and textual accounts of deities residing within the body. In other
words, locating deities on or within the body is a way of punctuating it, drawing attention
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
637
edited by Donn Welton (Malden, MA & Oxford: Blackwell Publishers, Ltd., 1998), 61-
70. Hekman provides an interesting comparison of Judith Butler and Susan’s Bordo’s
approaches to Foucault’s writings on the body.
638
Catherine Bell, Ritual Theory, Ritual Practice (Oxford: Oxford University Press,
2009).
639
Diamond 2013, 167.
640
Geoffrey Samuel, “Introduction to Part One” and “The Subtle body in India and
Beyond” in Religion and the Subtle Body in Asia and the West: Between Mind and Bod,
edited by Geoffrey Samuel and Jay Johnston (Routledge Studies in Asian Religion and
Philosophy) (London & New York: Routledge, 2013), 13-15 & 33-47. I am grateful also
to Lisa Brooks, my peer at UC Berkeley, for her insights on this issue during our May
2014 communication.
641
Samuel 2013, 34.
642
Samuel 2013, 34 and Larson 2009, 488 & 493.
! 210!
to sites of particular strength or vulnerability. Points on the body are marked for a
reason, be it to locate powerful forces of heat or generation that can facilitate an
experience of bliss or to locate places where the flow of breath can become trapped and
stagnate, prohibiting future spiritual advancement. Whether the nature of the potentiality
is strength or vulnerability, these bodily sites are marked as important objects of focus for
the ritual practitioner. To know the body is to be empowered.
When we search for precedents for visual and textual representations of deities
within the body and, more specifically, in the cakras, the question of the origins of such
practices comes to the forefront.643 The images discussed thusfar in this article
emphasize the microcosmic-macrocosmic correlation; the latter two images of the yogic
body [Fig. 30 & 31] reveal the hidden structures of more explicitly human rather than
divine bodies. Though we have a plenitude of Indian textual sources of the Vißñu
Vi!varüpa or mahåpurußa variety, most of the Hindu visual sources we have discussed
above do not predate the nineteenth century. The “cosmological Buddhas” from East and
Central Asia appear over a millennium earlier.
There are also Jain images of a cosmic man dating from the fifteenth to
seventeenth centuries. [Fig. 32 & Fig. 33] 644 John Cort describes how these images
were used by preachers to reveal the hidden order of the universe and to locate human
practitioners within it.645 Phyllis Granoff shows how such representations juxtapose the
mathematical order of the cosmos with the chaos and uncertainty of the round of rebirth.
Both of the Jain cosmic man representations featured in Granoff’s catalogue reveal an
underlying order to the cosmos as a means of promoting moral action. The cosmological
Buddha images described above are likely to have inspired a similar response in their
viewers. Granoff highlights how cosmic order is encoded in the images through the use
of numbers and labels to consolidate vast amounts of information.646 This detail is of
interest in demonstrating one technique available to artists for creating a totalistic vision
of the universe within the confines of the human form. It may be useful to keep this
method of encoding in mind later in this article as we proceed to examine the ways in
which the Nepalese painter of the body mandala painting at issue coped with the
challenges of spatial limitations.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
643
Mallinson has described how the cakra system itself is predated by the system of
granthi or knots within the subtle body. (Personal communication with Diamond cited in
Diamond 2013, p166 fn7) Therefore, one might begin by searching for references to
deities inhabiting these sites. For a visual representation, see Diamond 2013, Figure 11a,
a late seventeenth-century painting in which Brahma, Vißñu, and ¸iva are located at the
navel, heart and forehead of the body of Raja Mandhata. See Diamond 2013, p166.!
644 !These images were featured in the 2009-10 exhibition, Victorious Ones: Jain Images
of Perfection, curated by Phyllis Granoff at the Rubin Museum of Art.!

645
John E. Cort, “Contemplating the Jain Universe: Visions of Order and Chaos,” in
Victorious Ones: Jain Images of Perfection, edited by Phyllis E. Granoff (New York:
Rubin Museum of Art, 2009-2010), 34-37.
646
Phyllis Granoff, “The Mathematics of the Jain Cosmos,” in Victorious Ones: Jain
Images of Perfection, edited by Phyllis E. Granoff (New York: Rubin Museum of Art,
2009-2010), 54. !
! 211!
Turning now to focus upon representations of the subtle body, a body marked by
vital points and permeated by the flow of psycho-physical energies, we are lead to ask if
there are factors that may have made the body a more compelling subject for visual
representation during the nineteenth century and beyond than it had been in previous eras.
For example, might we find clues in the bhakti movement and its associated visual
tradition which emphasizes the overall breakdown of the boundaries between devotee and
god, with the human devotee partaking of and even making up the body of the divine?
With regard to representations of the subtle body that make visible the invisible centers
through which psycho-physical elements circulate, why do we suddenly find examples
like the folia from the Siddha Siddhanta Paddhati arising in the nineteenth century? In
discussing the representations of the subtle body featured in the Yoga exhibition, Curator
Debra Diamond remarked that in the Indian case, representations of the subtle body of
the yogic practitioner seem to proliferate in response to the demands of “new
audiences.”647 Remaining open to the utility of patronage as a model for interpreting the
connection between the production of visual representations and the demand for a means
of circulating systems of ritual or spiritual knowledge to audiences of often elite patrons,
we will investigate a few Tibetan images that reveal the secrets of the human form
invisible to the naked eye.
III. Visualizing the Invisible: Tantric and Medical Imagery in Tibet
The visual cultures of both tantra and medicine in Tibet employ modes of representation
for revealing hidden aspects of the structures and processes that lie beneath the surface of
the human body. This section takes a closer look at the two-sided cakra diagram
discussed in Chapter Five and compares it with a Tibetan medical painting. Heller 2010
theorizes that the cakra diagram is an eleventh-century creation from Western Tibet.
[Fig.23] Although this dating is not definitive, it is likely that this diagram is
significantly earlier than the Indian and Nepali examples discussed above. In line with
the theories of Robert Mayer, Heller suggests that this diagram provides evidence for the
transmission of Indic tantric knowledge to a Western Tibetan audience and perhaps even
of the incorporation of Hindu elements into Buddhist tantra.648
To review, the front side of the diagram depicts a body with some unusual cakras
and inscriptions. With regard to these inscriptions, Heller has observed that “rather than
give names for each cakra as in an anatomical system, the cakra are associated with
different ritual phases.”649 No deities are explicitly represented within the cakras or upon
any external points of the body. With this emphasis upon process vs. structure, the
diagram appears to emphasize the manipulation of psycho-physical energies now
associated with completion stage practices of yoginî tantra. Tsenshab Rinpoche suggested
to Heller that the uppermost cakra, composed of forty-nine squares, is connected with
progress through the bardo; considered in light of the inscription on the reverse side of
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
647
Personal communication, June 2013.
648
See Robert Mayer, A Scripture of the Ancient Tantra Collection: The Phur-pa bcu-
gnyis (Oxford: Kiscadale Publications, 1996), 135.
649
Heller 2010, 63.
! 212!
the image, Heller concludes that the diagram was likely created to provide instruction for
funerary rites.650
The reverse side of the image is a stack of eleven shapes marked by seed syllables
together symbolizing aspects of tantric and abhidharmic cosmology; the composite is
reminiscent of descriptions of the formation of the human body from the elements of the
universe. Such descriptions abound in tantric literature. There is an apparent resemblance
to the shapes and syllables used to manifest the basis for imagining the deities in the
generation stage body mandala practice of Cakrasaµvara.651 If this drawing is indeed
connected with funerary rites, it reveals an important aspect of Buddhist tantric
perspectives on death. Namely, only by understanding how the body is formed can one
properly understand how it dissolves in the death process; through this understanding,
one becomes empowered to harness the dissolution process to liberate oneself from the
cycle of rebirth.
Therefore, one way of understanding this diagram is as a representation of
invisible processes enacted by the tantric practitioner through ritual. Tantric sådhana
practice is often understood as comprised of two stages, the generation and completion
stages. A very basic distinction of these two stages can be articulated in terms of structure
and process. In the generation stage, the practitioner focuses upon the structure of the
subtle body; in the completion stage, the practitioner focuses upon manipulating
psychophysical energies like pråña along particular pathways within that structure.652 In
confronting visual representations of the subtle body, it is important to consider the
challenges posed to the artist in representing such processes.
Tibetan medical illustrators faced similar problems, as suggested by an
eighteenth-century Tibetan medical representation from the collection of the Rubin
Museum of Art. [Fig.34] This appears to be a later copy of a drawing from the set
commissioned by Desi Sangye Gyatso in the seventeenth century to illustrate his
commentary upon the Tibetan medical tantras, the Blue Beryl Treatise. 653 It should be
noted that much of the information contained in the Tibetan medical tantras is derived
from Indian Åyurvedic sources. Therefore, the medical tantras reflect certain shared
conceptualizations of the human form in both its subtle and coarse capacities. One
interesting aspect of this representation is that it presents different ways of viewing the
body side-by-side.654 We have more traditional or empirically verifiable anatomical
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
650
The chart-like geometry is reminiscent of the Jain cosmic man depictions discussed
above.
651
John C. Huntington and Dina Bangdel, The Circle of Bliss: Buddhist Meditational Art
(Chicago: Serindia Publications, 2003), 247.!!!
652
I am grateful to Dong Sung Shabdrung Rinpoche for his feedback on this topic during
our meetings in Fall 2013 at UC Berkeley.
653
See Meyer, Fernand with Yuri Parfionovich and Gyurme Dorje. 2002. Tibetan Medical
Paintings: Illustrations to the Blue Beryl Treatise of Sangye Gyamtso Vol.1 (London:
Serindia Publications, 1992)
654
On the modes of representation employed by Desi Sangye Gyatso’s medical paintings,
see: Dachille, R.E. 2007, ‘Modes of Representation and Meanings in the Blue Beryl
Paintings of Desi Sangye Gyatso,’ Unpublished masters thesis, University of Wisconsin,
Madison. & Dachille(-Hey). 2011. “The Case of the Disappearing Blue Woman:
! 213!
information abut the flesh and bones appearing alongside the central illustration of the
channels and cakras (aspects of the body known directly only by tantric practitioners).
Frances Garrett, Janet Gyatso, and Vincanne Adams have discussed the controversies that
have arisen over the course of Tibetan medical history with regard to representing the
subtle body.655 At the forefront is the issue of how to empirically verify the existence of
the channels within the human body. If only accomplished tantric practitioners can
access this understanding of the body, then how can it be measured, confirmed, and
depicted? And yet without a basic knowledge of how such invisible elements and
processes work, one cannot even grasp how the human body is formed. So an image like
this enables the viewer to imagine the human body in its most essential form, perhaps in
the hopes of one day experiencing it that way for oneself.
Both the tantric and medical illustrations seem to have been created for specialist
audiences. These were audiences who engaged in practices, whether spiritual or medical,
theoretical and/or practical, aimed at cultivating a way of seeing the human body that
transcends the obvious. Visual representations may serve to preserve this knowledge and
also to disseminate it by offering the audience a preview of a variety of perception of the
human body cultivated over time. The final portion of this article will consider how this
latter aspect of representation produced anxiety for one fifteenth-century Tibetan
commentarial writer on body mandala and provoked a response that might be labeled as a
variety of iconoclasm. This iconoclastic tendency, fueled by concerns that viewers might
mistake this ‘preview’ or simalcrum for the real thing may, in turn, be linked with other
motivations for refraining from representing the subtle body in fabricated material images
in India and the Himalayas.656 Before engaging these philosophical tensions, we will
pause to look closely at the Nepalese body mandala painting.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Understanding How Meaning is Made in Desi Sangye Gyatso’s Blue Beryl Paintings.”
Asian Medicine: Tradition and Modernity, Volume 7, Issue 1 ‘Gender, Health and
Medicine in Tibet’: Dr Heidi Fjeld and Theresia Hofer (eds), Autumn 2011.
655
Frances Garrett and Vincanne Adams, “The Three Channels in Tibetan Medicine, with
a translation of Tsultrim Gyaltsen’s ‘A clear explanation of the principal structure and
location of the circulatory channels as illustrated in the medical paintings,” Traditional
South Asian Medicine 8 (2008) 86-114. Janet Gyatso, “The Authority of Empiricism and
the Empiricism of Authority: Medicine and Buddhism in Tibet on the Eve of Modernity,”
Comparative Studies of South Asia, Africa and the Middle East, Volume 24, Number 2
(2004): 83-96. I look forward to reading Janet Gyatso’s recently published study of
Tibetan medical disourses : Gyatso, Janet. 2015. Being human in a Buddhist world: an
intellectual history of medicine in early modern Tibet. (New York: Columbia University
Press, 2015).
656
For example, if we subscribe to a Foucaultian view of tantra as a ritual technology
inextricably linked with power and control, much like the model embraced by Davidson
2003, then one motivation for the iconoclastic tendency to refrain from producing visual
representations of the subtle body could be the desire to control the spread of ritual
knowledge. For images, even more than texts seem to contain an extraordinary capacity
for replication and proliferation beyond the scope of their origin.!
! 214!
The Nepalese Body Mandala Painting: Challenges and Insights
A. Situating the body
Like the artists executing the tantric and medical representations discussed in the
previous section, the creator of the Nepalese body mandala painting faced the challenge
of representing invisible psycho-physical processes through visible forms. Just as a
medical illustrator may have struggled to devise a mode for representing the formation
and workings of the channels of the human body, the body mandala painter faced the
challenges of depicting the structure of the subtle body alongside the processes enacted
through it. On one hand, representing the Cakrasaµvara body mandala practice visually
required the artist to locate and demarcate essential points of the body as the abodes of
particular deities, a kind of parsing of the body’s form more clearly associated with
generation stage practices. On the other hand, the project demanded that the artist
simultaneously display the animation or manipulation of this form in processes associated
with the completion stage of the body mandala practice: the movement of psycho-
physical energies, the blazing of the inner fire, the melting of the drop of bodhicitta and
the repeated experiences of varieties of bliss.657 As the essential points on the subtle body
and the foci of such practices, the cakras represent these processes, processes like the
ones to be described below in terms of “piercing to the pith” of the body. They are,
however, represented together with what we might call the more structural aspects of the
practice; in other words, the techniques of mapping the deities of the mandala onto the
body are more strongly connected with the generation stage of tantric sådhana practice.
On a more basic level, we can begin exploring the Nepalese body mandala
painting by asking what kind of body is being represented. Is the body portrayed here
divine or human? The proportions of the body indicate some general iconometric
standard is being applied as one would find in the representation of a deity. There are
certainly no realistic or naturalistic elements in the sense in which those terms are
typically defined in an art historical context. There are none of the marks of a ‘mad yogi’
type either, but there are features beyond the deities inhabiting it that mark this body as
unusual or extraordinary. For example, the hair has been transformed into what appear to
be lotus petals, while the earlobes are elongated like those of a Buddha.658 Technically,
the body mandala practitioner first produces a vision of the self as Cakrasaµvara before
arraying the mandala deities upon the body.659 Therefore, the ritual practice itself uses
the form of the human body as a basis while simultaneously preserving a distinction
between the ordinary and enlightened body.660 The form of the body depicted in this
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
657
I am grateful to Professor Patricia Berger for suggesting this concept of “animation” to
me in our Fall 2013 meetings.
658
Meghan Howard, a peer of mine at UC Berkeley, suggested to me that perhaps the
stylized hair might be an attempt to represent the flaming and upright locks that often
appear in descriptions of fierce deities in tantric sådhana texts. [Personal communication,
Fall 2013].
659
There is a course a great deal of variety in the order and details of this practice as
articulated in different Indian and Tibetan sådhana texts.
660
I am grateful to Kurt Keutzer for his insights into the nature of the body projected in
body mandala practice.
! 215!
Nepalese painting seems to function in a similar way. Fusing elements of the human and
divine, this representation of the body may serve as a template for imagining the
transformation of an ordinary way of perceiving the body into an extraordinary or
enlightened perspective.
Another basic observation about this body is that, unlike the other representations
we have discussed above, it is seated. In my own attempts to diagram various body
mandala practices I have found the form seated in meditative posture to be the most
intuitive. Similarly, English 2002 observes: “Although in the Vajrayoginî tradition the
body mandala should be undertaken by the yogin who imagines himself as the goddess,
the correlations with the body points fit more naturally upon a figure seated in
meditation.”661 [See Fig.13] Therefore, the posture of the body emphasizes the
dimension of tantric practice, inviting comparison with depictions of renowned siddhas or
accomplished tantric practitioners.
The background is another feature to consider in evaluating the artist’s tools for
situating the body.662 While it lacks the specificity of rich landscape detail found in many
Tibetan thangka and Nepalese paubha paintings, the representation does locate the body
within a natural, hilly environment of the variety imagined as conducive to spiritual
practice in popular imagination. The upper portion is indeed the most descript, depicting
the heavens populated by gods resting upon clouds, with the cosmic features of sun and
moon as well as the sambhogakåya forms of Buddhas in union. On a more abstract level,
the tricolor scheme might suggest a correlation with the three realms to which the three
varieties of mandala deities of body, speech and mind relate. While it is indeed possible
to assume that the artist was merely replicating conventions of Buddhist painting in
situating the body of the practitioner within this landscape, it seems equally plausible that
the artist is using the landscape as a mode of referencing a particular feature of the
Cakrasaµvara body mandala practice. This feature typifies yet another variety of
macrocosmic-microcosmic correlation, between the human body and cosmos, on one
hand, and on the other, between the mandala and the Nepalese landscape.
The Cakrasaµvara body mandala practice associates the twenty-four goddesses of
the body, speech, and mind cakras with twenty-four sacred sites.663 These sites are
transposed onto the human body. English 2002 provides a diagram of the Cakrasaµvara
and Vajravårahî body mandala practices [See Fig. 35]; English eloquently summarizes
the structure and process behind this aspect of body mandala practice:
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
661
English 2002, p198.!
662
I am grateful to Orna Tsultrim for her suggestion to devote more attention to this
aspect of the painting during Reading Outside the Lines: A Workshop on the Intersection
of Buddhist Art and Texts September 13-15, 2013, Ludwig-Maximilians-Universitœt,
Munich.
663 !Sugiki 2009 deals extensively with the evolution of this dimension of the practice.
Tsunehiko Sugiki, “Five Types of Internal Mandala Described in the Cakrasaµvara

Buddhist Literature: Somatic Representations of One’s Innate Sacredness.” See “The
Structure and Traditions of the Systems of Holy Sites in the Buddhist Saµvara Cycle and
its Related Scriptural Cycles in Early Medieval South Asia: The Geography of Esoteric
Buddhism in the Eyes of the Compilers of the Scriptures,” Genesis and Development of
Tantrism, 2009.
! 216!
“...the site goddesses dwell ‘within’ the sites identified at a particular point on the
body...but they are imaginatively transformed into channels ‘within’ the body. Tantric
sources commonly refer to the goddesses ‘as’ the channels or veins...the twenty-four
male gods on the sites (consorts to the site goddesses) are said to ‘purify’ certain aspects
of the body. For example, Pracañ∂å’s consort, Khañ∂akapålin, becomes the nails and
teeth; the channel (Pracañ∂å herself) carries nourishment from the head (Pullîramalaya)
to the nails and teeth (Khañ∂akapålin).”664
Citing the work of Kalff, English 2002 also points out that this list of twenty-four
bodily constituents purified by the male consorts is the same list found in the Påli canon
as components of the human body.665 In the dissemination of tantric practice throughout
the Himalayan region from India, these sites were also mapped onto local landscapes, and
the meridians of this tantricized landscape were reinforced through pilgrimage
practices.666 Dina Bangdel has discussed the mapping of these sites onto the geography
of the Kathmandu valley.667 By depicting the body in this sort of background rather than
against the bare monochromatic backdrop found in many Indian representations of the
subtle body like those from the Siddha Siddhanta Paddhati [Fig. 30 & 31], the Nepalese
artist invokes a deep network of correlations. Embedding the body in multiple
frameworks of mandala, cosmos, and landscape, the artist succeeds in representing
defining elements of body mandala practice.
B. Locating the Mandala deities

Of course the more deities included in the mandala practice of a particular tantric
system, the more difficult it becomes to represent them in visual form within the space
not just of the body but also of a canvas or manuscript. In the case of the sixty-two deity
Cakrasaµvara body mandala, this is one obstacle representation poses for the artist.
Without the structure of the mandala palace to contain the deities, they are dispersed
throughout the canvas and connected by lines to their associated bodily sites, albeit in
potentially meaningful groupings. Though the artist does choose to represent a few
deities within the confines of the body, were he/she to attempt to include them all, the
body itself would be effaced by images of the deities. Ultimately this might be what
body mandala practice is doing, effacing the ordinary body by over-inscribing it with a
multiplicity of divine forms. However, for the purposes of representing a ritual process
grounded in the basis of the human form, the very foundation for tantric practice, we
must begin within the confines of the human form. Whether regarded as a tool or an
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
664
English 2002, 199 & 202.
665
English 2002 , p203. Michael Martin Kalff, “On Supreme Bliss: A Study of the
History and Interpretation of the Cakrasaµvara Tantra” (PhD diss., Columbia
University, 1979). Gray 2007 (p61) also makes this connection.
666
On the mapping of the mandala onto the Tibetan landscape, see Toni Huber, The Holy
Land Reborn: Pilgrimage and the Tibetan Reinvention of Buddhist India (Chicago:
University of Chicago Press, 2008).
667
I am indebted to Dina Bangdel for suggesting that I explore this correlation. Dee Dina
Bangdel, “Art in the Ritual Context: The Chakrasamvara Tradition in Newar Buddhism,”
Orientations No.8 Vol. 34 (2003).!
! 217!
obstacle, the human body is the ultimate framework for our situatedness in time and
space, the backdrop against which our stories are woven and the altar upon which our
rituals take place.
So how does how the method of mapping deities onto the human body depicted
here compare with methods of mapping them onto the space of a palace as frequently
depicted in art? There are many varieties of the Cakrasaµvara body mandala tradition,
the three main transmissions being those of the Indian siddhas Luipa, Ghantapada and
K®ßnapada.668 The anomalous nature of this painting together with certain inconsistencies
in the mode of representing the deities therein suggests that while certain details of
representation and inscription are of interest, it might not be wise to focus too literally
upon connecting this representation with a particular ritual text. Our focus here is instead
the mode of representing the practice employed by the painter. Therefore, we will begin
to approach the details through a comparison with a more familiar mode of representing
the Cakrasaµvara mandala. In comparing the body mandala practice with the format of a
fifteenth-century Nepalese paubha painting of a mandala palace, we will progress from
center to periphery, tracing the progression from the most subtle and profound to the
mundane; this progression is inscribed in the logic of the mandala. [See Fig.36] As
observed above, there are many different versions of the body mandala practice; this
variety of ritual forms of the Cakrasaµvara body mandala practice, in particular, have
been brilliantly outlined and compared by Sugiki 2009.669 Here we will simply make
some general observations about the relation of the form of the body mandala to that of
the mandala palace. Then, we will attempt to understand the mode of representation used
within this body mandala painting together with the peculiarities of some of its details.
I have inserted numbers into this image of the body mandala painting for the
purpose of explaining the positions and identities of the deities and provided a translation
of the inscriptions with corresponding numbers. [See Fig. 37 and corresponding list of
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
668
In their outline of the Cakrasaµvara body mandala practice, Huntington and Bangdel
2003 rely largely upon Geshe Kelsang Gyatso’s 1997 description.. That practice,
according to the Ghantapa system, was transmitted by the Sakya patriarchs and received
by Tsong kha pa. Huntington and Bangdel observe the overall similarity of the Newar
version of the practice to that propounded by Tsong kha pa. They also use “Newar
Sanskit ritual texts” including the “Trisamadhi puja vidhi, Samvarodaya dashami,
Abhidanottara tantra and Samvarodaya Tantra.” See Huntington and Bangdel 2003,
pp.243-250, especially 243. Geshe Kelsang Gyatso, Essence of Vajrayana: The Highest
Yoga Tantra Practice of Heruka Body Mandala (London: Tharpa Publications, 1997). On
the importance of the Samvarodaya Tantra within the Newar Cakrasaµvara tradition, see
David Gellner, Monk, Householder and Tantric Priest: Newar Buddhism and its
Hierarchy of Ritual (Cambridge: Cambridge University Press, 1992). More research is
indeed still needed to understand the nature and history of the Newar body mandala
practice.
!
669
See Tsunehiko Sugiki 2009. See also Tsunehiko Sugiki, review of The Cakrasaµvara
Tantra (The Discourse of ¸rî Heruka): A Study and Annotated Translation, by David
Gray, Journal of the International Association of Buddhist Studies, Vol. 31 No.1-2
(2010): 505-542.
! 218!
inscriptions and deity identifications]. We begin with Cakrasaµvara and consort at the
center of the mandala palace, surrounded by the four goddesses and four skullcups. In
some versions of the body mandala practice, these four attendant goddesses are imagined
to be the four channels of the heart. In the body mandala painting, the main deity and
consort appear upside-down within the cakra at the body’s heart center, and although no
retinue goddesses or vessels appear, they are likely symbolized by the syllables on the
eight petals of the cakra. The next tier of the mandala, often referred to as the ‘mind
cakra’ is embodied by eight deities plus their consorts who, in the case of the body
mandala , are generally positioned in the upper region of the body. The next tier is the
‘speech cakra,’ embodied by eight deities and their consorts positioned in the middle
region of the body mandala. Then, the ‘body cakra’ is embodied by eight deities and their
consorts positioned in the lower region of the body mandala. Therefore, the standard
hierarchy of body, speech and mind in reiterated here and translated from the radial axis
of center to periphery onto the vertical axis of the human body.
The majority of the deities of these three cakras are accounted for by the images
and inscriptions of the body mandala painting. Most of these deities of the three cakras
appear on the bodily site with which they are connected. In the rare case in which they
do not, as with the three couples in union that appear in the foreground of the painting
[#53-55], their association with the body part, in this case the anus and genitalia, is
indicated by a color-coded line. Red, white, and blue lines trace the courses of the right,
left and central channels respectively. Many of the deities associated with the three cakras
appear individually but are easily linked as they inhabit corresponding body parts (like
the right and left hand), exhibit the same flesh tone, and bear inscriptions with identical
seed syllables linking them to the same bodily site. Although the inscriptions are
generally in Devanågarî script and provide the familiar Sanskrit names with some minor
inconsistencies, there are a few cases in which the artist or scribe has provided the
Newari names for the part of the body that is being indicated, as in the case of the eye and
big toe.670 [See F & #49] In some instances, the goddesses appear in union with their
male consorts together with an inscription naming both and providing a syllable matching
the site with which the couple is connected. [See, for ex. #31-32,#39-40 & 53-55].
Twenty-eight additional deities or deity pairs hover around the main figure; gold,
red, and blue lines connect the majority of them with their corresponding bodily sites.
The eight pairs of deities that form the outermost layer of the painting are those often
referred to as the samaya-cakra goddesses (#1-4 & 23-26). They signify a set of eight
goddesses who typically inhabit the four corners and four gates of the mandala palace. In
the case of the body mandala, they are often imagined within the sense organs /apertures
of the body (the navel and the space between the eyebrows being perhaps less intuitive
members of this category). Their role in guarding the boundaries has been translated here
in the case of the body mandala painting by placing them in the outermost zones of the
painting and providing, in most cases, blue lines linking them to the bodily gates. What
is strange is that they are depicted with male consorts, a non-standard feature that does
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
670
I am grateful to Alexander von Rospatt for his guidance in identifying these
inscriptions.
! 219!
not appear to match textual descriptions. The artist/scribe even went so far as to supply
the names of these male consorts.671
Figures #7-10 appear to be four of the six armor goddesses together with their
consorts. Gold lines connect three of the four to the body of the main figure. These
goddesses are represented as a complete set of six in the lower register of the mandala
palace painting. [Fig. 36] Therefore, although they are not part of the body mandala
proper, these goddesses are included by virtue of association with preparatory practices
employed in body mandala; that are part of a larger set of auxiliary deities hovering
around the body of the main figure. Likewise, #’s 19, 20, 21, 27 & 28 are goddesses the
practitioner imagines in purifying the bodily elements; they are depicted individually
with gold lines that seem to connect them to the cakras of the main figure. All members
of this set are depicted with two arms except for #19, Padmajvålini, here referred to as
“Padmajvålini vajra våråhi,” who is associated with the purification of space and displays
four arms. These particular goddesses do not appear in the painting of the mandala palace
although that painting does include some auxiliary deities, like the armor goddesses, that
are not a part of the Cakrasaµvara mandala proper. Such deities appear in the corners
outside the mandala as well as in the top and bottom registers of the painting beyond the
outer rungs of the mandala palace, the protective circle and the cremation grounds..672
Both paintings, therefore, locate these deities in peripheral positions that indicate their
subsidiary status.
Deities numbered 11-13 and 16-18 are individual two-armed deities; red lines
connect them to apertures of the body such as eyes and ears. They appear to refer to the
set of six bodhisattvas invoked in the purification of the sense spheres. Numbers 14 and
15 depict deity couples that are somewhat more difficult to identify. While the former
lacks any color-coded line connecting it to a particular bodily site, #15 has a golden line
joined with that of #19 and connecting it to what appears to be the cakra at the forehead.
The inscription for #15 is illegible. However, the inscription for #14 reads “Raga Vajra
Raga vajrî.” The identity of this couple is unclear but shares qualities with couple #22.
Couple 22 also appears to be connected with the purification of the sense spheres and is
inscribed: “Ai!varyya vajra. Ai!varyya vajri.” Although Ai!varyavajra is connected
with the purification of the body in its totality in some practices, the invention of the
consort seems anomalous.
A basic observation can be made with regard to the number of deities represented
in the body mandala painting. Typically the sixty-two deities of the Cakrasaµvara body
mandala are counted as forty-eight deities of the body, speech and mind cakras (i.e. the
twenty-four pairs), eight deities of the samayacakra (the goddesses guarding the gates and
corners), plus the two main deities (Cakrasaµvara and Vajravårahî) and the four retinue
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
671
Further research is required to determine whether this detail is significant, but from
my own experience working with a proto-body mandala text from Dunhuang, consorts
(though in that case, female consorts) are sometimes added somewhat arbitrarily. We
have observed how inconsistencies like these often point to iconographic experimentation
and the attendant modification of existing structures; these themes illuminate important
aspects of the exegesis itself.
672
See Huntington and Bangdel 2003, p.262 for identifications of these deities.!
! 220!
goddesses. Here instead we have twenty-four deities basically within the bodily frame
and twenty-eight deities surrounding it as well as a host of deities inhabiting the cakras.
Some of these cakras hover beyond the confines of the body, but are connected to it by
color-coded lines. The web of deities and cakras is somewhat unclear but may supply
additional information about subtle body processes and structures. For example, the
sixteen, eight, and sixty-four petals or channels of which the throat, heart and navel
cakras are composed are represented together with syllables that may indicate the
goddesses residing in or as those petals/channels. The central deities in three of the six
cakras appear upside-down; in ritual visualization, it is common to imagine inverted seed
syllables at key bodily points.
More research is required to better understand the cakra representations in this
painting, but I would offer one preliminary observation. They reflect a focus upon the
cultivation and manipulation of psycho-physical energies that is often associated with
completion stage practices. Therefore, there is the potential for comparison with both the
Western Tibetan cakra diagram [Fig. 23] and the Tibetan medical painting depicting the
bodily cakras and channels [Fig.34] discussed above. As observed in the discussion of
those cases, the challenges of representing process and structure simultaneously require
the artist to innovate and to experiment with new modes of representation.
V. The Problem of Fabrication: Body and Image in the Body Mandala

Debate
Now that we have had the opportunity to compare the mode of representing the body in
this painting with other visual representations of the subtle body and of the mandala, we
can place the painting in dialogue with the body mandala debate texts. We are now
familiar with the basic distinction between “outer mandala,” the mandala of painted cloth
or powders) and “inner mandala,” the body mandala. In Chapter Four, we discussed the
notion of “fabrication” [bcos ma] as it is deployed by Mkhas grub to make sense of
tantric ritual acts of imagination and to mark body mandala practice as superior.
“Fabrication” proved to be a charged category of analysis imbued within philosophical,
linguistic, ritual, and even artistic dimensions of meaning. From a philosophical
perspective, fabrication is linked with the flawed tendencies of our mind to superimpose
false concepts upon reality. Fabrication is also tied to the “conventional designations”
characterizing the way we engage with the world through ideas and language, operating
on the conventional vs. ultimate level of truth. Mkhas grub uses the language of
fabrication to consider the primordial status of the body as mandala, tapping into issues
such as Buddha nature and the relationship of the human body to samsåric existence.
Tantric ritual acts of imagination are also forms of fabrication; however, not all
imaginings are equal. For example, in the case of sådhana, we encountered the language
of ‘similitudes,’ ‘fabrications,’ ‘imputations’ and the ‘contrived’ in descriptions of the
generation stage of practice completion stage as juxtaposed with the naturalness and
authenticity of the completion/perfection stage.673 Through its basis in the human body,
the body mandala possesses a naturalness lacking in other mandala, both imagined and
materially generated varieties. Mkhas grub rje distinguishes the body mandala as an
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
673
Bentor 2006 also discusses this distinction of the generation stage in terms of
“similitudes.” See Bentor 2006, p.186 fn4.
! 221!
“unfabricated mandala” [ma bcos pa'i dkyil 'khor] because it has been present since
human conception; other mandala supports are “newly fabricated” [gsar du bcos] using
paints and cloth and so forth.674 All mandala may function as supports for ritual practice,
but the body mandala provides a superior means of support for the body mandala, in turn,
rests upon the support of the human body.
Like an image on an altar or on cloth, the body functions as a support [Tib. rten]
for ritual practice.675 However, the texts of the body mandala debate suggest that the
body forms a different kind of support than others. How does the body serve as a basis
for generating deities, and how can it serve as a support for furthering the goals of tantric
practice? Mkhas grub argues the mechanics of visualization practice, specifying that only
imaginings based in valid cognition can properly transform the body to function as the
support for generating deities to become the body mandala. And through that connection
to the body, the body mandala is a distinct variety of tantric imagining for “piercing to
the pith” of that body. Through ritual practice, the true nature and potential of the human
body is discovered and actualized. The channels, winds, and drops of the subtle body or
“vajra body” play a vital role in this process. Mkhas grub emphasizes that “piercing to
the pith” in body mandala practice makes these subtle elements malleable. Ngor chen
justifies the naturalness and authenticity of the practice through its connection with these
inherent aspects of the body: “The five channel cakras are naturally established [rang
grub du yod pa] in the vajra body... Therefore, this is not a mental imputation [blos brtag
pa]”. (N1 560.4)
The body mandala debate demonstrates the delicate status of both visualization
practice as a method and of the human body as a basis for tantric practice. It also
suggests that representations, both material and imagined, poses problems of authenticity
for both philosophers and tantric exegetes. These problems extended to the domain of
textual authority as well; body mandala practices that are fabricated are critiqued as
apocryphal. Mkhas grub and Ngor chen link representation with embodiment through
classifying and relating texts, creative processes, and bodies.
Just as the body mandala painting presents a seated figure in meditation rather
than the body of a recognizable deity, the template of the ordinary human body is
necessary for practice. Despite the saµsåric ties of human embodiment, the broader
Buddhist perspective identifies the human form is the ideal form for soteriological
progress. Bearing these issues in mind, let us now briefly return to the body mandala
painting.
VI. Anomaly and Patronage: Further Reflections on the Body Mandala

Painting
Why would a nineteenth-century Newari painter ‘fabricate’ a body mandala if the
unfabricated quality of the human body is what distinguishes it from other supports for
mandala visualization practice? One possibility presented itself in a recent conversation
with Debra Diamond, Senior Curator at the Sackler Museum of Art in D.C. about their
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
674
See especially Mkhas grub 250.6-251.3.!
675
See Tsultemin (forthcoming) 2016 for a discussion of the concept of the “support” in
discussions of the artistic process in regard to the practice laying the ground for the
creation of a thangka painting.
! 222!
upcoming exhibition, Yoga: the Art of Transformation. We were discussing the
disproportionate amount of textual vs. visual representations of the ‘yogic body’ in India.
I commented upon the fact that while we find many depictions of tantric practitioners in
the art of Tibet and Nepal, in particular the famous Buddhist depictions of the
mahåsiddhas, illustrations of the subtle body including the channels, winds, and drops are
far from common. Now this has been somewhat changed of course by the modern tourist
art market in which low-quality reproductions of illustrations of the body copied by
artists from published sources abound. One example is reproductions of select paintings
of the anatomy and the subtle body from Desi Sangye Gyatso’s famed series of the
seventeenth- century medical illustrations of the Blue Beryl Treatise. Illustrations of the
cakras based in Hindu yoga also proliferate in such circles. However, when we search for
historical precedents, only a select few surface.
As mentioned above, when asked to speculate about the Indian case, Debra
Diamond remarked that many of the examples they’ve collected for the exhibition seem
to suggest the kind of “new audience” for yogic texts and practices.676 How might we
use this category of “new audiences” to think creatively about the patronage of the body
mandala painting? Unlike many Newari paubha paintings, this one does not include a
lower register depicting a donor and ritual specialist, so it seems less likely that the
painting was created to commemorate an initiation ceremony. Could there have been a
new audience for the body mandala practice in Nepal? Could this painting perhaps signal
the reintroduction or revival of the body mandala practice among the Newars? Was there
are disruption of the lineage of transmitting this practice that prompted an attempt at
revitalization? While we know the cult of Cakrasaµvara has been popular in Nepal and
that Newar artists have been prolific in the production of Cakrasaµvara statues and
mandala, it is difficult to trace the evolution of the body mandala practice there. Even if
manuscripts detailing the practice survive in Nepal, it’s hard to tell who was transmitting
these practices and to how large an audience. Scribal culture and ritual culture do not
always exist in clear relationship to one another. That said, considering existing
manuscripts and their colophons is one future avenue of research for connecting scribal
activity, ritual practice, and artistic production.
Another consideration is: what does the style of script used in the body mandala
painting tell us? Can it more accurately help us to date the painting and to determine its
patron or audience? Does it eliminate the possibility that the painting was created for a
Tibetan patron, for example? And finally, are there other new audiences we are
overlooking? Perhaps a European missionary or government official interested in
cataloguing the traditions of Nepal like Brian Houghton Hodgson?677 But who would
reveal such information to a non-initiate, and what sort of prototype would the artist be
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
676
Personal communication, June 2013.
677
See von Rospatt 2016 for an exploration of a painting commissioned by Hodgson.
Keynote Speech presented at Reading Outside the Lines: A Workshop on the Intersection
of Buddhist Art and Texts September 13-15, 2013, Ludwig-Maximilians-Universitœt,
Munich (Publication forthcoming, 2016). For more on Hodgson, see Waterhouse, David
M. 2004. The origins of Himalayan studies: Brian Houghton Hodgson in Nepal and
Darjeeling, 1820-1858. London: RoutledgeCurzon.
!
! 223!
relying upon? Without more information, we can’t identify the artist or patron of this
painting or their motives for representing the body mandala practice. However, in
presenting this host of questions, I hope I have succeeded in showing that there are some
interesting possibilities.
VII. Body, Image and the Problem of Matter
In highlighting the anomalous character of this painting, I have placed it in

dialogue with both visual representations that display the microcosm-macrocosm relation
embraced by body mandala practice as well as with those that reveal invisible aspects of
the structures and processes of the subtle body. Such comparisons have allowed me to
foreground questions regarding the modes of representing the subtle body displayed in
the body mandala painting. Thematic connections with Tibetan writings on body
mandala practice have likewise been explored to exhibit subtle tensions underlying both
the creation of material images as well as of simulations. In the process, I have begun to
uncover hidden aspects of the relationship of representations to ritual acts of imagination
and their respective roles as “supports” for achieving ritual goals. I have also explored the
category of fabrication itself as what some Tibetan perspectives might even label as the
curse of artwork: invention or even more harshly, imputation. Perhaps most importantly,
I have pointed to ways in which the body mandala practice is special because it relies on
a very different type of support, the body itself, naturally existing and unfabricated.
I will conclude by taking a moment to reflect upon the relationship of body,
image, and matter. From the Buddhist perspective, one of the defining qualities and
flaws of the body (Tib. lus, hon. sku, Skt. kåya/deha) is its status as a compounded entity,
an assemblage of psycho-physical components that assume material form largely by force
of the winds of karma. The etymology of the Sanskrit term deha derived from the root
dih, “to fashion or mold,” tells us that the body is, in a sense, fabricated.678 The
terminology for bodies, images, and forms seems to be virtually interchangeable in many
cases in both Sanskrit and Tibetan and emphasizes the shadowy or reflective nature of
images. We will recall that Mkhas grub’s issues with material images, articulated
through juxtaposing the “outer” and “fabricated” mandala with the superior
“unfabricated” body mandala, involve their identity as simalcra, as reflections or
imitations of a true perception of reality. Taking into account tensions surrounding the
human body within the larger context of his writings on body mandala, its conflicting
capacities as flaw and instrument for liberation, we might consider whether the category
of “matter” helps to illuminate the relationship of bodies and images.
Carolyn Walker Bynum’s Christian Materiality pushes back against the trend in
Medieval Studies to fixate upon “the body” and “visuality” as privileged categories for
interpreting Medieval Christian religiosity.679 Bynum demonstrates that what are often
interpreted as anxieties about bodies are actually anxieties about the status of “matter.”
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
678
Monier-Williams Sanskrit-English Dictionary, http://lexica.indica-et-
buddhica.org/dict/lexica
679
Caroline Walker Bynum, Christian Materiality: An Essay on Religion in Late
Medieval Europe (New York: Zone Books, 2011).
! 224!
The author cites the account of Theodulf of Orleans, the eighth- century author of the
Libri Carolini :
“...There is indeed a very great difference...between the sacrament...and images executed

by painters...The former is consecrated by the priest invoking the divine name, the latter
painted by a painter with the erudition of a human technique; the former is carried by
angels to the altar of God, the latter placed by human hands through their skill on walls to
startle the eyes of those looking at them.”680
While Bynum interprets this account in terms of the ‘hierarchy of matter,’ one
might likewise consider what it tells us about human creation and artistic fabrication. An
explicit link between bodies and images is found in Bynum’s recounting of
Bonaventure’s vision of St. Francis. St. Francis bears the “image [effigium] of the
crucified, which was not imprinted [figuratum] on tablets of stone or wood by the hands
of a craftsman, but marked [descriptum] into the members of his body by the finger of the
living god.”681). Bynum interprets this vision as follows: “Bonaventure’s language
makes clear the arguments of medieval theologians that only God can craft or enliven
flesh, that it is living men (not dead wood and stone) that are image and image
potential.”682
The tantric body mandala inscribes the body with the marks of divine enlightened
status, and this act of inscription is repeated through ritual practice as a catalyst for
change. This comparison between Buddhist and Medieval Christian attitudes towards the
creative process suggest that they share a sense that images, bodies have the potential to
inspire wonder and deception. They extol a quality Medieval Christians term divine and
Buddhists term natural or unfabricated. In terms of Buddhist theories of personhood,
bodies pose the ultimate threat of attachment to a stable and fixed notion of self and
other. They come into being and ooze and decay through forces beyond control. Through
intellectual and moral discipline, the Buddhist practitioner aspires to harness their
tendency to replicate in a multiplicity of forms through the stream of rebirth. However,
tantric practices like body mandala embrace bodies and images as powerful tools in the
project of liberation. Practitioners become both “image and image potential.” The body
mandala debate shows the resistance to reifying bodies and images within the very
context of generating images on the basis of the body. In struggling to make sense of this
paradox, we encounter what LaTour would call the “iconoclash”:
“But what if human hands were actually indispensable to reaching truth, to producing
objectivity, to fabricating divinities? What would happen if, when saying that some
image is human-made, you were increasing instead of decreasing its claim to truth? That
would be the closure of the critical mood, the end of anti-fetischism. We could say,
contrary to the critical urge, that the more human-work is shown, the better is their grasp
of reality, of sanctity, of worship. That the more images, mediations, intermediaries,
icons are multiplied and overtly fabricated, explicitly and publicly constructed, the more
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
680
Bynum 2011, 47.
681
cf 2 Cor 3:3; Exod 31:18, cited Bynum 2011, 112.
682
Bynum 2011, 112.!
! 225!
respect we have for their capacities to welcome, to gather, to recollect truth and sanctity
(“religere” is one of several etymologies for the word religion.)...Thus we can define an
iconoclash as what happens when there is uncertainty about the exact role of the hand at
work in the production of a mediator. It is a hand with a hammer ready to expose, to
denounce, to debunk, to show up, to disappoint, to disenchant, to dispel one’s illusions, to
let the air out? Or is it, on the contrary, a cautious and careful hand, palm turned as if to
catch, to elicit, to educe, to welcome, to generate, to entertain, to maintain, to collect truth
and sanctity?”683
Tantric practitioners are, themselves, reveling in the iconoclash, perpetually engaging

with images and bodies, mediators navigating the relationships between their creation and
destruction. To deny the relevance of representations and embodiment would be to deny
the importance of human life itself. What would it look like to continuously create
illusions and then “let the air out” of them? How could one adopt the “a cautious and
careful hand” necessary to navigate the fine line between attachment and aversion to
images and bodies? How might the “iconoclash” play a crucial role in actualizing a state
of mind in which in life’s final breath, we yearn neither for escape nor return, neither for
creation nor destruction?
VIII. Representation and Reality
We will conclude by reflecting on the relationship between images and bodies as

representations in Buddhist tantra, with a little help from LaTour. To review, we have
briefly described Buddhist philosophical attitudes toward representations. The Sakyapa
perspective, with which we will become more familiar through Ngor chen, has often been
described as “representationalism.” This term was defined above as a view which “asserts
that we know not external things, but the representations, or copies, of external things in
our own minds.”684 George Dreyfus describes the conflicting approaches to the
relationship between representation and reality developed by the Sakya tradition, of
which Ngor chen is a part, and Mkhas grub’s Gelukpa tradition :
“Dharmakîrti and his Sa-gya interpreters hold a position we will describe...as
representationalism, which postulates that awareness is directly in contact with only
representations, what Dharmakîrti call reflections or aspects. By contrast Kay-drup holds
a direct realist view, according to which mental episodes are in direct contact with
objects. Hence, the reflection of a thing in consciousness is not a representation but the
revelation of that thing itself.”685
The description of the Sakyapa position (as developed in response to the work of the
Indian philosopher Dharmakîrti) suggests that representations mediate our very
experience of the world. Mkhas grub’s position, on the other hand, proposes the
possibility of obliterating these representations. One way of describing these conflicting
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
683 !LaTour 2002: pp.18 &20!
684 !C.E.M.!Joad,!Guide*to*Philosophy,!I.ii.41!(1936)![OED!online]!
685
Dreyfus 1997, p.252. For more on the role of representations in Dharmakîrti’s
thought, see Dreyfus, p.220.
! 226!
positions on the nature of the mind is an iconoclash. If the Sakyapas are iconophiles, in
the sense that they accept the inevitability of representations, the Gelukpas are
iconoclasts, seeking to destroy the representations that mediate direct contact with reality.
LaTour describes the iconophile’s approach to images as attentiveness to the
“series of transformations for which each image is only a provisional frame.”686 In my
previous work, I described what it means to regard an image as a “provisional frame,” as
a process that:
“requires a viewer to anticipate that the way meaning is being conveyed by a single
image builds upon what precedes it as well as what follows. Viewing images in this way
entails accepting that one image possesses the capacity to both prefigure and refigure the
images after and before it. The undetermined quality of the segues between images
provides an opportunity for a viewer to become self-conscious of the ‘transformations’
operating through images because the ambiguous connections permit no definitive
interpretation of a single image.”687
The “iconophilic” attitude has potential applications for understanding Buddhist

approaches to both images and bodies. We have explored some of these applications,
such as in representations of the body in sådhana, in diagrams of the subtle body, and
even in terms of the karmic connections between births. In the context of meditative
practice, the Sakyas not only accept representations (images and ideas) as fundamental,
but they also, like LaTour’s iconophile, prioritize the movement between them. The
transitions between thoughts become a key focus of realization. How can iconophilia be
used to read the spaces between thoughts?
In our discussion of the Sakyapa tantric perspective late in Chapter Five, in
attempting to glean the attitude towards the body reflected by the Sakya Lam ‘bras
tradition, we discussed the framework of the three continua of cause, method or path, and
result. While formally, the body is identified with the method continuum, we observed a
degree of blurring the three categories. It was in this context that we were able to discern
just how the body, as the foundation for practice, and more specifically, for generating
the body mandala, could be endowed with a causal capacity. In other words, we
encountered the body as instrument, agent, and foundation for facilitating a realization of
the mind, the causal continuum. In a teaching on the “Nature of the Mind,” His Holiness
the Fourteenth Dalai Lama explains the Sakya Lam ‘bras approach in greater depth. He
references a meditative practice whereby one can experience the causal continuum,
defined as the “innate mind of clear light,” as a distinctly iconophilic process:
“Sakya Pandita says, ‘In between the arising of different thought processes, the arising of
a radiant clear light mind remains uninterrupted.’ So what he is saying is that if you
observe the mind, your own thought processes, one after another, in a kind of a sequence,
between the arising of one thought and the dissolving of that and the arising of another,
there are intervals; regardless of how short they may be, there are transitions between the
arising of different thought processes. The practice involves trying to tease out those
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
686
LaTour 1998, p.56
687
Dachille 2007, Unpublished Master’s thesis.!
! 227!
transitional periods so that one can recognize these gaps in the arising of another thought
and the dissolution and arising of another thought...So one refrains from both looking
backward into the past nor into the future through anticipation and hopes and so on, but
remains simply in the present moment. So initially what one experiences is simply a kind
of absence or a gap, but through experience, as one learns to prolong that period of this
absence, simply remaining focused on the present moment, at that time, one comes to
recognize the subtle thought processes. And although this is not exactly the clear light
state of mind, it is somewhat indicative of that basic clear light mind... ”
In navigating the space between thoughts, and even prolonging them, one is granted
access to a more subtle level of experience of the mind. His Holiness describes the
resultant mental state as one of “wonder”:
So in the Sakya tradition...the idea is to cultivate that kind of present moment awareness
with a sense of wonder, where one does not let one’s mind follow after the temptation of
looking into the past or into the future but simply remaining in that present moment of
consciousness. As one learns to remain simply focused upon the present moment of
consciousness, then one allows for the natural quality of the mind to express itself. And
the natural quality of the mind is clear light.688
Within the Sakyapa sources, we encountered the notion of the body itself as one
among many “confusing appearances.” In this sense, the body is a representation, an idea,
image, or thought that confounds clear definition. Like Baudrillard’s “image,” it inspires
wonder and confusion: “it is the reflection of a profound reality; it masks and denatures a
profound reality; it masks the absence of a profound reality; it has no relation to any
reality whatsoever; it is its own pure simalcrum.”689
IX. Piercing to the Pith of the Body Mandala Debate
The body mandala debate reveals profound philosophical tensions underlying tantric
exegesis and ritual practice in Tibet. This dissertation demonstrates how tantric polemics
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
688
His Holiness the Fourteenth Dalai Lama “Nature of the Mind.” UC Santa Barbara,
April 24. 2009 [video recording available on YouTube, 1:40-1:51:15]
Here, his Holiness draws a parallel between this Sakya practice and the Dzogchen
approach to experiencing rig pa awareness. I am grateful to my colleague Dan
McNamara for his insights on the connection between the Dzogchen and Sakya
approaches. [Personal communication, September 2015]
!
689 !JeanBaudrillard, “The Precession of Simalcra,” in Simalcra and Simulation,
translated by Sheila Faria Glaser cited inMargaret Miles, “Image,” in Critical Terms for
Religious Studies, Ed., Mark Taylor(Chicago: University of Chicago Press, 1998) pp.60-
72, p.66.
! 228!
function as a site for reconfiguring the boundaries of exegetical practice. Variations in
Mkhas grub and Ngor chen’s authorial styles, strategies of argumentation, citation
methods, and tone have revealed tensions underlying the intellectual climate of fifteenth-
century Tibet. Mkhas grub’s methods are informed explicitly by his philosophical
training as evidenced by his more aggressive polemical tone and his incorporation of
non-tantric discourses like pramåña. Despite the fact that his chapter on body mandala
forms part of an extensive treatise on the generation stage of the Guhyasamåja sådhana
practice, he can’t quite seem to get past the larger implications of core tantric ritual acts
of imagination like imagining oneself as a deity. Mkhas grub continually ‘grinds his
axe,’ to borrow van der Kuijp’s phrase, construing the connection of imagination and
fabrication as a potential threat to the efficacy of tantric ritual action. Mkhas grub works
hard to emphasize the centrality of proper causal basis for enlightenment, one that is not
corrupted by mental imputations and that complements the påramitånaya approach. Ngor
chen, on the other hand, relies more consistently upon the methods of tantric exegesis and
focuses upon navigating the relationships between texts, ‘aligning the connections,’ for
his readers. Yet he also plays off Mkhas grub’s reliance upon pramåña, taunting him
with comments like, “Jut because you don’t see it, doesn’t mean it isn’t there.” While we
have exercised caution to avoid attributing undue weight to the sectarian dimension of the
body mandala debate at the outset, working slowly through the materials has
substantiated some fundamental doctrinal conflicts informing Mkhas grub and Ngor
chen’s encounter.
In emphasizing the textual authority of the oral instructions and of the teachings
of the Sakyapa patriarchs, Ngor chen participates in a broader field of what Davidson has
termed “Sakyapa apologetics.” The nature of his defense of the Hevajra body mandala is
fueled by concerns over its legitimate basis in attested Indian sources. However,
factoring in his other polemical writings, we find that Ngor chen’s writings on body
mandala are also constructed explicitly as part of a larger defense against charges that the
Sakyapa tradition of interpreting the Hevajra Tantra (at the very core of the Lam ‘bras)
was Cittramatrin in orientation. Why precisely was this construed as problematic? In
repeated rhetorical acts of reifying the Prasangika Madhyamaka perspective through
polemical writing and debate practices, Tibetan authors have created an impasse. For
many, this philosophical orientation makes resolving the relationship of the påramitånaya
and mantranaya approach particularly challenging. The Sakyapa “representationalist”
perspective embraces the illusive nature of reality as an unavoidable dimension of
saµsåric existence. This approach lends itself more easily to explaining the efficacy of
tantric ritual action as part of a superior path leading to a higher level of realization, the
Thirteenth level of Vajradhara. However, Mkhas grub continues to ‘grind his axe’
precisely because the emerging Gelukpa tradition placed so much emphasis upon reason
in the soteriological quest. Regarding påramitånaya and mantranaya as two methods
leading to the same goal made it hard for Mkhas grub to avoid evaluating them according
to the same criteria. The emphasis upon successfully performing the Prasangika
Madhyamaka position came with the risk of arriving at a place of no return. Therefore,
Mkhas grub is more likely to have been “grinding his axe” precisely because of his
prowess (and that of his teacher) in executing a philosophical position. In emphasizing
the significance of valid cognition and the necessity of directing the mind towards the
apprehension of verifiable objects of knowledge, the Gelukpa stance limits the
! 229!
possibilities for explaining how the methods of tantra work. Tsong kha pa developed a
complex causal schema to explain how tantric ritual may be used to connect ordinary and
enlightened existence. Mkhas grub, likewise, emphasizes the “unfabricated” nature of
the body mandala through foundation on a basis that is not “newly fabricated.” Both
thinkers resolve the problems of “fabrication” by emphasizing the completion stage as the
essential link between defiled and enlightened embodiment. Yet, in a sense, the problems
they solve are problems that have created for themselves. The body mandala debate is
also, therefore, an example of the risks of taking philosophical debate too far as
emphasized in Mkhas grub’s attempt at a retraction in Reply to the Questions of the
Kalyånamîtra Kon ting gug !rî ba .
The tensions between gradualism and subitism expressed so potently by the iconic
debate of Kamala!îla and Mohoyen also continue to haunt the project of defining Tibetan
Buddhist identity. The polemical exchange between Mkhas grub and Ngor chen
demonstrates how fourteenth-century tensions over Buddha nature remain unresolved;
they have merely been combined with other philosophical tensions to take on a new form
in ritual discourses. The pull between determinism and agency colors the interpretation
of ritual methods and goals and their relationship to embodiment in compelling ways.
For example, the Gelukpa Guhysamåja body mandala practice creatively manipulates
different types of bodies, working through progressively more subtle varieties aimed at
controlling future embodiment. The Sakyapas, on the other hand, embrace the ambiguity
of cause, method, and result, experimenting with the very boundary between saµsåra and
nirvåña. Both experiment with the category of the body as a foundation or support for
ritual action in a fruitful interplay of instrumentality, agency, and transcendence.
In addition to ritual and philosophy, the body mandala debate suggests
connections between tantric ritual and other technologies for knowing the body. For
example, Mkhas grub’s emphasis upon the link between the goddesses and elemental
rlung suggests concerns with articulating the nature of the elements and the movement of
bodily winds shared by medical authors. Moreover, his emphasis upon ritual acts of
dissolution reflects his formative role in shaping and promoting the Guhyasamåja ritual
tradition as a technology for navigating the transition between death, the intermediate
state, and rebirth. Considering multiple technologies for knowing the body’s sites of
vulnerability and potentiality promises to improve understandings of the evolution of
tantric ritual and to meaningfully investigate tantric texts as bodily discourses.
What does it mean to “pierce to the pith” [gnad du bsnun] of the body mandala
debate, to locate the “main point” [gnad don], the very heart of the matter? This
dissertation embraces the analogy between corporeality and textuality in various ways. It
explores ritual technologies of inscribing the body as a surface to be effaced, protected,
and purified or a basis to be transformed. Ritual correlations of body and cosmos
suggest that to know the body is to know the world. It considers how in mapping and
manipulating vital points, the practitioner parses the body, making it intelligible. Both
Ngor chen and Mkhas grub agree that body mandala practice pierces to the body’s very
pith, granting access to its hidden potentialities. Soteriology and exegesis are therefore
analogous processes. The classification of texts and of bodies has proven to be volatile
sites for experimenting with different Buddhist ways of knowing. Sakyapa tantric
perspectives suggest that the body is an “explanatory tantra,” a unique resource for
accessing the true nature of the mind. In learning to “align the dependently arisen
! 230!
connections,” to connect the dots, the ritual practitioner and exegete both strive to get to
the heart of the matter. And yet the body is always pointing beyond itself. As JZ Smith
would remind us, in a sense, there is only commentary.
There are multiple ways in which the body is a text we have authored. It is the
creative product of our karma and of the workings of our mind. Looking at
representations of the body can help us to learn more about social, economic, and
doctrinal anxieties of fifteenth-century thinkers as well as about the interplay of different
intellectual discourses: ritual, medicine, exegesis, polemics. So, like Foucault, we might
regard the body as a “cultural text.” However, as Judith Butler points out, this blank slate
approach presupposes that there is something we can call a body really there. Foucault’s
disciplined body does indeed bleed; it resists the effacement of its materiality. So the
contradiction Butler finds within Foucault’s thought, between the constructed nature of
embodiment and of its resistance, might be similar to the conundrum Mkhas grub faces in
two respects: in his struggle to resolve the relationship of ordinary and enlightened bodies
as well as between determinism and ritual agency. We might even think of Butler in
dialogue with the Sakyapas. For them, the body is an appearance masquerading as
reality, an appearance we ourselves don’t realize has been projected or constructed by our
mind or (for Butler) by our culture.
How do we access the true nature of reality, the true nature of the mind? Is it
through reason or through embodied ritual action? Like the new historicists, do we need
to learn a new way of understanding representations to access what lies beneath? Or do
we need to consider the possibility that the ‘reality’ we seek beneath the surface is just
another expectation or construction? LaTour is an ideal conversation partner in the sense
that he tempers the poststructuralist conundrum in a way that accords with Buddhist
perspectives. He encourages the iconoclash, that moment where we are poised on the
brink between structuralist and poststructuralist thought, between creation and
destruction, eternalism and nihilism, emptiness and form. LaTour teaches us to regard
the tensions between these dualities as productive in a similar way that ritual is- playing
upon the tension between representation and reality.
! 231!
Fig. 1 Mkhas grug dge legs dpal bzang po Nineteenth-century 40.64 x 68.58 cm
(16 x 27 in) Collection of Shelley and Donald Rubin P1994.8.4 himalayanart.org
HAR#56
! 232!
Fig. 2
Mkhas grub rje and the bodhisattva Manju!rî
Eighteenth-century
30 in x 20 in, 76.2 cm x 50.1 cm (image); 58 1/2 in x 33 in, 148.6 cm x 83.8 cm (overall)
! 233!
Fig. 3
Mkhas grub rje and the great adept Dombhi Heruka
Eighteenth-century
29 3/4 in x W. 20 in, 75.6 cm x W. 50.1 cm (image); 58 in x W. 34 in. 147.3 cm x W.
86.4 cm (overall)
! 234!
Fig. 4
Ngor chen kun dga’ bzang po with two lineages
1430’s-1460
34 1/16 x 28 3/8 in. (86.5 x 72 cm)
Michael Henss Collection, Zurich
Published in Jackson 2010 himalayanart.org HAR#88708
! 235!
Fig. 5 Mandalas from the Vajråvalî Cycle, Painting Five in the Set of Fourteen
35 1/4 x 29 in (89.5 x 73.7 cm); Mount: 53 x 33 in (134.6 x 83.8 cm)
Philadephia Museum of Art, Stella Kramrisch Collection 1994-138-635
Published in Thurman and Rhie 1997 [Fig.21], Kossak and Singer 1998 [Fig.47c]
! 236!
! 237!
Fig. 6 and detail
Kålacakra Mandala from the Vajråvalî cycle, Painting 11 in the set of Fourteen
And painting detail of sådhaka
35 5/8 x 29 in (90.5 x 75.5 cm)
Published in Kossak and Singer 1998 [Fig.47b]
! 238!
! 239!
Fig. 7 and detail
Mandalas of the Vajråvalî and Kriya-samuccaya cycles and detail of patron, final
painting in the set of Fourteen
35 x 29 in (88.9 x 73.7 cm)
Kimbell Art Museum AP 2000.01
Published in Thurman and Rhie 1991 [Fig.73]
! 240!
Fig. 8. The Maitreya Temple, Glo smon thang [Exterior and interior of middle floor
restoration]
Mustang, Nepal
Photos by Luigi Fieni, published in Lo Bue 2010
! 241!
Fig.9 Sanderson 1986
! 242!
Fig. 10 Sanderson 1986
! 243!
Fig. 11 diagram of Proto-body mandala in IOL Tib J 576
Deity List based on IOL Tib J 576:
ba dzra be ro tsa na [Vairocana] #1

ba dzra shwa ra [3.3] dha tu
Comm:
On the head, (rnam par snang mdzad) [Vairocana] and
(thams cad bdag [3.3] nyid ma)
sa tva ba dzra [Sattvavajra] #2

ba dzra mu sti [Vajra-mußti]
Comm:
On the eyes(s): (sems rdo rje )[Sattva-vajra]
??(fem) [brtan ma rdo rje] ??
! 244!
rad na ba dzra [Ratnavajra] #3
ba dzra mu sti [Vajra-mußti]
Comm:
On the nose: (rin cen rdo rje) [Ratna-vajra] (påramitå? But male)
? fem) ['bar ma rdo rje]??
dar ma [3.4] ba dzra [Dharmavajra] #4

ba dzra mu sti [vajra-mußti]
Comm:
On the ear(s): (chos rdo rje) Dharmavajra (påramitå? But male)
? (fem)(dud ma rdo rje) 690 ??
kar ma ba dzra[Karmavajra] #5
ba dzra mu sti [Vajra-mußtî]
Comm:On the tongue: Karma-vajra (las rdo rje ) (påramitå? But male)and
?(fem) (bskyod ma rdo rje).??
ba dzra ag sho bya hung #6

[Akßobhya]
ba dzra [3.5] bu ta lo ca na hung
(Buddha-Locanå)
Comm:
On the middle finger of the right hand: (mi bskyod pa ) [Akßobhya] and
(yangs pa'i spyan) [Buddha-Locanå].
ba dzra sa tva [Vajrasattva] #7

ba dzra la sye [Vajra-låsyå]
Comm:
On the index finger: (rdo rje sems pa )[Vajrasattva] and
(fem) (rdo rje rgeg mo) [Vajralåsyå]. (inner offering)
ba dzra ra ja [Vajraråja] #8
ba dzra tu pe [Vajradhupå]
Comm:
On the 4th finger from the thumb: (rdo rje rgyal po ) [Vajraråja] and
(fem) (rdo rje bdug pa ma)[Vajradhüpa]. (outer offering)
Page 4:
[4.1] ba dzra ra ga[Vajra-råga] #9
a 'gu sha hung [A÷ku!a]
Comm:
On the little finger: (rdo rje chags pa ) [Vajraråga] and
(fem) (rdo rje gyo ga ma) 691
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
690
This is unclear and therefore my translation of this name is tentative.
!
! 245!
ba dzra sa tu [Vajra-sådhu] #10
ba dzra te sha ra ti [Teja-ratî/Dve!aratî]
Comm:
On the thumb,: (rdo rje legs pa ) [Vajrasådhu] and
(rdo rje sgril ma) rest in union.
sama ya sa tvam sa ma ya stvam

om om [4.2] hung aoum am hung
ba dzra rad na sam bha ba [Ratnasambhava] #11

bz dzra ma ma ki [Måmakî]
Comm:
On the middle finger of the left hand: (rin cen 'byung gnas) [Ratnasambhava] and
“Jewel-eye”( rin cen spyan)
bz dzra rad na [Vajra-ratna] #12

ba dzra [4.3] ma le [Vajra-målå]
Comm:
On the index finger: (rdo rje rin cen) [Vajaratna] and
(fem) (rdo rje 'phreng ba ma ) [Vajramålå] (inner offering)
ba dzra su rya [Vajra-sürya]692 #13

ba dzra pu spe[Vajra-pußpå]
Comm:
On the fourth finger from the thumb: (rdo rje nyi ma) [Vajra-surya]* and
(fem) (rdo rje me tog ma) [Vajrapußpå] (outer offering)
ba dzra ke tu [Vajra-ketu] #14

ba dzra pa sha [Vajra-på!a]
Comm:
On the little finger: rdo rje rgyal mtshan) [Vajraketu] and
(rdo rje zhags pa) [Vajrapå!a] (door guardian)
[4.4] ba dzra ha sa693 [Vajrahåsa] #15

ba dzra rad na rad ti[Vajra-ratnaratî]
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
691
Though the text reads “rdo rje gyo ga ma.” This seems a bit odd, translating as “vajra-
dishonesty.” Jake Dalton suggesting emendation to “vajra-hook” [rdo rje lcags kyu ma]
which matches the Sanskrit appearing in the mantra, A÷küßa. A÷küßa is a door guardian
of the Vajradhåtu mandala although in male vs. female form. The Sarvadurgati
Pari!odhana tantra has four female gatekeepers of whom rdo rje lcags kyu ma is
one.
692
In the standard Vajradhåtu arrangement, Vajrateja [rdo rje gzi brjid] vs Vajra-
sürya occupies this position. See Giebel, 2001
693
Note the irregularity of the style of writing ‘sa’ and ‘ma’ in this text.
! 246!
Comm:
On the thumb, cultivate (rdo rje bzhad pa) [Vajrahåsa] and
fem) (rdo rje bde ma) in union.
sa ma ya stvam sa ma ya stvam om am hung om am [4.5] hung
ba dzra a mi ta ba [Amitabha] #16

ba dzra ban da ra ba si ni [Pañ∂aravåsinî]*
Comm:
On the middle toe of the left foot: snang ba mtha' yas [Amitabha ] and
(fem) gos dkar ma[Påñ∂aravåsini]
ba dzra dar ma [Vajradharma] #17

ba dzra ki ti [Vajragîti].
Comm:
On the second toe from the big toe: rdo rje chos [Vajradharma] and
rdo rje dbyangs len ma 694 (fem) [Vajragîti] (outer offering)
Page 2:[2.1] ba dzra tig sna [Vajra-tikßña] #18

ba dzra a lo ke [Vajrålokå]
Comm:
On the fourth toe from the big toe: rdo rje rnon po [Vajratîkßña] and
rdo rje mye sgron ma 695 (fem) [Vakrålokå]] (inner offering)
ba dzra ca kra [Vajra-cakra]696* #19

ba dzra spo tha [Vajra-spho†a]
Comm:
On the little toe: rdo rje 'khor lo [Vajracakra] and
rdo rje lcags sgrog ma (fem)697 (door guardian)
ba dzra ba she698 [Vajra-bhåsa] #20

ra ga ra ti [2.2] [Råga-ratî]
Comm:
On the big toe,: rdo rje smra ba [Vajra-bhåsa] and
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
694 !Translated in Rgyud sde kun btus as glu ma, but the meaning in virtually the same.!
695 !Translated in Rgyud sde kun btus as same as mar me ma, but the meaning in virtually
the same.
696
According to the standard formulation of the Vajradhatu mandala as found in Giebel
and in the rgyud sde kun btus, the deity in this position should be Vajrahetu (i.e. rdo rje
rgyu )
697
The Tibetan translation of this name as a female deity is strange, as vajra-spho†a is
one of the gate-keepers of the mandala. One again we appear to have a case of gender
confusion.
698
Following the standard layout of the mandala and Vajra-bhåsa’s place here, I have
read this as “ba-sa” rather than as how it appears (pa-sa).
! 247!
gsal bkra ma 699 (fem)
sa ma ya stvam sa ma ma ya stvam aoum am hung aoum am hung
ba dzra a mo ga si ti #21
ba dzra sa ma ya [2.3] ta ra
Comm:
On middle toe of the right foot: don yod par grub pa [Amoghasiddhi] and
dam tsig sgrol ma (fem) [Samayatårå ].*
ba dzra gar ma [Vajra-karma] #22

ba dzra nir ti a [Vajra-nîrtyå]
Comm:
On the second toe from the big toe: rdo rje las [Vajra-karma] and
(rdo rje gar byod ma) [Vajran®tyå] (inner offering)
ba dzra rak sha [Vajra-rakßa] #23

ba dzra gan ti [Vajra-gantdhå]
Comm:
On the fourth toe from the big toe: (rdo rje srung ba)[Vajrarakßa] and
rdo rje byug pa ma 700 [Vajragandhå] (outer offering)
ba dzra yak sha [Vajra-yakßa] #24

ba dzra [2.4] gan ta [Vajra-ghañ†å]
Comm:
On the little toe: (rdo rje gnod spyin) [Vajra-yakßa]
and rdo rje dril 'sgrol ma
ba dzra san di [Vajra-sandhi] #25

ba dzra ra ti [Vajra-Ratî]
Comm:
On the big toe: (rdo rje mtshams sbyor) [Vajrasandhi] (*agrees w/ Geibel but
rgyud sde has khu tshur i.e. Vajramußti) and
fem) (rdo rje’i ‘bebs ma) 701 the two resting in union.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
699
More research is required to confirm whether Rdo rje gsal bkra ma, “Vajra-bright
effulgence” is a standard translation of Råga-ratî.
700
The translation in the rgyud sde kun btus is dri chab ma, but the meanings are
compatible.!
701
This word, which I am reading as ‘bebs” was unclear, and therefore I am uncertain of
the translation as Vajråve!a “vajra-descent” and of its relation to the Sanskrit, vajra-ratî.
In accordance with the pattern of the text, we would have expected Vajråve!a, one of the
door guardians, to appear in the previous pairing with Vajrasandhi as noted above.
! 248!
! 249!
Fig. 12. Diagram & deity list of the Vajradhatu mandala from Giebel 2001
! 250!
Fig. 13
Scroll painting of the Vajradhåtu mandala from Dunhuang
MG 17780 Recto
Le Musée Guimet
Site: Dunhuang Mogao Form: painting, liturgical Materials: ink and colours on silk
Size (h x w) cm: 101.5 x 61
! 251!
Fig 14 Mandala painting from Dunhuang with consorts
EO 1148 Recto
Le Musée Guimet
Site: Dunhuang Mogao Form: mandala Materials: ink and colours on silk
Size (h x w) cm: 67.2 x 68
! 252!
!
!""#$%&'()* +,-.,/,(0&12,/&3,4&5$1(
!"#" $%&'()%*%+,-./0-./%1.23.45-60%.%0&-4%./%+-./0-7(89()(-:()(;0--
$%&'%'# ()*+,-).# /%-,0#
1"# 2%.,&+3# $+4&+56%-.7#89:;)6<0+=#
>+',# $+-&)*+.+#
?)@,<# A+,.+'+B6<+C+#
1F#
D%',# 9B-,E6<+#
G)&,<# 9B)7<+'-HH<-#
?)@,<5%+',# I)<+&+,J#8()*+.E=#
?)@,<5K%',# /C%;+&+,J#8IEB+:J=#
1O#
G)&,<5K%',# AE7+&+,J##8LEMN+&+CE'-.J=#
G)&,<5%+',# $+4&+&+,J##8PE&E=#
?)@,<5%+',%&.#-.,%&B%H-+,%# AQR+C+4&E#
?)@,<5K%',%&.#-.,%&B%H-+,%# S+6H+C+4&ET#
VW# G)&,<5K%',%&.#-.,%&B%H-+,%# U+.H<+C+4&E#
G)&,<5%+',%&.#-.,%&B%H-+,%# A+'+C+4&ET#
2%.,&+3T#-.#@.-).#K-,<#9:;)6<0+# ?R+&X+C+4&E#
>+',%&.#%H7%# I+-,&%0+#+.H#Y;-,-7+&6<+#
VZ#
?)@,<%&.#%H7%# $+4&ER+.-#+.H#Y<+7+&6<+#
():%XC+&+#+.H##
D%',%&.#%H7%#
I+[4@7<);+#\-]%]#I+[4@X&J^#
V1#
?+&C+.JC+&+M+C-;:+B6-.##
G)&,<%&.#%H7%#
+.H#?+B+.,+6<+H&+#
>+',%&.#H))&K+0# _+BE.,+:+#
?)@,<%&.#H))&K+0# 9R+&E4-,+##
VV#
D%',%&.#H))&K+0# `+0+7&JC+##
G)&,<%&.#H))&K+0# 9Ba,+:@MN+3-#
?)@,<5%+',%&.#*)&.%&# 9*+3+#
?)@,<5K%',%&.#*)&.%&# P+::-&E4+#
Vb#
G)&,<5K%',%&.#*)&.%&# GJ3+.H+MN+#
G)&,<5%+',%&.#*)&.%&# I+<E6+3+#
c%3)K# ?@B6<+&E4+#
V!#
96)C%# d;MJ;+*+:&+C+&,J#
#
Fig. 15 Wright 2010
!"#
! 253!
!
!"#" $%&'()%*%+,-./0-1(%+-20%.3-4523-1(67()(---
$%&'%'# ()*+,-).# /%-,0# 1022+32%# 4)2)5&# 6+,5&%#
!7# 4&)8.# $+-&)*+.+# )9# 8:-,%# ;)&<#+==&%=+,%#
!># ?)5,:# @<-,A3:+# AB# &%C# C-'*&-<-.+,-).#+==&%=+,%#
!"# D%+&,E*%.,&%# @FG)3:0+# :H9# 2+I-'#2+J52-# *).'*-)5'.%''#+==&%=+,%#
!K# 6+L%2# M+,.+'+<3:+L+# 'LA# 0%22)8# ;%%2-.=#+==&%=+,%#
*)<I)'-,-).+2#;+*,)&'#
7N# O),:#;%%,# @<)=:+'-CC:-# :A# =&%%.#
+==&%=+,%#
P2+*%#);#%+&,:#
?):+&+,Q#R()*+.AS# E# E# ')2-C-,0#
%2%<%.,#
P2+*%#);#8+,%&# /L%G+&+,Q#
E# E# ;25-C-,0#
%2%<%.,# R?A<+FQS#
7V#
P2+*%#);#;-&%# MA=+&+,Q##
E# E# 8+&<,:#
%2%<%.,# RPATU+&+LA'-.QS#
P2+*%#);#+-&# $+W&+&+,Q###
E# E# +-&-.%''#
%2%<%.,# RXA&AS#
Y0%'# ZG-,-=+&3:+# ,:2Q9# E# E#
7[#
Y+&'# $+W&+IAT-\# )9# E# E#
6)'%# Z:+=+&3:+# )9# E# E#
7]#
X).=5%# ()F%^L+&+\# )9# E# E#
?+_W5=:)G+##
D%+&,# :H9# E# E#
`-\%\#?+_W5^&Qa#
7b#
1+&L+.QL+&+T+E
c:)2%#3)C0# )9 # E# E#
L-GF+<3-.#
D%+C# ?+-,&%0+# <+-9# E# E#
7!# @22#,:%#W)-.,'##
1+<+.,+3:+C&+# '+9# E# E#
R);#,:%#3)C0S#
M-=:,#:+.C# d+<A.,+F+# E# E# E#
(%;,#:+.C# @I+&AW-,+## E# E# E#
77#
?)5,:# D+0+=&QL+## E# E# E#
$+W&+# @<e,+F5TU+2-## E# E# E#
M-=:,#
@*+2+# E# E# E#
'-C%f':)52C%&#
#
(%;,#
7># X+FF-&AW+# E# E# E#
'-C%f':)52C%&#
M-=:,#F.%%# 6Q2+.C+TU+# E# E# E#
(%;,#F.%%# ?+:A3+2+# E# E# E#
7"# X)I#);#:%+C# 15<3:+&AW+# E# E# E#
O),,)<#);#;%%,# gGTQG+*+F&+L+&,Q# E# E# E#
Fig. 16 Wright 2010

!"#
! 254!
!
!"#" $%&'()%*%+,-./0-10%.2-34+&45.-6412-7(89()(--
$%&'%'# (%)*+# ,+--./-%# 012.*)13#
4$.)&12.3.5# 16# 7&183#
9:#
4;<)*=/>.5# =?# @1A*>#
4;BC1/>+.5# >D6# E%.&*F2%3*&%#
9!# 4G.*3.'.</>.H.5# 'H=# I.H%-#
4;<1J>.')KK>)5# >=# L1*>#M%%*#
012.3=# F# N-.2%#1M#%.&*>#%-%<%3*#
@=<.BO# F# 48.*%&#%-%<%3*5#
9!F9R#
N=PQ.&.H=')3O# F# 4M)&%#%-%<%3*5#
S=&=# F# 4.)&#%-%<%3*5#
GDT.H.U&=# F# F#
V./K.H.U&=W# F# F#
9R#
X.3K>.H.U&=# F# F#
G.'.H.U&=# F# F#
YC)*)J.&/>.# F# Z3)13#8)*>#GDT.H.U&=#
$.U&.T=P)# F# Z3)13#8)*>#V./K.H.U&=W#
Y>.J.&/>.# F# Z3)13#8)*>#X.3K>.H.U&=#
9\#
01B%[H.&.# F# Z3)13#8)*>#G.'.H.U&=#
$.U&.H%*=-O# F# G)J>*#>.3K#
;T.&=U)*=# F# 0%M*#>.3K#
L>]BA^)# F# @1A*>#
99# _B.U.^=# F# ,%2&%*#T-.2%#
$)[H.H.U&O# F# G)J>*##'>1A-K%&#
$)[H.&.*3=# F# 0%M*#'>1A-K%&#
`""#
$)[H.T.K<=# F# G)J>*#B3%%#
$)[H.B.&<=# F# 0%M*#B3%%#
`"`# ;B.[.H.U&)PO# F# S1T#1M#>%.K#
_.&*>F/%.&)3J#X1KK%''# F# L1**1<#1M#M%%*#
#
Fig. 17 Wright 2010
! 255!
!"#
Fig. 18 Main assembly hall at Tabo monastery in relation to the Vajradhåtu mandala
Luczanits 2004
! 256!
Fig. 19
Dhårañî/mandala
1919,0101,0.18
Institute&Copyright: British Museum
(Ch.xxii.0015)
Form: mandala, painting
Size (h x w) cm:
58.5 x 56.3
! 257!
Fig. 20
Dhårañî
Pelliot tibétain 4216
Site: Dunhuang Mogao Materials: ink on paper
! 258!
Fig. 21
Dhårañî/mandala
Pelliot Tibetain 389
ISite: Dunhuang Mogao Mateials: ink on paper Size (h x w) cm: 31 x 40
image: http://idp.bl.uk/ © Bibliothèque nationale de France
! 259!
Fig. 22
Mandala of Guhyasamåja body mandala based on Piñ∂ik®†a sådhana
! 260!
68 Amy Heller
Two Early Tibetan Ritual Diagrams for Cakra Meditations 69
Fig.
Fig. 23
Diagram
Diagram for Cakra
for Cakra Meditations
meditations (recto(recto)
& verso)
Diagram for Cakra Meditations (verso)
published in Heller 2010 and Pal 2007
! 261!
Fig. 24
Painting of the Cakrasaµvara body mandala
HAR# 5968
Nepal
Nineteenth or twentieth century
Ground Mineral pigment on cotton
Private Collection
! 262!
Fig. 25
HAR#100001; Nepal; 18th century
91.44 x 210.82 cm [36 x 83 in]
Ground mineral pigment on cotton
Collection of Shelley and Donald Rubin
! 263!
Fig 26
Vißñu Vi!varüpa. India, Rajasthan, Jaipur, ca. 1800-20. Opaque watercolor and gold on
paper, 38.5 x 28cm. Victoria and Albert Museum, London, Given by Mrs. Gerald Clark,
IS.33-2006
Published in Diamond 2013, fig. 10b
! 264!
Fig. 27
Published in Fernand Meyer with Yuri Parfionovich and Gyurme Dorje, Tibetan Medical
Paintings: Illustrations to the Blue Beryl Treatise of Sangye Gyamtso Vol.1 ( London:
Serindia Publications, 1992), Plate 6.
! 265!
Fig. 28 Published in Howard 1984, fig. 2 “Cosmological Buddha from a fresco in Cave
428 at Tun-Huang. Dated to the Northern Wei dynasty, circa 525 (After Tonkø
Makkøkutsu, vol.1, pl.162.)”
! 266!
Fig. 29 Published in Howard 1984, fig.1 “Cosmological Buddha. Dated to the Sui
Dynasty (581-618). Stone, 176.5 x 64.2 cm. Courtesy, Freer Gallery of Art, Smithsonian
Institution, Washington, D.C.”
! 267!
Fig. 30
Published in Diamond 2013 Fig.11B,
Bulaki
! 268!
Fig. 31
Published in Diamond 2013 Fig.10D, Folio 6 from the Siddha Siddhanta Paddhati
Bulaki
! 269!
Fig.32
Jain cosmos
fifteenth-seventeeth century
Published in Granoff Victorious Ones: Jain Images of Perfection 2009-2010 Fig 3.5
! 270!
Fig.33
“The Jain Universe in the Shape of a Cosmic Man or lokåpurußa”
Folia from loose leaf manuscript
Gujarat or Rajasthan, early seventeenth century
Ink and opaque water color on paper
Collection of Bina and Navin Kumar Jain
Published in Granoff Victorious Ones: Jain Images of Perfection 2009-2010 Fig.2.1
! 271!
Fig.34
HAR# 81836; Tibet; 18th century; Ground Mineral Pigment on Cotton; Collection of the
Rubin Museum of Art
! 272!
Fig, 35
Body mandala of Vajrayoginî
Published in English 2002 fig.33
! 273!
Fig. 36
Cakrasaµvara mandala
HAR #85813;
Nepal, 1490 Buddhist Lineage;
Ground Mineral Pigment on Cotton; Collection of the LA County Museum of Art;
Published in Huntington and Bangdel 2003 Fig.70; www.himalayanart.org
! 274!
Fig.37
Numbered diagram of the body mandala painting (Fig. 24 HAR# 5968)
! 275!
Numbered Inscription, Descriptions, and proposed deity identifications from the Body
Mandala painting
-Identifications have been supplemented with English 2002’s list of mandala deities;
suggested parallels with those deities and their associated sites are indicated by ‘**CKS.’
-Suggestions for Newarî terms and for resolving scribal idiosyncracies provided by
Alexander von Rospatt and indicated by ‘AVR.’
-Directions (North, Northwest...) have been abbreviated (N, NW).
-Illegegible akßaras indicated with ?
1. vajra kåla. Yama∂huti.

Blue/black line connects to ear (our left his right)
[flesh-colored 3-faced 6-armed male deity in union w/ two-tone red-orange female deity]
2. vajra ma?håbala. yama v®†h†hi

blue/black line connects to eye
[dark blue 3-faced 6-armed deity in union w/ two-tone red-green female]
**seems likely this refere to Yamadaµß†riñî of the ‘outer mandala’?meaning

samayacakra who is connected w/ eye- NW
3. vajråna?råraka urukåsyå.
Blue-black line connects to nostril
[yellow 3-faced 6-armed deity in union w/ blue/green female]
**CKS- Ulükåsyå- - ‘outer mandala’?meaning samayacakra- navel- N
4.praca?Na . kåkasayå.
Red line connects to mouth
[dark blue 3-faced 6-armed deity in union w/ blue-black female]
**CKS- Kåkåsyå- - ‘outer mandala’?meaning samayacakra- mouth- *matches-E
5. Vajrasatva; vajramahiteja
yellow line connects to navel cakra
[white one faced two armed deity in union w/ red ?4-armed? female]
6. Vajranila. Vajraheruka.
White line connects to secret place
[blue 1-faced 2-armed male deity in union w/ dark blue female]
7. vajra suryya. Saµ?vårinî/saµtråhinî [?Samtråsinî]

yellow line connects to ?crown cakra
[gold deity one-faced two-armed in union with green female]
8. !rî heruka saµcåri?ni [?Samcålinî]

yellow line connects to ?crown cakra
[dark blue deity one-faced two-armed in union with gold female]
! 276!
9. padma n®tyasvara. [?Padmanarte!vara] Mohinîdevî.
??no connecting line
[red deity one-faced two armed In union w/ green female]
10. Vail[?r]ocana.Yåminî.
Yellow line connects to axis between heart and throat cakra.
[flesh colored 1-faced 2-armed deity in union w/ red female]
**CKS Vairocana is ialso dentified w/ phlegm- nourished by goddess/vein Cakravartinî
11. mohå vajra

red line connects to eye
[flesh colored 1 faced 2 armed deity]
*Mohavajra is a name for the bodhisattva Kßitigarbha who is correlated with eye faculty
in the preparatory practices for purifying the body, speech and mind of the practitioner in
the Cakrasaµvara tradition. Mohavajrå is the goddess associated with vision in the
Hevajra system.
12.∂ho?ma vajra [does this refer to Dveßavajra?]

red line connects to ear
[dark blue 1-faced 2-armed deity]
13. ??cvaryya vajra [?wierd akßara- could this be ˆrßya vajra?]

red line connects w/ center of nose
[gold one-faced 2-armed deity]
14. raga vajra. Raga vajrî.

?no connecting line?
[red one-faced two-armed deity in union with gold female deity]
15. inscription difficult to read*

gold line connects w/19, ?crown cakra & by extension, 7
[blue-green yab yum, single-faced and two armed]
16. mohå vajrî

red line connects to eye
[flesh colored singled faced 2-armed]
As for Mohavajrî, Huntington and Bangdel 2003 place her in the SE of the mandala in
accord with the Candamahåro!ana Tantra. However, here she is more likely consort of
#11.
[See also inscription F to the left: mikhå laµke !varî- see below]
17. de?gha(/dha) vajrî

red line connects to ear
[blue single-faced 2 armed]
18.?dvaryya vajri
! 277!
red line connects w/ axis between eyebrows
[yellow single-faced 2-armed]
19. pa?(nv?)årinî vajra vårahi.

Gold line connects with ?crown cakra & ?15
[dark blue; 3-faced; 4-6 arms]
*We read the first word as Padmajvålinî, associated with the element of space.
20. pa?? (?padma?) n®tya!vari. Tårå.

yellow line connects to ??red deity at crown cakra
[blue-green; single faced; four armed]
*We read the first word as Narte!varî, associated with the karma family (as is Tårå) and
the purification of the wind element.
21.Åka?mani devi. Padmanî tårå

yellow line connects to ?throat cakra
[dark blue single faced four armed deity]
*We read the first word as Åkarßanî, associated with the lotus family [thus making more
sense of Padmanî tårå as well] and the purification of the fire element.
22. se!varyya vajra. se!varyya vajri.

Red line connects to mouth
[flesh colored yab yum each with one face and two arms]
***Aisvaryavajra- purification of sense fields
23.vajra mi†håna/ vajra mißh?åna Yamathanî.

Blue/black line connects w/eye
[3-faced 6-armed red deity in union w/ two-tone blue consort]
**CKS- Yamamathanî- - ‘outer mandala’?meaning samayacakra- nose- NE
24. vajrayaka. Yama∂ådhî.

Blue/black line connects w/ ear
[3-faced 6-armed red deity in union w/ two tone yellow& blue consort]
**CKS- Yamadådhî - ‘outer mandala’?meaning samayacakra -hair curl –SE
25. vajra kusuna/kusun sukasyå?/sukaråsyå

blue/black line connects to nostril
[3-faced 6-armed blue deity in union w/ yellow animal headed (?dog?) consort]
**CKS- sükaråsya- ‘outer mandala’? meaning samayacakra- anus- S
26. vajra ??riya. svånasya.

White line connects to ?hip/secret place
[3-faced 6-armed red deity in union w/ orange animal-headed (?lion) consort]
**CKS- ¸vånåsyå- ‘outer mandala’?meaning samayacakra- sexual organ- W
27. mårini devî. Måma kîtårå.
! 278!
Yellow line connects with heart cakra
[red single-faced 4-armed deity]
28. bhårinî devî. Lo??????rå

yellow line connects w/ ?navel cakra
[yellow single-faced 4-armed deity]
**perhaps this completes the set of 4 w. Locanå and Påñ∂ara, but again, why in the CKS
mandala? ***AVR- suggests Locanå Tårå could be Locanå-devî? Since there is a
inscriptions surrounding head:

?a few seem to include seed syllables
**some suggestion Newari may be used here
from top
our left (figure’s right):
A. diff to read: jå?˙/µ mahå?ke kå?? Can∂hå∂hî/ Can∂håkî
*CKS- This is likely referring to Cañ∂åkßi, a goddess of the citta cakra associated with
the topknot who nourishes the head and body hair as represented by Mahåkankåla.
[Assoc site: Jålandhara]
our right (his left)

B. diff to read??????????????????????
[could this be Pracañ∂å (maybe 4)Goddess/channel at head who nourishes the nails &
teeth in association with & Khañ∂akapålin? [Pullîramalaya]
Our left:
C. lå˙ [AVR*looks like anunåsika] håtikå ami tåhå
[AVR- Newar Above the east- AVR???]
*CKS?Amitåbha is associated with the nourishment of the bones nourished by the
goddess/channel Kharvarî of the citta cakra located at the point between the eyebrows.
[site: Råme!vara]
Our right:
D. Diff to read:mi??gå vikathavamkå ?mahå nåså
**CKS- Mahånåså is the goddess/channel of the cittacakra located at the back of the head
who nourishes the flesh in association with Vika†adaµß†rin [site: Arbuda]
E & F- SPLIT UP COUPLE?

Our left:
E. mikhå/ bhikhå ?vajra(/bala) prabhå deµ
**bala more likely
**As pointed out by AVR, mikha is the Newar word for eye. This may be a significant
point in understanding the script style used here.
! 279!
**CKS-Vajraprabha is indeed associated with the nourishment of the kidneys by the
goddess/channel LaNke!varî of the citta cakra who is located at the eyes. [site:
Devikeviko†a]
See inscription on the opposite side of the head : mikhå laµke !varî
Our right:
F. mikhå laµke !varî
*As noted above, AVR pointed out that mikha is the Newari term for eye.
**CKS-Lanke!varî is the goddess/channel of the citta cakra located at the eyes who
nourishes the kidney as Vajraprabha. See the inscription on the opposite side of the head:
mikhå ?vajra prabhå deµ
29. Our Left:

(at ear level) kaµ kålå pranåmatî
[deity on ear brown: ?1 face 4-armed deity in union w/ blue consort]
**CKS- Kankala is associated with the nourishment of the skin and filth by the
goddess/channel of the citta cakra, Prabhåvatî who is located at the right ear. See the
inscription on the opposite side of the head, which also matches up with the CKS system
in connection w/ the left ear.
30. Our right:

(at ear level) suvirå vîramati
[deity on ear flesh-colored 1 face ?4 arms in union w/ red/brown consort]
**CKS- Vîramatî is the goddess/channel of the cittacakra located at the left ear who
nourishes the sinew in association with Suråvairin. [site: Godåvarî]
31. Our left:

(at earlobe) subhadrå syåmadevî kå
[deity on earlobe: 1 face 4-armed deity in union w/orange consort]
**CKS-¸yåmadevî is the goddess/channel of the vakcakra who is located at the mouth
and nourishes the ‘coiled gut’ is association with Subhadra. [site: Kalinga]
Subhadrå is the goddess/channel of the vakcakra who is located at the throat and
nourishes the belly in association with Vajrabhadra.
So, perhaps we should read Subhadra., short a.
32. Our right:

(at earlobe) vaj(?r)aprabha subhadrå laµ
[deity on earlobe- 1 face?4 arms gold in union w/ red/brown consort]
**CKS- Subhadrå is the goddess/channel of the vakcakra who is located at the throat and
nourishes the belly in association with Vajrabhadra. [site: Lampåka]
??see also opposite side: subhadrå syåmadevî kå.
! 280!
trunk: ONLY 39 & 40 are yabyum deities- the rest are depicted
individually
*many include seed syllables
33. our left (his right):(at shoulder) vajra deha. Påµ/påh

[flesh colored 1 faced 4 armed]
**CKS- Vajradeha is associated with the nourishment of the heart by the
goddess/channel of the *citta cakra Drumacchåyå who is located at the shoulders. [site:
Målava]
34. our right (his left):(at shoulder) ?hemåchåyå mau[AVR- or anunåsika

mah***transcribe with colon: ma:]
[flesh colored 1 faced 4 armed]
??Himålaya is the sacred site associated with the penis within the vakcakra, so this does
not seem apt.*However, Målava is the sacred site connected w/ the shoulders within the
citta cakra, and Drumacchåyå is the goddess/channel nourishing the heart in association
with Vajradeha**
***Hemachaya -AVR- ?aspirated cha (AVr-/kßema]
35. Our left:(chest near armpit) kulîka. Kåµ

[flesh colored 1-faced 4-armed]
**CKS- Ankurika is associated with the nourishment of the eyes by the goddess/channel
Åiråvatî of the vakcakra who is located at the armpits. [site: Kåmarüpa]
36. Our right: (chest near armpit) seravatî. Phåµ (?/khåm)

[flesh-colored one face two arms]
**CKS- Airåvatî is the goddess/channel of the vakcakra located at the armpits (and
associated with the nourishment of the eyes in association with ANkurika).
37. Our left: (chest near heart) vajra ??? tilaµ/: vaµ
[gold 1 faced 4 armed]
**CKS- perhaps this is Vajraja†ila who is associated w/bile and nourished by the goddess
/channel Mahåbhairava of the vak cakra who is located at the breasts. [site: O∂ra]
See opposite side.
38. Our right: (chest near heart) mahåbhairava (seems to go w/ this)

[gold 1-face 4-armed deity]
**CKS- Mahåbhairava is the goddess/channel of the vak cakra who is located on the
breasts and nourishes the bile in association with Vajraja†ila. [site O∂ra]
! 281!
There is also a Mahåbhairava who is associated with feces and nourished by the
goddess/channel of the vak cakra Hayakarña who is located at the heart[site: Kå•cî]
39. Our left: (rib area) to right/center reads: ha?yakar?gha

**CKS-Hayakarñå is the goddess/channel of the vak cakra who is located at the heart and
nourishes the feces in association w/ Mahåbhairava. [site: Kå•cî]
[gold deity 1-faced 4-armed in union w/ red female]
?below reads: vajraprabha kåµ
?SO 38 and 39 could be Mahåbhairava and HayaHayakarñå respectively
40. Our right: (rib area) mahåvîra vayuvega

[red deity 4/6 armed in union with red consort]
**CKS- Våyuvegå is the goddess/channel of the vak caka who is located at the navel and
nourishes the lungs in association with Mahåvira. [site: Tri!akuni]
lower body
*seem to include seed syllables of a sort
these deities appear to be represented singly with their partner on the opposite side-
partners match in color
41. our left (his right): (on hand ) vailocarå sa. [blue deity ?1 face and 2/4 armed]
**CKS- Vairocana is associated with the phlegm nourished by the goddess/channel
Cakravartinî of the kåya cakra who is located at the thumbs & big toes. [site: Maru]
42. our right (his left): (on hand ) cakra vartî vaµ [blue deity 1 face ?2/4 arms]
**CKS- Cakravartinî of the kåya cakra who is located at the thumbs & big toes is the
goddess/channel who nourishes phlegm which is associated with Vairocana. [site:Maru]
See opposite for Vairocana.
43. our left: (thigh) hayagrîva. ?laµ [red deity 1 face 4 arms]
**CKS- Hayagrîva is associated by the blood which is nourished by the goddess/channel
Sauñ∂înî of the kåya cakra who is located at the thighs. [site: sauråß†®a]
44. our right: (thigh) saunvirî soµ/seµ [red deity 1 face 4 arms]
**CKS-Sauñ∂înî goddess/channel of the kåya cakra who is located at the thighs
nourishes the blood which I associated with Hayagrîva. See opp for Hayagrîva.
45. our left: (knee) Vajrasatva. ?kaµ [gold deity 1 face 4 arms]
**CKS- Vajrasattva is associated with snot which is nourished by the goddess/channel
Mahåvîryå of the kåya cakra who is located at the knees. [site : Kulatå]
46. our right: (knee) mahåviryya kuµ [gold deity 1 face ?4 arms]
! 282!
**CKS- Mahåvîryå is the goddess/channel of the kåya cakra who is located at the knees
who nourishes snot which is associated with Vajrasattva. [site : Kulatå]
See opp for Vajrasattva.
47. our left: (shin) åkåsagarbha suµ [red deity 1 face ?4 arms]
**CKS- Åkå!agarbga is associated with sweat which is nourished by the goddess/channel
Cakravarmiñî of the kåya cakra who is located at the shanks. [site: Suvarñadvipa]
48. our right: (shin) cakra varmi suµ [red deity 1 face ?4 arms]
**CKS-Cakravarmiñî is the goddess/channel of the kåya cakra who is located at the
shanks and nourishes the sweat which is associated with Åkå!agarbha. [site:
Suvarñadvipa]See opp for Åkå!agarbha.
49. our left: (big toe ) målyåli. ?naµ [flesh colored deity 1 face ? 4 arms]
**AVR-målpati- big toe
50. our right: (big toe ) suvirå naµ [flesh colored 1 face 4 arms]
**CKS-Suvirå is the goddess/channel of the kåyacakra who is located at the fingers and
toes and nourishes the fat which is associated with heruka. [site: Nagara] Should the opp
[49?] then be Heruka?
51. our left: (sole of foot) padman®tya ghara/dhara siµ [flesh colored deity 1 face ? 4
arms]
**CKS- Padmanarte!vara is associated with the tears which are nourished by the
goddess/channel Mahåbalå of the kåyacakra who is located at the back of the feet. [site
Sindhu]
52. our right: (sole of foot) mahå balå siµ [flesh colored 1 face 4 arms]
**CKS-Mahåbalå is the goddess/channel of the kåyacakra who is located at the back of
the feet and who nourishes the tears which are associated with Padmanarte!vara. [site
Sindhu]. See opposite for Padmanarte!vara.
foreground:
**these three are in union
53. (central-seems to be aligned w/ central axis) ratna vajra. ?cando rohå

[flesh-colored 1 face ?6 arms in union w/ flesh colored consort]
**CKS- Khañ∂arohå is the goddess/channel of the kåyacakra who is located at the anus
and who nourishes pus, which is associated with Ratnavajra. [site G®hadevatå]
(She is also one of the four goddesses in the retinue at the heart)[sh and kh exchanged -
AVR]
54. (flanking to our left his right) viru på?∂e [AVR suggests che/could this be kße
instead- ?consort of Khaganana] khagånanå hr: sipå.
White line connects to ?secret place
! 283!
***CKS- Khagånanå is the goddess/channel of the vak cakra located on the penis who
nourishes the hair part which is associated with Virüpakßa. [site: Himålaya]
[AVR- khagånanå is associated w/ Guhyesvari- genitals]
55. (flanking to our right his left) Mahå???? Cakra??gå/må peµ

[AVR suggests cakradega-?Me- could this be cakravega]????[maybe mahåbala]
(simha?)SMUDGED [could it match the other side- sipå? Like opposite side?]
White line connects to ?secret place
***CKS- Cakravega is the goddess/channel of the kåyacakra located at the penis who
nourishes the phlegm which is associated with Mahåbala. [site: Pretapurî]
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
! 284!
Bibliography
Tibetan Language Sources
(‘Phags pa) Blo gros rgyl mtshan. (1235-1280). Hevajra Body Mandala Sådhana. "Kyai
rdo rje lus dkyil gyi sgrub thabs/." In Sa skya bka' 'bum. Collected writings of the first
five great patriarchs of the Sakya order: Includes the three supplementary volumes
recently published by Khenpo Tsultrim Gyaltsen, listed separately [W20751]. Reprinted
from a set of Dege Parkhang prints. TBRC W22271. 13: 538 - 542. dehra dun: sakya
center, 1992-1993.
Bu ston Rin chen grub (1290-1364). Clarifying the Generation Stage: the extensive
explanation of the Mdor byas [Piñ∂ik®†a sådhana] Guhyasamåja sådhana. Dpal gsang ba
‘dus pa’i sgrub thabs mdor byas kyi rgya cher bshad pa bskyed rim gsal byed. In The
Collected Works of Bu-ston. New Delhi: International Academy of Indian Culture, 1967,
Vol. 9, 683-877. [cited in Bentor 2006]
Nag po pa’s Cakrasaµvara Sådhana, Free from Errors or Impurities. Bde mchog
nag po pa’i sgrub thabs ‘khrul ba’i dri bral. Toh 5049. In gsung 'bum/_rin chen grub
(zhol par khang). TBRC W1934. 7: 151 - 186. [lha sa]: [zhol par khang], [2000].
The Jewel Ornament of Tantric Classes: the Classification of the General Tantric
Classes. Rgyud sde spyi’i rnam par gzhag pa. Rgyud sde rin po che’i mdzes rgyan.
Lokesh Chandra (ed.) The Collected Works of Bu ston (vol.Ba). New Delhi: International
Academy of Indian Culture 1966. [cited in Szántó 2013]
Dol po pa Shes rab rgyal msthan. (1292-1361) The Ocean of Definitive Meaning of
Mountain Dharma. ( Jo nang ) Ri chos nges don rgya mtsho. Beijing: Mi rigs dpe skrun
khang, 1998. [cited in Mathes 2008]
Rays of Sunshine. Nyi ma’i ‘od zer. “Theg pa chen po rgyud bla ma’I bstan bcos
legs bshad nyi ma’i ‘od zer.” The ‘Dzam-thang Edition of the Collected Works of Kun
mkhyen Dol-po-pa Shes-rab rgyal-mtshan, vol.4 (ma), 883-1161. Delhi: Shedrup Books,
1992. [cited in Mathes 2008]
(Go rams pa) bsod nams seng ge (1429-1489). Illuminating the Pith: Dispelling
Objections to the Moonrays of the Pith. Gnad gyi zla zer la rtsod pa spong ba gnad kyi
gsal byed, In gsung 'bum/_bsod nams seng ge/. TBRC W1PD1725. 12: 655 - 812. [dkar
mdzes bod rigs rang skyong khul, sde dge rdzong, rdzong sar khams bye'i slob gling]:
rdzong sar khams bye'i slob gling, 2004-2014. [cited in Davidson 1992]
Glob bo mkhan chen bsod nams lhun grub. (1456-1532) Rje btsun bsod nams lhun grub
legs pa’i ‘byung gnas rgyal mtshan dpal bzang po’i rnam par mthar pa zhus lan. Tøyø
Bunko 41-683. [See Kramer 2008]
Grags pa rgyal mtshan. (1147-1216) Body mandala. Lus kyi dkyil ‘khor. Digital Sakya
lam ‘bras collection, Vol.10, p.140-143.
! 285!
For other editions, see:
Lus kyi dkyil 'khor, in Pod-ser-ma (Bhir:'Jam dbyangs lung-rtogs dpal-
bzang, 1970), pp. 169.3-173.4. [cited in Davidson 1992]
Lam la sogs pa’’chos nyi shu la) lus kyi dyil ‘khor, Sa skya Lam ‘bras
series Vol. 11, 68r-69v [Cited in Sobisch 2008]
The Wish-Fulfilling Tree. Mngon par rtogs pa rin po che'i ljon shing. In gsung
'bum ?dpe bsdur ma?/_grags pa rgyal mtshan/. TBRC W2DB4569. 1: 19 - 293. pe cin:
krung go'i bod rig pa dpe skrun khang, 2007. [See also edition cited in Davidson 1992 fn
47: Rgyud kyi mngon rtogs rin po che’i ljon shing [Complete works Vol. 3, pp.29.1.3ff.
Garrett 2008 also engages with Grags pa rgyal mtshan’s text] .
Mkhas grub rje dge legs dpal bzang. (1385-1438). new Zhol par khang edition of gsung
'bum/_mkhas grub rje (zhol). Reproduced from a set of prints from the 1897 Lhasa Old
Zhol (Ganden Puntso Ling) blocks. TOH 5481. New Delhi: Mongolian Lama Guru Deva.
1980-2. TBRC W384.
Dispelling Delusions regarding the Hevajra Sådhana. Kyai rdo rje’i sgrub thabs
‘khrul spong. TBRC W384.Vol. 8, pp.89-135
[Ocean of Attainment ] Gsang 'dus bskyed rim dngos grub rgya mtsho. TBRC
W384.Vol. 7 (ja), pp.5-384: See 233-262.
Reply to the Questions of the Kalyånamîtra Kon ting gug !rî ba. Dge
ba'i bshes gnyen kon ting gug shri ba'i dris lan. Thor-bu Collected Works, TBRC W384.
Vol. 9, 775-808 esp.776-7.
Some Difficult Points in the Generation Stage of the Ghañ†apa Body Mandala.
Dril bu lus dkyil gyi byang du byas pa’i bskyed rim gyi bka’ gnas ‘ga’ zhig. TBRC
W384. Vol. 6, pp.765-787.
The Illuminating Lamp for Traversing the Paths and Grounds of Mantra(naya)
and Påramitå(naya). Sngags dang pha rol tu phyin pa’i sa lam bsgrod tshul gsal sgron
dang sbyar ba. TBRC W384.Vol 8, pp.533-554.
The Extensive Explanation for Classifying the Divisions of Tantra. Rgyud sde
spyi’i rnam par bzhag pa rgyas par bshad pa. Collected Works [zhol], Vol.8, pp.443-630.
[See Lessing & Wayman 1968].
[Thunderbolt Wheel of Reply to Ngor] Phyin ci log gig tam gyi sbyor ba la
‘jug pa’i smra ba ngan pa rnam par ‘thag pa bstan bcos gnam lcags ‘khor lo. TBRC
W384. Vol. 2, pp7-100.
Secret Treasury of the Vajra-Dåkinîs Explanation of the Two Part (Hevajra
Tantra). Dpal brtag pa gnyis pa’i rnam par bshad pa rdo rje mkha’ ‘gro ma rnams kyi
gsang ba’i mdzod. In gsung 'bum/_mkhas grub rje/ ?zhol par ma/ ldi lir bskyar par brgyab
pa/?. TBRC W384. 7: 471 - 964.
----------- [Thunderbolt Wheel of Reply to Ngor] Ngor lan gnam lcags 'khor lo.
Dgag lan phyogs bsgrigs . Ch'eng tu: Si khron Mi rigs Dpe skrun khan, 1997. Chinese
colophon title: Pien lun wen hsuan pien. pp. 1-68.
! 286!
Biographies of Mkhas grub: [see Cabezón 1992 and Ary 2007 & 2015 for
translations and interpretations]:
Rje brtsun chos kyi rgyal mtshan. Secret Biography. Gsang ba’i rnam mthar (SNT)
Collected works, vol. a pp.421-493.
Gnas rnying ‘Jam dbyangs kun dga’ dge legs rin chen rgyal mtshan. Rnam mthar mkhas
pa’i yid ‘phrog (KYP), Collected works of mkhas grub rje Vol.ka, pp.1-22.
Chos ldan rab ‘byor Short Biography of Mkhas grub. Mkhas grub thams cad mkhyen
pa’i rnam thar bsdus pa. [see translation in Ary 2016 Appendix A]
Nag-dbang-chos-grags. 2009. Gsung ngag rin po che lam bras bu dang bcas pa i khrid
rim snying po i legs bshad bzhugs so. [Lhas-sa]: Bod-ljongs mi dmangs dpe skrun khang.
(Ngor chen) Kun dga’ bzang po (1382-1456). gsung 'bum/_Kun dga’ bzang po. Ngor
chen kun dga' bzang po'i bka' 'bum. Compiled by Bsod nams rgya mtsho and reproduced
from the Sde dge block prints. Dehra dun: photomechanical print from a set of prints
from the Sde dge dgon chen blocks. [W11577]
Commentary on the Ghantapa Body Mandala Practice. Dril bu pa'i lus dkyil gyi
bshad pa. TBRC W11577. 4: 735 - 766. [dehra dun]: [sakya centre]Vol.4 .[see also 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, pp. 398.1.1-405.4.1 (cited in Bentor 2015)].
Overcoming Objections to the Three Tantras. Rgyud gsum gnod ‘joms. Vol.1,
ff. 310r-312v (pp.625-630).
Commentary on Overcoming Objections to the Three Tantras. Rgyud gsum
gnod ‘joms kyi ‘grel pa. Vol.1, ff. 312v-327r (pp.630-659).
Translation of the Two-part (Hevajra) Root tantra. Rtsa rgyud brtag gnyis kyi
‘gyur. Collected Works Vol.2 pp.445-6
Extraordinary Ocean of Biographies of Lineage lamas and the Manner of
Arising of the Hevajra Tantra. Kye rdo rje’i byung tshul dang brgyad pa’i bla ma’i rnam
thar ngor mtshar rgya mtsho] Vol.2: 425 - 451.
Catalogue of Hevajra Commentaries. Kye rdo rje'i 'grel pa'i dkar chag. Vol. 2:
452 - 453.
-----------[Destroyer of the Proponents of Evil through Eliminating
Objections to the Body Mandala] . Kye'i rdo rje'i lus kyi dkyil 'khor la rtsod spong
smra ba ngan 'joms. Vol.1, 545-580.
-----------[Dispelling Evil View s]. Kye rdo rje'i lus kyi dkyil 'khor la rtsod spong lta
ba ngan sel. Vol.1, pp. 580-625.
-----------[Moonrays of the Pith] Kye rdo rje'i sgrub thabs kyi rgya cher bshad pa bskyed
rim gnad kyi zla zer. Vol. 2, ff.3r-211r (pp.5-418)
-----------Rdo rje phreng ba'i thig rnams phyogs gcig bsdus pa. (On drawing the
Vajråvalî-mañ∂alas) Vol. 3, pp. 851-861.
-----------Biography of Sa bzang 'phags pa gzhon nu blo gros [Sa bzang 'phags pa gzhon
nu blo gros kyi rnam thar]. Vol.1, pp.169-180.
-----------Thob yi rgya mtsho (Record of Received Teachings). Vol.1, pp.179-434.
! 287!
The Ocean of Attainment of the Sådhana of the Guhyasamåja Mandala. Gsang
‘dus dkyil ‘khor gyi sgrub thabs dngos grub rgya mtsho] TBRC W11577. 3: 369 - 410.
The Ocean of offering clouds: Rite of Offering to the Divine assembly of the
Guhyasamåja-mañ∂ala. Gsang ‘dus dkyil ‘khor gyi lha tshogs rnam mchod pa’i cho ga
mchod sprin rgya mtsho. TBRC W11577. 3: 420 - 453.
Eliminating Impurities in the Sådhana of completing the Saravavid-Mañ∂ala,

Kun rig kyi dkyil ‘khor yongs rdzogs kyi sgrub thabs sgrib pa rnam sel. TBRC W11577.
4: 7 - 39.
The Cluster of Siddhis: The sådhana of the Vajradhåtu-mañ∂ala. Rdo rje dbyings
kyi dkyil 'khor sgrub thabs dngos grub snye ma. In gsung 'bum/_kun dga' bzang po.
TBRC W11577. Vol. 2: 705 - 744.
The Cluster of the Two Accumulatios: The Rite of Offering for the Vajradhåtu-
mañ∂ala. Rdo rje dbyings kyi dkyil 'khor mchod pa'i cho ga tshogs gnyis snye ma. In
gsung 'bum/_kun dga' bzang po. TBRC W11577. Vol. 2: 754 - 777.
The Cluster of Empowermen: Rites of the Vajradhåtu-mañ∂ala. Rdo rje dbyings
kyi dkyil 'khor gyi cho ga dbang gi snye ma. In gsung 'bum/_kun dga' bzang po. TBRC
W11577. Vol.2: 777 - 866.
The Cluster of Activity: Homa Rites of the Vajradhåtu-mañ∂ala. Rdo rje dbyings
kyi sbyin sreg gi cho ga phrin las snye ma. In gsung 'bum/_kun dga' bzang po. TBRC
W11577. Vol. 2: 866 - 884.
The Red Book. Lam 'bras po ti dmar ma. “Collection also known as the Puti
Marchung or “The Little Red Volume.” Compiled by Ngorchen Kunga Zangpo. Index
was written by his nephew, Gyaltsab Kunga Wangchuk, who was the fourth abbot of
Ngor Monastery. Texts by Ngorchen and Muchen Konchok Gyaltsen were added at the
end of the original grouping of texts. Title page attribution of the collection to Marton
Chogyal is probably incorrect.” See tbrc.org W30149.
Lamp of Eloquent Explanation for Classifying the Krîyå Tantras. Spyod rgyud
spyi'i rnam gzhag legs bshad sgron me. In gsung 'bum/_kun dga' bzang po. TBRC
W11577. 4: 135 - 204.
Ocean of Eloquent Explanation of the Caryå Tantras. Bya rgyud spyi'i rnam
bshad legs bshad rgya mtsho.] In gsung 'bum/_kun dga' bzang po. TBRC W11577. 4: 204
- 418.
-Rdor phreng la gsungs pa’i sa yongs su gzung ba’i cho ga. Written at Sa skya
with Dge slong Bsod nams rin chen as scribe on the basis of the teachings of Sa bzang
gzhon nu blo gros. Ngor chen kun dga' bzang po'i bka' 'bum, compiled by Bsod nams
rgya mtsho. pp. 151b-159a4, v. 2 (ga). Tokyo : The Toyo Bunko, 1968.2 v. (Sa skya pa'i
bka' 'bum; v. 9-10).
Ngor chen’s Correct System, Ngor chen’s teachings of the Divisions of tantra as
a preliminary of the Hevajra teachings noted down by Nyag-re dPal gyi rgyal mtshan,
! 288!
edited by A med zhabs; collected works of A mes zhabs, vol.pa, no.7. [as cited in Sobisch
2008]
NOTES= (“Notes for the Correct Explication of How to Enter into the Writings
of the Venerable Sa skya pas: Opening Wide in a Hundred Directions the Dharma Gates
to which All Beings of Tibet are Guided”) Rje btsun sa skya pa’i gsung rab la ‘jug tshul
legs par bshad pa’i yi ge bod yul ‘gro kun bsgrod pa’i chos sgo phyogs brgyar ring du
phye ba. Edited by A-med-zhabs and originally composed by Chos dpal bzang po;
collected works, vol. kha, fols. 384r-393v. [as cited in Sobisch 2008] [See also alternate
version of A-med-zhabs’s collected works cited in Verrill 2012]
Sangs rgyas phun tshogs ( b. 1649 d. 1705 ), based on Dkon mchog lhun grub ( b. 1497 d.
1557).
History “Dharma origins” of Ngor. Ngor pa'i chos 'byung. Printed from blocks
carved at Sde dge par khang chen mo; copy made available from the library of Tai si tu
rin po che at Shes rab gling, Kangra, Himachal Pradesh, India; Sde dge: Sde dge par
khang chen mo.
----------The Source of the Wish Fulfilling Jewel, the Oceanic Qualities which Gather the
Rivers: the Biography (“Liberation Story”) of the Victorious Vajradhara Kun dga’ bzang
po. Rgyal ba rdo rje 'chang kun dga' bzang po'i rnam par thar pa legs bshad chu bo 'dus
pa'i rgya mtsho yon tan yid bzhin nor bu'i 'byung gnas. Scanned from the Sde dge blocks
carved at the time of Sa skyong tshe dbang rdo rje rig 'dzin, dkar mdzes khul: Sde dge.
[written 1688].
Additional Biographies of Ngor chen for further research:

Chos dpal bzang po ( b. 15th cent. ) Biography of Kun dga' bzang po'i . Kun dga' bzang
po'i rnam thar.
Dge legs phun tshogs (b.1981-). Essential Biography. Rnam thar gces btus. Sde dge:
Rab brtan dge legs phun tshogs, 2006.
Dkon mchog rgyal mtshan ( b. 1388 d. 1469 ). Biography of the Venerable Pure Lama
Kun dga' bzang po. Rje btsun bla ma dam pa kun dga' bzang po'i rnam par thar pa
[written 1457]. W2CZ7931. Dbu med manuscript.
Mus chen sems dpa’ chenpo dkon mchog rgyal mtshan (1388-1469) Biography of the
Dharma Lord Vajradhara. Chos rje rdo rje ‘chang gyi rnam thar. Sa-Skya-Pa Lam 'Bras
Bla Brgyud Kyi Rnam Thar: Selected Biographies of the Sa-skya-pa Lam 'Bras Masters
(Reproduction of the Golden Manuscript of the Sa-skya-Bka'-'bum). 1985. Dehra Dun:
Sakya Centre.
Nam mkha' blo gsal ( b. 15th cent.). Biography of Kun dga' bzang po. Kun dga' bzang
po'i rnam thar.
Tshe dbang rdo rje rig 'dzin (1786-?) & Sangs rgyas phun tshogs (1649-1705). The
Ocean of Eloquent Explanations: Clarifying the Inner and Outer Generation Stage of
Hevajra and the Biography of Ngor chen kun dga' bzang po. Kye rdor phyi nang bskyed
rim nyams len gsal byed dang ngor chen kun dga' bzang po'i rnam thar legs bshad rgya
mtsho. W18115.
! 289!
Sa chen kun dga’ snying po. Sras don ma. Lam ‘bras literature Series 12, foilios 1r-222v
(pp.1-446); NGMPP L 170/4, Lam ‘bras gzhung sras don ma (xylograph from
Tyangpoche), 22 fols. [As cited in Sobisch 2008, Title list # 283.] (See also contemporary
two volume series)
(Tsong kha pa) Blo bzang grags pa (1357-81). The Jewel Treasury: The Rite of
Empowerment of the Body Mandala of Ghanatapa, the Lord of Yoga. Rnal ‘byor dbang
phyug dril bu lugs bde mchog lus dkyil gyi dbang chog rin po che’i bang mdzod, Vol. 10
pps. 57-106 [Toh 5327]
The Clear Meaning of the Pith of the Guhyasamåja:An Exegesis of the Rnam
gzhag rim pa. Rnam gzhag rim pa’i rnam bshad dpal gsang ba ‘dus pa’i gnad kyi don
gsal ba. In gsung ‘bum/ Tsong kha pa blo bzang grag pa (Reproduced from Lhasa zhol
par khang blocks). New Delhi: Lama Guru Deva, 1978-9. Vol.6. pps. 5-166.
The Stage of Pure Yoga: the Guhyasamåja Sådhana. Dpal gsang ba ‘dus pa’i
sgrub thabs rnal ‘byor dag pa’i rim pa. In gsung 'bum/_tsong kha pa/?bla brang par ma/?.
TBRC W22273. 7: 623 - 678. [bla brang]: bla brang bkra shis 'khyil [T5303].
The Wish-Granting Extensive Explanation of the Cakrasaµvara Abhisamaya.
Bde mchog mngon rtogs rgya cher bshad pa 'dod pa 'jo ba/." In gsung 'bum/_tsong kha
pa/?bla brang par ma/?. TBRC W22273. 9: 195 - 592. [bla brang]: bla brang bkra shis
'khyil, [199?] [cited in Bentor 2015].
IOL Tib J 576/1. British Library: IOL Tib J 576

Site: Dunhuang Mogao (Ch.73.XIV.5 [25])
Language(s)/Script(s): Tibetan (lang.), Tibetan (script)
Materials: manuscript, ink on paper
Tibetan Kanjur:
Abhidhånottara Tantra. Mngon par brjod pa’p rgud bla ma. sde dge 369; snar thang 376;
stog 335; urga 372.
The Guhyasamåja Tantra. Sarvatathågata-kåyavåkcitta-rahasya-guhyasamåja-nåma-

mahåkalparåja; 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. Toh. 442, rgyud, vol.Ca, 90a-148a.
Hevajra Tantra (Hejavra-tantraråja-nåma’ Kye’i rdo rje zhes bya ba rgyud kyi rgyal po).
Toh. 417, rgyud, vol.Nga, 1b-13b.
Sampu†a Tantra. [Sampu†a-nåma-mahåtantra]. Chos kyi 'byung gnas. Yang dag par
sbyor ba'i rgyud chen po. In bka' 'gyur (sde dge par phud). TBRC W22084. 79: 148 - 318.
delhi: delhi karmapae chodhey gyalwae sungrab partun khang, 1976-1979. Toh 381 sde
dge
! 290!
Samvarodaya Tantra. Dpal bde mchog ‘byung ba zhes bya ba’i rgyud kyi rgyal po chen
po (Srimahasambhorodaya-tantraraja nama); 51 le’u & 24 rab byed; sde dge par phud
bka' 'gyur vol.78, ff,137r-264v (pp.273-528)(searchable 265a.1-311a.6), toh 373; snar
thang 379
[STTS]Sarvatathågata-tattvasamgraha-nåma-mahåyåna-sütra; De bzhin gshegs pa thams

cad kyi de kho na nyid bsdus pa zhes bya ba theg pa chen po’I mdo. Toh. 479, rgyud,
vol. Nya, (1b-142a7) 1-283.
The Vajramålå (Explanatory Tantra). ¸rî-vajramålå-abhidåna-mahåyogatantra-

sarvatantra-hdaya-rahasya-vibhanga-nåma: Rdo rje’i ‘phreng ba’i rgyud. Rnal ‘byor chen
po’i rgyud dpal rdo rje phreng ba mngon par brjod pa rgyod thams cad kyi snying po
gsang ba rnam par phye ba zhes bya ba. Toh 445, rgyud, vol. Ca, (208a-277b) 415-554.
Uttaratantra. Rgyud phyi ma. Toh. 443, rgyud, vol. Ca, 148a-157b.
Vajra h®daya alaµkåra tantra. 'Dus pa'i phyogs mthun kyi rgyud rdo rje snying po rgyan
gyi rgyud . bka' 'gyur : sde dge phar phud vol. 82 ff.36r-58v (pp.76-116); Toh 451; sto
g 413; snar thang 410; sde dge 449.
Vajrasattva-sådhana-nibandha. Bstan 'gyur gser bris ma Vol. 35 ff. 231r-237r (pp.462-

73)
Tibetan Tanjur [ bstan ‘gyur ]:
Abhayåkaragupta. Nißpannayogåvalî. Rdzogs pa'i rnal 'byor gyi phreng ba. (Toh 312 1)
sde dge bstan ‘gyur Vol.75 ff. 94v-151r (pp.188-301) 189.4-.6.
Dpal yang dag par sbor ba'i rgyud kyi rgyal po'i rgya cher 'grel pa man ngag gi
snye ma. Åmnåyama•jarî. Toh 1198 cha 1v-316r. In bstan 'gyur (snar thang). TBRC
W22704. 21: 5 - 708. [narthang]: [s.n.], [1800?].
Alaµkakala!a (Tshul khrims rin chen). Rnal 'byor chen po'i rgyud dpal rdo rje phreng
ba'i rgya cher 'grel pa zab mo'i don gyi 'grel pa. In bstan 'gyur (sde dge). TBRC W23703.
34: 4 - 442. delhi: delhi karmapae choedhey, gyalwae sungrab partun khang, 1982-1985.
Rnal 'byor chen po'i rgyud dpal rdo rje phreng b'i rgya cher 'grel pa zab mo'i don
gyi 'grel pa. In bstan 'gyur/?gser bris ma/. TBRC W23702. 34: 3 - 638. tibet: [snar thang],
[17-?]
Candrakîrti (zla ba grags pa). Pradîpoddyotana-nåma-tîkå. Sgron ma gsal bar byed pa

zhes bya ba’i rgya cher bshad pa sde dge bstan ‘gyur edition. Toh. 1785. Vol.29 (ha) p.2-
402 (ff.1v-201v).
! 291!
Darikapa. Dpal 'khor lo sdom pa'i sgrub thabs de kho na nyid bsdud pa (Sri
Cakrasaµvara-sådhana tattvasaµgraha). gser bris bstan 'gyur 147 pp531-428; snar thang
bstan 'gyur ff.207v-214r (pp414-427); Otani Beijing 2145.
Jetari (Slob dpon dgra las rnam rgyal). Caturmudrå-sådhana. Phyag rgya bzhi yi sgrub
thabs gser bris bstan 'gyur 2695 Vol 75 pp.387-393; snar thang bstan 'gyur vol.76,
ff.142v-144v (pp284-288); Otani Beijing edition: 4690.
Ghantapa (Rdo rje dril bu pa). Dpal ‘khor lo sdom pa’i dbang gi bya ba mdor bsdus pa.
Toh. 1431, D vol.21 438-444 (ie.219v-222v).
Någårjuna. Piñ∂î-k®ta. Mdor byas . sde dge bstan ‘gyur Vol.34 ff.1v-11r Toh 1796; see
Wright 2010 translation
Nag po pa. Saµvara-vyåkhyå. Sdom pa bshad pa. In bstan 'gyur (sde dge). TBRC
W23703. 22: 13 - 22. Delhi: delhi karmapae choedhey, gyalwae sungrab partun khang,
1982-1985.Toh:1460; bstan 'gyur gser bris 179 vol 14 ff. 8r-14v (pp.15-28); Otani
Beijing 2177.
?Nåropå. Dpal 'khor lo sdom pa'i sgrub thabs yid bzhin nor bu, translated by Marpa chos
kyi blo gros) in mkha' 'gro snyan brgyud kyi yig rnying pp. 299-320.
Piñ∂îk®ta-sådhanopåyikå-v®tti-ratnåvalî. Mdor bsdus thabs gyi ‘grel pa rin chen ‘phreng

ba. Ratnåkara!ånti. PTT, Vol. 62, snar thang bstan 'gyur Vol.35 ff.271r-375v (pp.541-
750); gser bris 698; Otani Beijing 2690
Samåja sådhana vyavasthole [sthåli ]. Rnam ghzag rim pa. Toh 1809. Sde sde bstan ‘gyur
Vol. 35 ff. 121a-131a ] (see partial edition by Kimiaki Tanaka)
¸rî guhyasamåja-mahåyogatantropattikrama-sådhana-sütra- melåpaka-nåma; Rnal 'byor

chen po'i rgyud dpal gsang ba 'dus pa'i bskyed p’i rim pa'i bsgom pa’i thabs mdo dang
bsres pa zhes bya ba; Translators: Dharma!ribhadra and Rin chen bzang po; P2662, Vol.
60-61; T1797.
Sanskrit Language Sources:
Abhayåkaragupta, Gudrun Bühnemann, and Musashi Tachikawa. 1991.

Nißpaññayogåvalî: two Sanskrit manuscripts from Nepal. Tokyo, Japan: Centre for East
Asian Cultural Studies.
Abhayåkaragupta, and Benoytosh Bhattacharyya. 1949. Nißpaññayogävalî of

Mahäpandita Abhayäkaragupta. Baroda: Oriental Institute.
Aiteryopanißad
B®ha∂aranyaka-Upanißad
! 292!
Editions and Translations:
Bsod-nams-grags-pa, and Giuseppe Tucci. 1971. Deb ther dmar po gsar ma Tibetan
chronicles. Roma: Istituto Italiano per il Medio ed Estremo Oriente. [as cited in Cabezón
2007].
Buddhaghosa, and Bhikkhu Nåñamoli. 2010. The Path of Purification: (Visuddhimagga).

Kandy: Buddhist Publication Society.
Buddhaguhya, and Stephen Hodge. 2003. The Mahå-Vairocana-Abhisambodhi Tantra.

New York: RoutledgeCurzon.
Candrakîrti, Hong Luo, and Toru Tomabechi. 2009. Candrakirti's

Vajrasattvanispadanasutra = Vajrasattvasadhana : Sanskrit and Tibetan texts. Beijing:
China Tibetology Publishing House.
Candrakîrti, Ngawang Samten and S.S. Bahulkar. 2010.

Guhyasamåjatantrapradîpodyotana†ikå Ía†ko†ivyåkhyå of Åcårya Candrakîrti. Dhîh:
Journal of Rare Buddhist Texts Research Department. Sarnath, Varanasi: Central
University of Tibetan Studies.
Dkon-mchog-lhun-grub, Lobsang Dagpa, and Jay Goldberg. 2002. The three visions:
fundamental teachings of the Sakya Lineage of Tibetan Buddhism. Ithaca, N.Y.: Snow
Lion Publications.
Eimer, Helmut, and Nor-chen Kun-dga -bzang-po. 1999. The early Mustang Kanjur
catalogue: a structured edition of the Mdo snags bka gyur dkar chag and of Ngor chen
Kun dga bzang po's bKa gyur ro cog gi dkar chag bstan pa gsal ba i sgron me. Wien:
Arbeitskreis für Tibetische und Buddhistische Studien, Universität Wien.
Elder, George Robert. 1978. The Samputa tantra: edition and translation, chapters I-IV.
PhD dissertation, Columbia University.
Fremantle, Francesca. 1971. “A critical study of the Guhyasamaja Tantra.” PhD

dissertation, University of London.
Giebel, Rolf W. 2001. Two esoteric sutras. Berkeley, Calif: Numata Center for Buddhist
Translation and Research.
Gos Lo-tsa-ba Gzon-nu-dpal, and George Roerich. 1976. The blue annals. Delhi: Motilal
Banarsidass.
! 293!
Gray, David B., and Thomas F. Yarnall. 2007. The Cakrasamvara Tantra: the discourse
of Sri Heruka (Sri heruka abhidhana). New York: American Institute of Buddhist Studies
at Columbia University.
Jackson, Roger R., and Rgyal-tshab Dar-ma-rin-chen. 1993. Is enlightenment possible?:

Dharmakîrti and rGyal tshab rje on knowledge, rebirth, no-self and liberation. Ithaca,
N.Y.: Snow Lion Publications
Kalff, Martin Michael. 1979. Selected chapters from the Abhidanottaratantra: the union
of female and male deities. [S.l.]: [s.n.].
Kittay, David. 2011. Interpreting the Vajra Rosary: Truth and Method Meets Wisdom
and Method. PhD dissertation, Columbia University.
2013. The vajra rosary tantra (Vajramalatantra). New York: American Institute of
Buddhist Studies.
Kong-sprul Blo-gros-mtha -yas, Elio Guarisco, and Ingrid McLeod. 2005. The treasury of
knowledge: book six, part four: Systems of Buddhist Tantra. Ithaca, N.Y.: Snow Lion
Publications.
Lee, Yong-Hyun. 2004.The Nißpaññayogåvalî by Abhayåkaragupta: A New Critical

Edition of the Sanskrit Text (Revised Edition). Seoul: Baegun Press, 2004.
Lessing, Ferdinand and Alex Wayman. Trans. of Mkhas grup rje’s Fundamentals of
Buddhist Tantra. Indo-Iranian Monographs v.8. The Hague, Mouton, 1968.
Losang Tsephel, Någårjuna, David Ross Komito, Andrew Fagan, and Dbyan s-can-
dga -ba i-blo-gros. 1995. Paths and Grounds of Guhyasamåja according to Arya
Nagarjuna. Dharamsala: Library of Tibetan Works and Archives.
Matsunaga, Yukei. 1978. The Guhyasamåja Tantra: a new critical edition. Osaka: Toho
Shuppan.
Mkhas-grub Dge-legs-dpal-bzang-po, and José Ignacio Cabezón. 1992. A dose of

emptiness: an annotated translation of the sTong thun chen mo of mKhas-grub dGe-legs-
dpal-bzang. Albany: State University of New York Press.
Mori, Masahide. 1997. “The Vajråvalî of Abhayåkåragupta: a critical study, Sanskrit

edition of select chapters and complete Tibetan version.” PhD dissertation. University of
London.
Någårjuna, and David J. Kalupahana. 1996. Mülamadhyamakakårikå of Någårjuna = The

philosophy of the middle way : introduction, Sanskrit text, English translation and
annotation. Delhi (India): Motilal Banarsidass Publ.
! 294!
Pandey, Janardan Shastri. 1998. Yoginîsañcåratantram. Åcåryatathågatarakßitaprañîtena
sañcåranibandhena. Pañ∂itåcårya-Alakakala!aprañîtayå ca upade!ånusåriñîvyåkhyayå
sahitam. Sarnath, Varanasi: Durlaba Bauddha Grantha ¸odha Yojanå, Kendrîya Ucca
Tibbatî ¸ikßå Samsthåna.
Sa-skya Pandi-ta Kun-dga -rgyal-mtshan, Jared Rhoton, and Victoria R. M. Scott. 2002.
A clear differentiation of the three codes: essential distinctions among the individual
liberation, great Vehicle, and Tantric Systems : the Sdom gsum rab dbye and six letters.
Albany: State University of New York Press.
Shendge, Malati J. Ía†såhasrikå-Hevajra-Tîka: A Critical Edition. Delhi: Paratibha

Prakashan.
Snellgrove, David L. 1959 (2010 Reprint). Hevajra Tantra: a critical study. Hong Kong:
Orchid.
Skorupski, Tadeusz. 1983. The Sarvadurgatipari!odhana tantra: elimination of all evil

destinies : Sanskrit and Tibetan texts. Delhi: Motilal Banarsidass. [SDPT]
Skorupski, Tadeusz, and Ulrich Pagel. 1990. (Samputå, partial edition). The Buddhist
forum. New Delhi: Heritage Publishers
Tanaka, Kimiaki. 2001-2002. Någabodhi’s ¸rî-guhyasamåja-mañ∂alopåikå-vim!atî-

vidhi , 8 askasutapatanavidhi. Journal of Indian and Buddhist Studies, Vol.50 No.1,
p.320-315.
2004. “Någabodhi’s ¸rî-guhyasamåja-mañ∂alopåikå-vim!atî-vidhi” The Tibetan
translation and Sanskrit text of Chapters 5 and 6. In Three mountains and seven rivers:
Prof. Musashi Tachikawa's felicitation volume. Delhi: Motilal Banarsidass Publishers.
2009. “Någabodhi’s ¸rî-guhyasamåja-mañ∂alopåikå-vim!atî-vidhi: the Sanskrit
text restored from the vajråcårånayotttam.” In Genesis and Development of Tantrism,
edited by Shingo Einoo. Tokyo: Institute of Oriental Culture, University of Tokyo,
Institute of Oriental Culture Special Series, 23, pp.425-434.
Tsemo, Sonam. 2013. The Yogini's Eye: Comprehensive Introduction to Buddhist Tantra.
[Place of publication not identified]: Xlibris.
Tsong-kha-pa Blo-bzang-grags-pa, and Gavin Kilty. 2013. A lamp to illuminate the five
stages teachings on Guhyasamåja Tantra. Boston : Wisdom Publications in association
with the Institute of Tibetan Classics.
Tsong-kha-pa Blo-bzang-grags-pa, and Glenn H. Mullin. 1996. Tsongkhapa's Six Yogas

of Naropa. Ithaca, N.Y.: Snow Lion Publications.
Tsuda, Shin ichi. 1974. The Samvarodaya-tantra: selected chapters. Tokyo: Hokuseido
Press.
! 295!
Tsong-kha-pa Blo-bzang-grags-pa, Robert A. F. Thurman, and Thomas F. Yarnall. 2010.
Brilliant Illumination of the Lamp of the Five Stages (Rim lnga rab tu gsal ba'i sgron me)
: Practical Instruction in the King of Tantras, The Glorious Esoteric Community. New
York: The American Institute of Buddhist Studies.
Wedemeyer, Christian K. 2007. Aryadeva's Lamp that integrates the practices

(Caryamelapakapradipa): the gradual path of Vajrayana Buddhism according to the
esoteric community noble tradition. New York: American Institute of Buddhist Studies at
Columbia University.
Wright, Roger. “The Guhyasamåja Piñ∂ik®ta-sådhana and its context.” Masters thesis,
School of Oriental and African Studies, London, 2010.
English Language Sources
Akira, Hirakawa, and Paul Groner. 1993. A History of Indian Buddhism: from
¸åkyamuni to early Mahåyana. Delhi: Motilal.
Aris, Michael. Hidden Treasures and Secret Lives: a study of Pemalingpa (1450-1521)
and the Sixth Dalai Lama (1683-1706). London; New York: Kegan Paul International,
1989.
Ary, Elijah Sacvan. “Logic and lineage: Jetsun Chokyi Gyaltsen’s Ascension and the
Secret Biography of Khedrup Geleg Pelzang.” Ph.D. Diss., Harvard University, 2007.
2015. Authorized lives: biography and the early formation of Geluk identity
Boston : Wisdom Publications, 2015.
Dina Bangdel, “Art in the Ritual Context: The Chakrasamvara Tradition in Newar
Buddhism,” Orientations No.8 Vol. 34 (2003).!
Jean Baudrillard, “The Precession of Simalcra,” in Simalcra and Simulation, translated by

Sheila Faria Glaser cited inMargaret Miles, “Image,” in Critical Terms for Religious
Studies, Ed., Mark Taylor(Chicago: University of Chicago Press, 1998) pp.60-72, p.66.
Bell, Catherine. 1992. Ritual Theory, Ritual Practice. Oxford University Press.
Benard, Elisabeth Anne. Chinnamastå: The Aweful Buddhist and Hindu Tantric
Goddesses. Delhi: Motilal Banarsidas, 1994
Bénisti, Mireille. 1960. Étude sur le stüpa dans l'Inde ancienne. Paris: École française
d'Extrême-Orient.
Bentor, Yael. 1996.Consecration of Images and Stüpas in Indo-Tibetan Tantric

Buddhism. Leiden: E.J. Brill.
1995. “On the Indian Origins of the Tibetan Practice of Depositing Relics and
! 296!
Dhårañîs in Stüpas and Images”. Journal of the American Oriental Society 115 (2).
American Oriental Society: 248-61.
----------------2000. “Interiorized Fire Rituals in India and Tibet,” Journal of the American
Oriental Society. Vol. 120, No. 4 (Oct.-Dec., 2000) 594-613.
2006. “Identifying the Unnamed Opponents of Tsong kha pa and Mkhas grub rje
Concerning the Transformation of Ordinary Birth, Death and the Intermediary State into
the Three Bodies.” Tibetan Buddhist Literature and Praxis: Studies in its Formative
Period, 900-1400. Ed. Ronald M. Davidson and Christian Wedemeyer. Leiden: Brill,
2006. 185-200.
2014. “The Body in Buddhist Tantric Meditations” conference paper presented at
“The Evolution of Tantric Ritual” held at UC Berkeley in March 2014.
2015. "Interpreting the Body Mandala: Tsongkhapa versus Later Geluk Scholars"
Revue d’Etudes Tibétaines. no.31, Feb 2015 pp 63-74
Berger, Patricia Ann. 2001. “Miracles in Nanjing: An Imperial Record of the Fifth
Karmapa’s Visit to the Chinese Capital,” in Marsha Weidner, ed., Cultural Intersections
in Later Chinese Buddhism, University of Hawai’i Press, Honolulu, 145-169.
2003. Empire of emptiness: Buddhist art and political authority in Qing China.
Stephen Beyer, The Cult of Tåra: Magic and Ritual in Tibet (Berkeley and Los Angeles:
University of California Press, 1978).
Bourdieu, Pierre. 1977. Outline of a Theory of Practice, Trans. Richard Nice (Cambridge:
Cambridge University Press, 1977).
1990. The logic of practice. Stanford, Calif: Stanford University Press.
1998. Practical reason: on the theory of action. Stanford, Calif: Stanford
University Press.
Brauen, Martin. The Mandala: Sacred Circle in Tibetan Buddhism. London: Serindia
Publications, 1997.
Broido, Michael M. 1988. “Killing, Lying, Stealing and Adultery: A Problem of

Interpretation in the Tantras,” in Buddhist Hermeneutics. Ed. Lopez, Donald S.
Bsod-nams-rgya-mtsho, and Musashi Tachikawa. 1991. The Ngor mandalas of Tibet:

listings of the Mandala deities. Tokyo: Centre for East Asian Cultural Studies.
Bstan- dzin-rgya-mtsho, The Fourteenth Dalai Lama. “Nature of the Mind”, Public
Teaching held at UC Santa Barbara, April 24. 2009 [video recording accessed via
YouTube].
Buhnemann 1992/3: “Some Remarks on the Text of the Nispannayogavali as Found in

Jagaddarpana's Kriyasamuccaya.: Zentralasiatische Studien 23. 1992/1993. 18-21
! 297!
Butler, Judith (1989). “Foucault and the paradox of bodily inscriptions.” Journal of
Philosophy 86 (11): 601-607.
Bynum, Caroline Walker. “Why All the Fuss About the Body? A Medievalist’s
Perspective.” Critical Inquiry 22 (Autumn 1995): 1-33.
2011. Christian materiality: an essay on religion in late medieval Europe.
New York: Zone Books.
Cabezón, José Ignacio. 1994. Buddhism and language a study of Indo-Tibetan

scholasticism. Albany, N.Y.: State University of New York Press.
Cabezón, José Ignacio and Geshe Lobsang Dargyay. 2007. Freedom from Extremes:
Gorampa’s ‘Distinguishing the Views’ and the Polemics of Emptiness. Boston: Wisdom
Publications.
Chandra, Lokesh. 1980. Borobudur: a new interpretation. [Wiesbaden]: [Steiner].

Chandra, Lokesh. “Borobu∂ur as a Monument of Esoteric Buddhism,” South East Asia
Review, V (1980).
Cooter, Roger. “The Turn of the Body: History and the Politics of the Corporeal.”
ARBOR CLXXXVI 743 (may-june 2010): 393-205.
Copp, Paul F. 2014. The body incantatory: spells and the ritual imagination in medieval
Chinese Buddhism. New York : Columbia University Press, [2014].
Cort, John E. “Contemplating the Jain Universe: Visions of Order and Chaos,” in
Victorious Ones: Jain Images of Perfection, edited by Phyllis E. Granoff (New York:
Rubin Museum of Art, 2009-2010), 34-37.
Cuevas, Bryan J. 2003. The Hidden History of The Tibetan Book of the Dead. New
York: Oxford University Press.
Dachille, R.E. 2007, ‘Modes of Representation and Meanings in the Blue Beryl Paintings
of Desi Sangye Gyatso,’ Unpublished masters thesis, University of Wisconsin, Madison.
Dachille(-Hey). “The Case of the Disappearing Blue Woman: Understanding
How Meaning is Made in Desi Sangye Gyatso’s Blue Beryl Paintings.” Asian Medicine:
Tradition and Modernity, Volume 7, Issue 1 ‘Gender, Health and Medicine in Tibet’: Dr
Heidi Fjeld and Theresia Hofer (eds), Autumn 2011.
“Iconoclasm and the Subtle Body: a Study of Art, Ritual and the Body in a Single
Nineteenth- Century Nepalese Painting.” Reading Outside the Lines: A Workshop on the
Universitœt, Munich (Contemporary Issues in Buddhist Studies) (Publication forthcoming
2015).
Dagyab, Loden Sherab. 1977. Tibetan religious art. Wiesbaden: Harrassowitz.
! 298!
Dalton, Jacob Paul, and Sam Van Schaik. 2006. Tibetan Tantric manuscripts from
Dunhuang: a descriptive catalogue of the Stein Collection at the British Library. Leiden:
Brill.
Dalton, Jacob Paul. 2011. The Taming of the demons: violence and liberation in Tibetan
Buddhism. New Haven: Yale University Press.
Davidson, R.M. “The Ngor-pa Tradition.” Wind Horse. Vol.1, 1981, pp.79-98.
1984. The Mollas of Mustang: historical, religious, and oratorical traditions of
the Nepalese-Tibetan borderland. Dharamsala: Library of Tibetan Works & Archives.
-------------- “Reﬂections on the Mahe!vara Subjugation Myth: Indic Materials, Sa-skya-
pa Apologetics, and the Birth of Heruka,” Journal of the International Association of
Buddhist Studies 14/2 (1991): 197-235.
--------------“Preliminary Studies on Hevajra's Abhisamaya and the Lam-'bras Tshogs-
bshad.” Tibetan Buddhism, Reason and Revelation, edited by Steven D. Goodman and
Ronald M. Davidson. Albany: State University of New York Press, 1992.
“Masquerading as Pramåña: Esoteric Buddhism and Epistemological
Nomenclature.” In International Dharmaki rti conference, and Shoryu Katsura. 1999.
Dharmaki rti's thought and its impact on Indian and Tibetan philosophy: proceedings of
the third International Dhamakîrti conference, Hiroshima, November 4-6, 1997. Wien:
Verlag der Österreichischen Akademie der Wissenschaften: pp.25-35.
--------------Tibetan Renaissance: Tantric Buddhism in the Rebirth of Tibetan Culture.

New York: Columbia University Press, 2005.
Davis, Richard H. 1991. Ritual in an oscillating universe: worshiping ¸iva in medieval

India. Princeton, N.J.: Princeton University Press.
Dehejia, Vidya. “On Modes of Visual Narration in Early Buddhist Art,” The Art Bulletin
Vol.72 No.3 (Sept. 1990): 374-392.
“Aniconism and the Multivalence of Emblems,” in Ars Orientalis (1992) 21,
pp,45-66.
“Rejoinder to Susan Huntington,” Ars Orientalis (1993), p.157.
Despeux, Catherine. “Visual Representations of the Body in Chinese Medical and Daost
Texts from the Song to the Qing period (Tenth to nineteenth Century).” Asian Medicine.
Vol.1, Issue 1: pp10-52.
Dhungel, Ramesh K. The Kingdom of Lo (Mustang): A Historical Study. Kathmandu:
Tashi Gephel Foundation, 2002.
Diamond, Debra, and Molly Emma Aitken. 2013. Yoga: the art of transformation.
Washington, D.C. : Arthur M. Sackler Gallery, Smithsonian Institution,
Dreyfus, Georges. 1997. Recognizing Reality: Dharmakîrti’s Pholosophy and Its Tibetan
Interpretations. Albany: SUNY Press, 1997.
! 299!
---------------------------The Sound of Two Hands Clapping: the Education of a Tibetan
Monk. Berkeley: University of California Press, 2003.
Eckel, Malcolm David. To See the Buddha (Princeton: Princeton University Press,
1994).
Ehrhard, Fraz-Karl. “Spiritual Relationships Between Rulers and Preceptors: The Three
Journeys of Vanaratna (1384-1468) to Tibet”. In The Relationship Between Religion and
State (chos srid zung 'brel) in Traditional Tibet. Proceedings of a Seminar held in
Lumbini, Nepal, in March 2000 (=LIRI Seminar Proceedings, 1). Lumbini 2004, S. 245-
265.
Einoo, Shingo. “Ritual Devices to Become a God in Vedic and Post-Vedic Rituals”
presented by at the conference on “The Evolution of Tantric Ritual” held at UC Berkeley
in March 2014.
English, Elizabeth. 2002. Vajrayoginî: Her Visualizations, Rituals, and Forms. Boston:
Wisdom Publications.
Flood, Gavin. 2000. “The Purification of the Body.” In Tantra in practice. Edited by
David Gordon White (Princeton, N.J.: Princeton University Press), pp.509-22.
2006. The tantric body: the secret tradition of Hindu religion. London: I.B. Tauris.
Furth, Charlotte. 1999. A flourishing Yi: gender in China's medical history, 960-1665.
Berkeley: University of California Press.
Garrett, Frances 2004, “Narratives of Embryology: Becoming Human in Tibetan

Literature.” PhD dissertation, University of Virginia.
Co-authored with Vincanne Adams.“The Three Channels in Tibetan Medical and
Religious Texts, including a translation of Tsultrim Gyaltsen’s ‘Treatise on the Three
Channels in Tibetan Medicine’.”Traditional South Asian Medicine8: 86-115.
---------------------- 2008, Religion, Medicine and the Human Embryo in Tibet, New York:
Routledge.
Gellner, David. Monk, Householder and Tantric Priest: Newar Buddhism and its
Hierarchy of Ritual (Cambridge: Cambridge University Press, 1992).
Granoff, P. E. 2009. Victorious ones: Jain images of perfection. New York: Rubin
Museum of Art.
Gluckman, Max. 1962. Essays on Ritual and Social Relations (Manchester: Manchester
University Press).
Goodman, Nelson. 1976. Languages of art: an approach to a theory of symbols.

Indianapolis: Hackett.
! 300!
Gray, Body. 2006. "Mandala of the Self: Embodiment, Practice, and Identity
Construction in the Cakrasamvara Tradition". Journal of Religious History. 30 (3): 294-
310.
Mandalas in the Yoginîtantras conference paper presented at “The Evolution of
Tantric Ritual” held at UC Berkeley in March 2014.
Gyatso, Janet. “Image as Presence: The Place of the Work of Art in Tibetan Religious
Thinking,” in The Newark Museum Tibetan Collection III. Sculpture and Painting ,
Valrae Reynolds, Amy Heller and Janet Gyatso (Newark: The Newark Museum, 1986),
pp.30-35
1992. “Letter Magic: A Peircean Perspective on the Semiotics of Rdo Grub-chen's
Dhårañî Memory.” In In the mirror of memory reflections on mindfulness and
remembrance in Indian and Tibetan Buddhism. Edited by Janet Gyatso. Albany: State
University of New York Press.
Gyatso, Janet 2004, ‘The Authority of Empiricism and the Empiricism of
Authority: Medicine and Buddhism in Tibet on the Eve of Modernity’, Comparative
Studies of South Asia, Africa and the Middle East, Volume 24, Number 2: 83-96.
Harding, Sarah, trans & ed. Machik’s Complete Explanation: Clarifying the Meaning of
Chod. Ithaca, NY: Snow Lion Publications, 2003.
Hartzell, James Francis. 1997. “Tantric yoga: a study of the Vedic precursors, historical
evolution, literatures, cultures, doctrines, and practices of the 11th century Ka!mîrî
¸aivite and Buddhist unexcelled Tantric yogas.” Ph. D. Columbia University, 1997.
Heimbel, Jorg. 2011. “Biographical Sources for Researching the Life of Ngor chen Kun
dga’ bzang po (1382-1456).” Revue d’Etudes Tibétaines 22, pp. 47-91.
Hekman, Susan. 1998. “Material Bodies,” in Body and Flesh: A Philosophical Reader,
edited by Donn Welton (Malden, MA & Oxford: Blackwell Publishers, Ltd.), 61-70.
Heller, Amy. “Two Early Tibetan Ritual Diagrams for Cakra Meditations,” The Tibet
Journal , Vol. XXXIV, No. 3 & 4/ Vol. XXXV, No.1 & 2 (Autumn and Winter
2009/Spring and Summer 2010): 59-70.
Holdrege, B. A.. (1998). Body Connections: Hindu Discourses of the Body and the Study
of Religion. International Journal of Hindu Studies, 2(3), 341-386.
Hopkins, Jeffrey, and Kevin Vose. 2009. Tantric Techniques. Ithaca, N.Y.: Snow Lion
Publications.
Howard, Angela Falco. 1981. “A New Chronology for the Kizil Mural Paintings,”
Archives of Asian Art, Vol. 44 : 68-83.
The Imagery of the Cosmological Buddha (Studies in South Asian Culture).
Leiden: Brill, 1986.
! 301!
Huber, Toni. The holy land reborn: pilgrimage & the Tibetan reinvention of Buddhist
India. Chicago: University of Chicago Press, 2008.
Huntington, John C. “The Iconography of Borobudur Revisited: The Concepts of Slesa

and Sarva[buddha]kaya.” In Klokke, Marijke J., and Pauline C. M. Lunsingh Scheurleer.
1994. Ancient Indonesian sculpture. Leiden: KITLV Press.
Huntington, John C. and Dina Bangdel, The Circle of Bliss: Buddhist Meditational Art
(Chicago: Serindia Publications, 2003)
Susan L. Huntington, “Early Buddhist Art and the Theory of Aniconism,” Art Journal
(Winter 1990) Vol. 49 No.4, pp.401-408.
Susan L. Huntington, “Aniconism and the Multivalence of Emblems: Another
Look,” Ars Orientalis (1992) 22, pp.111-156.
Jackson, David P. 1986 “A Painting of Sa skya pa Masers from an Old Ngor pa Series of
Lam ‘bras Thangkas,” in Berliner Indologische Studien, (2), pp.181-91.
1989. The Early Abbots of ‘Phan-po Ne-len-dra: The Vicissitudes of a Great
Tibetan Monastery in the 15th Century. Wien: Arbeitskreis für Tibetische und
Buddhistische Studien, Universität Wien.
1986 “The Identification of Individual Teachers in Paintings of Sa skya pa
Lineages,” in T. Skorupski, ed., Indo-Tibetan Studies, Spring, pp. 129-44.
A History of Tibetan Painting: The Great Tibetan Painters and Their
Traditions. Wien: Verlag der Osterreichischen Akademie der Wissenschaften,
1996. Chapter 2.
2003. “The Dating of Tibetan Paintings is Perfectly Possible – though not always
perfectly exact.” in Dating Tibetan Art: Essays on the Possibilities and Impossibilities of
Chronology from the Lempertz Symposium, Cologne, edited by Ingrid Kreide-Damani.
Wiesbaden: Ludwig Reichert Verlag. 91-103.
2005. “Lineages and Structure in Tibetan Buddhist Painting: Principles and
Practice of an Ancient Sacred Choreography. “ Journal of the International
Association of Tibetan Studies 1: 1-40.
2010.The Nepalese Legacy in Tibetan Painting. Seattle: University of
Washington Press.
C.E.M. Joad, Guide to Philosophy, I.ii.41 (1936) [OED online]
Kassor, Constance. “Thinking the Unthinkable / Unthinking the Thinkable: Conceptual

thought, nonconceptuality, and Gorampa Sonam Senge's Synopsis of Madhyamaka.”
Ph.D. Diss., Emory University, 2014.
Kapstein, Matthew. The Tibetans. Malden, MA; Oxford: Blackwell Publishers, 2006.
Khenpo Dr. Ngawang Jorden. “Buddha-nature: Through the eyes of Go rams pa Bsod
nams seng ge in fifteenth-century Tibet,” Ph.D dissertation, Harvard University, 2003.
! 302!
Kelsang Gyatso. 1997. The essence of Vajrayana: the highest yoga tantra practice of
Heruka body mandala. London: Tharpa Publications.
Kim, Bo-Young. 2007. “Indefinite boundaries: reconsidering the relationship between

Borobudur and Loro Jonggrong in Central Java. Ph.D Dissertation. University of
California, Los Angeles 2007
Kossak, Steven, Jane Casey Singer, and Robert Bruce-Gardner. 1998. Sacred visions:
early paintings from Central Tibet. New York: Metropolitan Museum of Art.
Kramer, Jowita. A Noble Abbot from Mustang. Life and Works of Glo-bo mKhan-chen
(1456- 1532) Wien: Arbeitskreis für tibetische und buddhistische Studien (Wiener
Studien zur Tibetologie und Buddhismuskunde 68), 2008.
Kvaerne, Per. 1975-6. “On the Concept of Sahaja in Indian Buddhist Tantric Literature,”
Temenos: Studies in Comparative Religion, vol.11-12: 88-135.
Lamotte, Etienne, 1988. “Assesment of Textual Interpretation in Buddhism,” in Buddhist

Hermeneutics. Ed, Lopez, Donald S. Honolulu: University of Hawaii Press.
Larson, G.J. “Differentiating the concepts of ‘yoga’ and ‘tantra’ in Sanskrit literary
history,” Journal of the American Oriental Society (2009)129: 487-98.
Latour, Bruno, and Peter Weibel. 2002. Iconoclash. Karlsruhe: ZKM.

Latour, Bruno, “How to be Iconophile in Science, Art and Religion” in Picturing Science,
Producing Art. Ed. Caroline A. Jones & Peter Galison. New York & London: Routledge,
1998
Lee, Yong-Hyun. “Synthesizing a Liturgical Heritage: Abhayåkaragupta’s Vajråvalî and

the Kalacakra Mandala.” Ph.D. Dissertation. Madison: The University of Wisconsin,
2003.
Leidy, Denise Patry, and Robert A. F. Thurman. 1997. Mandala: the architecture of
enlightenment. London: Thames & Hudson in association with Asia Society Galleries
and Tibet House.
Lopez, Donald “Polemical Literature [dgag lan]” in Tibetan Literature: Studies in Genre,
ed. Jose Ignacio Cabezon and Roger R. Jackson. Ithaca, NY: Snow Lion Publications,
1996.
Lo Bue, Erberto F. 2010. Wonders of Lo: the artistic heritage of Mustang.Mumbai :

Marg Foundation.
Luczanits, Christian. 2002. “On the Earliest Mandalas in a Buddhist Context.” In

Mahayana Buddhism. History and Culture. Sambhota Series XV, eds. Darrol Bryant &
! 303!
Susan Bryant. New Delhi: Tibet House: 111-136. 2008.
Review Article: Erberto Lo Bue (ed.) 2010. Wonders of Lo: The artistic heritage of
Mustang. Vol. 62 (2), Marg. Mumbai: Marg Foundation. The Tibet Journal XXXVIII, no.
3&4 (2013): 161-67.
“Siddhas, Hierarchs, and Lineages: Three Examples for Dating Tibetan Art” in
Jackson, David Paul, and Christian Luczanits. 2011. Mirror of the Buddha: early portraits
from Tibet : from the Masterworks of Tibetan painting series. New York: Rubin Museum
of Art.
“The Many Faces of Buddha Vairocana.” In The All-Knowing Buddha: A Secret
Guide, edited by Jan van Alphen. New York and Antwerp: Rubin Museum of Art and
BAI, MAS Books, 2013: 12-23, 163-166.
Mallmann, Marie-Thérèse de. 1975. Introduction à l'iconographie du tântrisme

bouddhique. Paris: [En vente, Librairie Adrien-Maisonneuve].
Mathes, Klaus-Dieter. A Direct Path to the Buddha Within: Gö Lotsåwa’s Mahåmudrå

Interpretation of the Ratnagotravibhåga. Boston: Wisdom Publications, 2008.
Matthiesson, Peter. East of Lo Monthang: In the Land of Mustang. Photos by Thomas

Laird. Ithaca: Snow Lion Publications, 1996.
Robert Mayer, A Scripture of the Ancient Tantra Collection: The Phur-pa bcu-gnyis
(Oxford: Kiscadale Publications, 1996)
Meisezahl, Richard O. 1985. Hastapüjåvidhi-Texte: Der Handritus im anuttarayogischen

Kult Götterpaares Cakrasambhara und Vajravårahî. Sankt Augustin, Germany: VGH
Wissenschaftsverlag.
Stephen Melville, “Deconstruction and the Limits of Interpretation,” The Art of Art
History: a Critical Anthology, edited by Donald Preziosi (ed) (Oxford & New York:
Oxford University Press, 1998 & 2009): 271-316.
Meyer, Fernand with Yuri Parfionovich and Gyurme Dorje, Tibetan Medical Paintings:
Illustrations to the Blue Beryl Treatise of Sangye Gyamtso Vol.1 (London: Serindia
Publications, 1992)
Miller, Willa Blythe. 2013. ‘Secrets of the Vajra Body: Dngos po'i gnas lugs and the
Apotheosis of the Body in the work of Rgyal ba Yang dgon pa.’ Ph.D. dissertation.
Harvard University.
Mori, Masahide. “The Vajråvalî of Abhayåkaragupta.” Ph.D. dissertation. London: The

School of Oriental and African Studies, University of London, 1997.
----------“The Vajråvalî Mandala Series in Tibet,” in Esoteric Buddhist Studies: Identity
in Diversity. Proceedings of the International Conference on Esoteric Buddhist Studies,
Koyasan University, 5 Sep.-8 Sept. 2006, Published 2008.
! 304!
Muller-Ortega, Paul Eduardo. 1989. The triadic Heart of ¸iva: Kaula tantricism of
Abhinavagupta in the non-dual Shaivism of Kashmir. Albany: State University of New
York Press.
Mus, Paul. 1998. Barabudur: sketch of a history of Buddhism based on archaeological

criticism of the texts. New Delhi: Indira Gandhi National Centre for the Arts.
Nag-dbang-blo-bzang-rgya-mtsho, and Zahiruddin Ahmad. 1995. The history of Tibet.

Bloomington, Ind: Indiana University, Research Institute for Inner Asian Studies.
Newland, Guy. “The Debate Manual (yig cha) in Tibetan Monastic Education,” in
Tibetan Literature: Studies in Genre, edited by José Ignacio Cabezón and Roger R.
Jackson. Ithaca: Snow Lion, 1996.
Owens, Bruce McCoy. 1995. “Human Agency and Divine Power: Transforming Images
and Recreating Gods Among the Newar”. History of Religions 34 (3). University of
Chicago Press: 201-40.
Padoux, Andre. “Contributions a L’etude du Mantrasastra. II: nyasa- l’imposition

rituelle des Mantra,” Bulletin de L’Ecole Francaise d’extreme Orient. 67 (1980): 59-102.
Pakhoutova, Elena. “Buddhist Practices and Rituals Centered on Buddha Vairocana in

Tibet,” in In The All-Knowing Buddha: A Secret Guide, edited by Jan van Alphen. New
York and Antwerp: Rubin Museum of Art and BAI, MAS Books, 2013: pp.38-45.
Patton, Laurie L. 2005. Bringing the gods to mind mantra and ritual in early Indian
sacrifice. Berkeley: University of California Press.
Payne, Richard. 2002. “ Tongues of Flame: Homologies in the Tantric Homa,” In The
Roots of Tantra. Edited by Katherine Anne Harper and Robert L. Brown. Albany: State
University of New York Press.
Petech, Luciano. 1990. Central Tibet and the Mongols: the Yüan Sa-Skya period of
Tibetan history. Rome: Instituto italiano per il Medio ed Estremo Oriente.
Radich, Michael David. 2007. ‘The somatics of liberation: Ideas about embodiment in
Buddhism from its origins to the fifth century C.E.’ PhD dissertation, Harvard University.
Robson, James. 2008. “Signs of Power: Talismanic Writing in Chinese Buddhism”.

History of Religions 48 (2). The University of Chicago Press: 130-69.
Rockhill, William Woodville. 1891. The land of the lamas: notes of a journey through
China, Mongolia and Tibet with maps and illustrations. New York: Century Co.
Said, Edward W. 2004, Humanism and Democratic Criticism, New York: Columbia
University Press.
! 305!
Ruegg, David Seyfort. “The Jo nang pas: A School of Buddhist Ontologists According to
the Grub Mtha’ shel gyi me long.” Journal of the American Oriental Society, Vol.83,
No.1 (Jan-Mar., 1963), pp.73-91.
1968/9. La théorie du tathågatagarbha et du gotra; études sur la sotériologie et la
gnoséologie du bouddhisme. Paris: École Française d'Extrême-Orient
1992. “Notes on Some Indian and Tibetan Reckonings of the Buddha’s Nirvåña
and the Duration of His Teachings.” In Bechert, Heinz.. The Dating of the historical
Buddha = Die Datierung des historischen Buddha. Part 2. Göttingen: Vandenhoeck &
Ruprecht: 263-90.
Ruegg, David Seyfort, and Rin-chen-rnam-rgyal. 1966. The life of Bu ston Rin po che,
with the Tibetan text of the Bu ston rNam thar. Roma: Instituto italiano per il Medio ed
Estremo Oriente.
Samuel, Geoffrey. 2008. The origins of yoga and tantra: Indic religions to the thirteenth
century. Cambridge, UK: Cambridge University Press.
Samuel, Geoffrey, and Jay Johnston. 2013. Religion and the subtle body in Asia and the
West: between mind and body. Abingdon, Oxon: Routledge.
Sanderson, Alexis. 1986. “Mandala and Ågamic identity in the Trika of

Kashmir.”In Mantras et Diagrammes Rituelles dans l'Hindouisme, ed. Andre Padoux.
Équipe no. 249 'L'hindouisme: textes, doctrines, pratiques.' Paris: Éditions du Centre
National de la Recherche Scientifique, pp. 169-214.
1993. “The Dependence of the Herukatantras on the ¸aiva Tantras of the
Vidyåpîtha.” (Lecture in Trinity Term, All Souls College, University of Oxford, 1993).
(Unpublished) cited in English 2002 fn 470 .
1995. “Vajrayåna: Origin and Function.” In Buddhism into the Year 2000.
International Conference Proceedings, Bangkok and Los Angeles: Dhammakāya
Foundation, pp. 89-102.
2009. “The ¸aiva Age: The Rise and Development of ¸aivism during the Early
Medieval Period.” In Genesis and Development of Tantrism, edited by Shingo Einoo.
Tokyo: Institute of Oriental Culture, University of Tokyo, 2009. Institute of Oriental
Culture Special Series, 23, pp. 41-350.
Schweiger, Peter. 2009/10. “Significance of Ming titles conferred upon the Phag mo gru
rulers: A reevaluation of Chinese-Tibetan relations during the Ming Dynasty,” The Tibet
Journal 34-35: 313-328.
Sharf, Robert H. "Experience," in Critical Terms for Religious Studies, edited by Mark C.
Taylor (Chicago: University of Chicago Press, 1998), pp. 94-116.
1999. “On the Allure of Buddhist Relics”. Representations, no. 66. University of
California Press: 75-99.
"Visualization and Mandala in Shingon Buddhism." in Living Images: Japanese
Buddhist icons in Context. Asian Religions and Cultures, no. 2, co-edited with Elizabeth
Horton Sharf. Stanford: Stanford University Press, 2001.
! 306!
----------------"Thinking through Shingon Ritual," Journal of the International Association
of Buddhist Studies 26, no. 1 (2003), pp. 51-96.
"Buddhist Modernism and the Rhetoric of Meditative Experience," Numen 42,
no. 3 (1995), pp. 228-283. Republished in Buddhism: Critical Concepts in Buddhist
Studies, edited by Paul Williams (London: Routledge, 2005), vol. 2, pp. 255-298.
"Ritual," in Critical Terms for the Study of Buddhism, edited by Donald S. Lopez,
Jr. (Chicago: University of Chicago Press, 2005), pp. 245-269.
----------------"The Buddha's Finger Bones at Famen-si and the Art of Chinese Esoteric
Buddhism." Art Bulletin 93, no. 1 (March, 2011), pp. 38-59.!
Shen Weirong. “On the History of Gling tshang principality mDo khams during the Yuan
and Ming Dynasties: Study on Sources Concerning Tibet in the Mind Shilu (I),” in
Schuh, Dieter, Petra H. Maurer, and Peter Schwieger. 2007. Tibetstudien: Festschrift für
Dieter Schuh zum 65. Geburtstag. Bonn: Bier'sche Verlagsanstalt.
Smith, E. Gene, and Kurtis R. Schaeffer. 2001. “Buddhist Literary and Practical Arts
According to Bo dong Pan chen Phyogs las rnam rgyal. “ Chapter Fourteen in Among
Tibetan texts: history and literature of the Himalayan Plateau. Boston: Wisdom
Publications: pp.179-208.
Smith, Jonathan Z. 1978. Map is not territory: studies in the history of religions. Leiden:
Brill.
“Religion, Religions, Religious,” in Critical Terms for Religious Studies. Edited
by Mark C Taylor. 1998. (Chicago: University of Chicago Press, 1998), 179-195.
Snellgrove, David. Four Lamas of Dolpo: Tibetan Biographies. Delhi: Motilal

Benarsidas, 1992.
2002. Indo-Tibetan Buddhism: Indian Buddhists and their Tibetan successors.
Boston: Shambhala.
Snodgrass, Adrian. 1992. The Symbolism of The stupa. Delhi: Mitilal Banarsidass.
Sobisch, Jan-Ulrich, and A-mes-zhabs. 2007. Life, transmissions, and works of A-mes-
zhabs Ngag-dbang-kun-dga'-bsod-nams, the great 17th century Sa-skya-pa bibliophile.
Stuttgart: F. Steiner.
Sobisch, Jan-Ulrich. 2008. Hevajra and Lam 'bras literature of India and Tibet as seen
through the eyes of A-mes-zhabs. Wiesbaden: Reichert.
Sogyal, Patrick Gaffney, and Andrew Harvey. 1992. The Tibetan Book of Living and
Dying. [San Francisco, Calif.]: Harper San Francisco.
Sperling, Eliot. 2000. "Si-tu Chos-gyi rgyal-mtshan and the Ming Court" Lungta XIII:23-
27.
! 307!
The 5th Karmapa and Some Aspects of the Relationship Between Tibet and the
Early Ming. in The History of Tibet, ed. McKay, Alex. London: Routledge Curzon,
2003.
Staal, Frits. 2008. Discovering the Vedas: origins, mantras, rituals, insights. New Delhi:
Penguin Books.
Stearns, Cyrus. Luminous Lives: The Story of the Masters of the Lam’Bras Tradition in
Tibet. Boston: Wisdom Publications, 2001.
-------------------,Trans. Taking the Result as the Path: Core Teachings of the Sakya
Lamdré Tradition. Boston: Wisdom Publications, 2006.
------------------ The Buddha from Dölpo: A Study of the Life and Thought of the Tibetan
Master Dölpopa Sherab Gyaltsen, Revised and Enlarged Edition. Ithaca, New York:
Snow Lion Publications, 2010.
Steinkellner, Ernst. 1983. “Tshad ma’i skyes bu: Meaning and Historical Significance of
the Term,” In Csoma de Korös Memorial Symposium, Ernst Steinkellner, and Helmut
Tauscher. 1983. Contributions on Tibetan language, history, and culture. Wien:
Arbeitskreis für Tibetische und Buddhistische Studien, Universität Wien: 275-283.
Sugiki, Tsunehiko, and 2003. Five Types of Internal Mandala Described in the
Cakrasamvara Buddhist Literature: Somatic Representations of One's Innate Sacredness.
144, 2003-12, P. 276-202.
2009. “The Structure and Traditions of the Systems of Holy Sites in the Buddhist
Saµvara Cycle and its Related Scriptural Cycles in Early Medieval South Asia: The
Geography of Esoteric Buddhism in the Eyes of the Compilers of the Scriptures,”
Genesis and Development of Tantrism. edited by Shingo Einoo. Tokyo: Institute of
Oriental Culture, University of Tokyo, Institute of Oriental Culture Special Series, 23:
515-562.!
Review of The Cakrasaµvara Tantra (The Discourse of ¸rî Heruka): A Study and
Annotated Translation, by David Gray, Journal of the International Association of
Buddhist Studies, Vol. 31 No.1-2 (2010): 505-542.
Szántó, Peter. “Before a Critical Edition of the Sampu†a Tantra,”
Tanaka, Kimiaki. 2013. Mitrayogin's 108 mandalas: an image database. Kathmandu:

Vajra Publications, 2013.
Teiser, Stephen F. 2007.Reinventing the Wheel: Paintings of Rebirth in Medieval

Buddhist Temples (Seattle: University of Washington Press).
Thubten Yeshe, and Jonathan Landaw. 2001. Introduction to Tantra: the transformation
of desire. Boston: Wisdom Publications.
Timalsina, Sthaneshwar. “Body, Self, and Healing in the Tantric Ritual Paradigm,: The
Journal of Hindu Studies, 2012
! 308!
Tsiang, Katherine R. 2010. “Buddhist Printed Images and Texts of the Eight-Tenth
Centuries: Typologies of Replication and Representation,” in Kapstein and van Schaik,
Esoteric Buddhism at Dunhuang, p. 201-52
Tsong-kha-pa Blo-bzang-grags-pa, Bstan- dzin-rgya-mtsho, and Jeffrey Hopkins. 1987.

Deity yoga: in action and performance tantra. Ithaca, N.Y., USA: Snow Lion
Publications.
Tsultemin, Uranchimeg. ‘Image and text in Mongolian Buddhist Art: Agwaankaidev on

Art and Artists,’ Paper presented at Reading Outside the Lines: A Workshop on the
Universitœt, Munich. (Publication forthcomung, 2016).
Tucci, Giuseppe. Report on Two Scientific Expeditions to Nepal, Serie Orientale Roma,
Vol.10 p16-17
van der Kuijp, Leonard W. J."A Text- Historical Note on Hevajratantra II: v:1-2." Journal
of the International Association for Buddhist Studies, Vol. 8, No. 1 1985, pp. 83-89.
(1985a).
"Studies in the Life and Thought of Mkhas grub rje I: Mkhas grub rje's
Epistemological Oeuvre and His Philological Remarks on Dignåga's
Pramåñasamuccaya," Berliner Indologische Studien 1 (1985b), 75-105. (1985b).
"Studies in the Life and Thought of Mkhas grub rje IV: Mkhas grub rje on
Regionalisms and Dialects," Berliner Indologische Studien 2 (1986), 23-49
"Ngor chen Kun dga' bzang po on the Posture of Hevajra: A Note on the
Relationship Between Text, Iconography and Spiritual Praxis," Investigating Indian Art.
Festschrift furH. Hertel, Berlin, 1987, 173-177.
von Rospatt, Alexander. 2010. "Remarks on the Consecration Ceremony in Kuladatta’s

Kriyåµgrahapaïjikå and its Development in Newar Buddhism." Hindu and Buddhist
Initiations in Nepal and India, eds. Astrid Zotter and Christof Zotter. Wiesbaden:
Harrassowitz: pp. 199-262.
2014. “Negotiating the Passage Beyond a Full Span of Life: old Age Rituals
Among the Newars.” South Asia: Journal of South Asian Studies. Vol.37. No.1, 104-129.
Keynote Speech presented at Reading Outside the Lines: A Workshop on the
Intersection of Buddhist Art and Texts September 13-15, 2013, Ludwig-
Maximilians-Universitœt, Munich (Publication forthcoming, 2016)
Vitali, Roberto, and Gu.ge Mkhan.chen Ngag.dbang grags.pa. 1996. The kingdoms of
Gu.ge Pu.hrang: according to mNga'.ris rgyal.rabs by Gu.ge mkhan.chen Ngag.dbang
grags.pa = [Gu.ge mkhan.chen Ngag.dbang grags.pas mdzad.pa'i mNga'.ris rgyal.rabs
bz ugf.lagf so]. Dharamsala, India: Tho.ling dpal.dpe med.lhun.gyis grub.pa'i
gtsug.lag.khang lo 1000 'khor.ba'i.rjes dran mdzad.sgo'i go.sgrig Tshogs.chung.gis
dpar.bskrun zhus.pa'o.
! 309!
Hodgson, see Waterhouse, David M. 2004. The origins of Himalayan studies: Brian
Houghton Hodgson in Nepal and Darjeeling, 1820-1858. London: RoutledgeCurzon.
Wayman, Alex. 2005 (Reprint of 1977 publication). Yoga of the Guhyasamåjatantra: the
arcane lore of forty verses : a Buddhist tantric commentary. Delhi: Motilal Banarsidass.
Weinberger, Steven Neal. 2003. “The significance of Yoga Tantra and the Compendium
of principles (Tattvasaµgraha tantra) within tantric Buddhism in India and Tibet.” Ph. D.
University of Virginia 2003.
Wilson, Liz. 1996. Charming cadavers: horrific figurations of the feminine in Indian
Buddhist hagiographic literature. Chicago, IL: University of Chicago Press
White, David Gordon. “Yoga in Transformation,” in Yoga: The Art of Transformation,

ed. Debra Diamond (Washington, D.C.: Freer Gallery of Art and Arthur M. Sackler
Gallery, 2013), 35-45.
Roger Wright, “A Survey of Explanatory Tantras and Commentaries of the Guhyasamåja

Tantra,” ?Unpublished paper. Accessed via:
http://www.wrighrp.pwp.blueyonder.co.uk/SOASMA2010/GuhyasamajaCommentarySur
vey-corrections-20100718.pdf
Wujastyk, D. (2009). Interpreting the Image of the Human Body in Premodern India.
International Journal of Hindu Studies, 13(2), 189-228.
Wylie, Turrell V. 1980. "Lama Tribute in the Ming Dynasty.” Tibetan Studies in Honour
of Hugh Richardson. (Warminster, England : Aris & Phillips, 1980).
“Monastic Patronage in Fifteenth-Century Tibet,” in The History of Tibet, ed.
McKay, Alex. London: Routledge Curzon, 2003.
Electronic Resources:
www.berzinarchives.com
www.himalayanart.org [Himalayan Art Resource]
http://idp.bl.uk [International Dunhuang Project]
Oxford English Dictionary Online
http://tibeto-logic.blogspot.com [Tibeto-Logic](Dan Martin)
https://sites.google.com/site/tiblical/home (Dan Martin)
www.tbrc.org [The Tibetan Buddhist Resource Center]
www.thlib.org [The Tibetan and Himalayan Library]
www.treasuryoflives.org
http://lexica.indica-et-buddhica.org
http://catalogue.ngmcp.uni-hamburg.dengmpp [Nepalese-German Manuscript
Cataloguing Project]
http://dl.lib.brown.edu/BuddhistTempleArt/history2.html [Tibetan Buddhist Wall
Paintings: A Photographic Survey] (Philip and Marcia R. Lieberman)
! 310!
!
! 311!

Body Mandala

Uploaded by

Document Informationclick to expand document information

Document Informationclick to expand document information

Copyright:

Available Formats

Body Mandala

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Body Mandala

Uploaded by

Copyright:

Available Formats

UC Berkeley

UC Berkeley Electronic Theses and Dissertations

eScholarship.org Powered by the California Digital Library

Rae Erin Dachille

A dissertation submitted in partial satisfaction of

University of California at Berkeley

Professor Alexander von Rospatt, Co-chair

Rae Erin Dachille

University of California at Berkeley

Professor Alexander von Rospatt, Co-chair

Professor Patricia Berger, Co-chair

Table of Contents__ pp.ii-iv

I. The Virtues and Limitations of Debate as a Tool for Interpretation

Chapter Two: What is the Body Mandala? pp.36-56

I. The Positions of others

Part II: Mkhas grub’s Own Perspective

Chapter Five: The Body Mandala Debate: Body as Explanatory Tantra

I. Ngor chen’s Introduction

V. Transforming the Basis

VI. Explanatory Continuum/ Explanatory Tantra [bshad rgyud]: Corporeality and

Chapter Six: In Defense of the Hevajra Body Mandala pp.174-202

I. Ngor chen’s Reply Part Three: Textual Authority

I. Modes of Representing the Subtle Body: The Microcosm-Macrocosm correlation

Bibliography pp. 285-310

Fig.1 Mkhas grug dge legs dpal bzang po

Fig.7 and detail

Fig.13 Scroll painting of the Vajradhåtu-mañ∂ala from Dunhuang

Fig.14 Mandala painting from Dunhuang with consorts

Fig.18 Main assembly hall in relation to the Vajradhåtu-mañ∂ala, Tabo monastery,

Whether regarded as a tool or as an obstacle, the human body is the ultimate

In engaging with the body mandala debate texts as examples of bodily

However, in translating rten as “support” rather than “container” or “receptacle,” we are

based upon his 2007 doctoral dissertation.

IV. Challenges and Support

Cakrasamvara Buddhist Literature: Somatic Representations of One's Innate Sacredness.

Boston: Wisdom Publications.

Focusing upon themes of ritual, embodiment, and representation, the introduction

To my teachers at UC Berkeley: To my advisor, Alexander von Rospatt, for giving me

The following chapter provides the necessary foundation to prepare us to engage

I. The Virtues and Limitations of ‘Debate’ as a Tool for Interpretation

II. Mkhas grubs rje’s Contributions to the ‘Body Mandala Debate’

It seems that this biographer envisions Mkhas grub’s patron as deliberately

Ary translates the relevant section from the biography as follows:

Of course in Mkhas grub’s case, his inherently polemical style as expressed by

tsun pa) 7b-8a by Ary 2015, p.128.!

III. Ngor chen’s Contributions to the Body Mandala Debate

IV. Additional Resources: Colophons and Collected Works

V. The Fifteenth-Century Context: Monastic Patronage, Scholasticism,

I. Parsing and Purifying the Body: the Early Buddhist context

Buddhaghosa’s Visuddhimagga (The Path of Purification) provides valuable insight into

140 !Nåñamoli 2010, Visuddhimagga 8.139.!

II. Purifying and Deifying the Body in Non-Buddhist Tantra

concept of becoming a god to worship a god: JS 12.1; MTV II (M®gendratantrav®††i) 29,

In beginning to approach the body mandala practices of the Guhyasamåja and in

as demonstrated by Muller-Ortega 1989.!

sensation, they are more obvious instruments of touch. !

I. An Outline of the Guhyasamåja Body Mandala Practice

Earth into water