Sunteți pe pagina 1din 170

J oumal of the International Association of

Buddhist Studies
Volume 16 It Number 2 Winter 1993
Lamas, Emperors, and Rituals:Political Implications in
Qing Imperial Ceremonies 243
Two Mongol Xylographs (Hor Par Ma) of the Tibetan
Text of Sa Skya Pal}<;lita's Work on Buddhist Logic and
Epistemology 279
Recent French Contributions to Himalayan and
Tibetan Studies 299
Contributions to the Study of Popular Buddhism: The
Newar Buddhist Festival of GumIa Dharma 309
A Re-examination of a K a n i ~ k a Period Tetradrachm
Coin Type with an Image of Mttrago/Maitreya on the
Reverse (Gobi 793.1) and a Brief Notice on the Impor-
tance of the Inscription Relative to Bactro-Gandharan
Buddhist Iconography of the Period 355
Reinterpreting the Jhiinas 375
The Journal of the interfUltional
Association of Buddhist Studies
(ISSN 0193-600XX) is the organ of
the International Association of
Buddhist Studies, Inc: It welcomes
scholarly contributions pertaining to
all facets of Buddhist Studies.
nABS is published twice yearly, in
the summer and winter.
Address manuscripts (two copies)
and books for review to Professor
Donald Lopez, Editor, nABS,
Department of Asian Languages
and Cultures, 3070 Frieze,
University of Michigan, Ann
Arbor MI 48109-1285 USA. A
style sheet for contributors is
available from the editor.
Address subscription orders and
dues, changes of address, and
business correspondence
(including advertising orders) to
Professor Lewis Lancaster
Treasurer lABS, of
East Asian Languages, 104 Durant
Hall, University of California
Berkeley CA 94720 USA
Subscriptions to nABS are $35
(or 55(0) per year for individuals
and $60 (or 9000) per year for
libraries and other institutions.
For infonnation on membership in
lABS, see back cover.
1993 by the International
Association of Buddhist Studies
Inc. The editor gratefully ,
acknowledges the support of the
Centerfor Chinese Studies and the
Department of Asian Languages
and Cultures of the University of
Donald S. Lopez, Jr.
Robert Buswell
Steven Collins
Collett Cox
Luis O. G6mez
OskarK.von Hinliber
Padmanabh S. J aini
Shoryu Katsura
Alexander Macdonald
D. Seyfort Ruegg
Ernst Steinkellner
Erik ZUrcher
Editorial Assistant
Alexander Vesey
Contributors to this issue:
RODERICK BUCKNELL is Senior Lecturer in Eastern Religions in
the Department of Studies in Religion at the University of Queensland,
Australia. His major research interest is Buddhist meditation.
JAMES L. HEVIA is Assistant Professor in the Department of History
at North Carolina A & T University, Greensboro, NC. He has pub-
lished articles on ritual and history, Euro-American representations of
China, and is currently completing a book on the 1793 Macartney em-
bassy to China entitled Cherishing MenfrOln Afar (Duke University
JOHN C. HUNTINGTON is Professor of Art History at The Ohio
State University. His primary research areas are Buddhist art and
iconography. He is currently working on the Gandhara, Indonesia, and
Himalayan/Tibetan volumes of his Encyclopedia of Buddhist
PER KV AERNE is Professor of History of Religions and Tibetology
at the University of Oslo. His is a member of the Norwegian Academy
of Science, Secretary-general of the International Association for
Tibetan Studies, and Editor of Acta Orientalia.
TODD T. LEWIS is Assistant Professor in the Department of
Religious Studies at the College of the Holy Cross in W orchester,
Massachusetts. He is the author of numerous articles on Buddhism in
the Kathmandu valley.
LEONARD W. J. V AN DER KUI.TP is Associate Professor in the
Department of Asian Languages and Literatures of the University of
Washington. He is currently preparing a history of the Yar rgyab
faillily and a study of the background of Bu ston's History.
Lamas, Emperors, and Rituals:
Political Implications in Qing Imperial Ceremonies
... meaning is always, to some extent, arbitrary and diffuse ... social
life everywhere rests on the imperfect ability to reduce ambiguity and
concentrate power.
... there is no basis to assume that the histories of the repressed, in
themselves, hold a key to revelation ... the discourses of the domi-
nant also yield vital insights into the contexts and processes of which
they were part.
Comaroff and Comaroff 1990, 11, 17.
In the "Lama Temple" of Peking, the Yonghe Palace, are two stone
inscriptions attributed to the Qianlong emperor, Hongli, each of which is
carved in Chinese, Manchu, Mongol, and Tibetan on the four sides of
large stone blocks. Their subject is Buddhism, but the difference and
discursive distance between them signify two poles of a contradiction
that animates much of the history of relations between Tibetan and
Mongol lamas and Qing emperors.
My purpose here is to consider
this contradiction as an historical artifact of extreme ambivalence, a
vacillation which, in the case of the Qianlong emperor, highlighted the
complexities of maintaining Manchu hegemony over much of Inner
The first of these inscriptions was written in 1744. It dedicates the
I wish to thank Ronald lnden, Mark Elliot, Stephen Shutt, Bruce Doar,
Evelyn Rawski, Susan Naquin, Chia Ning, Ruth Dunnell, Elliot Sperling,
and Judith Farquhar for comments and suggestions on earlier versions of this
l. For translations of each of the inscriptions see Lessing 1942, 9-12 and 58-
62. It should be noted that his translations appear to be composites of the
four versions. LeSSing does not reproduce original language versions of these
inscriptions, only a difficult to decipher photograph of the Lamashuo. How-
ever, photographs of each of the inscriptions in their multiple versions can be
found in Franke and Laufer 1914.
244 JIABS 16.2
Yonghe Palace as a Buddhist temple to HongH's father, the Yongzheng
emperor. In it the emperor speaks as a "filial son, the pious friend of
priests, a Chinese Asoka Dharmaraja" (Lessing 1942, 61-62), indicating
that his father has realized nirvaI).a and achieved the highest form of
enlightenment (Farquhar 1978, 32). From this inSCription and other
sources it is easy to develop an image of Qing emperors as "perfect
Buddhist monarchs, grand patrons of the True Law, and bOdhisattvas"
(Farquhar 1978,22).
The second inSCription is entitled Lamashuo (pronouncements on
Once characterized by LeSSing as the composition of a disap-
pointed old man "full of acrimony and acerbity" (1942, 62), the text was
written by the Qianlong emperor in the wake of the second campaign to
expel the Gurkhas from Tibet (1792). It conveys quite another message.
Speaking in a voice of authority,
the emperor presented his own ver-
sion of the historical relations between the Yuan, Ming, and Qing royal
houses and Tibetan and Mongol religious hierarchs, one that served to
justify his decision to impose a new selection process for "incarnated"
lamas such as the Dalai Lama and the Mongol Rje btsun dam pa
Those who have drawn attention to these stelae have tended to inter -
pret them in one of two ways. The first is to see them as indicating the
manipulative aspects of Qing treatment of the religious beliefs of others.
2. I have consulted two printed versions of the Lamashuo, one in the
Weizang tongzhi of Tibet, hereafter cited as WZ1Z),1, 23-26, one
found in the Shiwen shichuan ji (Poems and Prose on the Ten Great Cam-
paigns of the Qianlong Era), 1962, 674-676, as well as Lessing's translation
of the text from the stele. While there are clear variations between these ver-
sions, my use of the inscription focuses on their commonalties.
The possibility of different content appearing in the four languages in
which inscriptions like the Lamashuo are rendered has recently been discussed
by Elliot 1992 with respect to a variation noted by Lessing in the Manchu
version (1942, 61). According to Elliot, in the Manchu text the emperor de-
fended himself against criticism by some Chinese officials over his dabbling
with Tibetan Buddhism. Hongli added that if he had not studied Buddhism,
there would not now be peace in Inner Asia (1992, 26-27).
3. While Lessing's characterization of the tone of voice to be found in this
text is interesting, I would suggest that it is also consistent with the imperial
voice in texts discussed by Crossley 1990, and Zito 1987. Crossley finds
clear affinities to the kaozheng tradition that animated many of the other text
projects of the Qianlong era. Zito adds that in the case of the Hongli inscrip-
tion she deals with, the emperor positions himself, much as he does in the
Lamashuo, as the singular authority on the subject at hand. On kaozheng see
Elman 1984 and Guy 1987.
The second is to argue that in fact early Qing emperors were true believ-
ers in Tibetan Buddhism, but because of political exigencies in China,
had to hide their religious convictions.
While either of these interpreta-
. tions has certain explanatory power for understanding the degree to
which emperors were involved with Tibetan Buddhism, they each tend
to obscure a central contingency of Inner Asian politics. In order to
have any influence in the region, Manchu emperors had to address
Tibetan Buddhism whether they believed in it or not. Perhaps more
importantly, they had to do so in an idiom that was already well estab-
lished throughout the region. Their successes at incorporating Inner
Asia into their multi -ethnic empire were just as much a result of master-
ing this idiom, of "reducing ambiguity" and concentrating power in the
form of discursive authority, as it was of their military might and admin-
istrative acumen.
One possible way of explaining the forces at work in Tibetan-
Mongol-Manchu relations is to draw on the tribute system model of
"traditional Chinese foreign relations" (Fairbank 1942, 1948; Fairbank
and Teng 1941). The limitations of this approach have, however,
become increasingly clear. For example, in his now classic formulation
of the tribute system, Fairbank had argued that its "secret" lay in the fact
that it had become a "vehicle for trade" (1948, 132). Yet in the situa-
tions that will be discussed here, trade does not seem to be a factor.
Moreover, Fairbank's later modifications to the content and purpose of
the system (1968) offers very little guidance for or explanations of rela-
tions between Qing emperors and Tibetan and Mongol Buddhist hier-
archs. This is especially the case when we recognize that the Qing
emperors were not Chinese rulers, that their empire included more than
China, and that these same emperors appear to have been actively
engaged in Tibetan initiation rituals and in the cult of the emperor as
4. Lessing and Farquhar tend to come down on the side of the first argu-
ment. See Chia 1992, 207-208 for a review of this position and additional
sources. Grupper 1980 and 1984 and Sperling 1983 see the emperors they
discuss as believers in Buddhism. One might add parenthetically that while
it has long been a creed among historians that superior Chinese civilization
sinicized the Manchus, few have asked whether the Manchus were true be-
lievers or cynical manipulators of Confucianism.
5. On the military and administrative consolidation of Qing control in Inner
Asia see especially Fletcher 1978a-b as well as other sources on the region
cited in this section.
6. Scholars who have recently sought to retain some usefulness for the trib-
246 JIABS 16.2
Another way of proceecling might be to see these relationships as ones
in which the various parties attempted to encompass and include others
in their own cosmologies ti'rrough joint participation}n rituals of inclu-
sion and transformation. So, for example, Qing emperors frequently
attempted, through audience rituals, to establish with lamas relations of
the kind that obtained between a supreme lord (huangdi) and a lesser
lord (Janwang ),7 and thereby negate any claims by lamas to superiority.
But even this gesture was not without ambiguity. At the same moment
they attempted to include lamas in their emperorship as if the latter were
worldly lords, emperors also sometimes clistinguished them from the
category of fanwang (see below on the Khalkha submission to the
For their part, Tibetan lamas and Mongol hierarchs sought at various
times to assert a long-standing Buddhist view which placed the lama as
the intellectual/spiritual superior of a lord of the "mere" earth. In this
relationship, usually referred to as that of lama and patron (T. mchod
ute system model have at the same time questioned its applicability to spe-
cific historical circumstances; see, for example, Rossabi 1983 and Wills
1984, 1993, 102. Interpretations that do not exclusively rely on the tribute
system, but rather point to supposedly long-standing policies from one
dynasty to another, such as the principles of divide and rule, using barbarians
to manage barbarians, etc. would also have difficulty accounting for imperial
interest in Tibetan Buddhism and bodhisattvahood. See, for example, Yang
1968, 20-33. On Inner Asia see sources cited in Grupper 1984. Farquhar
1978, Jagchid 1974, and Rahul 1968-1969 tend to problematize the tribute
system hypothesis. Few, however, have been willing to abandon it com-
pletely. For other critical engagements with the tribute system model see
Hevia 1989; Hevia, forthcoming; and Farquhar and Hevia 1993.
On the problems the Manchus as well as other conquest dynasties pose for
transhistorical categories such as the tribute system it is perhaps worth recall-
ing that Lattimore (1962, 77) long ago drew attention to the propensity of
Euro-Americans to speak of the Chinese empire, when in fact under such
dynasties, China formed only a part of these empires. Much of the argument
for maintaining China-centered terminology pivots on the notion of siniciza-
tion. Crossley 1990 has raised serious questions about the sinicization of the
7. See Hevia 1989 on supreme-lesser lord relations and their link to the rela-
tionship between the cosmos and the emperor. In cosmological terms, the
relations between supreme lord and lesser lord were caught up in the produc-
tion of imperial virtue (de). The idea here seems to have been that as virtue
extended outward into the world, it resonated with attributes common to all
humans. In the lesser lords, it reoriented and attracted them toward the
Imperial source of virtue, to, as many sources have it, "sincerely face toward
transformation" (shanghua zhi cheng). For further elaboration see the section
below on imperial audiences.
yon), the lama claimed to command superior spiritual powers. As such
he could recognize a lord, including an emperor, as a cakravartin king,
instruct him in Buddhism, initiate him into tantric mysteries, and receive
offerings .from him for sustenance of the sect. The patron, in tum,
would be expected to accept a position as inferior, protect the lama, seek
his teachings, and promote Buddhism in his (the patron's) domain.8 In
either case-supreme-Iesser lord or lama-patron-the relationship was
hierarchical, with one party assuming the position of a superior, the
other of an inferior.
The differences evident here between the relations of supreme-lesser
lord and lama-patron draws attention to the multiple forms of power
present in the Qing empire. It also suggests that imperial hegemony was
itself a continuous undertaking; there was always that which resisted or
deflected Qing management and control, always counterdiscourses
emerging to either challenge or evade the projects of the Manchu
In a sense, therefore, Qing emperorship was itself a continu-
ing achievement, one that, among other things, involved the inclusion of
the strength of other rulers and significant personages into the powers of
the supreme lord.
In order to elucidate more clearly the stakes of these struggles for
lamas and emperors, I will focus on accounts of encounters between
them in various sites selected by the Qing court. Since the court hosted
these meetings, they were organized around principles of Guest Ritual
(binli) and imperial audiences (chaojian). Scholars have long acknowl-
8. Ishihama 1992, 507, notes that when granting titles the lama was the clear
superior to an eruthly lord. Much the same could be said of the other rela-
tions between lama and lord referred to here. Ruegg 1991 provides the most
detailed study of Tibetan lama-patron relations. He also argues that it is mis-
leading to see the relationship in terms of oppositions between secu-
lar/spiritual and profanelreligious (450), but as historically variable. For
example, the Dalai Lama might be conceived as a Ruler-Bodhisattva, and
Dharma-kings or Cakravartin-Sovereigns as manifestations of bodhisattvas.
For a discussion of cakravartin rulership see Tambiah 1976, 39-53 and La
Vallee Poussin 1988, 2, 484-487. Cakra or wheel refers to the king's chariot
rolling in the four directions defining the kingdom. Depictions of the
Qianlong emperor as the Bodhisattva Mafijusn have him holding a wheel in
his left hand. See Farquhar 1978, 7, Kahn 1971, 185, and Palace Museum
1983, 117.
9. Much of the work on popular culture in China directs attention to the
diversity of beliefs and practices in the late empire. See Naquin 1985 and
other articles in the same volume, Esherick 1987, and Kuhn 1990. For a
discussion and analysis of the construction of orthodoxy in Qing China see
Zito 1987.
248 JIABS 16.2
edged that imperial ritual and ceremony occupied an important place in
the establishment and continuation of Chinese dynasties. Just what the
role of ritual might have been in the reproduction of monarchical order
in imperial China is, however, far from evident, particularly in light of
recent critiques of instrumental, representational, symbolic, and theatrical
interpretations of ritual ( cf. Bell 1992).10 Recent research and theoretical
developments have also called into serious question methodologies
which separate "beliefs" from "reality" and then attempt to resolve the
resulting contradictions imputed to historical subjects in functional,
symbolic, or expressive terms (see Skorupski 1976; Sperber 1975;
Taussig 1987; and Thompson 1986). As a corrective to earlier treat-
ments of Qing imperial ritual, I intend to treat ceremonial audiences as
constitutive, rather than representative of, hierarchical political relations
between the Manchu imperium and Tibetan or Mongol religious
hierarchs. I I
Before proceeding to an exploration of the political work that ritual
does, however, a few words are in order about the nature of this study.
The methodology I adopt begins by reading across a disparate collection
of historical materials about meetings between lamas and emperors, and
about ritual practices. Rather than simply seeking facts about encounters
and ceremonies, I am concerned with the disjunctions, contradictions,
ambiguous presentations, claims, counter-claims, assertions, and refuta-
tions to be found in many of these materials. The differences and het-
erogeneity evident in these sources are, however, only one aspect of
their interest. Equally Significant are the many forms of signifying prac-
tice they embody: writings on paper, writings on stone, writings
informed by and sometimes commanding history, writings inspired by
the capacities attributed to incarnate beings, bodily practices in imperial
audience, bodily practices in meetings between incarnate beings and
others, and bodily practices in tantric initiation rituals. My purpose is
10. See for example the following, all of which treataudience ritual as func-
tional, symbolic, or expressive, Fairbank 1942; Jochim 1980; Mancall 1968, .
1971; Pritchard 1943; and Wills 1984.
11. These considerations of hierarchy and its connection to the ceremonial
construction of relations of power are not novel. They draw on more general
sociologies such as Dumont's classic study of hierarchy in India (1970),
Bourdieu's studies of social practice and bodily disposition (1977), and later
critical engagements with the work of each of these scholars (e. g., lnden
1990 and de Certeau 1984). These writings enable readings of rituals that are
sensitive to and find real political significance in the movement, location, and
concrete dispositions of time, space, and bodies in ritual practices.
not to present a comprehensive history of lama-emperor relations, nor is
it to open new sources on these subjects. Drawing on a variety of pre-
vious work, I offer what I hope will be understood as a nuanced reading
of things long known, things which to date have been separated by dis-
ciplinary boundaries and the division oflabor symptomatic of area stud-
ies. I claim no particular expertise beyond my own research on Qing
imperial ritual, only an ongoing fascination with power and its constitu-
tion in the practices considered here. If there is a virtue in this sort of
work it lies in its eschewing of historiographic naturalism; it is itself a
made object, one which draws attention to its own manufacture and to
the inventiveness of the sources it considers.
Qing Emperors and Tibetan Buddhism
Qing emperors were involved in Tibetan Buddhism to a degree that is
seldom acknowledged. This interest went beyond simply conceding the
importance of Buddhism for the empire's subjects and included, for
example, the construction of monasteries, the launching of military cam-
paigns that during Qianlong's reign helped to extend the dominion of
the Dge lugs pa sect (e. g., Martin 1990), and the participation of
emperors in tantric initiation rites. The depth of this involvement may be
accounted for by a variety of factors. As the Qianlong emperor pointed
out in his Pronouncements on Lamas, Qing interest in Tibetan
Buddhism had to do with the fact that important relations had previously
existed between the Yuan and Ming dynasties and Tibetan lamas from
Inner Asia. In the case of the Yuan, a lama-patron relationship was
forged between Khubilai Khan and the lama 'Phags pa of the Sa sky a pa
sect. During the early Ming period the fifth Karma pa Lama visited the
court of Ming Chengzu (the Yongle emperor) in 1407. In both cases
emperors bestowed titles on the lamas and lamas bestowed tantric initia-
tions on emperors. In the Ming case, Tibetan sources add that the lama
recognized the emperor and empress as the incarnations of Maiijusrt and
Tara. 12
In addition to these historical affiliations between Tibetan Buddhism
and the two dynasties that preceded the establishment of the Qing, the
Manchu ruling house was perennially concemed with the possibility of
12. On Yuan relations see Franke 1978 and 1981, Rossabi 1988, and
Richardson 1984, 34. On the Karma pa Lama's visit to Peking see Sperling
1983, especially 80-99 and Wylie 1980. On Tibetan incarnation see Wylie
1978. On Mafljusn see Lamotte 1960.
250 JIABS 16.2
the re-emergence of a Mongol kingdom in Inner Asia that might chal-
lenge its own pre-eminence (Rossabi 1975 and Petech 1950).1
concerns existed before the fonnal inception of the d),nasty in China and
were fueled by more than simply the fact that some Mongol Khans
refused to submit to Manchu overlordship. Among other things, only a
few decades before Nurhaci began to consolidate the Manchus, Altan
Khan and the third Dalai Lama had met in Mongolia and, invoking the
relationship between 'Phags pa and Khubilai Khan, forged a lama-
patron relationship (Bawden 1968, 29-30 and Rossabi 1975, 118).
Matters were further complicated when in 1639 the Tilsiyetti Khan,
Gombodorji, had his son, later entitled by the fifth Dalai Lama as the Rje
btsun dam pa Khutukhtu, accepted by the Khalkha Mongols as an incar-
nate lama. According to Bawden (1968, 53-54), the Khan's purpose
here may have been to provide a counter force to the power of the
Tibetan Dge lugs pa sect, while at the same time hedging against a
potential alliance between the Tibetans and the newly declared Qing
dynasty of Hung Taiji (Abahai). For their part, the Manchu rulers
seemed to have been intent on preventing either the Dge lugs pas or the
Khalkha khutukhtu from providing a focal pOint for Mongol restora-
tionists (Grupper 1984,51-52).
With the founding of the Qing dynasty the triangular relationship
between Manchus, Mongols, and Tibetans became more elaborate. The
Dalai Lama and occasionally the Rje btsun dam pa Khutukhtu acted as if
they themselves were rival lords. They invested, entitled, and provided
seals for Mongol Khans, arbitrated disputes between Khans, and, like
emperors and Khans, received and dispatched embassies, commanded
populations, and in some cases even armies (Bawden 1968, 31, 34,48-
50, and 63-69; Ishihama 1992; Rossabi 1975, 112-114, 119; and Ruegg
1991,450). In addition, each of these lamas was regarded as an incar-
nate bodhisattva, the Rje btsun dam pa Khutukhtu, Vajrapa1).i and the
Dalai Lama, A valokitesvara, two bodhisattvas who, with MafijusrI,
fonned atriumvirate. These celestial bodhisattvas embodied the univer-
sal totality of the three aspects of the Buddha-power (Vajrapa1).i), com-
passion (AvalokiteSvara), and wisdom (MafijusrI). It is perhaps not so
surprising, therefore, that a cult of the emperor as the bodhisattva
13. Here it is useful to follow Crossley's distinction (1990) between the
dynastic house and the Manchu clans in general. This is particularly the case
in the Qianlong era when Manchu-ness was literally constituted by order of
the emperor. See Crossley 1987, which admittedly does not draw the same
conclusion I have here.
Mafijusn would emerge under the early Qing emperors.
At the same time that Manchu emperors showed concern over the
activities of lamas and khutukhtus, they also demonstrated a keen inter-
est in Tibetan Buddhism. Beginning with Nurhaci, emperors promoted
the cult of specific deities, such as Mahakala, in Shenyang and later
Peking. Hung Taiji built temples to MahakaIa and KaIacakra in
Shenyang. The Kangxi emperor constructed Buddhist temples at Rehe,
and his grandson the Qianlong emperor built reproductions of the Potala
and the Panchen's residence at Tashilhunpo at the same site. Qing
emperors also altered or embellished existing structures in Peking, of
which the Yonghe Palace is only one example. 14 In addition, the
Qianlong emperor authorized monumental translation and text editing
projects of the Buddhist canon.
Qing emperors also joined with
Tibetan and Mongol Buddhist hierarchs in the promotion of the cult of
Mafijusrlon Mount Wutai.
It also seems significant that emperors
were willing to accept names and titles such as the bodhisattva Mafijusri
and cakravartin king (Farquhar 1978) and receive consecrations from
Tibetan Buddhist lamas (Grupper 1980 and 1984).
Emperors may also have been drawn to Tibetan Buddhism because.
lamas possessed extraordinary magical powers. At one end of the spec-
14. Franke 1981, 308 and Grupper 1980 and 1984 discuss the link between
Mongol rulership and MahakaIa. On temple construction at Rehe see Qi
1985. Other examples of temple restoration and patronage include those to
Mahakala and Yamantaka which bracket the imperial palaces in Peking, see
Arlington and Lewisohn 1935, 82, 127-128. The Kangxi emperor also recon-
structed temples on Mount Wutai, see Gimello 1992, 134. The promotion of
Buddhism at the center of Qing power in China may account for the curious
claim made by Buddhist monks to Lessing in the 1930s. They told him that
the Supreme Harmony hall (Taihe dian), the fIrst of the audience halls in the
"Forbidden City," was a maI)<;lala for Yamantaka (1976,89-90).
15. Other examples of Manchu involvement in Tibetan Buddhism from
Nurhaci to Hongli abound. According to Grupper (1984, 57, 73), the Kangxi
emperor acknowledged the close tie between the Qing royal house and Tibetan
Buddhism when he enfeoffed the fIrst Lcang skya Khutukhtu. The site of the
fIef was appropriately enough Dolonnor; the enfeofflng document asserted a
relationship between Khalkha submission and the patronage of Tibetan
Buddhism by Nurhaci and Hung Taiji. Also see Jagchid 1974, 44. The
Yongzheng emperor also patronized Buddhism, although according to
Hummel (1943, 918), he was more interested in Chan.
16. On Wutai see Farquhar 1978, 12-16; on lamas and emperors at Wutai see
Bawden 1961,58, Hopkins 1987,28-29, and Pozdneyev 1977, 336. On the
basis of these and other examples, Grupper argues that the early Manchu
kingdom was "indistinguishable" from those of Mongol Khans (1984, 52-54,
67-68). For the earlier history ofWutai and MafijuSri see Gimello 1992.
252 JIABS 16.2
trum of such magical capacities were levitation, self -dismemberment and
fe-union (Das 1881, 159), and supernatural powers of perception.
Mongol sources record, for example, that when the Rje btsun dam pa
Khutukhtu visited the court of the Kangxi emperor, cthe emperor repeat-
edly tested and attempted to trick him. The Khutukhtu saw through each
of these ruses, revealed the subterfuge, and delighted the emperor with
his powers in the process (Bawden 1961, 51-56, Pozdneyev 1977, 333-
334; on the powers of enlightened beings see Bhattacharyya 1980, 88-
At the other end of the spectrum was the ability oflamas to command
the powers of celestial beings in order to influence events on earth. For
example, the biography of the Lcang skya Rol pa'i rje (1717-1786)
records how he performed a ritual on Mount Wutai that launched bolts
of fire onto a battlefield hundreds of miles away where Qing forces
were engaged in a campaign against the Jinchuan "rebels." Lcang
skya's intervention not only carried the day for the Qing, but aided in
the eventual suppression of the uprising and spread the Dge lugs pa sect
into re gions where it had hitherto been marginal or non-existent (Martin
The supernatural powers of lamas might have had other significance
as well. Some sources note that Tibetan lamas of the Sa skya pa sect
vied at displays of magical power with shamans at the Yuan court (e. g.,
Heissig 1980,24,36 and 1953,514). These particular powers appear to
have been closely aligned with lamas' medical knowledge, a factor that.
may also have brought them into confrontation with Mongol shamans.
Whatever the case, in the famous second conversion of the Mongols to
Buddhism in 1578, shamanism was reported to have been forbidden and
shamanic idols replaced by images of the Buddha and various other
deities. I? Given the potential for conflict in matters where the powers of
Tibetan lamas and those of shamans overlapped, one might well ask if
lamas provided a convenient counter-balance to shamans at the Yuan
and Qing courts. 18 Perhaps the promotion of the cult of Mahakala by
17. Heissig 1980, 27, 36, Bawden 1968, 32-33, and Ahmad 1970, 88-99.
Shamanism did not, of course, disappear among Mongol groups as a result of
this meeting between Altan Khan and the Dalai Lama.
18. Recent work in China among contemporary Manchu shamans is very
suggestive on this count. In interviews conducted by Wulaxichun (1986,
104-106), a story about the Nudan shaman indicates rivalry between lamas
and shamanism, with the shaman triumphing over the lama. Here I follow
the text of a story in Chinese and a translation by Shi Kun (n.d.). For
Qing emperors not only incorporated aspects of Tibetan Buddhism and
Mongol rulership into Manchu emperorship, but undermined the
powers of shamans within the Manchu clans.
One of the more obvious of such incorporations which relates directly
. to the question of emperorship was the promotion of the Manchu ruler
as the bodhisattva MafijusrL 19 Various Tibetan works, for example,
"urged consecrated sovereigns to adopt the twin goals of
Bodhisattvahood and universal dominion" (Grupper 1984, 49-50).
Equally compelling are those aspects of Buddhist notions of divine
rulership which seem to make a link between the bodhisattva Mafijusri
and a cakravartin king. According to Snellgrove, there had been from
very early on in Buddhism an association of rulership with Mafijusri.
The Mafijusrimulakalpa, for example, notes that in constructing a
maJ;l<;lala for the deity" ... the great cakravartin-chief is placed at the
center. He has the colour of saffron and is like the rising-sun. He holds
a great wheel which is turning ... He is like a great king with his palace
and his decorations, a great being who is crowned and adorned with all
adornments" (Snellgrove 1959, 207). While this description may be
usefully compared to the various pictorial representations of the
Qianlong emperor as a bodhisattva (color reproductions show him in
saffron robes holding the wheel), it extends, more importantly, the range
of possible meanings for imperial interest in Tibetan Buddhism.
For example, consider some of the implications of daims that Manchu
emperors were involved in Tibetan initiation rituals (see below for
further discussion). This issue is especially important because it seems
just as plausible to assume that emperors could have achieved the sort of
political manipulations of Buddhist populations with which they are
accounts that demonstrate a lama's superior powers to those of shamans, see
Reissig 1953, 521-526.
Whether or not Tibetan Buddhism was used to check the power of
shamans, it is interesting to note that by the time of Qianlong's reign
shamanism seems to have been in serious decline. See Crossley 1987 and
1990 for imperial sponsored efforts to revive it. Rossabi (1975, 114, 118)
has argued that championing the spread of Buddhism and the suppression of
shamanism was a device used by various khans to achieve hegemony over
other Mongols. Bawden (1968, 178-179) notes a continued opposition
between Buddhism and shamanism among Mongol groups in the early part of
this century.
19. While it seems to be the case, at least in Chinese sources, that Qing
emperors did not claim to be the incarnate bodhisattva Mafijusrl, they also
seem to have done little to discourage others from making the claim on their
behalf. See Farquhar 1978.
254 JIABS 16.2
often charged simply by patronizing Buddhism from a distance. It was
not, inother words, necessary for them to participate in these rituals to
benefit from being identified with Buddhism. 20
What then could have been the motive of Manchu emperors? One
explanation may have to do with the promises implicit in the ritual tech-
nologies of some tantIic teachings. They offered the possibility of
achieving buddhahood in a single lifetime, rather than through eons of
rebirth (Snellgrove 1987,236). Of great significance in this regard was
the knowledge certain lamas commanded for the construction of
m ~ < ; l a l a s and for the initiation of others into rites that allowed them to
achieve buddhahood. Seen from this position, HongH's paean to his
father seems less problematic and later reports, such as those by Lord
Macartney, that Hongli himself had achieved buddhahood less peculiar
(see Cranmer-Byng 1963, 136,232).
By the time of the reign of the Qianlong emperor certain changes in
lama-emperor relations had occurred. The Sa skya pa sect that had close
affiliations with Nurhaci and Hung Taiji seems to have been down-
graded; in its stead was the Dge lugs pa sect. Of particular interest in
this respect was the association between the Qianlong emperor and the
Mongolian scholar and Dge lugs pa adept, Lcang skya.
The latter's
career seems worth reviewing both because of his association with tre
emperor and because it spans the period that separates the young from
the old Hongli, the earlier and later inscriptions at the Yonghe Palace.
Lcang skya studied Manchu, Chinese, and Mongolian at the court of
the Yongzheng emperor, where he became close friends with a class -
mate, the emperor's fourth son, HongH. In the early 1730s, he jour-
neyed to Tibet, studied with the Dalai Lama, and was ordained by the
Panchen Lama in 1735. In addition to placing his magical powers at the
20. In his generalizations about the nature of Buddhist rulership, Tambiah
points out that the major responsibility of the patron in a lama-patron rela-
tionship was to preserve and nourish the Three Jewels, thus creating a field in
which merit could be made by all living beings (1976, 41). This suggests
that one attribute of a cakravartin was his ability to constitute such a field in a
relationship with a lama.
21. He appears in Qianlong era Chinese sources as Zhangjia Hutuketu and
was the second incarnation, the first having been enfeoffed by the Kangxi
emperor. In some English language sources he is referred to as the "Grand
Lama of Peking." Cammann (1949-50, 10-11) says he was commonly
. known as Lalitavajra, Sanskrit for the Tibetan, Rol pa'i rdo rje. Rockhill
(1910, 47) presents Leang skya as an agent of the Panchen Lama. Also see
Turner 1800, Appendix 4; Hedin 1933,94-127; and Das 1882; 29-43.
servite of the Qianlong emperor, Lcang skya was also involved in
translating Indian commentaries and tantras from Tibetan into Mongol
and Manchu; teaching Hongli Tibetan and Sanskrit; establishing
. colleges (1744) for the teaching of philosophy, tantra, and medicine at
the YonghePalace; transmitting the fifth Dalai Lama's Sacred Word of
Mafijusri ('Jam dpal zhal lung); and acting as mediator between
Tibetans, Mongols,and Manchus. 22 Finally and perhaps inost signifi-
cantly for the subject of this study, Lcang skya bestowed upon the
Qianlong emperor tantric initiations. According to his Tibetan bio-
graphy of the Khutukhtu, on one such initiation occasion, the emperor
relinquished the highest seat to Lcang skya, knelt before him during the
consecration, and later bowed the top of his head (tingli) to the lama's
Leang skya's activities on behalf of the Qing court distinguished him.
from other incarnated beings with whom the court had dealings. The
emperor noted as much in his Pronouncements on Lamas, indicating
that Leang skya was the only lama ever entitled by the court as "Teacher
of the Kingdom" (WZTZ 1,23). The many duties and achievements of
Lcang skya, as well as his special role as the bestower of tantric initia-
tions on the emperor, highlights the degree to which Hongli was
involved in Tibetan Buddhism. Through the agency of the Lcang skya,
the emperor apparently sought to center Tibetan Buddhism within his
own rulership and patronize it with the wealth Qing emperors drew
from the Chinese part of their empire.
This brief review of Manchu affiliations with Tibetan Buddhist hier-
archs suggests a connection between such relations and the constitution
and reproduction of Qing emperorship. Far from being discrete aspects
22. On Lcang skya's life I draw primarily from Hopkins 1987, 15-35, 448-
449, The Collected Works oj Thu'u bkwan blo bzang chos gyi nyi TlUl (1969),
and Grupper 1984, who relies on Karnpfe's (1976) German translation of
Lcang skya's Tibetan biography, and Chen 1991. Also see Bawden 1968,
70, 85, 121; Chia 1992, 220-232; Jagchid 1974, 43-44, 53-54; Pozdneyev
1977,320,351-352; and RahuI1968-69, 220-221.
23. The brief description provided here of the Qianlong emperor's initiation
comes from the Tibetan chronicle of Lcang skya's life, portions of which are
cited (in Chinese translation) by Wang 1990,57-58. For a full Chinese trans-
lation of the Tibetan chronicle see Chen and Ma 1988. I am indebted to
Evelyn Rawski for bringing these sources to my attention.
24. On centering see Hevia 1986, 251-256. Chia (1992, 224-227) has argued
that the Qing court attempted to make Peking a center of Tibetan Buddhism.
I concur, but as will be discussed below with respect to the Pronouncements
on LaTlUlS, I believe Hongli's ambitions were even grander.
256 JIABS 16.2
or images of rulership, politics and religion appear to have been fused
both embedded within cosmologies. What was at issue between lamas
and emperors might be explored, therefore, in terms of both the incom-
patibilities and overlaps between competing cosmologies. To address
these issues, I want to point out certain connections between host-guest
protocols (audiences) and Tibetan initiation rites. Then, in conjunction
with the history presented above, I will reconsider a few of the enCOun-
ters between lamas and emperors in order to draw some initial conclu-
sions about the nature of relations among these personages in the 17th
and 18th centuries.
Imperial Audiences, Lama Audiences, and Tibetan Initiations
While generally treated by Euro-American historians as the site of
highly formalistic performances which merely acted out pre-existing
relations, imperial audiences, ranging from the routines of empire to the
spectacular celebrations of an emperor's birthday, might better be
thought of as constitutive of a host of relations of power which were
organized around the emperor as the pivot between the cosmos and the
Among these various audiences are those described in the
Guest Ritual (binli) section of the Da Qing tongli (juan 45-46).26 Guest
Ritual was the formal idiom or medium through which interdomainal
relations (i. e., kingdom-to-kingdom) of the kind implied in the
foregoing discussion were conducted. Like other audience rituals, it was
organized on the principle that virtuous superiors attracted to themselves
virtuous inferiors, i. e., powerful others who demonstrated Sincerity
(cheng) in the form of reverence (gong), obedience (shun), earnestness
(geng), and faithfulness (zhi), and in so doing fashioned a complex
25. Although the form and content changed over time, from at least the Tang
period audiences were codified in imperial ritual manuals (see the Da Tang
Kaiyuan li and the Ming jili). These manuals elaborated the details of an
annual cycle of rites performed by a sage ruler and his court (Zito 1984). By
the 1760's a clearly defined set of protocols differing in a number of ways
from those of the Ming period had been organized under two sections and
four chapters of the Da Qing tongli (Comprehensive Rites of the Great Qing,
1756, hereafter DQ1L). Audiences included routines of empire (changchao or
regular audience) and spectacular celebrations on the solstices, the first day of
the year, and imperial birthdays (dachao or grand audience). See DQ1L,juan
18-19 and Hevia 1986,231-250.
26. I use here the 1883 reprint of the 1824 edition of the DQ1L. The 1824
edition is especially useful for making the point that rites change (see Zito
1984, 77). Various forms of audience were usually followed at some point
by feasting, see juan 40.
imperial sovereignty (see Hevia 1989). In such a scheme, the position
of the emperor was impossible unless he acted as a completer of cosmic
initiatives; his position was similarly impossible to sustain without loyal
inferiors who actively completed his initiatives. Hence no subject
position could be constructed without the recognition and collaboration
of others. In the human world it was in and through audiences that such
political subjects were made.
Audience was one part of the routine of embassies. 27 Prior to the
audience proper, officials made certain preparations. They established
places for participants in the hall designated for audience, set out impe-
rial regalia such as banners, umbrellas, and chariots, and rehearsed par-
tiCipants. On the day of audience, the emperor took his throne, and the
guest (a lesser lord or his ambassador) and his entourage were led to a
position on the west side of the courtyard outside the audience hall
proper. There the guest performed three kneeUngs and nine head-
Ascending the west stairs of the hall, the guest proceeded
to the threshold and knelt. The emperor asked questions (usually about
the guest's health) which were transmitted by the Director of the Board
of Rites to a translator who addressed them to the guest. The guest's
response followed the same path in the reverse direction, with the
Director of the Board of Rites molding them into a memorial (daizou)
addressed orally to the emperor. Once the conversation was complete,
the audience ended. The guest and his entourage then retraced their
In special cases the emperor deemed appropriate a guest might partici-
pate in an additional rite which took place inside the audience hall. The
procedures closely parallel those outlined above, with the exception that
the guest not only entered the hall but might also be given a seat. Then
the emperor might call for a bestowal of tea. First the emperor drank the
tea, while all knelt and knocked their heads to the floor. Then the tea
was circulated, and the guest knelt in acceptance, performed one head
knocking, sat, drank, and performed an additional head knocking in
27. On the embassy routine see Wills 1984. For specific reference to the
routines of Mongol embassies during the Ming and Qing see Serruys 1967
and Chia 1992.
28. The action in question here is also referred to as kotow or kowtow in
Euro-American literature on China. While there are many forms of kneeling
and bowing the head indicated in Chinese ritual texts, the English language
usage usually indicates the one John Fairbank dubbed the full kotow, which,
as in the case here, involved three kneelings and nine head knockings.
258 JIABS 16.2
thanks for the bestovval. The emperor might then question the guest
much as before.
TIlls brief outline draws attention to the movement, placement, and
position of participants in ritual space. For example, the guest is always
oriented to the west side of courtyards, gates, and stairways leading to
and from halls. Second, the katow, when performed, OCGurs outSide
and at the foot of the stairs leading to the western door of the audience
Third, sitting in the emperor's presence was a speCial privilege
which could be enhanced by the bestowal of a tea ceremony and addi-
tional bowing, kneeling, and knocking the head to the ground within the
audience hall.
The Comprehensive Rites also contains details for lesser and varied
host-guest protocols pertaining to persons from the level of high ranking
imperial princes down to that of commoners. What is striking about
these protocols is the rigorous application of principles of movement,
placement, and bodily activities of participants, all of which form an
ensemble of actions that highlight differences between grades of people.
This section begins with meetings between imperial princes and various
ranks of princes of outer dependencies (waifanwang) enfeoffed to the
empire, and each has sections which mimic those of imperial audience.
However, as the rank of the imperial prince reduces in relation to the
rank of the outer prince, the imperial prince moves ever farther out of the
hall to greet the guest. Placement within the hall as well as the spatial
location at which the host sees off the guest also varies depending upon
differences in rank.
At the same time, imperial superiority is maintained throughout by
carefully managing the locations of partiCipants and their actions at vari-
ous moments in the rite. So, for example, when a third rank imperial
prince hosts an outer prince of the first rank, their seating positions in
the hall are reversed (host on west, guest on east). However, in the
. opposite case (first rank imperial prince and third rank outer prince), the
host takes up a position like that of the emperor, in the center of the hall,
with the guest on the west facing east (see DQTL,juan 46).
Although there do not seem to be comparable protocol manuals on
29. Numerous sources provide diagrams and pictures of regalia and its layout
for audiences. See, for example, DQHDT, juan 19-20 and Wan Yi et al.
1985, 30-45. Texts on Guest Ritual do not include the establishment of
places and other forms of preparation. Instead the reader is referred back to
the Felicitous Rites (Jiali,juan 18-19). The description of Guest Ritual pro-
vided here is from the DQTL,juan 45, 1a-4a.
audience in either Mongol or Tibetan, accounts of meetings between
lamas and others also seem concerned with the management of bodies in
ritual time and space. In his discussion of meetings between the Rje
btsun dam pa Khutukhtu and various guests, Pozdneyev reported that
the positions taken up by the Khutukhtu were predicated on the rank of
the guest. So, for example, the Khutukhtu would come out of his audi-
encehall further to greet a more senior person and they would sit facing
each other in the hall itself, while commoners prostrated at the entrance
to thehall (1978, 348-349). Similar patterns emerge in the accounts of
audiences with the Dalai Lama (Turner 1800, 333-334). Further, as in
imperial audiences, when the host takes up a position in the center of the
hall, the guest is positioned below the host and on the west side of the
hall or at the right hand of the host.
References to the actions of ritual participants in audiences with lamas
,and khu!nkhtus indicates other affinities between Tibetan and Mongol
practices and Qing audience rituals as well. Of particular interest is the
use of spatial placement to indicate differences in rank. These examples
also suggest that the ritual codes of audience were concerned with the
task of establishing seniority across discursive domains. In accomplish-
ing this task, the bodily actions and positions of ritual participants at
various times appear to be a crucial index for constituting differences
between them and establishing superiors and inferiors.
Much the same could be said about certain aspects of Tibetan initiation
Like Qing audience rituals, initiations included a period of
preparation in which the master of initiations ritually constructed a
m ~ Q a l a . i. e., a figure of an idealized palace with entryways, hallways,
throne rooms, and thrones upon which buddhas, bodhi sattv as , and
d.eities sit. Spatially, initiates are like guests in the presence of the
master of initiation who, led by him, move in and out of the m ~ Q a 1 a and
make offerings to the master and the deities. At the time of the
ceremony, the master of the initiation deploys his superior spiritual
powers in summoning the buddhas and bodhisattvas from their various
abodes in the universe and then fixing them in the m ~ Q a l a by mantra
and mudra (see Snellgrove 1987, 1,216-217 and 1959, 66-68).
Throughout the rite, bodily actions further clarify and constitute a rela-
tion of superior and inferior between master and initiate. Initiates move
through the rite prostrating and touching variously their whole bodies,
30. Space prevents a thorough explication of initiation practices. Here I draw
for my main points from Snellgrove 1987, 1, 213-303 and 1959.
260 JIABS 16.2
faces, foreheads, the tops of their heads, and their mouths to the ground,
during which they pronounce mantras, and before and after which they
perform specific mudra learned from their master. At one point, the
master becomes a buddha charging the initiate to secrecy (Snellgrove
An example of a portion of an initiation may help to clarify the power-
ful relationships organized by these rites. An especially pertinent ritual
reported by Snellgrove involves the consecration of universal
sovereignty found in the MafijusrimalakaZpa. The relevant passage
notes that beginning from a position on the west "facing east and look-
ing toward the mal).<;1ala," pupils render their master royal honors, treat-
ing him as if he were MafijusrI the prince (or perhaps a supreme lord
like an emperor?). They spread a great canopy, set out flags, banners of
victory, and hold a white parasol over his head, while waving white fly-
whisks. They then ask the master if they too may become a buddha
(Snellgrove 1987, 1,226-227).
Here we see mimetic relationships established between worldly
sovereignty and cosmic sovereignty. As such this initiation rite makes
claims for establishing a relationship between cosmic infinitude and
transient human life as strong as those made in the ensemble of imperial
rites that fashion a relationship between the cosmos (tian), the son of the
cosmos (tianzi), and his kingdom (guo). As he guides a pupil through
initiations, the master as a buddha binds the pupil to himself and to the
cosmic technologies that make the latter's transformation into a buddha
possible. These are powers which ought to awe lords of the mere earth.
That initiation masters were supposed to possess such powers should be
borne in mind when we learn, for example, that lamas and khutukhtus
initiated Qing emperors from Nurhaci to Qianlong into tantras. Put
simply, when the cosmological logic of ritual is borne in mind, the pow-
ers of the lama rival those deployed in the imperial Grand Sacrifice to
the Cosmos or in court audience for the constitution of imperial
Encounters and Inscriptions
I began this paper with an assertion that the two inSCriptions of the
Qianlong emperor at the Yonghe Palace marked two poles of a contra-
diction, signifying ambivalence on the part of early Qing emperors
toward Tibetan Buddhism. The brief history of imperial involvement in
Tibetan Buddhism and the discussion of ritual presented above suggest
some potential problems in lama-emperor relations. All parties made
various claims to pre-eminence; no one could completely ignore the
claims of the others. Nothing highlights these political realities more
than the contradictory accounts of meetings between Qing emperors and
various Buddhist hierarchs from Inner Asia. What these accounts tend
to show is that while the Qing court did at times defer to Tibetan
Buddhist hierarchs, increasingly over the course of the eighteenth-
century Manchu emperors asserted supreme-lesser lord, rather than
lama-patron relations in their intercourse with Tibetan lamas and
Mongol khutukhtus. In the face of these Qing hegemonic gestures,
lamas and khutukhtus attempted to retain the high ground of spiritual
After the establishment of the Qing dynasty in China in 1644, and
well before the Manchus asserted hegemony over Tibet, the first signifi-
cant encounter between a Qing emperor and a lama occurred when the
fifth Dalai Lama journeyed to Peking in 1653. The court of the Shunzhi
emperor was split over where the lama should be received. Thinking that
it might be a useful way for winning over Mongol groups who had yet
to submit to Manchu overlordship, the emperor's Manchu advisors
thought it wise to meet the lama in Mongolia. His Chinese councilors
objected, arguing that cosmic portents indicated that the lama sought to
challenge the emperor's supremacy. In keeping with the spatial princi-
ples of imperial ritual, therefore, ifhe left his capital and went to
MongOlia, he would be acknowledging the lama's superiority (SZZSL,
68, Ib-3a, 31b). The emperor decided to give audience in Peking, but
with certain modifications that vary from guidelines to be found in ritual
The Veritable Records (Shilu) of the Shunzhi emperor of
January 14, 1653 notes that the Dalai Lama arrived and visited (ye) the
emperor who was in the South Park. The emperor bestowed on him a
seat and a feast. The lama brought forward a horse and local products
and offered them to the emperor (SZZSL 70, 20a-b).
The differences in question include the holding of the audience in the
large park to the south of Peking rather than in one of the outer palaces
of the imperial city32 (ritual manuals suggest the Supreme Harmony
31. The DQ1L would, of course, not be collected and edited for another cen-
tury, so it may seem odd to s p e ~ here of deviations. The point, however, is
that this particular encounter is different from those outlined in other ritual
manuals as well as various accounts of audiences at the Qing court that also
occurred before the above text was compiled.
32. The South Park referred to here is probably the Nanhaizi or Nanyuan
262 nABS 16.2
Hall) and the fact that the audience was characterized as a visit (ye),
rather than as a "summons to court" (zhaojian), the usual form for
recording such events in the Veritable Records. In the latter case, while
ye connotes a visit from an inferior to a superior, I believe it suggests
some sense of deference in this context. On the other hand, certain
things were done in accordance with imperial audience as outlined in
other sources such as the Ming and Qing ritual manuals. The emperor
bestowed a seat and a feast on the lama. The lama, like other loyal infe -
riors, made offerings of local products (fangwu).
If this entry on audience appears anomalous when compared to impe -
rial audience protocols, the account of the same audience in the auto-
biography of the fifth Dalai Lama is even more unusual. While he does
not mention the site at which the audience took place, the lama claims
that the emperor descended from his throne, advanced for a distance of
. ten fathoms and took his hand. The lama also reports that he sat in audi-
ence on a seat that was both close to the emperor and almost the same
height. When tea was offered, the emperor insisted that the lama drink
first, but the lama thought it more proper that they drink together. On
this occasion and over the following days the lama recorded that the
emperor gave him numerous gifts fit for a "Teacher of the Emperor"
(dishi). The emperor is also said to have requested that the Dalai Lama
resolve a dispute between two other lamas. On his return trip through
Mongolia to Tibet, the lama displayed the presents given by the emperor
and appears to have distributed some of them along the route. 33
haizi, located outside the south wall of Peking. Apparently used as a hunting
park by the Manchu court, it can still be seen on maps from the early part of
this century (see Clunas 1991, 46). I am indebted to Susan Naquin for this
One cannot help wondering if the solution to the problem posed by the
lama's visit might help to explain the use of other sites around Peking to
address relations with Inner Asian lords. The example of the Ziguang pavil-
ion to the west of the main audience halls of the "Forbidden City" is well
known, but audiences and feasts also might take place at the Yuanming yuan.
TheDQHDT, 1818 edition,juan 21, 6a-7a, diagrams a feast in a round tent
at the Yuanming yuan.
Holding audiences outside the main halls of the palace for problematic
guests continued through the end of the dynasty. Between 1870 and 1900,
no American, or Japanese ambassador was received in the Supreme
Harmony Hall. They were hosted at the Ziguang pavilion or other balls, see
Rockhill 1905.
33. The Dalai Lama's account is taken from Ahmad 1970,175-183, who
relies on the autobiography of the fifth Dalai Lama, v. 1, 197a-198b. See his
reference to this source on p. 340. On the dispute the lama was asked to
What is especially interesting about these two accounts is not simply
that they differ, but that the dimensions along which they diverge
involves ritual practice. The imperial records mention the lama's offer-
. ings to the emperor, all of which may be construed as his acceptance of
a position of inferiority. The lama's account emphasizes offerings made
by the emperor to him and includes many examples of the emperor
deferring to the lama as a person of superior spiritual insight. The
imperial records solved the problem of a meeting with an important and
potentially dangerous personage by shifting the location to one outside
the imperial audience hall complex proper. The lama's account empha-
sizes that the emperor came down from his throne to greet him, an act of
considerable deference.
A similar pattern of divergent accounts emerges in connection with
meetings between the Shunzhi emperor's successors and the Rje btsun
dam pa Khutukhtu. Here too the court seemed willing to accord a
degree of deference to the Khutukhtu, while still working to establish a
supreme lord-lesser lord bond. So, for example, at the famous submis-
sion of the Khalkha Mongols to the Qing at Dolonnor in 1691, the
Veritable Records indicates that when the Kangxi emperor received the
Khutukhtu in an audience on May 29, the Khutukhtu knelt (gui) before
the emperor. The emperor bestowed tea and other gifts on the
Khutukhtu. The next day another audience was held for other members
of the Khalkha nobility; they performed three-kneelings and nine-head-
knockings (sangui jiukou). 34 At the same time, all of the activities that
occurred at Dolonnor were catalogued under the general rubric for clas-
sifying relations between the supreme lord and lesser lords, i. e.,
"cherishing men from afar" (huairou yuanren, see SZRSL 151, 23a). It
appears, therefore, that the Khutukhtu assumed the position of a loyal
inferior, but one who was in some way differentiated from the
remainder of the Khalkha nobility.
On its side, Mongol versions of encounters between the emperor and
the Rje btsun dam pa Khutukhtu closely parallel in form the Dalai
Lama's version of his meeting with the Shunzhi emperor, a pattern
which continued into the Qianlong era (see Bawden 1961,49-60 and
resolve see Heissig 1953, 528.
34. The Khutukhtu appeared in the first audience with the Tiisiyetii Khan
who was also recorded as kneeling. In the entry for the following day, how-
ever, the Khutukhtu is not mentioned, only Khalkha Khans and ranks of
nobles, which would presumably include the Tiisiyetii Khan, see SZRSL,
151, 8a, lOa.
264 nABS 16.2
Pozdneyev 1977, 332-336). In 1737, for example, the second Rje btsun
dam pa Khutukhtu journeyed to Peking, where he was met and honored
by high officials and lamas at the Anding Gate. When he arrived at his
quarters, the Qianlong emperor met him. Upon seeing the emperor, the
Khutukhtu knelt, but the emperor insisted he not do so. Later in an
audience that included a tea bestowal, the emperor asked the Khutukhtu
to sit closer and higher than other guests (Bawden '1961, 71 and
Pozdneyev 1977,341). In addition, the Qianlong emperor lavished gifts
on him and acknowledged his powers. 35
Much the same sort of conflicting presentation occurred when the
Panchen Lama visited Rehe and Peking in 1780. According to the
lama's account, the emperor left the throne and greeted him at the door
to the reception hall. Taking his hand, the emperor led him to the throne,
where the two sat facing each other and "conversed as intimate friends."
Later the emperor visited the lama at the special residence that had been
prepared for him, a reproduction of the Panchen's palace at
Tashilhunpo, and sought his teachings. Banquets and gift giving
followed over the next several days. Various sources claim that during
his stay the lama initiated the emperor into the Mahakala and
Cakrasarp.vara tantras.
Here again the lama is cast as teacher, the
emperor as patron and pupil.
The Veritable Records provides quite another point of view, one that
differs from both the Tibetan account and the Veritable Records' version
of the visit of the fifth Dalai Lama discussed above. In these records the
emperor summoned the lama to audience (zJ1aojian) in the Yiqingkuang
Hall at Rehe. Three days later the lama was again summoned to the
round tent in the Garden of Ten-thousand Trees (W anshou yuan), where
Inner Asian lords of various ranks looked on while the emperor
bestowed caps, gowns, gold, silver, and silk on the lama (GZCSL 1111,
35. Pozdneyev dates the visit as summer 1736, Bawden, 1737. I have found
no indication of an audience for either year in the Veritable Records.
However, there is an entry for a banquet held on the ninth day of the first
lunar month (February 27, 1738) in a large tent at the Fertile Abundance
Garden (Fengze yuan), located between the Middle and Southern Lakes in
Peking with the Rje btsun dam pa and Lcang skya Khutukhtus in attendance,
see GZCSL, 60, 8a. As Susan Naquin pointed out to me, the date of the
banquet is significant, because special receptions of foreign "tributaries" took
place during the new year celebrations.
36. I follow Das's translation from an abridged version of the Panchen
Lama's life, see 1882, 39c42. On the initiations see Das and Grupper 1984,
59. Also see Cammann 1949-1950 on the lama's visit.
4a arid lOa-b). 37 While these audiences constitute the encounter as one
between the supreme lord and lesser lords, the lama was differentiated
from the various Inner Asian lords looking on, much' as the Rje btsun
dam pa Khutukhtu had been at Dolonnor. According to a directive in
the Rehe zJzi (Rehe Gazetteer) the lama was allowed to kneel (gui) before
the emperor instead of bowing (bai), provided he was sincere (cheng). 38
There was another sort of deference that may have occurred at Rehe as
well. According to a diagram to be found in the 1818 edition of Da
Qing hui dian tu (Diagrams of the Collected Statutes of the Great Qing,
Qereafier DQHDT,juan 21, 7a), during feasts held at the round tent in
the Garden of Ten-thousand trees, khutUkhtus and lamas were seated
closer to the emperor than Mongol nobles. 39 This is the sort of distinc -
tion a supreme lord could make when cherishing men from afar.
These records indicate that conflicting and contradictory accounts of
the signifying practices (i. e., movement in time along east-west and
high-low axes, as well as bowing, kneeling, and enunciating) of ritual
participants were not uncommon when lamas and emperors met. Such
differential presentations of bodily practices tell us much about tre
efforts of Manchu emperors and Buddhist hierarchs to incorporate each
other as sub lords, patrons, or pupils. Even when honoring lamas and
"alteitng audience protocols for them, the Qing court insisted that they
were recipients of imperial grace (en), making it quite clear, at least by
37. Other occasions of feasting and bestowal followed, including one in the
Preserving Harmony ball (Baohe dian) at Peking on October 29, 1780
(GZCSL, 1112, 17b-18a and 1116, 4a).
38. Rehe zhi, 24, lOb. The reason given in this case for allowing the lama
to kneel was that it was customary in Buddbism to bow (bai) only to the
Buddha. This particular reference to respect for the customs of others was not
unusual. It is evident, for example, in the negotiations over the form of audi-
ence during the Macartney embassy to Cbina, wbere again the issue was the
sincerity of the act (see ZGCB, 3, 20b and Hevia 1989). It is also present in
the instructions to the imperial envoy, Songyun, before his departure to Tibet
in 1795. In order to accord with the teachings of the Yellow sect, be was
ordered not to bow his bead to the ground (koubai) before the Dalai Lama, see
GZCSL, 1458, 34b-35a. Tbese various examples suggest that interpretations
of the "kowtow" ougbt to be re-evaluated, beginning with the tribute system
version and Levinson's modification ofit (see 1968, 2, 68-69).
39. Also see Wan et al. 1985,76-77 and 288 for two paintings of banquets,
one at the Ziguang pavilion and the other at the Garden of Ten-thousand Trees
in Rebe. Tbese pictures show lamas positioned bigber and closer to the
emperor than other Inner Asian dignitaries. Space prevents a more thorougb
consideration of feasts, but clearly seating at imperial banquets was another
way in wbicb bierarchy was constituted.
266 JIABS 16.2
the time of the Qianlong reign, that the lama was a loyal inferior of the
supreme lord. In contrast, Tibetan and Mongol accounts seem con-
cerned with the superior knowledge or expertise oft!Ie lamas relative to
that of their imperial hosts as well as with specific acts of bodily practice
that differ from those described in imperial ritual manuals. They also
tend to construct the emperor as an offerer of gifts, and hence as a
devotee/pupil, and the lama as receiver of alms.
The Pronouncements on Lamas and a SenSe Of an Ending
It is within the context of these competing accounts of meetings between
lamas and emperors that we might now consider the second of the two
inscriptions at the YonghePalace, Hongli's Pronouncements on Lamas.
Crucial to an understanding of this essay is the fact that the emperor
poSitions himself as the ultimate authority on matters involving lamas
and khutukhtus. After reviewing the history of interaction between
Tibetan lamas and previous dynasties, Hongli asserted that the Qing had
never used the title "Teacher of the Emperor" (dishi) , 40 only the title
"Teacher of the Kingdom" (guoshi),. and that, as noted above, exclu-
sively in the case oftheLcang skya Khutukhtu (WZTZ 1,23).
Second, in spite of evidence to the contrary such as that discussed
above, Hongli pointed out that while the Qing dynasty acknowledged
the importance of the Dalai and Panchen Lamas, it did so only because
the Yellow sect (Dge lugs pa) was important to the Mongols. He added
that the two lamas had submitted to Manchu lordship by making offer-
ings of local products to the court in 1642 and had subsequently
received titles and seals from the Qing dynasty. For their part, emperors
accepted the submission of the lamas as they did any other lords-they
were bound to obey the injunction of the cosmos (tian) to follow the
path of cherishing and showing kindness to men from afar (huairou zhi
dao, WZTZ 1,24) .
. Third, after questioning the very notion of incarnate (zhuanshi) lamas,
Hongll proceeded to reorganize the selection process of the Dalai Lama
and the Rje btsun dam pa Khutukhtu. Thoroughly criticizing what he
saw as a selfish (si) monopoly of certain Tibetan and Mongol clans over
succession, the emperor decreed in the name of his own disinterested-
ness (gong) that in the future the names of potential incarnates would be
40. On the origins of this tenn and for citations of the relevant bibliography
concerning its history see Dunnell 1992. On the Yuan use of the tenn see
Franke 1981.
placed in a golden urn and, under the eyes Qfhis observers, determined
by lot (WZIZ 1,24-25).41 The emperor had Tibetan, Mongol, Manchu,
and Chinese versions of the Lamashuo produced. Apparently they
were widely disseminated.
The importance of the Pronouncements on Lamas to issues raised
throughout this essay can be summarized as follows. By establishing
that the Qing dynasty had never considered lamas to be teachers or
emperors their students, the Qianlong emperor effectively rejected any
claims of spiritual superiority lamas might make in a relationship with a
reigning emperor. In invoking an emperor's responsibilities to the cos ~
mos, Hongli displa<;ed any claims lamas made concerning their own
constitutive agency as "teachers" onto the reality of an immanent cosmos
and its earthly son, the emperor. Finally, in casting the relationship in
terms of a hierarchy of rulership, the emperor effectively refuted
Buddhist constructions of interdomainal relations as ones between lamas
and patrons. In this context, the Lamashuo may be interpreted not only
as a defense of imperial policy (Lessing 1942, 62), but as a gesture at
closure; an attempt to halt Tibetan or Mongol Buddhist statements that
privileged lama-patron relations over any other kind. The efforts of
Qing emperors to end "assertions by lamas of superiority are epitomized
best perhaps in the refusal by Hongli and his predecessors to award the
title "Teacher of the Emperor" to a lama. They are also embodied in the
imperial-sponsored construction of alternative sites for the practice of
Tibetan Buddhism and in Hongli' s insistence that the two most impor-
tant lamas of Tibetan Buddhism had submitted to Manchu ()verlordship
41. It is difficult to discern conclusively if in fact the policy initiated by
Hongli was carried out. For an account of the selection process in Tibet see
Waddell 1895, 248-251. If it had been employed to select the Dalai Lama, it
would have affected the ninth through the thirteenth incarnations. Goldstein
(1989, 44) notes that the thirteenth was not selected this way in 1879, but
makes no reference to previous incarnations. According to Shakabpa, the sys-
tem was not used for the selection of the ninth Dalai Lama (1806); it seems
to have been used for the selection of the eleventh (1841) and twelfth (1858);
and may have been used for the selection of the tenth Dalai Lama (1822); see
1967, 172, 174-76, and 183. It is also clear from Shakabpa's account that the
use of the system was a volatile political issue in Tibet (186). Richardson
(1984, 71) notes only that the decree was ignored in 1808, but insisted upon
by 1818 when the tenth Dalai Lama was being sought. In the latter case the
child already selected by the Tibetans was subsequently chosen by lot.
Hongli's alteration of the process by which incarnations of the Dalai Lama
and the Rje btsun dam pa Khutukhtu were selected continued a pattern of
imperial intervention in such processes. See Bawden 1968, 132-133 and
Rossabi 1975, 156-157.
268 JIABS 16.2
before tlle dynasty was established in China ..
Concluding Comments
Like the stone inscriptions at the Yonghe Palace with which I began, the
discussion presented here concerning audiences, initiation rituals, and
encounters between emperors and Buddhist hierarchs maps difference
and ambiguity, challenging efforts to reduce complicatedpriliticalrela-
tionships to thetirheless regularities of cultural essences. How are we to
reconcile or histOrically address the contrary accounts of lama-emperor
engagement? What are we to make of audience protocols and alterations
to them; of the interest of Manchu emperors in Tibetan initiations and in
titles such as cakravartin king; of the occasional imperial endorsement of
the Sa skya pa as opposed to the Dge lugs pa sect of Tibetan Buddhism;
or of imperial action which seemed at times to countenance arid at other
times oppose Manchu shamanism? What of the triangular relationship
between the Qing emperor, the Dalai Lama, and the Rje btsun dam pa
Khutukhtu, in which each was presented as a human incarnation of a
celestial bodhisattva? There are no clear and easy answers to these
questions because, among other things, to do them justice would require
a wider reading in Mongol, Tibetan, and Manchu sources, as well as in
archival materials still extant in China, than has yet been attempted.
Moreover, there are simply too many instances of ideas, concepts, prac-
. tices, and terminologies flowing among these groups and being appro-
priated and re-deployed within political struggles to continue to warrant
dealing with Manchus, Tibetans, and Mongols as unitary and exclusive
national entities. It may be necessary to reconceptualize sovereignty in
terms other than those which map ethnicity and culture over territory
(thus producing the requisite ingredients in a nation-state construction).
Such an imperative is only intensified by the fact that Manchu emperors,
lamas, and khutukhtus appear to have vied with one another for
supremacy on the basis of cosmological principles, ones which could
neither be ignored nor completely captured and incorporated into one
over-arching cosmology.
The Dalai Lama's account of his meeting with the Shunzhi emperor
acknowledges, for instance, that the lama was still given a seat below the
height of .that of the emperor. This statement would seem to clearly sig-
nal an acceptance of an i ~ e r i o r status. Yet at the same time, there is
little doubt that the lama also asserted superiority in that he suggested a
lama-patronrelationshlp. Such subtle negotiations continued right into
the Qianlong era and only appear to end with the Pronouncements on
Lamas. Indeed, the fact that it is so difficult to pin down whether
Tibetans and Mongols ever actually complied with Hongli's alteration of
the selection process of reincarnates suggests as much (see note 41).
Problems posed by such indeterminacy are only partly a function of
conflicting accounts. They are also a product of the very formation of
imperial sovereignty. Earlier I argued that the main feature of audience
ritual was the constitution of such sovereignty, which by definition was
hierarchical in nature. In the past, there has been a tendency to fix atten-
tion on the superior in this relationship, assuming a notion of power that
is fundamentally oppressive and instrumental. It seems to me, however,
thatit makes just as much sense, after Foucault (1980), to view imperial
power as productive, to see the specific relationship formed in superior-
inferior relations as a joint construction which empowers the latter as
well as the former. Audiences below the level of the emperor particu-
larly emphasize the constitutive nature of power. Assuming the position
of a superior, the emperor's servants themselves became "hosts" to vari-
ous "guests" in audiences, addressing and forging relationships with
others of the imperial polity. In this way the Manchu imperium
extended the emperor's virtue (de, Waley's "power of the exemplar")
globally, ordering the world in a specific way. For their part, lamas
appear to have spread Tibetan Buddhism through recognizing powerful
others (particularly military powers) as cakravartins and incarnate bod-
hisattvas, as well as through the performance of initiation rituals. In
these ways they asserted their superiority over lords of the world.
These considerations of power draw on certain insights which emerge
from reading across the sources considered. The first of these is that
"common sense" divisions between religion and politics and between
ritual and "bureaucratic" routines obscure rather than explain these polit-
ical relations. Second, a functionalist understanding of ritual as integra -
tive and productive of community solidarity cannot adequately account
for conflict or contradictory gestures within ritual action.42 Third, ritual
action is as historically contingent and politically significant as the
Qianlong emperor's stone inscriptions at the Yonghe Palace and ought
to be treated as such, rather than as an aspect of the residual category of
culture within political and social histories (see Farquhar and Hevia
42. The .classic articulation of this notion of ritual can be found in van
Gennep 1960, first published in 1909.
270 JIABS 16.2
The Tibetan and Mongol accounts cited here may also be viewed as
specific resistances to and strategies for deflecting the hegemonizing
practices of the Qing imperium. They nevertheless seem to be organiZed
through different metaphysical assumptions, different views of the
nature of reality, and, more than likely, differing views of just how
bodies can have signifying capabilities. Yet what they share is equally
important. For both lamas and emperors, meetings appear to have been
a kind of pivot at which asymmetrical hierarchies were faShioned, in
which the present and future were significantly addressed, and in which
bodily action constructed highly consequential relationships.
The disposition of bodies and the organization of ritual space were
about who was actually submitting to whom, with the mutual recogni-
tion that such submission had wide political consequences. Yet, since
participants vied to hierarchize each other in audiences, submission was
a complicated affair. On their side, Manchu emperors wanted lamas to
offer themselves sincerely to the emperor; that is, to accept loyally a
position of an inferior in a relationship with the supreme lord. For their
part, lamas wanted emperors to humbly accept a position as patron and
pupil of the lama. I do not think it would make much sense to either
party for submission in such relations to be coerced. I suggest, there-
fore, that at least on the Qing court's side, meetings between lamas and
emperors were about constructing scales of sincere (cheng) loyalty. 43
Participants scrutinized the bodily movements of others as outward
signs of inner conditions in an effort to determine whether verbal state-
ments or other kinds of action (such as gift giving), all of which pre-
sumably manifested loyalty and submission, were indeed sincere.
It is not clear to me the extent to which the parties involved in the
encounters I have presented here were aware of each other's construc-
tion of events, but it seems highly unlikely that the Qianlong emperor,
for example, did not have some inkling of how lamas might present their
meetings with him to others. In this respect, the Pronouncements on
Larrws might be read as an assertion that lamas could not be counted on
to be loyal inferiors, i. e., to bring to completion the emperor's initia-
tives. The form the Qianlong emperor selected for determining incarnate
lamas makes this point-it is a parody of a pre-existing Tibetan selection
43. In accounts of embassies found in various imperial court records, the
emperor invariably reminded his officials to evaluate the sincerity of em-
bassies. See, for example, the ZGCB on the Macartney embassy and Hevia
1986 and 1989.
process which looked for various signs on a child indicating advanced
progress on the path to buddhahood. By introducing a lottery, the
emperor declared that the Tibetan Buddhist beliefs associated with
human incarnations of enlightened masters were about as conclusive as
a game of chance.
The Qianlong emperor's casting of the relationship between the Qing
court and the Yellow sect in terms that privilege hierarchies of lords
over hierarchies of spiritual powers makes, I would argue, the concerns
of the Manchu court easier to understand. Lama hierarchs posed a threat
because they challenged the very premises upon which an encompassing
imperial sovereignty was grounded. That is, they embodied a competing
and equally powerful hierarchical view of the cosmos that placed them
above the multitude of earthly lords, even if the latter be patrons.
Moreover, if Tibetan lamas had been able reliably and consistently to
incorporate Manchu emperors as pupils, then any claims emperors made
in Inner Asia to supreme lordship could be challenged on cosmological
grounds. Lamas were also dangerous because they had the potential for
confusing the loyalties oflesser lords, such as Mongol Khans. Yet the
problems lamas posed to imperial sovereignty were not easily resolved
(cf. Ruegg 1991,451). While the Qianlong emperor might have paro-
died incarnation, he remained profoundly interested in Tibetan
Buddhism, patronized it, and seemed to have had little trouble with
being treated by many as a incarnated bodhisattva. And if Hongli and
other emperors were interested in tantric initiations, who is to say that
they might not have seen them as one among other ways of fulfilling
their cosmological responsibilities in a Manchu (as opposed to a
Chinese) empire?
Seen from this position, the temporal and discursive distance posited
earlier between the stone inSCriptions at the Yonghe Palace seems less
dramatic. For the point is that the Pronouncements on Lamas is not so
much directed at Tibetan Buddhism per se, as it challenges what the
emperor saw as abuses by the Tibetan monastic nobility. These efforts at
reform were given material density by the placement of the Lamashuo
inscription. Rather than flanking the central path running south-north
through the Yonghe Palace, as the 1744 inscription does, the
Pronouncements on Lamas was placed at a more inner location directly
on the center path. Hongli here centers and encompasses his authorita-
tive version of Tibetan Buddhism deep within the Qing imperium.
44. See the diagram of the palace complex in Lessing 1942.
272 JIABS 16.2
Ahmad, Zahiruddin. 1970. Sino-Tibetan Relations in the Seventeenth
Century. Rome: Istituto Italiano Per il ed Estremo Oriente.
Arlington, L. c., and W. Lewisohn. 1935. In Search of Old Peking. Reprint
ed. Hong Kong: Oxford University Press, 1991.
Bhattacharyya, Benoytosh. 1980. An Introduction to Buddhist Esoterism.
Delhi: Motilal Banarsidass.
Bawden, C. R., ed. and trans. 1961. The Jebtsundamba Khutukhtus of Urga.
Wiesbaden: Otto Harrassowitz.
_____ . 1968. The Modern History of Mongolia. London:
Weidenfeld and Nicolson.
Bell, Catherine. 1992. Ritual Theory, Ritual Practice. Oxford: Oxford
University Press.
Bourdieu, Pierre. 1977. Outline of a Theory of Practice. Cambridge:
Cambridge University Press.
Cammann, Schuyler. 1949-50. "The Panchen Lama's Visit to China in 1780:
An Episode in Anglo-Tibetan Relations." Far Eastern Quarterly 9:3-19.
Certeau, Michel de. 1984. The Practice of Everyday Life. Translated by
Steven Rendall. Berkeley: University of California Press.
Chia Ning. 1992. The Li-fan Yuan in the Early Ch'ing Dynasty. Diss. Johns
Hopkins University.
Chen Qingying. 1991. "Lcang-skya Rolpavi-rdorje and Emperor Qian Long."
Theses on Tibetology in China. Compiled by Hu Tan. Beijing: China
Tibetology Publishing House. 67-90.
Chen Qingying, and Ma Lianlong. 1988. Zhangjia guoshi Robidoji zhuan
(The Biography of "Teacher of the Kingdom" Lcang skya Rol pa'i rdo Ije).
Peking: Minzu chubanshe.
Clunas, Craig. 1991. "Whose Throne Is It Anyway? The Qianlong Throne in
the T. T. Tsui Gallery." Orientations 22.7: 44-50.
The Collected Works of Thu'u bkwan blo bzang chos gyi nyi rna. 1969.
Introduction by E. Gene Smith. Vol. 1. New Delhi: Ngawang Gelek
Comaroff, John, and Jean Comaroff. 1992. Ethnography and the Historical
Imagination. Boulder: Westview Press.
Cranmer-Byng, J. L. 1963. An Embassy to China. Hamden, CT: Archon
Crossley, Pamela. 1987. "Manzhou yuanliu kao and the Formalization of the
Manchu Heritage." Journal of Asian Studies 46.4: 761-790.
______ . 1990. Orphan Warriors. Princeton: Princeton University
Da Tang Kaiyuan li (The Rites of Kaiyuan Court of the Great Tang). 732.
Reprinted in Siku quanshu (Complete Library in Four Branches of
Literature). Series 8. Vols. 99-108. Taipei: Shengyu yinshuguan, 1978.
DQHDT. Da Qing huidian tu (Diagrams of the Collected Statutes of the
Great Qing). 1818. Ed. Tuo Jin et al. Peking: Palace Edition.
DQTI... Da Qing tongli (Comprehensive Rites of the Great Qing). 1883.
Reprint of 1824 re-edited edition. Peking: Palace Edition.
Das, Sarat Chandra. 188l. Contributions on the Religion and History of
Tibet. Reprinted. New Delhi: Manjusri Publishing House, 1970.
_-----. 1882. "Contributions on Tibet." Journal of the Asiatic
Society of Bengal 51: 29-43.
Dumont, Louis. 1970. Homo Hierarchicus. Chicago: University of Chicago
Dunnell, Ruth. 1991. "The Hsia Origins of the Yuan Institution of Imperial
Preceptor." Asia Major 4.2: 85-111.
Elliot, Mark. 1992. "Turning a Phrase: Translation in the Early Qing through
a Temple Inscription of 1645." Aetas Manjurica. Ed. Gimm, Martin,
Giovanni Stary and Michael Weiers. Vol. 3. Wiesbaden: Otto
Harrassowitz. 12-41.
Elman, Benjamin. 1984. From Philosophy to Philology. Cambridge, MA:
Harvard University Press.
Esherick, Joseph. 1987. The Origins of the Boxer Uprising. Berkeley: Uni-
versity of California Press.
Fairbank, John K. 1942. "Tributary Trade and China's Relations with the
West." The Far Eastern Quarterly 1: 129-149.
______ . 1948. The United States and China. Cambridge: Harvard
University Press.
______ . 1968. "A Preliminary Framework." The Chinese World
Order. Ed. John K. Fairbank. Cambridge: Harvard University Press. 1-19.
Fairbank, John K., and Teng, S. Y. 1941. "On the Ch'ing Tributary Sys-
tem." Harvard Journal of Asiatic Studies 6: l35-246.
Farquhar, David M. 1978. "Emperor as Bodhisattva in the Governance of the
Ch'ing Empire." Harvard Journal of Asiatic Studies 38: 5-34.
Farquhar, Judith, and James Hevia. 1993. "Culture and Postwar American
Historiography of China." Positions 1.2.
Fletcher, Joseph. 1978a. "Ch'ing Inner Asia c. 1800." The Cambridge
History of China. Ed. D. Twitchett and J. K. Fairbank. London:
Cambridge University Press. 10, 1: 35-106.
______ . 1978b. "The Heyday of the Ch'ing order in Mongolia,
Sinkiang and Tibet." The Cambridge History of China. Ed. Twitchett and
Fairbank. 10, 1: 351-406.
Foucault, Michel. 1980. Power/Knowledge. New York: Pantheon.
274 JIABS 16.2
Franke, Herbert. 1978. From Tribal Chieftain to Universal Emperor and
God.. Munich: Verlag Der Bayerischen Akademie Der Wissenschaften.
______ . 1981. "Tibetans in Yiian China." China Under Mongol
Rule. Ed. John D. Langlois, Jr.. Princeton: Princeton University Press.
Franke, Otto and Berth()ld Laufer. 1914. Epigraphische Denkmtiler aus
China 1. LamaistischeKlosterinschriften aus Peking, lehol, and Si-ngan.
Berlin: n. p.
GZCSL. Gaozong chun huangdi shilu (The Veritable Records of the
QianlongEmperor). 1964.Taipei: Huawen shuju.
Gimello, Robert M. 1992. "Chang Shang-ying on Wu-t'ai Shan." Pilgrims
and Sacred Sites in China. Ed. Susan Naquin and Chiin-fang Yii.
Berkeley: University of California Press. 89-149.
Goldstein, Melvyn. 1989. A History of Tibet, 1913-1951. Berkeley:
University of Califomia Press.
Grupper, Samuel M. 1980. The Manchu Imperial Cult of the Early Ch'ing
Dynasty. Diss. Indiana University.
______ .1984. "Manchu Patronage and Tibetan Buddhism DUring
the First Half of the Ch'ing Dynasty." The lournal of the Tibetan Society
4: 47-75. .
Guy, R. Kent. 1987. The Emperor's Four Treasuries: Scholars and the State
in the Late Ch'ien-lung Era. Cambridge: Harvard University Press.
Hedin, Sven. 1933. lehol City of Emperors. New York: E. P. Dutton.
Heissig, Walter. 1953. "A Mongolian Source to the Lamaist Suppression of
Shamanism in the 17th Century." Anthropos 48: 1-30; 493-537.
__ -,-___ . 1980. The Religions of Mongolia. Trans. Geoffrey
Samuel. Berkeley: University of Califomia Press.
Hevia, James. 1986. Guest Ritual and Interdomainal Relations in the Late
Qing. Diss. University of Chicago.
______ . 1989. "A Multitude of Lords: Qing Court Ritual and the
Macartney Embassy of 1793." Late Imperial China. 10.2: 72-105.
______ " Forthcoming. "Sovereignty and Subject: Constructing
Relations of Power in Qing Imperial Ritual." Body, Subjectivity, and
Power in China. Ed. Angela Zito and Tani Barlow. Chicago: University of
Chicago Press.
Hopkins, Jeffrey. 1987. Emptiness Yoga. Ithaca: Snow Lion Publications.
Hummel,. Arthur. 1943. Eminent Chinese of the Ch'ing Period. Washington:
United States Govemment Printing Office.
Inden, Ronald. 1990. Imagining India. Oxford: Blackwell.
Ishimama Yumiko. 1992. "A Study of the Seals and Titles Conferred by the
Dalai Lamas." Tibetan Studies: Proceedings of the 5th Seminar of the
. International Association for Tibetan Studies, Narita 1989. Narita:
Naritasan Shinshoji. 2: 501-514.
Jagchid, Sechin. 1974. "Mongolian Lamaist Quasi-Feudalism During the
Period of Manchu Domination." Mongolian Studies 1: 27-54.
Jochim, Christian. 1980. Imperial Audience Ceremonies of the Ch'ing
Dynasty. Diss. University of Southern California.
Kahn, Harold. 1971. Monarchy in the Emperor's Eyes. Cambridge: Harvard
University Press. .
Kampfe, Hans-Rainer. 1976. Ni ma'i'od zerlNaragan-u gerel: Die Bio-
graphie des 2. Pekinger Leait skya-Qutuqtu Rol pa'i rdo rge (1717-1786).
Monumenta Tibetica Historica, Abt. II, Bd. 1. Sankt Augustin:
Kuhn, Phillip. 1990. Soulstealers. Cambridge: Harvard University Press.
Lamotte, Etienne. 1960. "Manjusrl." T'oung Pao 48.1-3: 1-96.
Lattimore, Owen. 1962. Studies in Frontier History. London: Oxford Uni-
versity Press.
La Vallee Po us sin, Louis de. 1988. Abhidharmakosabhlisyam. 2 vols.
Berkeley: Asian Humanities Press.
Lessing, Ferdinand. 1942. Yong-ho-kung: An Iconography of the Lamaist
Cathedral in Peking with Notes on Lamaist Mythology and Cult. 18. Vol.
1. Stockholm: Sino-Swedish Expedition Publications.
____ . 1976. Ritual and Symbol. Taipei: Oriental Cultural
Levenson, Joseph. 1968. Confucian China andlts Modern Fate: A Trilogy.
Berkeley: University of California Press.
Mancall, Mark. 1968. "The .Ch'ing Tribute System: An Interpretive Essay."
The Chinese World Order. Ed. J. K. Fairbank. 63-89.
-'-_____ . 1971. Russia and China. Cambridge: Harvard University
Martin, Dan. 1990. "Bonpo Canons and Jesuit Cannons: On Sectarian
Factors Involved in the Ch'ien-lung Emperor's Second Goldstream
Expedition of 1771-1776 Based Primarily on Some Tibetan Sources." The
Tibetan 10urna1152: 3-28.
Ming jili. (Collected Rites of the Ming Dynasty). 1530. Reprinted in Siku
quanshu (Complete Library in Four Branches of Literature). Series 8. Vols.
113-124. Taipei: Shengyu yinshuguan, 1978.
Naquin, Susan. 1985. "The Transmission of White Lotus Sectarianism in
Late Imperial China." Popular Culture in Late Imperial China. Ed. D.
Johnson, A. J. Nathan, and E. Rawski. Berkeley: University of California
Press. 255-291.
Palace Museum, eds. 1983. Zijincheng dihou shenghuo, (Life of the
Emperors and Empresses in the Forbidden City). Peking: China Travel and
Tourism Press.
276 JIABS 16.2
Petech, Luciano. 1950. China and Tibet in the Early 18th Century. Leiden:
E. J. Brill.
Pozdneyev, Aleksei. 1977. Mongolia and the Mongols. Bloomington:
Indiana University Press.
1978 (1887). Religion and Ritual in Society.
Bloomington: The Mongolia Society.
Pritchard, E. H. 1943. "The Kowtow in the Macartney Embassy to China in
1793." Far Eastern Quarterly 2.2: 163-201.
Qi Jingzhi. 1985. Waibamiao beiwen zhuyi (Stelae of the Eight Outer
Temples [at Rehe] with Commentary). Beijing: Palace Museum Press.
Rahul, R. 1968-1969. "The Role of Lamas in Central Asian Politics."
Central Asiatic Journal 12: 207-227.
Rehe zhi (Rehe Gazetteer). 1781. Reprint ed. Dalfan, Liaodong: Leihai
shushe, 1934.
Richardson, Hugh. 1962. Tibet and Its History. London: Shambhala. 2nd
ed., 1984.
Rockhill, William W. 1905. Diplomatic Audiences at the Court of China.
Reprint ed., Taipei: Ch'eng-wen Publishing Co., 1971.
______ . 1910. "The Dalai Lamas of Lhasa and Their Relationship
with the Manchu Emperors of China, 1644-1908." T'oung Pao 11:1-104.
Rossabi, Morris. 1975. China and Inner Asia. New York: Pica Books.
______ . ed. 1983. China Among Equals. Berkeley: University of
California Press.
______ . 1988. Khubilai Khan: His Life and Times. Berkeley:
University of California Press.
Ruegg, D. Seyfort. 1991. "Mchod yon, yon mchod, and mchod gnas/yon
gnas: On the Historiography and Semantics of a Tibetan Religio-Social
and Religio-Political Concept." Tibetan History and Language. Ed. Ernst
Steinkellner. Vienna: Arbeitskreis Flir Tibetische und Buddhistische
Studien Universitat Wien. 441-453.
Serruys, Henry. 1967. Sino-Mongol Relations During the Ming II: The
Tribute System and Diplomatic Missions (1400-1600). Melanges Chinois
et Buddiques 14. Brussels: Institut Beige Des Hautes Etudes Chinoises.
Shakabpa, Tsepon W. D. 1967. Tibet: A Political History. New Haven: Yale
University Press.
Shi Kun. N. d. "Ny Dan Shamaness." Unpublished translation. Department
of Anthropology, Ohio State University.
Shiwen shichuan ji (Poems and Prose on the Ten Great Campaigns of the
Qianlong Reign). 1962. Taiwan: Geda shuju.
Skorupski, John. 1976. Symbol and Theory. Cambridge: Cambridge
University Press.
Snellgrove, David L. 1959. "The Notion of Divine Kingship in Tantric
Buddhism." The Sacral Kingship. Leiden: E. 1. Brill. 204-218.
---c.......,---. 1987. Indo-Tibetan Buddhism. 2 vol. Boston: Shambala.
Sperling, Elliot. 1983. Early Ming Policy Toward Tibet. Diss. Indiana
Sperber, Dan. 1975. Rethinking Symbolism. Cambridge: Cambridge
University Press.
SZZSL. Shizu zhang huangdi shilu (The Veritable Records of the Shunzhi
Emperor). 1964. Taipei: Huawen shuju.
SZRSL. Shengzu ren huangdi shilu (The Veritable Records of the Kangxi
Emperor). 1964. Taipei: Huawen shuju.
Tambiah, S. J. 1976. World Conqueror and World Renouncer. London:
Cambridge University Press.
Taussig, Michael. 1987. Shamanism, Colonialism, and the Wild Man.
Chicago: University of Chicago Press.
Thompson, Keith. 1986. Beliefs and Ideology. London: Tavistock
Turner, Samuel. 1800. An Account of an Embassy to the Court of the Teshoo
Lama in Tibet. Reprinted New Delhi: Manjusri Publishing House, 1971.
Van Gennep, Arnold. 1960. The Rites of Passage. Chicago: University of
Chicago Press.
Waddell, L. Austine. 1895. Tibetan Buddhism. Reprint ed. New York: Dover
Publications, 1972.
Wan Yi, Wang Suqing, and Lu Yanzheng, compilers. 1985. Qingdai
gongting shenghua. (Life in the Forbidden City). Hong Kong: The
Commercial Press. . .
Wang Jiapeng. 1990. "Gugong Yuhuage tan yuan." Gugong bowuyuan
yuankan 1: 50-62.
Wills, John E., Jr. 1984. Embassies & Illusions. Cambridge: Harvard
University Press
______ . 1993. "Maritime Asia, 1500-1800: The Interactive Emer-
gence of European Domination." The American Historical Review. 98.1:
WZTZ. Weizang Tongzhi (Gazetteer of Tibet). 1937. Shanghai: Shengyuyin-
Wulaxichun. 1986. Manzu gushenhua .(Ancient Myths of the ManchU).
Huhot: Inner Mongolian People's Publishing House.
Wylie, Turrell V. 1978. "Reincarnation: A Political Innovation in Tibetan
Buddhism." Proceedings of the Csoma De Koras Memorial Symposium.
Ed. Louis Ligeti. Budapest: Academiai Kiad6. 579-586.
______ .. 1980. "Lama Tribute in the Ming Dynasty." Tibetan
Studies in Honour of Hugh Richardson. Ed. Michael Aris and Aung San
Suu Kyi. Warminster, England: Aris and Phillips Ltd. 335-340.
278 nABS 16.2
Yang, Lien-sheng. 1968. "Historical Notes on the Chinese World Order." The
Chinese World Order. Eel. 1. K. Fairbank. Cambridge: Harvard University
Press. 20-33.
ZGCB. Zhanggu congbian (Collected Historical Documents). 1928-1930.
Peking: Palace Museum.
Zito, Angela. 1984. "Re-presenting Sacrifice: Cosmology and the Editing of
Texts." Ch'ing-shih wen-t'i 5.2: 47-78:
______ . 1987. "City Gods, Filiality, and Hegemony in Late Impe-
rial China." Modem China 13.3: 333-371.
Two Mongol Xylographs (Hor Par Ma) of the
Tibetan Text of Sa Skya PaI].<Jita's Work on
Buddhist Logic and Epistemology
The history of printing in Central Tibet has received the attention ofsev-
eral scholars in recent years. David P. Jackson, who has written the
most about early Tibetan prints, has pointed out that Tibetan authors are
generally of the opinion that the printing of books by way of xylographs
started in this area sometime during the beginning of the fifteenth cen-
tury.I In addition, he has made significant contributions to our know-
This paper is one of the results obtained during my stay in Beijing from
October to December 1992 that was made possible by a generous grant from
the Committee on Scholarly Communication with the People's Republic of
China, Washington D. C. My thanks are owed to the Chinese Center of
Tibetological Research, Beijing, for their kind hospitality and assistance. I
should also wish to express my gratitude to Messrs. Li Jiuqi, Chief Librar-
ian, Shao Guoxian, Deputy Librarian, and Ngag dbang nor bu, Assistant
Researcher, of the Library of the Cultural Palace of Nationalities for the warm
cooperation I received, one which made it possible for me to survey a slight
portion of the enonnous collection of Tibetan texts in their library. My
thanks are likewise extended to Prof. Chen Jianjian of the Central Institute for
Nationalities, Beijing, for her kind help, and last but not least, I am also very
much indebted to Ms. Wu Wei of the Chinese Center for Tibetological
Research for all the support given to me.
1. See Jackson 1983, and his "The Earliest Printings of Tsong-kha-pa's
Works: The Old Dga' -ldan Editions," Reflections on Tibetan Culture. Essays
in Memory of Turrell V. Wylie, eds. L. Epstein and R. Sherburne (Lewiston:
E. Mellen Press, 1990) 107-116 and, "More on the Old Dga' ldan and Gong
dkar ba Xylographic Editions," Studies in Central and East Asian Religions
2 (1989): 1e18. For a survey of a number of early (and not so early) printing
projects in Tibet, see Tshe ring ph un tshogs, "Snga rabs bod kyi par skrun
dang par khang skor mdo tsam gleng ba," Bod rig pa'i ched rtsom gees btus,
ed. Ngag dbang (Lhasa: Bod ljongs mi rigs dpe skrun khang, 1987) 345-
375, and Klu tshang Rdo Ije rin chen, "Rtsom rigs kyi rnam bzhag las par
skrun gyi skor bshad pa," Bod ljongs zhib 'jug 1 (1989): 103-117. See also
Chab spel Tshe brtan ph un tshogs and Nor brang U rgyan, Bod kyi 10 rgyus
rags rim g.yu yi phreng ba, Bar Cha [Part 2] (Lhasa: Bod ljongs dpe rnying
280 JIABS 16.2
ledge of the corpus of texts that were committed to the printing block
during that time. Heather Karmay was, to my knowledge, the first to
provide evidelice that Tibetan texts were being printed as early as some-
time around the year 1306, although not in the Tibetan cultural area, but
rather in China proper.
Elsewhere, we have examined various notices
of Mongol-sponsored printing projects anent the Kalacakra literature and
the dissemination of its esoteric doctrines, first in China and then in
Mongolia. The earliest evidence for the preparation ofaxylograph of its
main tantra, most likely in China, dates from either the last decade of the
thirteenth or the first decade of the fourteenth century. 3 The present
paper has to do in part with an even earlier xylograph of a Tibetan text
that was to all appearances prepared in China as well.
Among the writings for which Sa skya Pa1)Qita Kun dga' rgyal
mtshan (1182-1251), the fourth patriarch of the Sa skya school and one
of the finest scholars of his (or any other) era, is justly famous by any
standards, his most widely studied work, was the Tshad ma rigs pa'i
gter and autocommentary, which he completed sometime in the 1220s.
This study of Buddhist logic and epistemology (tshad rna) soon became
a classic and went into numerous printings, in addition to eliciting an
enormous commentarialliterature. 4 The Tibetan library of the Cultural
Palace of Nationalities in Beijing contains two different xylographs of
this work, one of which dates from the beginning, the other from the
middle of the Yuan dynasty. The first of these is the earliest known
blockprint of this work in particular and, perhaps, also constitutes the
earliest Tibetan blockprint as such. An essential description of these and
a reproduction of their print-colophons (par byang) can be found in
Appendix One.
The first of these two xylographs was initially sponsored by empress
Cabi (?-1284),5 or Ca[m]bui, the dpon mo chen mo, "Grand Lady,"
dpe skrun khang, 1990) 294. The allegation in the Life and Teachings of
Tsong khapa, ed. R. Thunmin (Dharamsala: Library of Tibetan Works and
Archives, 1982) 11, that Snar thang housed the printing blocks for the canon
during Tsong kha pa's stay there is of course unacceptable.
2. See her Early Sino-Tibetan Art (Warminster: Aris and Phillips Ltd.,
1975) 44.
3. See my "Fourteenth Century Tibetan Cultural History II: The Mongol
Imperial Family and Tibetan Kalacakra Texts" which is forthcoming in Asia
4. For this, see D. P. Jackson, "Commentaries on the Writings of Sa skya
PaI,1qita: ABibliographical Sketch," The Tibet Journal 8.3 (1983): 8-12.
5. For the literature on her and the previous problems of dating her passing,
senibr wife of Qubilai and mother of prince Zhenjin (1243-5 January
1286), the heir apparent who passed away before his father. The cutting
of the blocks apparently commenced under her patronage at the instiga-
tion of a "Chos kyi rgyal po bzang po," whom I am unable to identify.
The work was evidently left unfinished, presumably due to Cabi's
passing, and the project was brought to completion by order of her
daughter-in-law Go go cin ?-1300), "the wife 'of [her] sup-
reme son" (sras mchog gi btsun mo) Zhenjin, and like her husband a
major sponsor of 'Phags pa Blo gros rgyal mtshan (1235-1280), Sa
skya nephew. The date given for this in the colophon is the
eighth day of the rgyaZ ba lunar month of the wood-male-
monkey year. This would therefore be the oldest known xylograph of a
Tibetan text Unfortunately, no place name is given for where the blocks
were originally carved. Towards the very end of the colophon we read
that a certain Dpal mo 'Bog gan, presumably a lady, realized two hun-
dred [copies?] of the text (for purposes of acquiring good karma). Of
further interest is the fact that its pagination, like that of the second
blockprint, is given in Tibetan and Chinese on the left of the "a" side,
and orlIy in Chinese on the "b" side of the folio. The only reason for
having a Chinese pagination that I can think of is that it was added lest
the Chinese block-carvers would be confused about the woodblock-
order. Of equal interest is the fact that it has a marginal notation of
"KA" (= Vol. 1) which could very well indicate that it was the first vol-
ume of a projected printed edition of Sa skya s collected oeuvre,
one that presumably never materialized. The various catalogues of his
writings known so far do not list the Tshad rna rigs pa'i gter autocom-
mentary as comprising or being part of a first volume.
The colophon of the second xylograph is not altogether unproblematic.
It basically begins with stating that the view of Vasubandhu and tre
author of the 'Bum tik (= Satasahasrikiiprajfiapararnita'b.rhattlka),
Gdang ta se na (sic), that is, is that the Buddhist doctrine
will remain for five thousand years. This then lays the foundation for
dating this print to the earth-female-hare (sa rno yos) year, 1339, which,
it is alleged, completes three thousand four hundred and forty-nine years
see note two of my aforementioned paper.
6. For these, see Jackson 1987, 507-523. It may be that the "Mongol block-
print" of Sa skya Pm:t4ita's Sdom gsum Tab tu dbye ba, alluded to in Jackson
1983,6 and 22, note 17, belongs to this same period. If so, then we could
assume that the Mongol court was involved in having, if not his entire
corpus, then at least some of his major writings published in print.
282 nABS 16.2
since the passing of the historical Buddha. However, the Sa skya schOOl
dates the passing of the Buddha to the year 2133 B. C. E. so that the
year 1339 does not come close to this chronology. The fact that imme-
diately thereafter the colophon continues by stating that: "Henceforth, we
claim that the Buddha's Teaching will remain for one thousand five hun-
dred and fifty-one years," and that the earth-female-hare year is men-
tioned once again below,leads us to suspect that the colophon's author
either fell victim to an error in arithmetic, or that, more drastically, we
may have to correct sa rno yos to shing rno yos, "wood-female-hare,"
which would be the year 1315. The plausibility of this correction is
strengthened by the next reference to emperor Buyantu (Renzong, r. 7
April 1311 to 1 March 1320); note the use of the present stem of the
verb skyong! This, in turn, is followed by the mention of a Orand
Empress-Dowager (tha' i hu, Ch. taihou). If we allow for an error in the
year of the colophon, then she must be identified as Hong-gu-la-shi-da-
ji/gu, and not as Mai-lai-di, the mother of emperor Toyon Temiir
(Shundi, China 19 July 1333 to 14 September 1368; r. in MongOlia
to 23 May 1370). The colophon then notes two other individuals,
namely a cleric by the uninformative name of "Kun dga'" who had pro-
posed the project to a Sha-zin- a-kho-che who then made the formal
request (zhits) to have the print "established."7 The editor-in-chief of 1re
text was the elderly(?) 'Jam snyeg,8 who was assisted by Zla ba and
7. The last line of this passage is not easy to interpret. It reads rnam 'grel
rigs gter stong phrag par du bsgrubs II, which means "Established a thou-
sand [copies of?] the roam 'grel rigs gter as a print," where the phrase roam
'grel rigs gter can be interpreted a dvandva compound meaning rnam 'grel
and rigs gter, that is, a print of the Tibetan translation of Dharmakirti's
[Prama!la1varttika and the Tshad rna rigs gter. Alternatively, we can also
take it in the more likely sense of roam 'gre/ gyi rigs gter, that is to say, a
"treasury of tshad ma anent the Varttika," bearing in mind the meaning of
"tshad ma rigs pa'i gter." Thus far, a Yuan, "Mongol print" of
Dharmakirti's work is not referred to in the known Tibetan literature.
8. A '''Jam nyeg"--nyeg and snyeg are homophones--is mentioned as one of
the scribes of 'Phags pa's Kyai rdo rje'i bdag 'jug gi cho ga dbang la 'jug
pa, Sa skya pa'i bka' 'bum, Vol. 6, compo Bsod nams rgya mtsho (Tokyo:
The Toyo Bunko, 1968), no.47, 118.4.3. There this man is styled "won-
drous scribe" Cphru/ gyi yig mkhan) and it is also said that he was among
those who petitioned 'Phags pa to write it. The work in question is dated to
the year 1266. Seng ge bzang po's biography of his master Dka' bzhi pa
Rig[s] pa'i seng ge (1287-1375) of 1418 notes a Slob dpon 'Jam nyag in
Khro phu monastery who taught him the Prama!lavinifcaya and a Summary
[of logic and epistemology?] sometime early in 1322; see SENG 22 [SENG
Thog dpon. The ones who were actually responsible for the printing
process were the official (mi chen) Sar du, Ta'i hyo and Peb ha duo The
preparation of the printing blocks was begun on the fourth day of the
eighth month and completed on the fifteenth day of the eleventh month
of the wood-female-hare year, that is, again if the correction is in order,
they were cut and edited from 2 August to 14 November of 1315 (or
from 9 August to 14 November 1339). Lastly, the place where the
blocks were prepared is stated to be the monastery of "Ka'u lang ho." 9
It is of course tempting to hold that this print is simply a clone of the
first one due to the good possibility that the original blocks of 1284 had
worn out. This does not appear to be case, however. The differences
between these two prints in terms of pagination are substantial enough to
warrant the view that this second print derives from newly carved
The first indication that all was not well with the transmission of ~
Tshad rna rigs pa'i gter's verse-text and autocommentary in which the
verses are also reproduced is found in a passage of Glo bo Mkhan chen
Bsod nams lhun grub's (1456-1532) exegesis of Sa skya Pa1).<;lita's
auto commentary of 1482, where he quotes from the mid-fourteenth
century commentator Gnas drug pa BIo gros mtshungs med to the effect
that there were conflicting readings in at least two manuscripts of the
text's tenth chapter. 10 Other important notices of conflicting readings in
the eleventh and last chapter are alluded to expressis verbis in Bo dong
P ~ c h e n 'Jigs med grags pa's (1375-1451) biography of 1453 by his
disciple and patron' Jigs med 'bangs, that is, the Sna dkar rtse scion
Nam mkha' bzang po of Yar 'brog.
The issue at hand was Sa skya
9. This could refer to a monastery at Gaolang river, (at the time) south of
Dadu. My thanks to my colleague A. Yue-Hashimoto for drawing my atten-
tion to this possibility. This may be confirmed by a passage in one of the
Wang Guowei writings to which F. W. Cleaves has drawn attention in his
"The Bodistw-a c:ari-a awatar-un tayilbur of 1312 by Cosgi Odsir," Harvard
Journal of Asiatic Studies 17 (1954): 15, 33, note 18. The "monastery at
Gaoliang" river is none other than the well-known monastery of Da Renwang
huguo, built by Qubilai, which was located on a site that is right behind the
Beijing National Library. One wonders if this be an alternate name for what
the Yuanshi refers to as the Southern Monastery (nansi) in YS 2, [20] 434
and 8, [90] 2284. This was the locale where Rong po Rdo Ije rgyal mtshan
(1283-1325) had the KalacakramUlatantra printed sometime between 1310
and 1325.
10. GLO 352 [GLOI 358, GL02 223].
11. See the Dpalldan bla rna thams cad mkhyen pa phyogs thams cad las
rnam par rgyal ba'i zhabs kyi rnam par thar pa ngo mtshar gyi dga' ston,
284 HABS 16.2
s numerical determination of two kinds of a refutation by way of
a reductio ad absurdum (that' gyur, * prasa?zga), one which in its con-
traposed form implies a proof of a similar logical type and structure
(bzlog/ldog pa rang rigs 'phen pa) and one which implies a dissimilar
one (gzhan rigs 'phen pa). 12 Apart from the intellectual satisfaction of
establishing the correct reading, the textual problems raised by these
divergent manuscript (and blockprint) traditions were not exclusively a
scholastic exercise in philology. On the contrary, it was something that
had obvious practical Significance when we bear in mind that the art of
disputation was one that was very well developed in the Tibetan monas-
tic environment.
It is also precisely at this juncture that the Tshad rna rigs pa'i gter
commentaries of Go rams pa Bsod nams seng ge (1429-1489) of 1471
and Gser mdog Pal). chen Shakya mchog ldan (1428-1507) of 1474 refer
to the readings of a Mongol print (hal' splpar rna) which had a classifi-
cation of these two in, respectively, four and fourteen types. The read-
ing they assert of this print is confirmed by both blockprints, 13 so that
we may conclude that the hor par rna (singular!) to which they refer .
indicates one or both of these. Whereas Go rams pa accepts the hor par
rna's reading, Gser mdog Pal). chen does not with some vehemence and
argues instead for accepting a four by sixteen classification, for which,
Encyclopedia Tibetica. The Collected Works of Bo-don(sic) Pan(sic)-chen
Phyogs-las rnam-rgyal, Vol. I, (New Delhi: The Tibet House, 1981) 283 [=
Bo dong phyogs las rnam rgyal gyi rnam thar (Chengdu: Si khron mi rigs
dpe skrun khang, 1990) 203]. . .
12. The textual conundrums and philosophical issues were first discussed in
S .. Onoda, "Phya pa Chos kyi seng ge's Classification of ThaI 'gyur,"
Berliner Indologische Studien 2 (1986): 65-85, and the points made in this
paper are now made with more precision and somewhat expanded in his excel-
lentOnoda 1992: 71-86; see now also the involved T. Tani, "Rang rgyud
'phen pa'i thaI 'gyur [Hypothetical Negative/Indirect Reasoning (prasa1Jga)
with the Implication of Independent Direct Proof (svatantra)] [Tibetan
Commentators' Meta-Interpretations on DharmakIrti's Interpretation of
prasa1Jga];" Tibetan Studies. Proceedings of the 5th Seminar of the
International Association for Buddhist Studies, Narita 1989, Vol. 1, eds.
ShOren Ihara and ZuihO Yamaguchi (Narita: Naritasan Shinshoji, 1992) 281-
301. For the relevant passage in Sa skya PaJ.l4ita's work, see the Sde dge
print of the Tshad ma rigs pa'i gter gyi rang 'grel in the Sa skya pa'i bka'
'bum, Vol. 5, compo Bsod nams rgya mtsho (Tokyo: The Toyo Bunko,
1968), no. 20, 263.1.4-5 and
13. Onoda 1992, 80; blockprint no.1 reads on fol. 187a: bzlog pa rang rigs
'phen pa bzhi II . . gzhan rigs 'phen pa bcu bzhi yod II, and blockprint no.2
has on fol. 187a-b: bzlog pa rang rigs 'phen pa bzhi II .. gzhan rigs 'phen
pa bcu bzhi yod II. .
as indicated by Glo bo Mkhan chen, there is already a thirteenth century
precedent by way of the Sde bdun gsaZ ba'i rgyan of Lho pa Kun
mkhyen Rin chen dpal, a disciple and biographer of Sa skya PaJ)<;lita. 14
It is of course likely that, when earlier fourteenth century exegeses of the
text come to light-I am thinking here particularly of the works by
Byams mgon dpal, alias Phyogs glang gsar rna, and his student Dka'
bzhi pa Rigs pa'i seng ge IS-we shall have further evidence that these
early xylographs of the Tshad rna rigs pa'i gter autocommentary were
put to use in other interpretations of Sa skya Pa9-Qita's work. Aside
from the fact that these blockprints are of undoubted historical signifi-
cance, they also underscore the methodological imperative that textual
criticism should precede translation (and interpretation), something that
is all too often and easily forgotten these days in the study of Tibet's
vast literary heritage. 16 Indeed, at least three different xylograph-edi-
tions of the auto commentary were prepared during the fifteenth century.
We now have located the GIang ri thang monastery blockprint, as
ordered by Kun dga' rgyal mtshan dpal bzang po, that was completed on
the fourteenth day of the third month of the wood-female-ox year, which
can only refer to 22 March 1445.1
Other prints that have yet to come to
light are the one prepared in Thub chen mngon par dg[y]es pa temple in
Glo bo Smon thang in 1474, for which Gser mdog PaI.1 chen wrote a
notice,18 and the one of Dpal rdo rje gdan mi 'gyur Me ba chen po
14. GLO 404 [GLOI 412, GL02 256]:
sgrub byed 'phen pa nyi shu la II rang rigs 'phen pa rnams bzhi II gzhan
rigs 'phen pa bcu drug ste II tsha reg gsum gyis grang reg gsum II 'gog pa'i
spyod pa roam dgu las II rang rigs 'phen pa rnam gsum bri II [hag rna drug
dag rang bzhin dang II 'bras bu'i thaI ba rnam gnyis te II brgyad las phye
ba'i bcu drug nyid II
Glo bo Mkhan chen concurs with this numerical determination.
15. SENG 20-21, 25, 36 [SENG 1 56, 60, 71]. Some of their views are cited
severally in Glo bo Mkhan chen's exegesis. To be sure, Seng ge bzang po
does not state that Phyogs glang gsar rna wrote a Tshad ma rigs pa'i gter
commentary. Although he does mention that he enjoyed special renown in
Sa skya for his expert know ledge of this difficult work.
16. This holds already for the serious inconsistencies in the most wide-
spread available texts of both by way of the Sde dge print of 1736, for which
see the useful tabulation of variant readings in the verse texts in the Tshad rna
rigs pa'i gter, ed. Rdo tje rgyal po (Pe cin: Mi rigs dpe skrun khang, 1989)
395-40l. For the particulars of this print, see Jackson 1987,232-236.
17. This blockprint is housed in the Tibetan library of the Cultural Palace of
Nationalities where it is catalogued under no. 004783(2). It consists of one
hundred and fifteen folios and bears the marginal notation "Ka" (= Vol. 1).
18. See the Rigs gter gyi gzhung par du bsgrubs pa'i dkar chag in GSER
286 nABS 16.2
monastery that was effected through the financial patronage of Gong
dkar Rdo rje gdan pa Kun dga' mam rgyal (1432-1496). 19 It may be
that the Sa skya xylograph of the verse-text of the Tshad rna rigs pa'i
gter as sllch also belongs to this century. 20 '
In his discussion of these arguments, S. Onoda points out that the four
by sixteen scenario is al-so met with in the later Dga' ldan pa and Dge
lugs pa bsdus grwa texts, the earliest of which he signalled is the work
by Mchog lha 'od zer (1429-1500). He tentatively suggests that its ori-
gin might be sought in Rgyal tshab Dar rna rin chen's (1364-1432)
Tshad rna rigs pa'i gter commentary. This is perhaps unlikely given the
fact that it was conSistently surpressed by later e(litors of his collected
oeuvre, and is so far only extant by way of a Bla brang Bkra shis 'khyil
blockprint. Like their mastyr Tsong kha pa, Mkhas grub Dge legs dpal
biang po is but content with giving a very general descrip-
tion of both types, adding that one should look elsewhere for a detailed
subdivision of their typology. 21 Lastly, their disciple Dge 'dun grub pa
(1391-1474) is so far the earliest known Dga' ldan pa scholar to have
argued for a four by sixteen scenario, one which we encounter in his
survey of Buddhist logic and epistemology of 1437.
In addition to the dossier provided by S. Onoda, we may also refer to
the discussion in the anonymous, undated, and hitherto unknown work
on Buddhist logic and epistemology entitled the Tshad rna shes rab
sgron rna, a beautifully calligraphied handwritten dbu rned manuscript
of which we located in the Tibetan library of the Cultural Palace of
Nationalities, which also accepts the four by sixteen classification with-
220"222. It was sponsored by her ruler Rkra shis mgon (?-1489).
19. Rgya stan Byang chub dban rgyal's biography of Gong dkar ba, the
Chos kyi rje thams cad mkhyen pa rdo rje gdan chen po kun dga' rnam rgyal
dpal bzang po'i mam par thar pa ngo mtshar rin po ehe'i gter mdzod, (ols.
45a, 49b. Two identical blockprints (on different size paper) of this
biography are housed in the Tibetan library of the Cultural Palace of
Nationalities where they are catalogued under nos. 002655(1), (2). -
20. This xylograph is briefly noted in D. P. Jackson, "Sources for the Study
of Tibetan Prami17J.a Traditions Preserved at the Bihar Research Society,
Patna," Studies in the Epistemological Tradition, ed. E.
Steinkellner (Wien: Verlag der Osterreichischen Akademie der Wissen-
schaften, 1991) 104.
2!. See his Tshad rna sde bdun gyi rgyan yid kyi mun sel [based on the Bkra
shis !hun po print], ed. Rdo Ije rgyal po (Pe cin: Mi rigs dpe skrun kbang,
1984) 372.
22. See his Tshad ma'ibstan bcos chen po rigs pa'i rgyan, Collected
Works, Vol. 4 (Gangtok: n. p., 1979) 367-370. -
out further comment. 23 Moreover, the recently published survey of
. Indo-Tibetan Buddhist epistemology and logic by Bcom ldan Rigs pa'i
ral gri (ca. 1235-1314) makes it clear that this great controversialist
accepted a total of seventeen different types of prasat.zgas which he
classified by means of a four by thirteen division.24 Lastly, Bo dong
chen's monumental Tshad rna rigs pa'i snang ba
refers rather
critically to three different positions which are not attributed to anyone
specifically. These are a five by sixteen, a four by fourteen and a three
by ten classification, the first two of which are attested in different
Tshad rna rigs pa'i gter manuscripts (and other xylographs?).26 He
. himself argues for a three by eleven scenario. These issues deserve.
further exploration.
We may take this opportunity to furnish some additional remarks on
the Tshad ma shes rab sgron mao Its colophon does not shed any light
on its authorship, the title does somewhat resemble the so-called Gzhal
bya shes rab sgron ma'i phreng ba--the phreng ba is somewhat pecu-
liar in this title-of unknown authorship which, according to Zhang
Rgyal ba dpal, yet another biographer and disciple of Sa skya
the latter had studied under Mtshur Gzhon nu seng ge around the years
1200 to 1201. 27 This work has not been located so far. However,we
can unequivocally say that the intellectual environment of the Tshad ma
shes rab sgron ma is clearly the exegetical traditions that had their
inception in Gsang phu ne'u thog monastery. With some modifications,
it accepts its rather distinctive five-fold typology of the so-called non-
valid means of cognition (tshad min) against which Sa skya
reacted so critically.28 For example, its discussion of the notion of
23. AN fols. 65b-66a. For three other manuscripts of early tshad ma texts,
see Appendix Two.
24. See his Tshad ma'i bstan bcos sde bdun rgyan gyi me tog, handwritten
dbu med manuscript of the Tibetan library of the Cultural Palace of
Nationalities, catalogue no. 002468(2), fo1s. 88b-89a [Ibid., ed. Rdo rje rgyal
po (Pe dn: Krung go'i bod kyi shes rig dpe skrun khang, 1991): 128-129].
25. See the text in Encyclopedia Tibetica. The Collected Works ofBo dong
PaT} chen Phyogs las rnam rgyal, Vol. 8 (New Delhi: The Tibet House,
1970) 533-543. S. Onoda did not examine Bo dong PaI,1 chen's remarks.
26. GLO 404 [GL01 412, GL02256].
27. Jackson 1987, 106.
28. For these, see the preliminary (and sometimes erroneous) remarks in my
"Phya-pa Chos-kyi seng-ge's Impact on Tibetan Epistemological Theory,"
Journal of Indian Philosophy 5 (1978): 355-369. For Phya pa Chos kyi seng
ge (1109-1169) there were a total of five types of invalid cognitions, a deter-
mination that was followed in the PramaT}aviniicaya commentary by his dis-
288 nABS 16.2
reflection (yid dpyod),29 an epistemological type which Sa skya PaJ).<;lita
is loathe to accept, begins with a negative reference to an earlier defini-
tion in which it was maintained that itis a type of cognition that ascer-
tains an object independent of either a direct experience of it, or a logical
justification on which basis the object could be deduced. We find
something very similarto this in Gtsangnag pa's PramartaviniScaya
commentary30 which belongs to the second half of the twelfth century,
not to mention the fact that such a standpoint is also more or less
attributed to Phya pa. The author's own position is that reflection con-
sists of an ascertainment of an imperceptible object, on par with an
[actual] object, that has not been previously cognized and which is,
however, independent of a logical indicator (nags) or justification. The
text is polemical and argumentative, but none of its references to other
opinions are identified by their owners-an exception is made in its cita-
tion of Indian Buddhist philosophers-the author being content with
prefixing these opinions by the somewhat exasperating kha cig na re,
"some say." However, several glosses in an unknown hand do iden-
tify-future research will have to show whether rightly or wrongly-
several such kha cigs. Among the first of these is a sublinear "Rgya"
anent a definition of a cognition which does not ascertain what is present
[to it] (snang la rna nges pa) with which the author disagreed. 31 It is
quite probable that "Rgya" refers here not to an Indian (rgya[-gar])
scholar, but rather to Rgya dmar pa B yang chub grags, a diSCiple of
Gangs pa She'u BIo gros byang chub and Khyung Rin chen grags and
one ofPhya pa's masters, for an epistemological type of the snang la
rna nges pa variety never seems to have been conceptualised in India. 32
ciple Gtsang nag pa Brtson 'grus seng ge, for which see GTSANG fols. 22b-
24a. For these men, see also van der Kuijp 1983, 60-84, Jackson 1987
index, 590, 603, my Introduction to GTSANG 1-5, E. Steinkellner, "Early
Tibetan Ideas on the Ascertainment of Validity (nges byed kyi tshad rna),"
Tibetan Studies. Proceedings of the 5th Seminar of the International Asso-
ciation for Tibetan Studies, Narita 1989, eds. ShOren Ihara and Zuiho
Yamaguchi (Narita: " Naritasan Shinshoji, 1992) 257-273, the numerous con-
tributions by S. Onoda referred to in Onoda 1992, 226-227. and his "Phya pa
Chos kyi seng ge's Theory of 'gal ba," Tibetan Studies. Proceedings of the
5th Seminar'of the International Association for Tibetan Studies, Narita
1989, eds. ShOren Ihara and Zuiho Yamaguchi (Narita: Naritasan Shinshoji,
1992) 197-202.
29. AN fo1. 6a.
30. GTSANG fo1. 23a.
31. AN fo1. 5b.
32. For him, see van der Kuijp 1983, 60, 293, notes 213-215. The remain-
Other klla cig na res are glossed by "Lo," "Gangs pa," and "Khyung,"
which would refer to possibly Rngog Lo tsa ba BIo ldan shes rab (1059-
1109) and his disciples Gangs pa She'u and Khyung.
On the strength
of the glosses of "Rtsang nag pa," that is, "Gtsang nag pa," we can
argue that it postdates Gtsang nag pa's logical oeuvre.
Moreover, we
may have to place the author at sometime between Gtsang nag pa and the
Tshad rna rigs pa'i gter, although we can only adduce negative evidence
for this inasmuch as, for instance, he does not take issue with Sa skya
s criticism of the earlier five-fold classification of the tshad min.
Only a detailed study of our text will reveal whether or not there is
evidence for the author having known Sa skya s work. A po-
tential snag in a pre-Tshad ma rigs pa'i gter dating of this work may be
the two glosses that attribute a pOSition to a "S[R?]tag pa" or "S[R?]
tag."35 The only individual who might be identified by "Stag" is Stag
lung Lo tsa ba Shakya bzang po (1322-1404). The chronicle of the Stag
lung Bka' brgyud sect written by Stag lung pa Ngag dbang mam rgyal
(1571-1626) in 1609 states that he had written a Pramli{lavlirttika
commentary which carried the subtitle of RigIs?] tshul snang chen.36
However, it seems unlikely that the Tshad ma shes rab sgron ma is to
be placed sometime in the latter part of fourteenth century, or even
Excursus: The Rtses thang Print of Dignaga's Prama1).asamuccaya
Aside from the numerous printing projects that were undertaken anent
Tibet's indigenous literature, the fifteenth century also knows of various
xylograph editions of canonical texts, including those of the Tibetan
renditions of the writings of Dignaga and DharmakIrti. One such xrlo-
graph is the Rtses thang print of the Tibetan rendition of Dignaga's
ing "Rgya" glosses are found in AN fols. 23a, 26a, 51b, and 62b, but these
do not have parallel attributibns in other texts examined so far. AN fol. lOb
does have an "insert" of rgya dag na re ... "Indians say .... " Other candi-
dates for "Rgya" may also be Rgya MI' chims ru ba-GSERa 452 writes
"Rgya stan Phying ru ba"-, the founder of Bde ba can and a onetime abbot
of Gsang phu ne'u thog monastery's Gling stod college-for the literature on
this monastery, see below note 43-0r any of the more or less contemporane-
ous Rgya ston-s. .
33. For instance, AN fols. 21a, 23a, 31a.
34. See, for instance, AN fol. 22a.
35. See, for instance, AN fols. 5b, I5b.
36. See the Chos 'byung ngo mtshar rgya mtsho,Vol. I (Tashijong, 1972)
290 JIABS 16.2
blockprint specifies that it is the translation of
andSeng ge rgyal mtshan-which is housed in the
Tibetan library ofthe Cultural Palace of Nationalities where it is cata-
logued under no. 004806(2). It consists of ten folios with seven lines
per folio side and measures 9 x 61 cm. The upper center of the title page
has a marking in red inkplJyi, zha, 18. Phyi, "external" would suggest
that it was accessible to the "public," whereas nang in other manuscripts
would mean that it was for "internal" circulation only. There is a mini a _
ture on either side of fol. 1 b; on the left MafijusrI and on the right
Dignaga. The printer's colophon states that it was prepared by Bsod
nams bkra shis in Rtses thang monastery for the fulfillment of [his] pre-
ceptor's final wishes and the longevity of the Phag mo gru/Sne'u gdong
ruler Grags pa 'byung gnas (1414-1445). The person who may have
been in charge of the actual printing, or who had particular expertise in
this text-I have difficulty in interpreting the phrase chos 'di mkhas
pa-was a Lama Rin chen dga'.
Text of the Printer's Colophon
bstan bcos kun gyi mi gcig thos pa'i mchod II thub pa'i bstan pa dngos
stobs legs sgrubs pa'i II rigs lam dri rna med pa'i gzhung lugs 'di II dpal
ldan bIa ma'ithugs dgongs rdzogs pa'i phyir dang I [here the text has a
rin spungs shad] rgyal ba'i sras grags pa 'byung gnas kyi /1 sku tshe
brtan cing chab srid brgyas bya'i phyir II chos gra chen po dpal gyi rtses
thang du II bsod nams bkra shis bdag gis par du sgrubs II de' ang ?byung
pa'i bde ba gang thob pas II rgyal ba'i dkyil 'khor bzang zhing dga' ba
der II padmo dam pa shin du mdzes las skyes II snang ba mtha' yas rgyal
bas mngon sum du II lung bstan pa yang bdag sogs der thob shog II
chos 'di mkhas pa bla rna rin chen dga'?o I
Lastly, Gser mdog PaJ) chen signals the existence ofaxylograph of
the Tibetan version of the Prama!wvarttika in Thub chen mngon par
dgyes pa monastery in his notice Of this "publication" dated 1474.37
Like that ofthe Tshad ma rigs pa'i gter autocommentary, it too was
sponsored by Bkra shis mgon. The scribe who prepared the manuscript
was Shes rab grags and the carverlprinter was Dge legs seng ge.
37. See his Rnam 'greZ gyi gzung par du bsgrubs pa'i dkar chag in GSER
216-220. .
Note: "No." refers to the catalogue number of the Tibetan library of the
Cultural Palace of Nationalities, Beijing.
1. No title page
No. 004817.
Folios: 2a-190a; six lines per folio side.
Dimension: 12.5 x 64 cm:
Incomplete: fols. 1 and 66 are missing.
Marginal notation: KA
Tibetan/Chinese pagination.
Incipit: [2a] ... pa mams kyang log par rtog pa du ina mthong pas de
sun dbyung pa dang yang dag pa'i don gtan la dbab pa'i phyir 'di
brtsam moll
Printer's Colophon: [189b-190a] tshad mar gyur ba [read: pa] rgyal ba'i
gsung rab kun II tshad rna nyid du nges byed gzhung kun gyi II tshad
mas grub pa'i rigs mchog kun bsdus pa'i II tshad ma'i bstan bcos 'di ni
rigs pa'i gter II'di ni shes bya'i gnas la mkhyen pa rab gsal zhing II 'gro
ba kun la kun nas brtse ba'i thugs mnga' ba II chos kyi rgyal bo [read:
po] bzang bo'i [read: po'i] zhabs kyi zha! snga nas II bstan pa dri med
mam par spel phyir rab tu mdzad II gzhung lugs [190a] [gzhung lugs;
reduplication] 'di yidri med 'thad pa mthong gyur zhing II gzhung rtsom
de yi ngo mtshar dpag yas shes gyur nas II gzhung mchog 'di ni blo gsa!
mams la rgyas pa'i phyir II gzhung 'di'ipar gi gter chen legs par phye
ba yin II 'di ni rigs gzugs dge mtshan ldan zhing II dad dang brtse dang
dpal 'byor phun sum tshogs II sna tshogs bkod pa'i tshulla rab mkhas
pa II dpon mo chen mo cha bus mgo btsugs nas II de sras mchog gi
btsun mo dpalldan pa II bstan pa'i rtsa lag skye rgu'i rna Ita bu II rigs
gzugs yon tan phun sum tshogs pa yi I/go go cin gi lung gis rdzogs par
bsgrubs II rgyal ba mams la dam chos mi zad Itar II par ' di las kyang
chos tshulzad mi shes II de 'dra sgmb par gang zhig bka' sgo ba II de'i
bsod nams tshul yang zad mi shes II de las byung ba'i dge ba rgya
mtsho dang II rgyal sras rgya mtsho'i spyod pa 'dres gyur nas II sems
can rgya mtsho'i tshogs mams smin byed cing II rgyal ba'i ye shes rgya
mtshor 'jug par shog II gnas skabs su yang rgyal po yab sras dang II
btsun mo sras dang brgyud par bcas pa mams I I sku khams bzang zhing
292 JIABS 16.2
sku tS.he ring ba dang II chos dang zang zing dpal 'byor phun tshogs
shog 1/ 'di 'dra gang gis dran bskul byas pa dang II yi ger 'bri dang par
du rko padang II yo byad sgrub pa'i gnyer pa la stsogs kun II gnas
skabs mthar thug don rnams lhun grub shog II phyogs dus thams cad du
bkra shis par gyur cig II shing phospre'u 10 I rgyal gyi zla ba yi I yar
ngo'i tshis Jread: tshes] brgyad la /legs par grub pa yin I
II [figure comprising a circle plus two semicircles underneath it] btsun
mo dam pa mtho tis kyi II yon tan kun kyi brgyan pa can II dpal mo 'bol
gan zhes bya bas II rigs gter brgya phrag gnyis bsgrubs te II bshad nyan
spel phyir phul ba yis II rgyal po sku tshe legs brt:m zhing II kun gyi
sangs rgyas thob par shog II bkra shis par gyur cig I .
2. No title page.
No. 004796.
Folios: 1-190b; six lines per folio side.
Dimension: 12.5 x 65.2 cm.
Tibetan/Chinese pagination.
Fol. Ib has two miniatures in black and white: '''Jam dpal [Mafijusn]"
on the left and "Chos rje Sa skya Palf <;li ta" on the right side.
Fol. 190a has seven lines.
Printer's colophon: [190a-b] tshogs gnyis rgyas pa'i kling [read: gling]
las legs 'khrungs shing I I legs spyad Ihun mdzes phan bde'i 10 'dab can
II sku gsum 'bras Idan phrin las bsil grib kyis II nyon mongs gdung sel
dkon mchog gsum la 'dud II me tog u dum stong gi Inga ltas las II gnas
gtsarig Iha yis bzang po zhes bsgrags pa'i II bskal bzang 'di la bde
gshegs stong 'byon pa'i II 'khor ba 'jig dang gser thub 'od bsrung 'das
II brgyapa'i dus 'dir shakya'i [read: shakya'i] rgyal po byonllrgyal ba'i
bstan 'di Inga stong gnas so zhes II sangs rgyas gnyis pa Itabu'i dbyig
gnyen dang II 'bum tig mkhan po gdang a se na bzhed II de la thub
bstan 10 ni sum stong dang II bzhi brgya zhe dgu, sa mo yos los rdzogs II
da phyis rgyal ba'i bstan pa 10 stong dang II lnga brgya Inga bcu nga
gcig gnas par 'dod II de lta'i thub pa'i bstan pa'i rtsa lag mchog Illnga
brgya bdun par 'dzam kling [read: gling] byang phyogs su II stobs kyi
'khor lobsgyur ba brgyad par ni II sngon bsags bsod nams mtsho chen
'khyil pa lall dam 'byor chu skyes yon tan ge sar 'khrigs II sa chen 'di
na sa la mnga' mdzad pa II bu yan du gan rgyal srid skyong pa [read: ba]
na II bsod nams stobs kyis sa la mnga' bsgyur zhing II dad pa'i stobs
kyis dbu sde rnam gnyis mchod II snying rje'i stobs kyis mnga' ris chos
bzhin skyong II rigs gzigs 'byor ldan tha'i hu chen mo la II shes bya'i
mIcha' la mkhyen pa'i chu 'dzin 'khrigs II shes rab klog 'khyug stong
nyid 'brug sgra sgrogs II legs spy ad char gyis nyon mongs tshad gdung
selll kun oga'i mtshan can bla ma'i gsung bzhin du II sna tshogs shes
bya'i nam mkha' yangs pa la II dri med blo gros 'od zer rab 'phro zhing
Ilthub bstan gsal byed bstan pa'i khur 'dzin pa'i II sha zin a ko che yis
zhus nas ni II bstan pa yun du gnas zhing dar ba dang II yangs pa'i 'ro la
bde skyid 'byung ba dang II 'gro kun byang chub chen po thob bya'i
phyir II mam ggrel [read: 'gre!] rigs gter stong phrag par du bsgrubs II
'eli'i [read: 'eli yi] dran bskullag len zhus dag pa II mang thos sde snod
'dzin pa 'jam snyeg yin II lung rigs dbang phyug smra ba'i zla ba dang II
thog dpon gnyis kyis zhus dag grogs ldan byas II dam chosdpar 'eli'i
bya ba bsgrub byed pa II dad brtson ldan ba'i [read: pa'i] mi chen sar du
dang II ta'i hyo dang peb ha du yis bsgrubs II de las byung ba'i dge ba
rgya mtsho des II tshogs gnyis rgya mtsho yongs su rdzogs byas nas II
sems can rgya mtsho yongs su smin byed [190b] cing : 'gro kun rgya
mtsho'j tshogs dang bcas pa mams II sku gsum rgya mtsho'i kIing
[read: gling] du phyin par shog II gnas skabs su yang rgyal po yab sras
dang II btsun mo sras dang brgyud par bcas pa mams II sku khams
bzang zhing sku tshe ring ba dang II chos dang zang zing dpal 'byor
phun tshogs shog II 'di 'dra gang gis dran bskul byas pa dang II yi ger
'bri dang par du rko pa dang II yo byad sgrub pa'j gnyer pa la stsogs
kun II gnas skabs mthar thug don mams lhun 'grub shog II sarno yos
bu'i 10 I zla ba brgyad pa'i I tshes bzhi nas dbu btsugs nas I zla ba bcu
gcig pa'i tshes bcwo Inga'i nang du ka'u lang ho'i sde chen por I legs
par grub pa yin I phyogs dang dus dang gnas skabs thams cad du I bkra
shis dang bde legs chen pos khyab par gyur cig II
The Tibetan library of the Cultural Palace of Nationalities houses a large
number of other early treatises on tshad rna. Among these there are the
following three manuscripts which I was able to inspect only very
briefly due to exigencies of time:
l.Title: Bod snga rabs pa'i tshad ma'i spyi don
No. 004783(1)
Folios 1-97 a
294 JIABS 16.2
Upper center of the title page has Phyi, Zha, 17, for which see the
remarks in the excursus.
Incipit: [lb] 'phags pa 'jam dpalla phyag 'tshallofl dag gsal mthu'
ldan byams pa'i dkyil 'hor canl I gsung gi 'd zer 'gro ba'i lam ston pas I
I rna rig mun se1 kun gyi sgron mer gyur I I rdzogs sangs nyi ma'i sku la
phyag 'tshallo II
Colophon: [97a] dad mos brtson pas yang dag don rtogs nas II rtog ge
rigs pa'i brgyan gyi snying po brtsams II de [b]rtsams bdag gis dge ba
gang thob pa II 'gro kun bde gshegs snying po myur thob shog II dgag
dang sgrub pa bya ba'i gnas 'di la II lha ngan 'chal pa'i wa tshogs zil
rnnan nas II rtsalldan smra ba'i seng ger 'gyur 'dod na II mtha' ldan rtog
ge tshulla 'jug par gyis II rna rig mun gdongs srid pa'i lam dub la II lam
ston rigs shes sgrone [= sgron me] par gyur pa yis II phan bde gtan gyi
gnas mchog 'gyur 'dod na II gsal byed rtog ge'i tshulla 'jug par gyis II
sdug bsngal rgyur gyi rtog ngan dra ba 'di II bden rtogs shes rab mtshon
gyis mam bcad nas II zhi ba byang chub dam pa thob 'dod na II mam
dpyod rtog ge'i tshulla 'jug par gyis II II rtog ge rigs pa'i brgyan [=
rgyan] gyi snying po zhes bya ba I shag kya'i dge slong dha rma rad na
zhes bya bas sbyar ba II rdzogs shyo II II rtog ge rigs pa'i snying po
'di II kho bo zhang gis gzabs nas bris II bshes gnyen se'i thugs mdzodll
Despite the title page, the actual title of this work by the monk Dharma
rad[read: t]na was Rtog ge rigs pa'i [b]rgyan gyi snying po. It was
carefully written out by a certain Zhang. Who was Dharmaratna, that is,
Chos kyi rin chen!dkon mchog or Dar rna rin chen! dkon mchog? Ifmy
suspicion is correct, then he may be identified as the late twelfth and
early thirteenth century scholar signalled by Tshal pa Kun dga' rdo rje
(1309-1364) as "Dar dkon."38 This is no doubt an abbreviation of "Dar
rna dkon mchog." The bearer of this name, either a native of Phu thang,
or one who was associated with this place on a professional basis, had
been one of the disciples of the very influential Gnyal zhig 'Jam pa'i rdo
Jje,39 himself a student of 'Dan bag pa Smra ba'i seng ge who in tum
had studied with Phya pa. Dar rna dkon mchog is said to have been
active in Yar lung and Mtsho smad temples. 'U yug pa Bsod nams seng
ge, alias Rigs pa'i seng ge, the author of the first Tibetan commentary on
38. TSHAL 69-70 [TSHAL fol. 29a-b, Inaba-SaW 1963, ISO, Chen-Zhou
1988, 62].
39. On him, see van der Kuijp 1983,117,314, note 356.
the Prama7}avarttika, had been one of Gnyal zhig's disciples prior to
joining Sa skya Pal!9ita in Sa skya. It now appears that he wrote two
other texts on tshad mao The first WQuld be the Bsdus pa text on tshad
rna-it is entitled Rigs pa [s]grub pa-which might have been written
when he was still with Gnyal zhig; he refers to it towards the end of his
Pramil7}avarttika exegesis. 40 The other text would be his short Tshad
rna mam 'greZ gyi bsdus don rigs pa'i sdom. I was ableto inspect a
late nineteenth century Sde dge print of the latter through the kind offices
of Mr. Rgya rong BIo bzang of the department of nationalities' literature
of the Beijing National Library where it is catalogued under no. 3154-l.
This little text is nothing but a topical outline of his Prama7}avarttika
exegesis which he possibly wrote while at Sa skya, so that it must be
assigned to a later period of his life.
2. No title page
Incipit: [Ib] thugs rje chen po la phyag 'tshal 10 II sangs rgyas gzhan
phan dgongs pa can I I kun bzang rab zhi gnyis myed cing I I thams cad
khyab pa'i sku ldan pa I I skyob pa mchog la phyag 'tshallo II
Colophon: [IS2a] II rgya shes rab dbang phyug dang 'jang shag kya
dpal dang s/rkyel grags pa seng ge la sogs pa chen po mams kyi zhabs
kyi rdul spyi bos blangs te mang du thos pa rgya mtsho'i pha r01 du
phyin pa'i don I slob dpon chos mchog gis mdzad pa'iti ka 'thad ldan
dang I kha che dznya na shri'i ti ka la sogs pa mthong ba'i don I bsams
pa las byung ba'i shes rab kyis legs par gtan la phab pa'i don ston par
byed pa I blo la zin pa tsam gyis mkhas pa'i mchog tu 'gyur zhing
mkhas pa rtag tu mi nyams par byed pa'i ti ka chen po 'di I shag kya'i
dge slong gtsang chu mig pa seng ge dpal gyis gsang phu ne'u thog gi
gtsug lag khang du brtsams te yar lung kha 'brug gi gtsug lag khang,du
yi ger bkod pa'o II,
This is a commentary which, as is to be expected,
takes to some extent the Indian exegeses of Dharmottara and JfianaSrI as
its point of departure, was begun by Gtsang Chu mig pa Seng ge dpal
in the monastery of Gsang phu ne'u thog, and committed to writing in
40. van der Kuijp 1983, 117; Jackson 1987, 153, note 40.
41. For later references to him, see van der Kuijp 1983, 117,314, note 356,
296 JIABS 16.2
the temple of Kha 'brug in the Yarlung river valley. His masters were
evidently Rgya Shes rab dbang phyug, 'Jang Shakya dpal and Skyel
Grags pa seng ge. Tshal pa writes that Skyel nag G ~ a g s [pal seng [ge]
had been a student of Gnyal zhig and DanaSila; a member of
SakyaSribhadra'S party, that he was responsible for setting up a tradition
of tshad ma study in SIiar thang monastery and that this attracted such
men as Skyo stonGrags [pa] 'bum and Chu mig pa Seng ge dpal.
seems therefore rather likely that this Skyel nag is none other than the
colophon's "Skyel Grags pa seng ge." Most sources have it that Chu
mig pa succeeded Rgya MI' chims ru ba as abbot of Gsang phu ne'u
thog's Gling stod college, a post which he allegedly occupied for eigh-
teen years. 43
3. No title page.
No. 004827(1)
Folios 1-68a
Upper center of the first page has phyi, zha, 9.
Incipit: [lb] tshad rna bde bdun gyi-phyogs cig du bsdus pa gzhan gyi
phyogs thams cad las mam par rgyal ba I ..
Colophon: [68a] II gzhan gyi phyogs thams cad las mam par rgyal ba
zhes bya ba I rigs par smra ba'i dge slong sing ha shris gsang phu ne'u
thog gi gtsug lag khang du logs par brtsams te I dpal rtse dkar gyi gtsug
lag khangdu yi ger bkod pa'o II rdzogs sho II
42. TSHAL 63, 71, 73 [TSHAL 1 fols. 26b, 30a-31a, Inaba-SatO 1963, 142,
152, 154, Chen-Zhou 198857,64-65]. Dpa' bo Gtsug lag phreng ba (1504-
1566) relates that although Chu mig pa was a scholar at Snar thang, he had
come to Rgya MI' ching ru ba [in the Gling stod college of Gsang phu ne'u
thog] to debate with him. Unable to defeat him, he became his disciple; see
his Coos 'byung mkhas pa'i dga' ston, Vol. 1 (New Delhi: Delhi Katmapae
chodey gyalwae sungrab partun khang, 1981) 749 [Ibid., Stod-cha, ed. Rdo
tje rgyal po (Pe cin: Mi rigs dpe skrun khang, 1986) 734].
43. For Gsang phu [s]ne'u thog, see my "The Monastery of Gsang phu ne'u
thog and Its Abbatial Succession from ca. 1073 to 1250," Berliner Indolo-
gische Studien 3 (1987): 103-127, and the studies by S. Onoda in his "The
Chronology of the Abbatial Successions of the Gsang phu sne'u thog
Monastery," Wiener Zeitschrift flir die Kunde Sudasiens 8 (1989): 203-213,
and "Abbatial Succession of the Colleges of Gsang phu sne'u thog
Monastery," Bulletin of the National Museum of Ethnology, Osaka 15.4
(1990): 1049-1071. GSERa 453 is unique in stating that Chu mig pa, Gnyal
zhig's nying slob, "grand-disciple," was never abbot of [the Gling stod col-
lege otT Gsang phu, but that he did teach there for a long time.
This work was evidently entitled the Gzhan gyi phyogs thams cad las
mam par rgyal ba, which is described a summary of the purport of
DharmakIrti's tshad rna writings. The colophon gives the author's name
. in Sanskrit as "Sing ha shri" which in Tibetan would read "Seng ge
dpal." He may therefore have to be identified as Chu mig pa as well.
The text was begun in Gsang phu ne'u thog, and completely committed
to writing in the monastery of Dpal rtse dkar.
Anonymous. Tshad ma shes rab sgron mao Handwritten
dbu med manuscript in fols. 67. Tibetan library of the Cul-
tural Palace of Nationalities, catalogue no. 004827(5).
Glo bo Mkhan chen Bsod nams !hun grub. 1985. Sde bdun
mdo dang bcas pa'i dgongs 'grel tshad ma rig[s] pa'i gter
gyi 'grel pa'i rnam bshad rig[s] lam gsal ba'i nyi rna [Sde
dge print]. Dehra Dun: Pal Evam Chodan Ngorpa Centre.
Ibid. 1970. Tshad rna rigs pa'i gter gyi rnarn par bshad pa
rigs pa rna ius pa la 'jug pa'i sgo. Gangtok.
Ibid. 1988. Tshad ma rigs gter gyi 'greZ pa [based on the
Sde dge print]. Ed. Rdo Ije rgyal po. Pe cin: Krung go bod
kyi shes rig dpe skrun khang.
Gser mdog P a ~ chen Shakya mchog ldan. 1975. Glo bo
phar phyin gyi par bzhengs pa'i dkar chag tu gnang ba,
Collected Works. Vol. 17. Thimphu. 213-242.
Ibid. 1975. Rngog Lo tstsha ba chen pos bstan pa ji ltar
bskyangs pa'i tshul mdo tsam du bya ba ngo mtshar gtam
gyi rol mtsho, Collected Works. Vol. 16. Thimphu. 443-
Gtsang nag pa Brtson 'grus seng ge. 1989. Tshad ma main
par nges pa'i ,tl ka legs bshad bsdus pa. An Ancient
Commentary on Dharmakirti's Pramtir:tavinifctiya. Otani
University Collection No. 13971, in Otani University
Tibetan Works Series. Vol. 2. Kyoto: Rinsen Book Co.
Seng ge bzang po. 1983. Mkhan chen bka' bzhi pa chen po
rig pa'i seng ge'i rnam par thar pa yon tan rin po che'i
rgya mtsho. Dehradun: Sakya Centre.
298 JIABS 16.2
Ibid. 1986. Mi nyag mkhas dbang lnga'i rnam thar.
Chengdu: Si khron mi rigs dpe skrun khang. 37-115.
Tshal pa Kun dga' rdo Ije. 1981. Del? ther dmar po. Ed.
Dung dkar Blo bzang 'phrin las. Pe cin: Mi rigs dpe skrun
Ibid. 1961. Gangtok: Namgyal Institute of Tibetology.
Song Lian, et al. 1976. Yuanshi. 15 Vols. Beijing:
Zhonghua shuqu.
Other References
Chen Qingying, and Runnian Zhou, trans. 1988. Hongshi. Lhasa: Xizang
renmin chubanshe; translation of TSHAL.
Inaba ShOju, and Hisashi SaW, trans. 1964. Hu lan deb ther. Ancient
Chronicle of Tibet [in Japanese]. Kyoto: Hozokan.
Jackson, D. P. 1983. "Notes on Two Early Printed Editions of Sa-skya-pa
Works." The Tibet Journal 8 (1983): 5-24.
______ . 1987. The Entrance Gate for the Wise (Section III). Sa
skya Par;ljita on Indian and Tibetan Traditions of Pramar;a and
Philosophical Debate. 2 Vols. Wiener Studien zur Tibetologie und
Buddhismuskunde. Heft 17,2. Wien: Arbeitskreis fUr Tibetische und
Buddhistische Studien Universillit Wien.
van der Kuijp, L. W. J. 1983. Contributions to the Development of Tibetan
Buddhist Epistemology-from the Eleventh to the Thirteenth Century. Alt-
und Neu-Indische Studien 26. Wiesbaden: Franz Steiner.
Onoda, S. 1992. Monastic Debate in Tibet. A Study on the History and
Structures of Bsdus gnva Logic. Wiener Studien zur Tibetologie und
Buddhismuskunde. Heft 27. Wien: Arbeitskreis fUr Tibetische und
Buddhistische Studien der Universillit Wien.
. Recent French Contributions to Himalayan and
Tibetan Studies
Alexander W. Macdonald, ed. Rituels himalayens. L'Ethnographie 83
(special volume). Paris: Societe d'ethnographie de Paris, 1986. Pp.
The journal L' Ethnographie has devoted a special volume to
Himalayan ritual reflecting not only the diversity of "Himalayan"
studies in France in the 1980's, but also the renewed emphasis placed
by anthropologists as well as historians of religion on ritual as a cru-
cially important theme of study in order to understand the foundations
of any social entity, whether local community or regional civilization.
Not only have studies of ritual in the Himalayan area (understood in
the broadest geographical sense so as to include the trans-Himalayan
Tibetan plateau as well) been scarce, but, as the editor points out in his
introduction to the volume, "Few comparative analyses of ritual have
been attempted. Each specialist had a tendency to shut himself up in
his own field: Ladakh, Nepal, Sikkim, Bhutan continue to be studied
separately" (p. 5). The present volume attempts to remedy this, not by
proposing far-flung comparative vistas (which would be premature),
but by juxtaposing eleven substantial articles having ritual in one form
or another as their common theme. Unfortunately, there are no con-
tributions dealing with Bhutan; here a vast and complex field awaits
The contributions to this volume can, in a very rough and ready way,
be divided into three groups: those dealing with Tibet; Tibeto-Burman
speaking peoples ("tribes") of the Himalayas, subject to Tibetan cul-
tural and religious influence; peoples not subjected to Tibetan
A very limited number of anthropologists have been able to do field-
work in Tibet itself in recent years, but only after the present volume
was edited. The Tibetological contributions to this volume are there-
300 JIABS 16.2
fore either philological, i. e., based on textual studies, or anthropologi_
cal, but based on field-work outside Tibet itself.
There are two contributions based on textual material: KaHa
Buffetrille, "Un rituel de mariage tibetain", and Samten G. Karmay,
"L' arne et Ia turquoise: un ritual tibetain." The former contains the
text and translation of a text detailing the rituals to be performed by the
lama at the moment when the bride arrives in front of the door of the
house of her future husband. It was composed by Kong-sprul Blo-
gros-mtha' -yas (1813-1899) on the occasion of the marriage of the
prince of Derge in 1870. As Kong-sprul was one of the founders of
the eclectic (ris-med) movement in eastern Tibet, it is not surprising
that the text is characterized by the inclusion of both Bonpo and
Buddhist elements. The same text had already been translated by D.
Schuh, "Die Darlegungen des tibetischen Enzykloptidisten Koil-sprul
Blo-gros mtha' -yas tiber osttibetische Hochzeitsbrauche," in R.
Kaschewsky (ed.), Serta Tibeto-Mongolica (Wiesbaden, 1973) 295-
349. This was only brought to the author's notice after she had fin-
ished her own translation, but differences in translation are commented
on in the notes.
Karrnay's article deals with the concept of bIa, "soul," on the basis
of beliefs attested from the period of the royal dynasty as well as in
later texts (Padma gling-pa 1450-1521, and a Bonpo text from 1852).
The concept of the "soul"-which of course runs counter to orthodox
Buddhist doctrine-is a fundamental one in Tibetan "popular" religion,
and Karmay convincingly shows that rituals of "calling" or
"ransoming" the soul (believed to have temporarily left the body, thus
exposing the latter to illness and misfortune) are not of Indian origin,
but perpetuate very ancient indigenous Tibetan beliefs.
Ladakh is still largely a land of Tibetan culture. Pascale Dollfus,
"Lo-Gsar, Ie Nouvel An populaire au Ladakh," provides a detailed
day-by-day description of the so-caned so-nam lo-gsar, "agricultural
New Year," which is celebrated two months before the official New
Year, not only in Ladakh, but also in certain communities in Nepal
(Dolpo, Langtang, and Yolmo), in Sikkim, and in Tibet. Dollfus, who
has done extensive fieldwork in Ladakh, especially points out the co-
existence of Buddhist and non-Buddhist elements in the Ladakhi rit-
uals, thus joining, thematically, the focus of the articles by Buffetrille
and Karmay.
The contribution of Charles Ramble likewise employs anthropologi-
cal method and perspective, but attention is now shifted to a population
of Tibetan culture in Nepal, the villages of the Baragaon region of the
Upper Kali ,Gandaki: "The Muktinath Yartung: A Tibetan Harvest Fes-
tival in its Social and Historical Context." The Yartung (dbyar-ston)
festival is one of the three annual seasonal festivals in the. area, and as
such related to the "agricultural New Year" of Ladakh. It takes place
in the seventh month, corresponding to September. Ramble not only
gives a description of the festival, but shows that it discloses an obso-
lete,social order. The festival is described with special emphasis on
tracing the course of events from the point of view of one of the three
principal communities involved, the Bonpo village of Lubra. (An in-
teresting pOint for further investigation and reflection is the apparently
double significance of the term bon-po in this area, i.e., both as a f o l ~
lower of the Bon religion in general, and as a specific religious special-
ist, p. 240). Although there is no enmity between Bonpos and
Buddhists in the Baragaon area today, this may not always have been
so, as a closer scrutiny of the Muktinath Yartung festival reveals.
Four articles deal 'with ethnic groups iI?- Nepal speaking Tibeto-
Burman languages and subject in varying degrees to cultural influence
from the Tibetan (and largely Buddhist) world to the north and the
Indian (and Hindu) world to the south (the latter augmented by politi-
cal domination by the Nepali-Hindu government in Kathmandu). As
N. J. Allen pOints out in "Thulung Weddings: the Hinduization of a
ritual cycle in East Nepal," one must be wary of simplifications: "The
Hindu-tribal dichotomy provides a model that is exceedingly crude"
(p. 33), among other reasons because "it risks attributing to the
Thulung a changelessness which .. '. is certainly unrea1." Further, "it
risks attributing to Hinduism a homogeneity that is possibly even more
unrea1.". The goal should be to "look for an underlying unity and
homogeneity in Himalayan society" (p. 16), for" ... eventually the
social history of the various peoples of Nepal will need to be studied as
a unitary field" (p. 33). Allen seeks to reconstruct the traditional style
of wedding among the Thulung, based on the accounts of informants
as well as comparative linguistic data; he sees it "as the result of Hindu
influence operating of a culture that was essentially tribal" (p. 15).
Philippe Sagant, "La cure du chamane et 1'interpretation des laIcs,"
deals with shamanism among another people of East Nepal, the Limbu.
As indicated by the title, his chief concern is to study the position of
302 JIABS 16.2
the shaman in a particular community, and the understanding among
his patients and clients of the cures he is believed to effect. Numerous
aspects of the role of the shaman are examined within the framework
of the Limbu world-view; no comparative thrusts are attempted, but
the article is fundamental for the comparative study of Himalayan
The Magar of the village of Sankh at first sight seems to present an
extreme case of acculturation to Indo-Nepalese culture. Among other
things, they have forgotten their Tibeto-Burmese language and only
speak Nepali. Nevertheless, they are proud of their Magar identity
(defined as caste identity), and in her contribution to this volume,
"Papini bitha; Ie mariage de la mauvaise fille: Essai d'identification
d'une fete magar," Anne de Sales analyses a ritual which is unique to
this village and which expresses the Magar identity of its inhabitants.
Although it reflects influences not only from the dominant Indo-
Nepalese culture but also from Tibet; it is essentially the result of an
active, creative process of ''bricolage'' in which the Magar have made
use of whatever "symbolic objects" that could serve their purpose.
Brigitte Steinmann deals with an aspect of the ritual life of the
Tamang: "Le culte des dieux du clan les Tamang: la terre, Ie livre et la
lignee." Steinmann has discussed Tamang rituals in several recent
articles as well as in a monograph, Les tamang du Nepal (Paris 1987).
In the present article she deals with the cult of the clan deities of the
Tamang of Eastern Nepal. The Tamang are not strongly influenced by
Hinduism; on the contrary, they are in a certain sense Tibetan
Buddhists, and have married lamas belonging to the Rnying-ma-pa
order. However, they have put Tibetan Buddhism to their own use
within the general framework of their complex religious institutions.
This leads to the typical duality of Tamang religion, in which archaic,
indigenous beliefs and institutions coexist with--or, as the case may
be, are opposed to-the lamaist interpretation of those beliefs. The
cult of the clan deities is no exception to this pattern. Thus the lamas
officiate at the cult of the deities of Tibetan origin associated with cer -.
tain clans, and the non-Buddhists priests, the labon, venerate the ances-
tors of the lineages (p.304).
The remaining three articles deal with very different communities.
None of them, however, are influenced by Tibetan culture. These arti-
cles are mentioned briefly here, not because they are of less interest or
Significance than the others, but because the present reviewer is less
qualified to discuss them. "N aissance d'un village tharu: a propos de
rites de claustration villageoise" by Gisele Krauskopff "provides an
analysis of Tharu representations of village territory and its boundaries
through two rituals: the first being the village-shrine foundation and
the second one a village Gloistering ritual" (p. 131). Such rituals were
regularly performed in earlier times when there was less. pressure on
forest and empty land, allowing a village to move to a new site; how-
ever, the great influx of hill people and subsequent increased ecologi-
cal degradation of the Terai had rendered these rituals practically obso-
lete. The author describes a ritual of demarcation, performed in 1983,
twelve years after disastrous floods forced an entire village to buy land
from a neighboring village in order to move to a new site on higher
John K: Locke, S. J., deals with a Buddhist Tantric ritual performed
by the Newars,"The U p o ~ a d h a Vrata of Amoghapasa LokeSvara in
Nepal." He traces the history and the development of the ritual and
provides a detailed description of the ritual as performed today, in a
specific form recorded at one of the twelve pilgrimage sites
(GuhYeSvarI) associated with this deity in the Kathmandu Valley. He
also summarizes the legends connected with each of the twelve pil-
grimage places.
The last article to be noted deals with the Kalash of the Hindu Kush;
Jean-Yves Loude and Viviane Lievre, "Fetes d' ete chez les Kalash du
Nord-Pakistan." The Kalash are the only remaining non-Muslim peo-
ple in Pakistan, and although their freedom of religion is protected by
the authorities, motorable roads have opened their hitherto inaccessible
valleys in the Hindu Kush to modernization. It is therefore of great
importance to record their rituals while there yet is time.
Fernand Meyer, ed. Tibet. Civilisation et societe. Paris: Editions de la
Fondation Singer-Polignac! Editions de la Maison des Sciences de
I'Homme, 1990. Pp. 204 + 53 plates.
This volume contains papers delivered at the seminar organized in
April 1987 by the Fondation Singer-Polignac in Paris. With the ex-
ception of three very short papers contributed by scholars from the
Chinese People's Republic, the papers are all substantial and original
contributions to Tibetan studies, including studies of related population
in Nepal.
304 JIABS 16.2
Several papers may be grouped under the heading of art, architecture
and iconography. The first is that of Gilles Beguin, "Remarques Con-
cernant les influences new ares dans la peinture tiMtaine a l'epoque des
Phag-mo-gru-pa." Beguin deals in turn with mural painting and with
thankas. The former is studied principally on the of the early
fifteenth century Gyantse sku-'bum (stiipa), and while pointing out the
conspicuous presence of Newar influence, Beguin stresses that the
murals also show a considerable degree of originality. As for thankas,
Newar influence is traced from the fifteenth to the early seventeenth
century, and a preliminary fourfold stylistic classification (depending
upon the degree of Newar influence) is attempted.
The extremely complex historical evolution of the iconography of
two protective deities is studied by Amy Heller in "Remarques prelim-
in aires sur les divinites protectrices Srung-ma dmar-nag du Potala."
The paper does not offer any final identification of the 'Red and Black
Deities' who since the Third Dalai Lama have been the protectors of
the Dalai Lama, but explores their fluctuating and complex relation-
ship to other deities, principally Beg-tse. .
Anne Chayet writes on the architectural history of Bsam-yas, the
first monastery to be founded, in the eighth century AD, in Tibet:
"Contribution aux recherches sur les etats successifs du monastere du
bSam-yas." Having to some extent escaped the destruction wrought
by the Chinese on religious buildings in Tibet, Bsam-yas still presents
a group of edifices which can provide the historian of architecture with
much invaluable information. The author concludes (p.114) that sev-
eral sources of inspiration may have influenced the lay-out and shape
of Bsam-yas: the Indian monastery of Otantapuri (or perhaps Nalanda)
as well as, on a deeper level, Mount Meru and, corresponding to it, the
non-Buddhist models such as the royal encampment of the
Tibetan kings, the ritual edifice of the mdos, and even the mingtang
palace of ancient China. The study of the origins of Bsam-yas thus
opens perspectives pointing back to a very early period in Tibetan
A fundamental study of Tibetan paintings is presented by Fernand
Meyer: "Introduction a l'etude d'une serie de peintures medicales creee
a Lhasa au XVIIe siecle." Starting with L. A. Waddell, Western schol-
arS have published a number of Tibetan anatomical paintings which
remained unconnected until the existence of a set of seventeen paint-
ings, preserved in the Ethnographic Museum of Ulan Ude, was made
known in 1979. The following year a similar set was reported from the
Institute of Medicine and Astrology in Lhasa (Sman-rtsis-khang). The
latter set was published in Lhasa in 1986. In this very substantial
. article, Mexer studies the impressive set of no less than seventy-nine
medical and anatomical paintings prepared by order of the regent
Sangs-rgyas-rgya-rtltsho (1653-1705) in order to illustrate the medical
text Vaidurya sngon po of which he was the author.
This vast series of paintings, completed, at the latest, by 1703 (p.
33), reflects the great interest of the regent in the science of medicine.
The set preserved in the Sman-rtsis-khang, which contains several
thankas that may belong to the original set dating from the late seven-
teenth century. was restored and completed by the famous medical
scholar Mkhyen-rab-nor-bu (1883-1962) on the order ofthe Thirteenth
Dalai Lama. A second set, still preserved in Lhasa, was likewise
assembled by the Thirteenth Dalai Lama and is said to have been pre-
viously kept in the Norbulinka. The Ulan Ude set was copied in Lhasa
for the Buryat monk Dorjiev (1849-1938). Meyer's conclusion (p. 48)
is worth quoting in full:
The series of medical paintings, created in Lhasa between 1687 and 1703,
under the auspices of the regent Sangs-rgyas-rgya-mtsho, constitutes, with
regard to its amplitude and originality of conception, an extraordinary
iconographical whole, not only for Tibet, but in the general history of
medicine. It is both the result of various earlier traditions and the expres-
sion of the genius of a man who, continuing the work of the Fifth Dalai
Lama, strove to construct the vast institutional and cultural system which'
was to unify Tibet under the authority of the dge-lugs-pa school. In addi-
tion to its technical purpose, the iconography was also conceived of as a
prestigious aesthetic work and a political act. '
Recently the whole set has been magnificently reproduced in two vol-
umes: Anthony Aris, ed., Tibetan Medical Paintings, (London:
Serindia, 1992).
Another field within Tibetan studies which is dominated by French
scholars, is that of Tibetan music, especially liturgic music. To an al-
ready impressive series of musicological articles, published in various
journals, Mireille Helffer now adds "Recherches. recentes concernant
l' emploi des notations musicales dans la tradition tibetaine." In this
extremely important article, the various systems of notation are pre-
sented and analyzed according to a double grid. The types of notation
are first of all classified according to instrument: human voice, drums
and cymbals, and wind instruments. Within each class, the traditions
- - - - - - - - ~ - - - - ~ ~ - - - - - - - - - - - - . ~ ~ -
3'06 nABS 16.2
of each major school. of Tibetan Buddhism are discussed, in some
cases taking different monastic traditions within a single school into
account. For reasons which are not mentioned, systems of notation
used by adherents of the Bon religion are not dealt with, but Helffer
has studied Bon traditions extensively in other articles.
Pre-Buddhist and popular religious traditions is a field of study com-
pletely dominated. by French Tibetologists: one need only invoke
names such as Lalou, Stein, Macdonald, Spanien, Karmay, and
Blondeau. In the present volume anew and fascinating contribution to
the study of symbolism and ritual is the article by Anne-Marie
Blondeau, "Questions pnSliminaires sur les rituels mdos." Blondeau
does away, once and for all, with the widespread misunderstanding in
Western Tibetologicalliterature that mdos signifies the same as the
nam-mkha' ("thread-crosses") used in various rituals. Blondeau traces
this misunderstanding to the dictionaries of Jaschke (1881) and Das
(1902), the latter inspired by G. Sandberg (Hand-book of Colloquial
Tibetan, 1894). In fact, the term mdos has a much wider implication,
being the "totality of ritual constructions and ritual objects and
materials which are assembled, including, as the case may be, nam,
mkha'." (p. 95). Blondeau stresses that the question of the nature and
function (in relation to the glud, "ransom"), and origin (Buddhist or
non-Buddhist) of the mdos can only be decided by having recourse to
relevant Tibetan texts. In fact, the Tibetan ritual literature dealing with
mdos is extremely abundant, and Blondeau concludes that the mdos
rituals are performed with a great variety of purposes, and have, in a
general way, the function of removing spiritual as well as material
obstacles. Finally, two rituals involving mdos are analyzed, one of
which is integrated into a Buddhist framework, while the other is
practically free of Buddhist associations. Blondeau takes care to avoid
hasty conclusions, but accepts a non-Buddhist origin for the mdos
rituals as probable, and citing a large number of parallel traits, suggests
a link with the ancient funerary rites.
Two historical studies are included in this volume: a short note by
Samten G. Karmay,"A propos d'un sceau en or offert par l'empereur
Shunzi," and Patrick Mansier, "La guerre du Jinchuan (rGyal-rong):
son Contexte politico-religieux." Gyarong is a region in which Tibetan
culture predominates, and which remains little studied. Mansier's
article is of considerable scope and interest, as it gives a broad
overview of the political and religious situation in the Gyarong region
in the 18th century. The enforcing of direct imperial rule, which only
succeeded after overcoming fierce resistance, had aspects which fore-
shadowed the occupation of Tibet itself two centuries later, including
the manipulation of religious rivalries between Dge-lugs-pa, Karma-
pa, and Bon-po factions.
Two articles within lfterature and linguistics round off the collection
of articles directly concerned with Tibet: Alexander W. Macdonald,
"Cendrillon au Tibet," which discusses a Tibetan version of the
Cinderella story (the text was published by Macdonald already in
1967), and Nicolas Tournadre, "Presentation de la grammaire tradi-
tionelle et des cas du tiMtain. Approche classique et analyse moderne."
The latter constitutes a welcome clarification of the nature of the
Tibetan "case" system.
Finally, two substantial articles deal with populations in Nepal. The
contribution of Philippe Sagant, "Les tambours de Nyi-shang (Nepal).
Rituel et centralization politi que" demonstrates the importance of the
small Tibetan communities of northern Nepal for a proper understand-
ing of Tibetan civilization. On the basis of an analysis of a communal
ritual in the Nyi-shang area, Sagant discloses two mutually opposed
ideologies: an archruc, decentralized pOlitical system where personal
prowess of those "elected" by the local mountain deity constitutes the
basis of power, and a centralized system, based on the primacy of age,
heredity, and clan privilege. The latter system was adopted by the
Yar-lung dynasty of the ancient Tibetan kings and later sanctioned by
Buddhism. In the case of the ritual in question, one can observe the
shift of emphasis from the first to the second.
Finally, Brigitte Steinmann, the foremost French expert on Tamang
culture and religion, has contributed to this volume: "Interpretation de
concepts tiMtains par des lamas tamang rnying-ma-pa du Nepal, dans
Ie ritual funeraire." The article discusses the extraordinary complexity
of Tamang religion, in which various elements--Buddhism, local
shamanism, the cult of clan deities-have retained their distinctive
profiles. This is strikingly brought out by the fact that each strand in
Tamang religion has its own class of priests: the lhabon are in charge
of the cult of the clan deities; the tamba recites cosmogonic myths; the
bombo are the shamans; and the lamas represent a non-monastic
Tibetan Buddhist tradition of the Rnying-ma-pa order. Steinmann
does not undertake a chronological analysis of these various elements
of Tamang religion, but instead raises the question as to which
308 JIABS 16.2
elements of Tibetan auddhism correspond to indigenous (and already
heterogeneous) Tamang beliefs, and thus have been appropriated by
the Tamang. This is the perspective in which she a n a l y ~ e s the Tamang
funerary ritual, leading to the conclusion that it may be more fruitful to
regard Tibetan texts such as the Bar do thos sgol ("The Book of Lib-
eration from the Intermediate State through Hearing") as learned
elaborations of existing popular rituals (corresponding to the basic
structure of the Tamang ritual) rather than to seek an explanation of
Tamang ritual in the Tibetan texts which are actually reGited.
Contributions to the Study of Popular Buddhism:
The Newar Buddhist Festival of Gumla Dharma
Buddhist monasticism arose to provide refuge and support for renun-
ciates seeking enlightenment, but the tradition survived by building
multifaceted relationships with lay patron communities that provided
for the monks' subsistence. Solidifying the loyalty of a cross-section
of society's economic classes, Buddhism evolved to espouse the basic
foundations of spiritually-centered civilization. Centered on high
moral standards and attuned to daily life, local Buddhist traditions en-
compassed a broad range of intellectual discourse and ritual perfor-
mances. Over the first millennium, the s a ~ g h a ' s role developed as
monks taught a variety of audiences, provided ritual assistance, and
participated in a yearly festival agenda, adapting to myriad local tradi-
tions in the process.!
The author would like to thank the family of Karkot Man Tuladhar, Subama
Tuladhar, and Sanu Bajracarya for the always generous assistance they
showed to me while I researched Buddhism in their community. My studies
would not have been possible without the kind toleration and guidance of
countless other individuals in Kathmandu's Buddhist community and I thank
the beloved Newar uptlsakas. I am also grateful to acknowledge research
funding from several sources: for 1979-1982, the Fulbright-Hays Dissertation
Program; throughout 1987, the Faculty Fulbright Program; and in summer,
1991, a College of the Holy Cross Batchelor-Ford Fellowship. A final thanks
to Gregory Schopen and Robin Lewis who offered helpful suggestions in the
drafting of this manuscript.
1. The norm of Buddhist pluralism is a striking feature in the tradition's
socio-historical profile. The Vinayas all show an early sensitivity to the
greatly varying ecological, social, and cultural contexts that monks had to
face. Recognizing this legitimate "malleability to contextual adaptation"
helps explain the great differences in praxis seen even in the early sources. G.
Schopenhas begun to articulate this central historical variable, noting how the
saI!lgha must have adapted amidst varying societies and cultures, from tribal
rain forest dwellers, to highlander nomadic pastoralists, to highly brahmanical
societies. This is an emic Buddhist perception: a Vinaya distinction between
situations in central places where the rules must be strictly observed (madhya-
310 nABS 16.2
Since exchange is the basis of social life (Murphy 1971; Harris
1989), studies of Buddhist cultural history must account for exchanges
central to the tradition's community life and specify how cultural per-
formances served this fundamental relationship. The general ideal is
well-outlined in the monastic literature. The monastic community
served the world through its example of renunciation and meditation
(Wijayaratna 1989), by performing rituals (Gombrich 1971, 201ff;
Lewis 1993b) and providing medical service (Zysk 1991). As pre-
servers and transmitters of the Dharma, the sarpgha's duty was to
attract the Buddhist lay community's merit-making donations by being
spiritually worthy (Lamotte 1984); complementing this, monks were
explicitly taught to seek outprasaditas ("dedicated sympathizers") and
danapatis ("generous donors") (Lamotte 1988, 78) so as to insure the
Buddhist sasana's existence amidst lay society.2
Guided by a missionary ethos, Buddhist monks and laity adapted
practices to diverse ecological, linguistic, and cultural circumstances.
Inclusive and practical, this tradition spread across Asia as a rich
multi-stranded fabric, carefully adapted to the logic of local life.
Future ethnographic studies and textual historical research is needed to
provide the data for understanding the diversity of Buddhist domesti-
cations and the transcultural logic of Buddhism's global adaptations.
This process of applying the universal Buddhist teachings and textual
norms to the logic of local life includes human contexts as diverse as
settled farming villages, nomadic pastoralists, and urban communities.
The "genius" of Buddhism is evident, in part, in its acceptance of plu-
ralism, especially in its many textual voices; it is also found in the
accommodation of multipraxis while still retaining a strong center that
could "re-form" indigenous ancestral religions.
Indic Buddhism shows an array of evolutionary trajectories sharing
common traits: srupas as centers of community ritual (Lewis 1993c);
viharas as refuges for meditation, study, and material resources;
sarpgha members who assume leadership of the community's ritual
defa "middle country," i. e. the Buddhist homeland) and the far-off areas
where less strict standards were tolerated (pratyantajanapada, "frontier prin-
cipalities") (Lamotte 1988, 8).
2. The specific benefits of being a generous Buddhist donor are extolled in a
Pali text: appreciation by everyone, loved by worthy individuals, renowned
everywhere, fearless in any company, rebirth in heavenly realm (Lamotte
1988,415). The literatures of all schools in all periods extol the great worth
and rewards for dana.
. Yet Buddhism in practice so encompassed the elaboration of
myriad distinctive lifestyles and cultures that even by Fa-Hsien's time
(400 CE) there was seemingly' indescribable diversity: "Practices of
. the sramaJ}as are so various and have increased so that they cannot be
recorded." (Beal1970, xxix).
This study fills in a small portion of the mosaic regarding the
Mahayana-Vajrayana calendrical orientation. Grounded in a prelimi-
nary overview of ancient Indic practices and comparative precedents, it
documents and analyzes the Mahayana Buddhist observances still
undertaken during the summer festival month of GUlp1a Dharma (or
just GUI!lla), a yearly time of intensive observances for the Newar
Buddhist community of the Kathmandu Valley.
Among the many positions represented in the Pali Nikayas, there is
one (doubtless monastic) voice that would regard the rich festiv.allife
of ancient and modern Buddhist countries as a wrong turn. Passages
from the S01J1yutta (V. 4) defining the "Stream-Entering" stage declare
"belief in the efficacy of rituals" (silabbata-parlimlisa) as detrimental
(Dutt 1945a, 181). This puritanical virtuoso voice is echoed in one
portion of the well-known Sigolavlida Sutta where Sakyamuni preach-
es restraint from attending dramatic performances (Dutt 1945a, 170).
Yet such references, which have been used to posit a pure, antiseptic
"primitive" tradition, must be connected to a context of very advanced
laity and monks, a group never constituting more than a small minority
in any Buddhist society. The judgment thus was not a blanket procla-
mation regarding society at large or in opposition to other textual
Many other early textual discourses present rationales for activities
that make positive contributions toward serving the tradition and the
Buddhist community: texts note simple mantras or textual passages
3. An early 6-fold division of monastic specialization gives clear evidence for
the sarp.gha's engaged orientation toward society: 1. instructors (dharma-
kathika);2. meditators (dhylInin); 3. folklorists (tira1cakathika); 4. sutra spe-
cialists (sutradhara); 5. Vinaya specialists (vinayadhara); 6. catecheticists
(Lamotte 1988, 149). Another specific designation often men-
tioned in inscriptions is the reciter (bhlir.zaka), which also suggests popular
312 JIABS 16.2
given bySakyamuni that could be effective in repelling negative influ-
ences in any environment; in the Milindapaiiha, the laity are instructed
to listen to the Dharma and make exertions to resist its decline (Dutt
1945a, 175). The Dlghanikaya speaks of the upasaka's duty to, "Help
others in increasing faith, moral virtues, knowledge, charity" (ibid.,
169). An early proof text in the tradition is, again, the Sigolavada
Sutta, which specifically enjoins the layman to "maintain .. ; the tradi-
tions of family and lineage; make himself worthy of his heritage; and
he should make offerings to the spirits of the departed" (de Bary 1972,
43). Still other voices (quoted below) speak about the merit of
tual celebrations in the presence of the Buddha's relics.
The early formulation called "the graded teaching': (anupurvlkatha)
established pUT}ya/dana as the foundation for Buddhism in practice,
while also legitimating a Buddhist community's diverse cultural activ-
ity. The anupurvlkatha are: 1. danalpuT}ya; 2. silalsvarga; 3. evils of
papa/kama; 4. value of renunciation; 5. Four Noble Truths (Lamotte
1988, 77). This hierarchy oflegitimate, progressive activities defines a
"syllabus" for advanCing in spiritual attainment.
As pUT}ya provided the chief theoretical orientation in the Buddhist
layman's world view and ethos, dana has always been the dOminant,
starting practice and life-long foundation of spiritual advancement.
Merit-making remained the universal, integrating transaction of
Buddhism in practice (Dargyay 1986, 180), regardless of the res-
pective intellectual elite's orientation toward competing Theravada,
Mahayana, or Vajrayana doctrinal formulations or spiritual disciplines.
PUT}ya has soteriological as well as practical, worldly consequences.
Pali suttas urge all disciples, monastic and lay, to cultivate the five
cardinal precepts to maximize pUT}ya and so the course
of spiritual advancement: 1. sraddhii ("faith"); 2. Slla ("moral obser-
vances"); 3. tyaga ("generosity"); 4., sruta ("listening"); 5. prajfia
("insight") (Lamotte 1988, 70). The Indic sources thus implicitly
authorized festival possibilities through which Buddhists could accom-
plish the precepts: venerating images, taking precepts and fasting,
arranging public recitations of sutras, and encouraging meditation
(Conze 1967, 47-55; Warder 1970, 191). It will be evident in this
essay how all five cardinal precepts are amply encouraged during
Gul!lia Dharma in the Kathmandu Valley.
The most constant expression of lay Buddhist faith and generosity
occurred through donations: dana. Dana's "investment" is described
and celebrated in the vast jataka and avadana literature and in the
great Mahayana s11tras. The Mahayana fully sustained the e.arly
framework with dana as the foundation for householder bodhisattva
. practice. Generosity to all beings is applauded, although the best
"pUl}ya return" accrues to gifts made to the buddhas, bodhisattvas, and
the srupgha. Passages in the sl1tras discussing the paramitas develop a
host of possibilities, emphasizing both the value of dana to the indi-
vidual, as an expression of karuI.1a, and its value as renunciatory prac-
tice (Dayal 1932, 165-193). Some texts make quite specific recom-
mendations to the laity on the best pUlJ.ya investments.
It was for regularizing dana presentations and valued pu!Zya-making
that monks and laity doubtless developed standard ritual procedures
and calendrical norms. Orthoprax rituals evolved that complemented
meditation and study; employing medical terms, specific rituals were
seen as compassionate action (Stablein 1973; 1978) that could achieve
specific results for suffering humanity. (For the Mahayana writers, rit-
ual (pllja) was quintessentially an expression of upaya, a disciplined
act that aids the spiritual destiny of all beings, self and others [Pye
1978,58-9, 98]. )
Buddhist rituals include spoken words and Simple deeds. The
paritta of the Pali Canon are one early manifestation (Skilling 1992);
the earliest Mahayana elaboration is on the bodhisattva's ritual service
emphasizing dharaI.11 mastery (Dayal 1934, 267-69). These long
mantras can be spoken to protect both the speaker, the srupgha, and
entire settlements. Resort to these formulae was one of the divisions in
Buddhist medicine (Zysk 1991, 66). Later Buddhist ritualism furthers
the foundations of spiritual practice and provides infusions of good
karma and radiant auspiciousness for towns, residences, and at
moments of individual life-cycle passage or crisis.
Ritual service dominated mature Indic Buddhism in its missionary
program. This is clear in early East Asian Buddhist history, where
cumulative dhara traditions were instrumental in the successful mis-
sionization of China (Strickmann 1989). Myriad other Buddhist
householder rituals evolved to insure the regular performance of such
mantra recitations that both expressed and, through recitation, orches-
4. The arranges the following hierarchy: 1. donating land to
the srup.gha; 2. building a vihara on it; 3. furnishing it; 4. allocating revenue
for it; 5. assisting strangers; 6. tending the sick; 7. in cold weather or famine,
giving food to the saI!lgha (Lamotte 1988, 72).
314 JIABS 16.2
trated the attempt to actualize such ideals. The Paficaraqa is an early
example of such traditions (Skilling 1992; Lewis 1994a).
The concern shown by the smpgha in the early Vinayas to please the
sensitivities of local society, especially monarchs and rich merchants
(Schopen 1992a), suggests that ritual service was always a part of the
saIpgha's mission. Monastic rituals and dana "events" sponsored by
notable individuals likely set precedents for later traditions (Beal 1970
ed., xxxvii). The wider Mahayana rationale for later Buddhism's lux-
uriant ritualism is succinctly expressed in the guru maJ.1<;lala puja,
which includes the repetition of the three refuges, six paramitas, the
bodhisattva vow, and the eightfold path (Gellner 1992). This trend
toward ritual service continued in great elaboration with the Vajrayana
(Skilling 1992; Lewis 1993b).
In South Asia, including Nepal, it was the competing brahmaJ.1a
priesthood and the distinctive caste-ordered society
of South Asia that shaped Mahayana-Vajrayana ritualism. The later
Buddhist adaptation of pollution-purity norms, formal life-cycle rites
(saIpskaras), procedures for image veneration, and calendarical organi-
zation all represent, within the faith, the upaya of helping the lay soci-
ety survive amidst the subcontinent's Hindu cultures. Likewise, later
Buddhist mantra collections for puja, a Buddhist homa, Buddhist
saIpskaras (Lewis 1993b), Buddhist vratas (Locke 1987; Lewis 1989a)
and festivals, ideals of Buddhist royalty (Reynolds 1972; Gokhale
1966), etc. all reflect the Indic cultural contextS and the Buddhist ethos
of adaptation.
Like other great world religions, Buddhism shaped cultures that
ordered time through regular yearly festivals. Some festivals orches-
trated the reliving of classical Buddhist events in illo tempore (Eliade
1959, 70): celebrations of the Buddha's birth, enlightenment, and
5. This state of Hindu-Buddhist competition has continued for at least 1500
years and endures in modern Nepal today. Another area in which Newar
religious study provides insight on greater lndic history (Gellner 1992;
Lienhard 1977), I will explore this topic in a future essay and further develop
the argument that the destiny of Buddhist tradition was much less contingent
upon doctrine and much more hinging upon competing modes of ritual service
and the economics of monasticism.
parinirvlil}a are universal, although differing as to season (Swearer
1987); other more regional sacred events likewise mark the year
(Strong 1992), as different communities were free to assign their own
definitions for these "auspicious daYs."6 These include: Sakyamuni's
descent from heaven where he preached to his mother, or events mark-
ing a key point in a bodhisattva's life, be it Vessantara (Cone and
Gombrich 1977), Avalokitesvara, or the death anniversary of a local
saint (Tambiah 1984; Strong 1992). Across Asia, local communities
have domesticated stories of visits by buddhas or bodhisattvas, often
explaining the ordering of the local pantheon and sacred geography
through conquest and conversion. Hsuan Tsang also notes that Indic
rituals performed at srupas dedicated to specific early saints were
based upon the individual monk's (or nun's) "school" or specialization
(BeaI197.D ed.,
Each fortnight on the new and full moon days, all early Indic saIp.gha
members had to recite the a summary of the community's
Vinaya regulations. This recitation came after any transgressions were
confessed (alocanli) in private to the monk's superior. Uposatha
became the regular occasions to review, correct, and certify the proper
standards of monastery discipline (Prebish 1975; Wijayaratna 1989).
(Based upon the Indic lunar calendar [Das 1928], uposatha includes
the overnight of the full-moon and no-moon period, hence each can
span two solar days each month [Lamotte 1988, 70]. )
Emphasizing the fundamental interdependence between saIp.gha and
lay community, householders were encouraged to visit their viharas on
the uposatha days to make offerings (dana). On these uposatha days,
devout lay folk (upasakas) have the regular opportunity to observe
eight of the ten monastic rules while residing continuously on the
vihara grounds. (The frequent lay observance of fasting after mid-day
led to their being commonly referred to as "fasting days" [Beal1970
ed., lxxiv].) In many places across India, upl(sakas donned white
robes while living under their extended vows (Dutt 1945a, 176).
Another common uposatha custom was for lay folk to remain in the
vihara to hear monks preach the Dharma. Thus, the lunar fortnight
rhythm clearly dominated. the early Buddhist festival year: each year's
passing had the absolutely regular succession of uposathas. (The
6. In China, themonastic tradition did not follow lndic precedents rigorously.
See Welch 1967, 110. Welch does note several Chinese monasteries that did -
follow the lndic norm exactingly.
316 JIABS 16.2
Aryii.tal}ga Uposatha Vrata, a later Newar Mahayana tradition
cussed below, elaborates upon this tradition and focuses the laity's
bhakti upon Amogapasa Lokesvara [Locke 1987].)
The Indic Buddhist calendar also utilized the -eighth lunar day
of each fortnight as another auspicious time for pious actions
and vow-taking. In the Pall canon (Mahavagga II, 1),asin I-Tsing's
time, these also were called "fasting days." seems to have been
the common day chosen for the early festivals outside the viharas:
is also mentioned (around 400 CE) by Fa-Hsien as
the day when a great ratha yiitriiis celebrated in Pa!aliputra (Legge
1965, 79; Dutt 1977, 39). These, too, have remained focal days for
Newar Mahayana festivais and special GUlpla observances. Hsuan
Tsang records that there were also three months each year-PhaJguna,
Kartika-when Buddhists observed "long fasts" (Beal 1970
ed., 180), another Gmpla custom that is now nearly obsolete.
Meshed with the lunar month system, the most prominent yearly
Indic Buddhist observance was the monsoon rain retreat called
varsiivilsa (paIi: vassa or vassiiviisa) (Dutt 1962, 54). Dating from
. ,
pre-Buddhist and adopted by Sakyamuni, the rain
practice required by the Vinaya cUrtailed monks' mobility outside the
monastery and encouraged meditation and study for its three-month
duration (Wijayaratna 1989). In most of South Asia, this period coin-
cides with the slack agricultural season (between planting and harvest);
it was likewise a time for intensive lay devotional exertions, as it is
until today in Thailand (Tambiah 1970, 155). Hsuan Tsang notes that
the time for retreat in India could be either A.sadha, 15 to A.svina, 15 or
Sravana, 15 to Kartika, 15 (Beal I, 72-3), variation allowed in the
PaJi Vinaya (Warren 1922,412). His account also suggests that monks
could alter the time for retreat to suit local conditions: in Baluka
(Central Asia) monks retreated during the winter-spring rainy season
Special ceremonies for came to mark the beginning
formal ending (paviiral:zii), and new robe donations (ka.thina) to monks
who completed the retreat. The paviiral}ii ceremony is much like the
7. Special ceremonies were developed by the community around the monastic
initiations for novices (pabbajjii) and full monks (upasampadii). Along with
the custom of adolescent, premarital short-term monasticism evolved in cer-
tain contexts-in Theravadin Burma (Spiro 1970) and Thailand (Tambiah
1970) and in modem Mahayana Nepal (Gellner 1992)-there also developed
the preference for initiation right before
bi-weekly uposatha for the saI!igha, but for the lay community the
emphasis is on a grander scale of merit-making, as the texts specify
that dana made on this day would be more fruitful than at other times
(Dutt 194.5b, 249). The presentation of new robes by the
laity-some traditions also evolved to have the laity sew them in spe-
. cia! ways-likewise garners special karmic rewards.
Pavara!la, the day marking the completion of the rain retreat, be-
came the year's merit-making landmark for the early community
(Beal, xxxix), a tradition that endures across South and Southeast Asia.
(Tambiah 1970, 154-160). Ancient Indic caitya veneration recounted
by 1-Tsing on this day in 690 CE exactly resembles the City-wide
Newar GUlpla: rituals, especially Mataya in Patan (Gellner 1992,
218-9) ..
[On the PavaraIJa day] the assembly should invite a preceptor to mount a
high seat and recite a Buddhist sutra, when lay devotees as well as priests
throng together like clouds of mist They light lamps continually, and offer
incense and flowers. The following morning they all go out round villages
or towns and worship all the caityas with sincere minds. They bring storied
carriages, images in sedan chairs, drums, and other music resounding in the
sky, banners and canopies hoisted high ... At this time, laymen present
gifts ... (1896, 87)
All Buddhist lineages applaud the great pU!lya accruing to those who
build viharas. Ancient texts (Lamotte 1988, 72) and modern belief
(Spiro 1986,458; Tambiah 1970, 147ff; Welch 1967, 210ft) assert this
as the greatest possible dana. There are indications that ancient yearly
festivals were established locally to celebrate each shrine's anniversary
of dedication, and these became thereby its yearly "birthday" when
donor families should refurbish, clean, and ritually renew it (Beal,
xxxix). (This is a widespread practice in the modern Newar Buddhist
community [Lewis 1984, 394]. )
Just as the vihara was the institution that ordered and sustained the
saIpgha's communal life, so, too, were there institutions that advanced
the Buddhist interests of the laity. Some inscriptions indicate the
coordinated pious activities of craft guilds (sre!li); more common are
the go:\:thi-"assemblies, associations, fellowships" (Monier Williams
1956 ed., 367)-that coordinate large donations or regular rituals.
These institutions are ancient, as the Pa:li jatakas cite subscription
plans among upasakas (Rbys-Davids 1901, 886). Such groups were
often formed to complete caityas or meditation caves, or for ren-
318 ]lABS 16.2
novation projects (Dehejia 1972; Kosambi 1965, 182). There are very
old Newar traditions which organize regular rituals, pilgrimage,
restorations, even shrine cleanings (Toffin 1975; Lewis 1984, 179-
182). Another even more important aspect of Newar practice should
be underlined: hold collective properties, including money;
most include some provision for increasing the group treasury by
lending these funds at interest serially through the membership. Thus,
not only underwrote pious Buddhist practice: such institutions
became important sources of community investment capital.
Stopa veneration constituted the earliest ritual focus for both monks
(Schopen 1987; 1989; 1991a,b,c) and laity. Rules and regulations for
their establishment and maintenance doubtless followed, as did the
custom of celebrating the monument's foundation "birthday." For all
Buddhist schools, the srupa became a focal pOint, the singular land-
mark denoting the tradition's spiritual presence on the landscape
(Dallapiccola 1980; Harvey 1984; Snodgrass 1985; Lewis 1993c).
Buddhist writers advanced many understandings of srupas, but for pre-
sent purposes let us observe that srupas marking events associated with
the tathagata(s) were the natural sites for Buddhist festivals of
remembrance and veneration.
I-Tsing's recounting of a ritual at an
Indian vihara around 685 CE shows the centering of traditional cele-
brations at caityas:
The ... priests perform worship of a caitya and the ordinary service late in
the afternoon or at evening twilight. All the assembled priests come out of
the gate of their monastery, and walk three times around the Stiipa, offering
incense and flowers. They all kneel down, and one of them who sings well
begins to chant hymns describing the virtues of the Great Teacher with a
8. Symbiotically, great regional stUpas were pivotal in the social history of
Buddhism: these monuments became magnets attracting vihara building and
votive construction, for local piija and pilgrimage. The economics of
Buddhist devotionalism at these centers generated income for local
artisans, and merchants (Liu 1987), an alliance basic to Buddhism throughout
its history (Dehejia 1972; Lewis 1993d). At these geographical centers
arrayed arqund the symbolic monument, diverse devotional exertions, tex-
tual/doctrinal studies, and devotees' mercantile pursuits could all prosper in
synergistic style. The regional Mahacaitya complexes, with their interlinked
components-viharas with land endowments, votive/pilgrimage centers, mar-
kets, state support, etc.-represent central fixtures in Buddhist civilization.
For local communities, such stiipas were also focal points in the yearly festi-
val round, drawing Buddhists toward the sacred precincts. Empowered votive
artifacts bought by the pilgrims at key lndic sites were likely used in the
establishment of caityas and buddha images in frontier settlements.
melodious, pure, and sonorous voice, and continues to sing ten or twenty
slokas. They in succession return to the place in the monastery ... when ...
a Siitra-reciter, mounting the Lion-Seat, reads a short siitra ... among the
scriptures which are to be read, [is] the "Service in Three Parts" ... by the
venerable Asvaghosha ... from which its name is derived. (Takakusu 1896,
Once the making of Buddha images became accepted (Dehejia 1989,
Lancaster 1974), their construction, consecration, and upkeep must
likewise have entailed ritual observances. I-Tsing makes the funda-
mental case for their role in Mahayana Buddhism:
Tnere is no more reverent worship than that of the Three Jewels, and there
is no higher road to perfect understanding than meditation on the Four
Noble Truths. But the meaning of the Truths is so profound that it is a mat-
ter beyond the comprehension of vulgar minds, while the ablution of the
Holy Image is practicable to all. Though the Great Teacher has entered
Nirvlll}a, yet his image exists, and we should worship it with zeal as though
in his very presence. Those who constantly offer incense and flowers to it
are enabled to purify their thoughts, and also those who perpetually bathe
his image are enabled to overcome their sins ... receive rewards, and those
who advise others to perform it are doing good to themselves as well as to
others ... (Takakusu 1896, 147)
Such were the sentiments that by Gupta times legitimated the fun
elaboration of Buddhist ritual and festival traditions, and this historical
observation is matched by texts such as the Bodhicaryavatara that laud
precisely these activities. As we have seen in I -Tsing' s account,
Buddhist puja was practiced by entire viharas in conjunction with the
lay community and by individual monks with their private icons. He
mentions detailed procedures, including rites with
anointed water, repainting, pOlishing; accompanied by music, the icon
would then be reinstated in the temple, with offerings of incense and
flowers. The water used for this ritual is likewise described as medic-
inal (Takakusu 1896, 147).
Another example of Buddhist ritualism is the "bathing the Buddha
Image" puja that commemorated Sakyamuni's birthday in the month
Vaisakha. As found in the Kashrniri NilamatapuralJa: "In the bright
fortnight, the image of the Buddha is to be bathed with water contain-
ing all herbs, jewels, and scents and by uttering the words of the
Buddha. The place is to be carefully besmeared with honey; the tem-
ple and stilpa must have frescoes, and there should be dancing and

320 JIABS 16.2
amusements" (Dutt 1977, 14). This ritual spread across Asia (Lessing
1976). -
I-Tsing underlines the immense pU1J.ya earned by Buddhist pujas:
"The washing of the holy image is a meritorious deed which leads to a
meeting with the Buddha in every birth, and the offering of incense
- and flowers is a cause of riches and joy in every life to come. Do it
yourself, and teach others to do the same, then you will gain immea-
surable blessings" (Takakusu, 151-2). A popular -Khotanese Maha-
yana text concurs, stating that to make a buddha image is to guarantee
rebirth in Maitreya's era (Emmerick 1968,321); in another verse, wor-
shipping an image is said to be equal in merit to worshipping the
Buddha himself, as both emanate from the dharmakaya: "Whoever in
my presence should perform merits, or whoever should produce faith
equally before an image, equal will be his many, innumerable, great
merits. There is really no difference between them" (Emmerick 1968,
201). Thus, many Mahayana sutras, in agreement with the Pali
Parinibbana Sutta's description of relic veneration, laud as especially
meritorious offerings of incense and flowers to images, encouraging
the presentation with musical accompaniment. Sites identified with
bodhisattvas were also centers of Mahayana worship: "Whatever
Bodhisattvas for the sake of bodhi have performed difficult tasks such
as giving, this place I worship" (Emmerick 1968, 163).
Travel to venerate the stiipas and caityas marking important events
in the Buddha's life also defined early Buddhist pilgrimage (Lamotte
1988, 665; Gokhale -1980). This meritorious veneration of the
Buddha's "sacred traces" (Falk 1977) was organized into extended
processional rituals. The development of pilgrimage traditions shaped
the composition of site-coordinated biographies of Sakyamuni
(Lamotte 1988, 669; Foucher 1949) and likely did so for some of the
jataka and avadana compilations. -Such texts promised the laity vast
improvements in their karma as wen as such mundane benefits as
rewards for undertaking pilgrimage. The Chinese reports on the
notable sites visited seem to rely on such texts for the information
provided; their accounts also give clear testimony to the sense of the
wide-ranging benefits (pwJ.ya, blessings, health, etc.) that the pilgrim
The traditional deSignation of Buddhist sites specified first four, and
then eight centers marked by monuments (Bagchi 1941; Tucci 1988).
By the time of the Asokavadana's composition, thirty-two pilgrimage
centers existed in the Gangetic basin visited by devotees (Strong 1983,
119ft). There was also a circuit in northwest India (Lamotte 1988,
335). Such religious travel had important economic effects, and local
economies developed around the great caityas. By 400 CE, the world
of Mahayana Buddhist pilgrimage had long transcended the Gangetic
culture hearth to include stOpa sites in Khotan, Sri Lanka, Srlvijaya
Java, Funan, and China. Monks, pilgrims, and traders traveled the
same routes (Takakusu 1896; Birnbaum 1989- 90, 115-120).
Relatively little is known about the history of regional Buddhist pil-
grimage traditions in India. The Chinese accounts give a sense of a
regular roughly guided by a clockwise pradak#rzli path
(Lamotte 1988, 665). If modem customs reported in the ethnographic
literature (e. g. Keyes 1975) have ancient precedents, then uplisakas
from localities likely organized group outings, read from the textual
guidebooks, followed the lead. of local guides in visiting the many
places linked to Sakyamuni's biography. Therewere probably sea-
sonal preferences for viSiting certain sites, such as being in Bodh Gaya
for the full moon in Vaisakha, when Sakyamuni's birth, and
parinirvlirza were celebrated. Hsuan Tsang notes a festival of bathing
the Bodhi tree at Gaya (Beal II, 117; cf. Strong 1983, 125-127), of
ascending a nearby mountain overnight before the dlina presentations
are made after (Beal II, 115), and the belief that circumambulat-
ing the tree secures power of knowing former births (Beal II, 124).
Another Mahayana festival focused on the "cult of the book"
(Schopen 1975). According to the early Prajfilipliramitli texts, ven-
eration of the Buddha's Dharma was vastly superior to worshipping his
bodily relics. A section of the, Saddhannapurztj.arika describes the
superior ritual in which a Mahayana text is venerated (Kern 1884, 96)
(and in the Chinese version is carried on the devotees' heads [Hurvitz
1976, 82]). As will be seen, such traditions are still evident in the
Newar GUl!lla observances.
The most extraordinary form of Buddha image veneration noted in
numerous locations was the ratha ylitrli ("chariot festival"). The
Chinese pilgrim Fa-IDen noted that in Pataliputra, there were images
ofbuddhas and bodhisattvas placed on twenty four-wheeled, five-story
rathas made of wood and bamboo. Beginning on an day and
continuing for two nights, the local vaisyas are said to have made vast'
donations from speCially-erected dwellings along the path; in Khotan,
too, there was a fourteen-day event that was attended by the entire city,
------ -------------
322 JIABS 16.2
for which each monastery constructed a different four-wheeled ratha
(Legge .1965, 18-19) .. Although ratha yatriis are not held during
GUlpla, the greatest yearly festivals of Newar Buddhism are the ratha
yatriis of Pat an and Kathmandu dedicated to A valokitesvara each
spring (Locke 1980; Owens 1988).
The early texts also mention an extraordinary quinquennial festival
called pancavarsika (Strong 1987,91-97; Strong 1990), pancavarsika
(Beal I, 50), or by the Chinese pilgrims (Beal I, 214).
Although there is no clear consensus as to its origins (e. g. Lamotte
1988, 66; Edgerton 1953), pancavarsika was clearly a time when vast
royal donations were made to the saIpgha, other deserving ascetics,
and the destitute. The Chinese accounts and the avadana
citations point to the custom of a king giving all material goods he
owned to the saIpgha, followed by his ministers buying it all back with
gold from the treasury. There are a number of these celebrations in .
Central Asia and India, several during the autumnal equinox.
Pancavarsika was also a time for displaying extraordinary images or
renowned relics during festivities organized by kings'and merchants,
while witnessed by a huge social gathering. This "Five-Year Ja:tra:"
was the most dramatic single cultural performance that contributed to
the saIpgha's material existence.
According to Newar Buddhist tradition, the pumima (full moon) of
GUIpla: commemorates the day Sakyamuni Budctha attained enlighten-
ment and defeated Mara. The two weeks surrounding this day are thus
9. There is some evidence that th,e name itself did not stem from the time
period. Hsuan Tsang's report of SIladitya's Paficavarsika (Beal, 233) sug-
gests that the name may derive from the view that five years' wealth is
expended in bestowing the necessary dana. Other sources support this, imply-
ing that the festival could be done at any time by a donor with the requisites
for feeding all the local srup.gha and others who were needy. The Sanskrit tale
of Sanavasa, a merchant who returns home and "celebrates a paficavarsika,"
suggests this (Lamotte 1988, 207), as does a passage from the
(See Dutt 1977, 51). In both Patan and Kathmandu, five-year and twelve-year
Samyaka dana festivals (respectively) doubtless represent a continuity of this
ancient custom (Sakya 1979; Lewis 1984; Lewis 1994b).
the natural focus of the local Buddhist year.
Although no direct
identification of Gumla (ninth month) as varsli exists, the activities in
. .
the Kathmandu Valley communities, the monsoon time of year, and
the Pafica .Dana custom (discussed below) all invite the supposition
that some historical association must exist.
This section provides a
minimal background sketch of modem Newar Buddhism necessary for
appreciating the details of festival practice. 12
The riches from trans-Himalayan trade, the fertility of valley soils,
and relative political isolation all endowed the Kathmandu Valley
(until 1769, the defining area of all "Nepal") with the ability to support
a rich, artistic, and predominantly Indicized civilization. Although
mentioned in passing references across earlier Indian literature, no
epigraphic evidence has been found in Nepal before 464 CE when
Sanskrit inscriptions attest to the Kathmandu Valley as an Indic fron-
tier zone ruled by a rlijavamsa calling itself Licchavi. Diverse Hindu
and Buddhist traditions existed in close proximity (Beal II, 80-81),
with the most mentioned srup.gha that of the Mahasarp.ghikas. Among
over 200 recorded inscriptions, there are references to caityas, land-
owning viharas, and and patronage by caravan lead-
ers. A few hints of Vajrayana practice are discernible, but Mahayana
themes predOminate: votive praises are addressed to and
other tathagatas as well as to the bodhisattvas Mafijusrl,
Samantabhadra, and, most frequently, Arya Avalokitesvara (pal 1974).
10. The Newar lunar month is divided into a waxing fortnight. including the
full moon day (punhi), is indicated by the suffix thva (hence,
and a waning fortnight, indicated by the suffix gli As pro-
moted by the modernist Theravada movement (Kloppenberg 1978; Tewari
1983), most Newar Buddhists have also adopted the Buddha Jayanti festival
(in mid-spring), which offers an earlier and contradictory date for commemo-
rating these same events. For a summary of the Newar Hindu festival year in
Bhaktapur, see R. Levy (1990).
11. As the rain retreat is an inauspicious time for weddings in Tberavada
countries (Spiro 1970; Tambiah 1970), so, too, is an inauspicious
month for Newar weddings.
12. In accounts of Buddhist history generally, a socio-culturally informed
depiction of Mahayana Buddhism in practice has been neglected. Such lacu-
nae have been often noted and recent anthropological studies on Himalayan
Buddhism have begun to illuminate northern Buddhism in practice (e.g.
Ortner 1989; Holmberg 1989; Mumford 1989). On Newar Buddhism, the
great Indologist Sylvain Levi's classic study of Nepal (1905-8) is still valu-
able. Important recent studies of the Newar Buddhist context are listed in the
bibliography under Allen, Lienhard, Locke, Toffin and Owens. David
Gellner's monograph (1992) is a recently published landmark study.
324 JIABS 16.2
Although little has been published on Nepalese Buddhism in the
post-Licchavi period (900 AD-1200) or the MalIa era (1200-1769);
there is ample evidence of continuing archaic texts and artistic tradi- .
tions (Macclonald and Stahl 1979; Slusser 1982). Unconquered by the
Muslim or British empires that ruled South Asia, Kathmandu Valley
civilization still preserves many ancient Indic traditions tbat endure in
the distinctive urban society and culture of the Newars (Lienhard
1984), who speak a Tibeto-Burman language. This is true of both the
Hindu and Buddhist religious traditions which are observed in rich
multiplicity .
The former city-states of the Valley-Bhaktapur, Patan, and
Kathmandu-all evolved in parallel form according to the caturvan:za
model, though differing in details of their caste nomenclature. Caste
defines the social order and dominates socio-cultural discourse with
Hindu or Buddhist identity a boundary marker at the highest levels.
The Newar Buddhist community consists entirely or house-holders
(Locke 1985; Gellner 1992). The saIpgha has for centuries married
and now a two-section endogamous caste group with surnames
Vajracarya and Sakya maintain the "monastic traditions." They still .
inhabit dwellings referred to as vihara (New. bllhll) and over three
hundred viharas exist in the Valley today (Gellner 1987). This domes-
tication evolved centuries ago, as no vestige of fortnightly uposatha
endures and Vinaya texts are rare in Newar manuscript collections
(Takaoka 1981; Novak 1986; Mitra 1971 ed.). For over five centuries,
however, Newars desiring the classical celibate monastic diSCipline
could take ordination in the local Tibetan viharas (Lewis and Jamspal
1988; Lewis 1989c).
Like married Tibetan lamas of the Rnying-ma-pa order, the house-
holder vajrllcllryas take training and serve the community's ritual
needs (Gellner 1989), with some among them specializing in textual
study, medicine, astrology, and meditation. The Newari ,Acllrya-
defends the evolved role of the householder llcllrya
as superior to the celibate monk according to Mahayana-Vajrayana
ideology that individuals should engage with their society and culture.
This llcllrya bodhisattva is similar to the textual bodhisattva
13. Tibetan version in the Tanjur is entitled Vajracaryakriyasamuccaya
(Shukla 1975, 129).
The real teacher is he who apart from other qualities, does not live like a
monk, does not shave his head and puts on good clothes and beautiful
ornaments. Amongst other qualities of a teacher are counted his knowledge
about purificatory rites, his kind disposition, pleasing humor, maintenance
of all theacaryas, insight into the art of architecture and in the science of
mantras, skill as a profound astronomer ... and his capability to select an
auspicious plot for the construction of a stupa and for the installation of the
idol of the Buddha. (Shukla 1975, 127-8)
The Newar Buddhist spiritual elite still passes on vajrayana initia-
tions (Skt. New. dekka) through guru-chela ("teacher-disci-
ple") lineages. Only those born as vajrilcilryas may take formal initia-
tions into the householder sarpgha as each male must be ritually
empowered to be eligible to perform basic rituals for laymen (Gellner
1988). The traditional line of this is in the main vihara (New.
rna bilhil) of the patrilineage. The folklore about great Newar tantric
masters describes special retreats called purascaral} cwanegu under-
taken to build upon this foundation to pursue spiritual insight and
supernormal powers. As only select groups may take the esoteric
vajrayana initiations, the Newar Buddhist tradition is formally two-
tiered, with only high-caste Vajracaryas, Sakyas, and Uray (merchant
and artisan subcastes) eligible for the that direct meditation and
ritual to tantric deities such as Sarpvara, Hevajra, and their consorts.
Most Newar Buddhists, including all from the lower castes, partici-
pate exclusively in the exoteric level of Mahayana devotionalism.
They direct their devotions to caityas (especially the great srupas such
as SvayambhU) and make regular offerings at temples dedicated to the
celestial bodhisattvas. Lay folk created hundreds of voluntary organi-
zations (New. guthi, from Skt. some with land endowments,
that have supported devotional practices for centuries. Buddhist insti-
tutions and devotions in Nepal have been underwritten by guthis since
Licchavi times (Riccardi 1979).
Despite the anomaly of a caste-delimited sarpgha, Newar Buddhist
laymen closely resemble co-religionists in other countries. They sup-
port their local vajrilcilrya sarpgha who, in return, help them look after
their spiritual destiny in this world and beyond. A vast and complex
web of ritual relations link laymen to their vajrilcilrya sarpgha who
perform life cycle rituals (Lewis 1993b), festival pOjas, textual recita-
tions, healing rites, site consecration ceremonies (Gellner 1992;
Slusser 1982,420-1). We now survey the specific observances of the
Gurpla festival to note how spiritual service exchanges (dilna) , fun-
326 JIABS 16.2
damental to all Buddhist communities, have been domesticated in the
Newar community.
As was common throughout Asian history, merchants and artisans are
prominent among Newar Buddhist laymen and have been the major
patrons of stiipas and viharas across the Kathmandu Valley. Like all
Newar Buddhists, they have formal ties to the Newar sarpgha through
their family vajrllcllrya purohit: the ritual traditions that came to de-
fine Newar Buddhist identity involve frequent yearly and life-long
resort to this purohit's pujas.
But Kathmandu Vaney Buddhism has for a millennium been an
international phenomenon: one segment among Newar Buddhists has
for centuries sustained alliances with Tibetan Buddhism (Lewis ,1989;
Lewis and Jamspal 1988; Lewis 1993a; Lewis 1994a); more recently
(since the 1920's), another faction has helped transplant Theravada
modernism from Sri Lanka and Southeast Asia (Kloppenberg 1977;
Tewari 1983; Lewis 1984). The Uray merchants of Kathmandu (who
derive their name from the Sanskrit upllsaka) have been'leaders in
supporting both of these camps as well as in performing the special
devotions that define the greatest Newar Buddhist festivals. Based
upon continuing research among one prominent Uray subcaste, the
Tuladhars ("Scale Holders"), this section describes the many aspects of
Gumla as it is celebrated in Kathmandu's old bazaar.
We must at this juncture introduce the focal point of Valley
Buddhism and Gurpla celebrations in Kathmandu city: Svayambhu
The Mahayana history of Nepal, claiming origins in
earlier yugas, recounts Svayambhu's origins-and the entire Nepal
Valley's establishment-as the product of Mahayana hierophony and
the compassionate actions of bodhisattva MafijusrI. The epigraphic
evidence is that "SvayambhU Mahacaitya" was founded in the
Licchavi period (400-879 CE) in the early fifth century (Slusser 1982,
14. In a remarkable final section appended around 1830 by a Newar
vajracaryapandit to the Sanskrit version of the Buddhqcarita (Cowell 1969
ed.), Svayambbi:i is described as the last site visited by Sakyamuni before his
parinirva!}il. Another Mahayana source on the origins of this focal stupa in
GmpJa observances is the SvayambhU Pura1}a (Shastri 1894), a late text
known only in Nepal and Tibet.
174; Riccardi 1979). Today, this hilltop stupa over twenty-five meters
in diameter is surrounded by five monasteries and quite regularly
linked to the Buddhist festival year of Kathmandu City (Lewis 1984).
It is also tied to all Buddhists in the Kathmandu Valley and surround-
ing regions through the twelve-year Kathmandu Samyaka festival
(Sakya 1979; Lewis 1994b). Svayambhil once had extensive land
endowments traditionally dedicated to its upkeep (Sakya 1978) and in
many respects the history of regional Buddhism is embedded in the
layers of thissrupa's successive iconography, patronage, and restora-
tion. Thus, Svayambhil is a chief point of reference for the GUIP.Ia
festival as wen for the upasakas of Kathmandu City: living proximate
to it is thought to be so great a blessing that Buddhists use the term
("merit-field") as a synonym for their Valley.
The Month-long GUlJ1lil Activities
The saIp.gha members of a few large viharas recite Buddhist texts each
day during GUlp.Hi. There is an old practice of reading out the entire
Nava Grantha (Nine Tomes) distinctive to Newar Buddhist tradition.l
Throughout, many vajrilcilrya priests go in early morning to
read siltras at the most frequented centers of Newar Buddhism:
Kanakacaitya Mahavihara (colloquially, "Jana Baha"), Svayambhl1,
and Santighata Mahavihara (SrI gha: baha). They most often :recite
texts that confer protection (e. g., [Lewis 1994a] ) and, at a
layman's request, chant protective mantras which are sealed by his
touching a leaf of the text to the person's bowed head. For this service
he receives a small payment in money and/or rice.
Among Newar lay folk, there are several guthis (Skt. orga-
nized for the purpose of text recitation daily during in Jana
Baha, Asan Baha, and in Uray courtyards. The most common text
chosen is the Mafijusrl NilmasalJ1glti (Bajracarya 1991; Davidson
1981; Wayman 1985). To accommodate those who go to Svayambhil
individually or with the biijan (see below), the tutaJ:t bvanegu
("siltra recitation") begins about the time the biijan returns to the
bazaar, i.e. about 7 AM, or else is done in the evenings.
15. The Nava Dharma or Nava Grantha are: Prajfiaparamita; GaTJ4avyilha;
Dasabhilmi; Samtidhiraja; Lankiivatara; Saddharmapw:ujarika; Lalitavistara;
SuvanJaprabhtisa; Tathligatagilhya (Hodgson 1874). No scholastic or philo-
sophical tradition ordering these works has been discerned; more likely is
their grouping for ritual purposes where they are arranged in the guru m31)<;Iala
piija within the dharma illaIf4ala, with Prajfiaparamita at the center.
328 HABS 16.2
Although month-long text lecture series (dharmade.anil) by the
Newar srupgha were until the last decade common at important viharas
in Kathmandu, from 1979-82 only one such series was done during
GurpIa. '
Offerings of pleasant sounds at stiipas and before images of the
buddhas and bodhisattvaS are advocated in many Mahayana sutras.
Generations of Newar musicians have responded with devotional fer-
vor to the Mahayana's resounding encouragement of such ritual cele-
brations. The bajan's morning serenade reminds city Buddhists each
morning of it being Gurpla (Lewis 1989b).
The bajan is best defined as a "music playing group" and there are
. many bajan forms in Newar culture (Wegner 1986; 1987; 1988), each
with the religiOUS purpose of adding a musical component to devo-
tional processions. During Gurpla, the Tuladhar Gurpla bajan plays
two kinds of portable drums, the naykhilJ1. and the dhaJ:!, and two kinds
of cymbals, the tab and the chusya. In modern times, the group has
employed low-caste Damai musicians who play the melody lines using
Western instruments: clarinets and trumpets.
The master of the taJ:!
cymbals, who leads the bajan's playing, is the role of highest musical
expertise since he signals the other musicians and leads the orchestra
in playing together.
In every Newar Buddhist caste, leadership of the bajan's devotional
activities rotates through internal sub-groups,l7 Every year a new
senior leader (pala) with his sub-group leaders are in charge of the
actual performance of the bajan's devotions, both during GurpIa and
for the other activities. IS
16. High caste Newars traditionally never played the flute or other wind
. instruments because these connote low caste status; until recent decades, they
used to employ Jyapus (agriculturalists, ranked as siidras in caste lists) who
still play (among themselves) a variety of flutes and maintain many other old
musical traditions.
17. The Newar procession tradition defines every major institution in local
Buddhist organization (Lewis 1984; M. Allen, n. d.) and orchestrates all
important "cultural performances" (Singer 1972, 70). Newar Buddhist mer-
chants today form eight different kinds of devotional processions.
18. The yearly schedule of GmpIa activities in the Asan Tuladhar community
actually begins before the month commences. For up to a whole month
beforehand, . the paWs organize informal instructional classes for younger
members. The first required gathering occurs several days before the start of
GU:qlla: the group employs the senio011ost Vajracarya from the neighborhood
to perform a piija to Nasa dyalf (Siva-Nataraja), the deity whom Newar
Buddhists must worship before devotional playing of any sort can officially
A standard day19 for the Asan bajan during GUIpla goes as follows: a
group of bajan drummers leave Asan just after sun-rise and proceed
directly to the Svayambhii hilltop. Once they reach the great srupa,
they go around it once in pradalqi1Ja. After this, the group settles at a
traditional site just north of the Arnitabha niche. Others who left Asan
after the bajan arrive for the next half hour. The palas also arrange for
the delivery of the extra drums, usually by employing a hired worker
(most often still a member of the Jyapu farmer caste) to carry them.
Some stand around and talk, others make private devotional rounds,
some may practice drumming. At a publicized time (1991: 6:30 AM),
a group-gathering drumbeat is played on the naykhilJ'L, and the head
paW takes roll to insure that all of the year's pala-comrnittee house:.
holds are represented. By this time, the Damais have appeared and the
bajan's daily piija offerings, already prepared by the women of pala-
committee households, is distributed among those present. These
items are offered at the shrines around the hilltop as the group pro-
The Svayambhii complex pradalqil}a begins by all present standing
before the elaborate Amitabha shrine on the western side of the srupa.
As the piija plate is presented to the shrine attendant, the group stands,
hands in "namaskara," and sings one or more of the standard Newar
Mahayana devotions: Da.sabala Stotra, the Saptaparamita Stotra, the
Bandesi or the Bhadra-cari.
Once the piija plate is returnel;i, the
begin (Ellingson 1990). The bajan group tllen proceeds to the pala's house
where the fIrst offIcial drumming is ceremonially begun and all take prasad
from the puja, and then the group is served a snack of beaten rice, meat, and
other vegetables, plus aylaJ;" "distilled spirits."
19. This description is of a typical day in 1981; in this as in many other
devotional activities among the Tuladhars, the individuals involved have a
wide range of acceptable alternatives they can introduce. Minor variations
from this description occur, based upon the leadership of that year's pala
20. These texts have been reproduced in many modem published devotional
texts in Kathmandu. The Dasabala ("Ten Strengths") celebrates the ten pow-
ers of the Buddhas (e. g. Shrestha 1983). ,
Bandeirl is a dharaI)I dedicated to Vajrasattva and praises: Bodhisattvas
Amoghapasa, Lokanatha, and Samantabhadra; dharmadhatu caityas; the bud-
dha consorts (Tara, MamakI, LocanI, PadmanI); and the Saddharma-
pU1y;!afika. _
The full name of the Ne_wari Bhadracari is the Sanskrit Aryabhadra-
caripraT} idhana raja , verses oliginally appended to ,the GaI)c;lavyUha (Beyer
1973, 478; translation 188-9). These litually repeated Jines date from the
early centuries CE; the dharaI)I verses praise buddhas and bodhisattvas,
---- - - ~ - - - - - - - - - -
330 JIABS 16.2
group follows a route that completes a double prada10i(la of the stUpa
and then visits, in a clockwise order, all of the important shrines on the
hilltop including: the Bka' rgyud Tibetan gompa, Basundhara, the
"Sikkimese gompa," HarIti Ajima, and the tantriC shrine Santipur.
When this cycle is complete, the group returns to its "base." After a
short rest, the bajan again takes up the drums and cymbals, plays the
assembly drumbeat and then launches into the prime instrumental
composition played by Buddhist GUl11la blijans: "Bhagavan Gvara."
All Kathmandu bajans play this (Sakya 1971). On a good day, the
group will have swelled to sixty or seventy with more late arrivals for
the final of the stupa and the descent down the main stairs.
On the way down and back to town, the group is careful to move in a
clockwise manner around any shrine or stupa it passes. The older gen-
eration knows that different compositions and drumming riffs should
be played at different pOints along the route back to town, but such
strict details are not usually followed these days.
A key stop on the way back is at Bijesvar!, a yogini temple promi-
nent in the Newar Vajrayana tradition. Here the bajan pauses for
darsan and, if desired, individual pl1ja. Some of the elderly Uray men
from the Asan community come out to meet the bajan only here where
the final group attendance record is noted. After crossing the river and
entering back into town, all the hajan proceeds to the lana BaM
Avalokitesvara temple for prada10i(la, exits around the Kel Tol Ajima
temple just outside the entrance, then goes around the three main tem-
ples (Annapl1f1,1a Dev!, GaI,1esa) in'central Asan.
To complete the morning procession, the group proceeds to Asan
Baha for a prada10i(la of the tall Asoka Caitya there, navigates the
aUeys for a closing darsan of Siva Na!araja (Newari: Nasa
[Ellingson 1990]) outside Ta Che BaM, then returns to the plila's
house where it plays a closing drumbeat. The palli takes the drums for
safekeeping and the bajan' s daily round is complete.
The GUl11la bajan provides a rich and satisfying devotional channel
for Newar Buddhists who value the old musical traditions and who
wish to venerate the sacred caityas, viMras, bodhisatvas, and deities
who protect their locality (Harm, G3I,1esa, and Siva-Nataraja).
delineate bfferings, confess transgressions, rejoice in merit, request teachings,
and share the pUl}ya earned from personal devotion. The bodhisattva
Samantabhadra, invoked in these texts, finds little other devotional attention
in Newar tradition (Tuladhar 1986, 8-20).
The ability to play the drums and cymbals is one traditional measure
of the culturally-accomplished Newar Buddhist layman. Likewise,
each Newar caste's musical performance of the bajan during GUl!1la
. reflects .upon the sub-community's status and solidarity .. During
GUI!11a, this-is evident during the morning excursions to Svayambhll
and on other outings discussed below. The Buddhist bajan remains a
vibrant tradition, as there is still a core of Tuladhars-including many
young men-who enjoy heartily rapping out drumbeats, clanging cym-
bals, and singing in praise of buddhas, bodhisattvas, and other deities.
Svayambha Jfilinamlila Bhajan
Before leaving Svayambhl1, we must visit another musical expression
of Buddhist devotion prominent there: the Jfianamala Bhajan. Until
the later Rana period (1846-1950), the bhajan style of devotional
music was practiced only by Hindu devotees in the Kathmandu Valley.
Newar Buddhists have in this century also incorporated this type of
musical playing into their own tradition, taking up instruments (tabla,
sitar, harmonium, violin) and composition style imported from India.
Once the Buddhists learned the instruments, they soon composed
songs to their divinities and formed groups that now organize regular
devotional singing. .
Bhajans play each night in many city neighborhood rest houses, after
many shops have closed and they are open to all who wish to join in.
Composed of many Newar Buddhist castes from all over Kathmandu,
the Jfianamala Bhajan at Svayambhl1 must be noted for many men also
participate there and contribute financially. The bhajan plays in the
rest house adjoining the main stairs at the hilltop several hours each
morning on all important days of the lunar year and daily during
A typical playing session begins with offerings to the gods of the
bhajan site, a hymn to Nasa dyaJ:1, G a ~ e s a , and then moves to
"Govinda," in which all of the important gods' names are recited and
during which a conch is blown. The balance of the morning's songs
are devotional hymns to the great Buddhist divinities (Avalokitesvara,
Sakyamuni Buddha, Basundhara, etc.). The conch is blown again to
signal the last sequence of hymns as an oil wick lamp is lit and shown
to all of the bhajan house deities. (This offering light is called arati
after it is offered to the gods.) The men hold their fingers close to the
332 nABS 16.2
flame, then touch them to their eyes and foreheads. A final salute to
Nasa dyal). ends the playing.
When the iarge bhajan convenes, enthusiastically-rendered songs of
praise and supplication abound. As the bhajan has grown in popular-
ity, it has emphasized the bhakti dimension of Mahayana lay
Buddhism. This trend also reflects the extent of Indian influences on
modem Newar life, a fact that is illustrated by the popular use of Hindi
film melodies for new bhajan songs. However, Newar uplisakas feel
pride in their adaptation of this musical genre but with songs of their
own composition with Buddhist content. (New compositions are still
being composed.) The words and ethos expressed in the Newar bhajan
capture the spirit of modem Newar devotionalism. As one Tuladhar
layman said to me: "If you want to seek the rasa (taste) of our
Dharma, you must listen to the bhajan."
Household Devotions
Individual families may call a vajriiciirya priest into their homes for
the daily reading of privately-owned Buddhist texts. The families pay
him a daily stipend and then make a special offering at the end of the
month. The traditional ideal is that every text in the household collec-.
tion should be read during GUlpH. Where texts are still read, for most
Uray families it is a mechanical ritual nQt attended to for content since
few Vajracarya readers (or laity) can understand the Sanskrit or archaic
Several generations ago, especially devout individuals would retreat
to a vihara for the entire period to meditate, study, and fast (Locke
1980,235). Today, some individuals try to set aside during the month
a period for modest textual study or more extended periods of medita-
tion. Another type of ritual observance is abstention from different
foods which is sealed by a- vow at the beginning of the month. Uray
laymen may choose to avoid meat, onions, garlic, and/or alcohol for
the entire month. Several informants asserted that some individuals
used to "fast" for the entire month, but I saw no evidence of this prac-
tice in Kathmandu.
Many families light the fixed votive lamps around the caityas in their
local town vihara every night during GUlpH. Although the wealthiest
neighbors were the most active, every family usually contributes oil to
the wick lamps there. On the main days of Gurpla-full moon. and
21. Gellner 1987, 359.
eighth days (a:S'.tami)-the lamps are also lit at sunrise as well. Family
offerings to the main vihara shrine (Newari: dyaJ:t) are more
elaborate and individuals make a greater point than at other times in
. the year to. do pradalqi1)li of the caitya complex in the courtyard. In a:
Kathmandublihli courtyard where I lived in 1982 and 1987, the early
evenings throughout GUIpHi were alive with devotions and socializing
by most who lived nearby.
Gurpla is also the special season for families to participate in vratas
dedicated to the bodhisattvas. As Locke (1987) and I (1989a) have
published separate accounts of several of these one- or two-day long
periods of fasting, ritual, and textual recitation, no further remarks on
these traditions will be made except to note that observing the
vratas is one of the most demanding .forms of Newar devotional obser-
After a purificatory ritual for the family at the start of the month
(which may involve the abstention from meat and alcohol), individuals
sit together each morning to make as many caitya images as they can
using special black clay and small molds. Women are usually the most
active in the usually indoor and private work of dya[! thliyegu;22 their
role here complements the mens' participation in the public blijan.
(Menstruating women, however, must abstain from this task.)
There are a variety of molds, but most are ofcaityas of varying sizes.
To make an individual image to "high Buddhist standards"
-a tradition known only to Dray, Sakyas, and Vajracaryas-entailsa
twelve-step process marked at each turn by a mantra recitation.
image also gets a grain of rice that gives jiva (life) to it.
22. The Newar use of this tenn (Skt. deva) here for caityas and images
(and in other traditions, as below) matches the Khotanese Mahayana tradition
as recorded in the Book of Zambasta which also uses the epithet "deva" to
refer to buddhas and bodhisattvas (Emmerick 1968).
23. The steps are given here with the accompanying mantras: 1. o'!l basudhe
svahil, taking the clay; 2. 01Jl vajra bhilvay sVlihil, shaping the clay; 3. o'!l arje
viraje svahil, putting oil in the mold; 4. o'!l vajra dhtltu garbhe svahil, putting
the clay in the mold; 5. 01Jl vajra klrti chedaya hU1Jl phat svahil, removing the
excess clay; 6. o'!l dharma dhiltu garbhe svahil, putting in a paddy grain; 7.
o'!l vajra mungaratko hu'!l phat svahil, covering up the paddy grain; 8. o'!l
vajra dharma rate svahil, putting on extra clay to remove the image; 9. 0'!l.
supratisthata vajre svahil, putting the image with others already made; 10. O'!l
mani sata dipte svahil, after placing it. Tucci (1988 ed., 57-60) has given two
other sets of mantras for this process based upon Sanskrit and Tibetan
334 nABS 16.2
In some families, the thliyegu is actually the work of a formal
rituaJcalled the vrata (Newari: luchidyatt vrata), and as
the name implies, 100,000 images must be molded-to fulfill the vow
made at the outset. At the end of the month, the sum: of clay images is
molded into a large three-dimensional maI,1Qala and worshipped by the
fasting family with a kalasa pujli performed by a vajrliclirya. The
entire mass having been consecrated with offerings, it may be used to
build a family votive caitya or (more commonly) be taken out and
deposited at a tirtha along the River, along with offerings to
the nligas .24 .
Special Community Observances During GWJ1.lli
In several Kathmandu viharas, there is a tradition of vajrlicliryas
exhibiting gold-lettered Prajfitipliramitli texts. (The text itself is read
each mOrning and evening.) Characteristic of early Mahayana tradi-
tion (Schopen 1975, 168), this "cult of the book" is seen each Gmpla
morning: sa:rp.gha members cover the large text in brocade, and to
receive blessings lay folk make an offering for the privilege of a short
darsan and having one leaf of the manuscript touched to the forehead.
Viewing is supposed to. confer health and protection.
For five days after the middle of, the owners of notable
Buddhist images (in most cases, Dlpa:rp.kara Buddha) display them in
public, usually in the ground floor area of their houses. The sa:rp.ghas
of some viharas also display images, texts, and art that they own and
some also hang out long scroll paintings that illustrate the important
avadlin(1,S emphasized by Newar tradition.
This period of display is
the time when Mahayana Buddhists set out in public their non-tantric
religious treasures in what was once the greatest yearly Buddhist dis-
play. Such "cultural performances" (Singer 1972, 67ff) invariably
draw large crowds who come for daily puja and darsan throughout the
24. The dyaly, thliyegu rituals I observed in Newar homes closely conformed
to the guidelines in the Vrata text (New. luchi dyaly,) translated by
Tissa (1974). Not all Kathmandu families who make these images do so as
part of the vrata.
25. The following four scrolls were displayed at viharas visited by the GUIp.Ia
blijans: 1. SiI!ilialasartabahu in Bhagawan Baba, Thamel; 2. Story of Kesa
Chandra in Itum Baba (stolen in 1980); 3. }fabasattva Raja Kumar, Naradevi
courtyard; 4. Painting of Buddhist hells in Sri gha: Baba.
Although over twenty-five Kathmandu households own Samyaka
images, in 1987 only three still chose to display them in public for
Bah! Bvayegu. For the others, their images are now simply
arranged upstairs where only family members and invited guests can
view them and make offerings. Owners who elect not to display their
images cite fear of thievery as the main deterrent. Even the few who
still put out their Baht recount attempted robberies and the
actual theft of many ornaments and smaller objects.
Even with the vast reduction in scale, Baht Bvayegu provides
an opportunity for the Asan Buddhist community to experience the
grandeur of its highest art traditions, visit the town's viharas, and rec-
ognize the prestige of its greatest lay patrons. This yearly spectacle
shows the classical role of merchants as leaders among upasakas who
underwrite local Buddhist tradition and draw together the community
through their pious display of wealth.
The display attracts a day-long procession by Gurpla bajans that visit
many of the displayed deities in uptown, midtown, and downtown
Kathmandu. Many caste groups form processions to make the rounds
for darsan across the important landmarks in the entire city's Buddhist
The visitation starts during the morning return from Svayambhil
when the bajan visits the viharas in the far uptown: Bhagawan Baha
26. In this domain, contact with international art networks has had a clear
detrimental effect on Newar Buddhist culture. The "security provisions" that
local groups have put in place to safeguard temple art from thieves have often
distorted the original architectural-artistic order of the shrines and limited the
laity's contact with their sacred icons: images and paintings must be locked
away from everyone. This commoditization has corrupted the communities
from within, too: the temptation for an impoverished palZi to sell off guthi art
has been another cause for loss and breakdown. The display of empowered
Buddhist treasures has been a fundamental mechanism of Buddhist cultural
transmission since antiquity, and thievery has seriously undermined this in the
Kathmandu Valley (Sassoon 1989).
Some scholars, particularly art historians, have been unwilling to acknowl-
edge the cultural consequences of these "art transactions." The language of
art connoisseurship suppresses the reality of their data's path from Newar
neighborhood, to thief, to smuggler, to American "collection." For example,
" ... [The painting] was last displayed in August 1967 on the occasion of
bvayegu ... Then, like so many other Nepalese paintings it passed
into a private collection" (Slusser 1987, 20). This scholarship of "pieces"
does not acknowledge the problematic tradeoff: undermining living traditions
in order to, purportedly, understand isolated objects. An excellent recent pub-
lication documenting this legacy is found in L. S. Bangdel1991.
336 JIABS 16.2
(Bajracarya 1979), Chusya Baha, Musya Balla, SrI gha: Balla, Nhu
Balla, Jana Baha. Although the morning return that starts the day's
Bahl Dya!l Svayegu usually attracts more participants than a regular
day during (1981, 45), it is the afternoon-evening procession
that draws the larger crowd (1981, 250), the maximum number of
Tuladhar drummers (1981, 10), and the most hired Damai musicians
(1981: 3 trumpets, 2 clarinets). The only women who come along are
very small girls brought by their fathers, mother's brothers, or grand-
fathers. During this five-hour procession, the group in 1981 visited
about 25 different viharas, stopping approximately midway for a pala-
coordinated snack. In summary, the Baht DyaJ:! Svayegu procession
attracts the best musicians and is the time when Buddhist laymen join
en masse to have darsan of and worship the greatest treasures of their
city's cumulative Buddhist tradition.
On the thirteenth day of the waning moon near the end of is
Pafica Dana,27 when all but the poorest Buddhist laymen open their
storerooms and engage their kitchens for the purpose of making offer-
ings to samgha members who come to receive alms in their house-
holds. Vajracaryas and Sakyas from the city of Patan also come to
Kathmandu as their town's Pafica Dana day is held several weeks
earlier (Gellner 1992).
On this day, the bazaar streets become crowded a bit more than usual
as Vajracaryas bearing bags for offerings make their way from house
to house. Individuals do so either on an individual basis or for their
entire vihara. (If the latter, they must carry a large brass bowl and
wear a special cap.) Most lay Buddhists give five measures of paddy
for group collections and two to individuals.
There are a number of small guthis in Uray neighborhoods whose
endowment is for the purpose of making paficadana offerings to the
sarpgha. One guthiyar (member) stays in a store-front or rest house
27. Hem Raj Sakya (1979, 78ft) has noted an alternative derivation for this
name: nadam (that I never heard among Uray in Kathmandu) that he relates to
the navadana of the Bodhisattvabht1mi. These are: l. svabhava-danam; 2.
sarvadanam; 3. dana; 4. sarvato-mukha1J1. danam; 5.
danam; 6. sarvakt'ira1J1. danam; 7. vigMtarthika1J1. danam; 8. iMmutra-sukha1J1.
danam; 9. Visuddha1J1. danam (Dayal 1970, 173ft). Another possible deriva-
tive usage preceding paiicadana mentioned by Gellner (1987, 294) is pur;ya-
ja ("merit boiled rice") that refers to the gift of khlr. Manandhar (1986, 141)
lists pa1J1.jaram as an alternative rendering.
that is'often decorated with guthi-owned Buddhist paintings or images
to give paddy to any who appears.
In many courtyards where Buddhists live, householders decorate
their house walls with paintings hung from first-floor windows. Some
families distribute rice from large bowls placed outside the front door
while others invite their closer sarpgha acquaintances or family purohit
to come upstairs where their baht dyaJ:rs and other images are displayed
and the offerings are more elaborate.
When the sarpgha member enters the room, he immediately sits on
the straw mats arranged for this purpose. A woman of the house offers
purificatory water to his right hand and applies a red tikli to his fore-
head, prasad from the morning bodhisattva pOja offerings.
Depending on the family's preference for that year, the household's
designated ritual leader will present the vajracarya with various food-
stuffs and gifts: although the name paiica dana ("five gifts") implies a
set number of offerings, many more possibilities exist (R. K.
Bajracarya 19'80, 100-1). Paddy grains presented in an offering bowl
(pinda patra) are one essential donation. Other grains such as de-
husked rice, wheat, soybeans, chickpeas may be donated; gifts such as
fruit, sweets, and money are also made.
Khtr (rice pudding; Skt. is another indispensable offering and.
it is served on a leaf plate. For the Uray, this acceptance of cooked
rice signifies social equality with the vajracaryas, an issue of con-
tention in recent years in the caste context of the modern Hindu polity
of Nepal (Rosser 1964; Lewis 1989c). The reason Newar household-
ers give for presenting this is that they are imitating the textual figure
Sujata, the woman who gave khtr to the Buddha on the eve of his
Buddhist lay folk also offer distilled spirits, to
the Vajracaryas they know.
Newar Buddhists also have "Special Paiica Dana" traditions as well:
a single family makes offerings to the entire Newar saIpgha, including
a parcel of land and a vihara, with a portable image from Svayambho
brought down as witness. This can be on the appropriate day
during GUIpla or at another auspicious time during the year. For these
28. For those versed in Vajrayana symbolism, the khir symbolizes bodhicitta
and the esoteric conviction that without the woman's contribution of prajfia,
enlightenment is impossible.
29. The Newar Mahayana-Vajrayana norm does not proscribe liquor con-
sump !:ion as it is essential in tantric ritual. Many families feel ambiguity here,
however, distinguishing ritual use from profane drunkeness.
338 JIABS 16.2
day-long events, other householders can join in making their own diina
to the smp.gha members who pass, sharing thereby in the great PU1}ya
The Newar Pafica Dana offering by a single family Ukely
has ancient precedents: a similar ritual is described by 1-TSing in the
Snvijaya region around 690 CE (Takakusu 1896, 45-6).
Other aspects of Pafica Dana must be considered in the context of
cross-cultural Buddhist studies. On this day, householder vajracaryas
and sakyas take up the occupation of begging, the classical occupation
of the celibate monk. 31 Here, as in the modern Newar sarp.gha initia-
tion, a connection is made with the classical ludic norm and the
Mahayana claim that the bodhisattva's life need not be bounded by
celibate monasticism. Thus, each year the vajracliryas reiterate the
implicit claim that they are Buddhist masters as worthy as to
receive dana that produces great pU1}ya for the giver.
Another connection made in the Newar Pafica Dana tradition is with
DIparp.kara Buddha. This former Buddha, before whom Sakyamuni
began his bodhisattva career and who was popular across Asia as
"Calmer of Waters," protector of merchants (Coedes 1971,21), has in
Nepal been adopted according to later Mahayana Buddhist cosmology:
as the embodiment of the Adibuddha's power and the dharmakaya,
sharing this manifestation presence with SvayambhO Mahacaitya
(Shakya 1979, 75). There is a yearly spring festival to Dlparpkara in
Kathmandu (where he has the colloquial name "Cakan Dyal,l"
[Lienhard 1985] ) and his images are requisite for the great Samyaka
festivals. Newar tradition has also domesticated Dlparpkara as their
special figure who receives and witnesses great dana ceremonies. At
the time of presentations to the sarpgha, individuals receiving dana
30. See Lewis 1984,252 for the curious permutations that govern the choice
of the special sarp.gha recipients: the greatest dana gifts are presented to the
first in line, not according to scholastic accomplishment or spiritual mastery.
This apparently has been true for over 150 years (Wright 1877, 36).
31. The alms round is now rare even in Theravada countries (Spiro 1970;
Tarnbiah 1970). The only other occasion for this type of association in Newar
tradition is during each young vajracarya's initiation ceremony, the acarya
Before their installment into the vajracarya caste, each must take up
the monastic life for four living on alms (Locke 1975). After this
period, the boys renounce the "Sravakayana" and embrace the "Bodhisattva-
yana" as householders pursuing the career of Vajrayana ritual hierophant.
Each does so saying that the monk's life is "too difficult" (Gellner 1988).
The Tibeto-Burman Tarnangs living northwest of the Kathmandu Valley
also have a yearly one-day alms round by their householder monks (Holmberg
1984, 700). This custom could be in iinitation of Newar GmpJa practice.
first chant, "OT}'L nama':t sri dipaT}'Lkaraya" (Shakya 1979, 75). As
Gellner has shown, the Newar understanding of DipaIp.kara is
expressed in the popular story collection, the Kapisavadana.
In these
_ works, King Sarvananda gives lavishly to the saIp.gha headed by
DipaIp.kara -and is propelled toward buddhahood. Modern Newar
upasakas evoke this royal figure to frame their offerings' same pur-
pose (Gellner 1987, 298).
Modern-Vray and Vajracaryas understand the connection between
Pai'ica Dana and pU1:zya explicitly. As one Vajracarya said, "DipaIp.kara
Buddha _provided a great service to humanity by establishing Pai'ica
The devout can make great pU1:zya that protects households,
rescues beings from bad fates, and may secure a high rebirth state,
even Arnitabha's paradise, Sukhavati, for those who give generously."
As one layman stated during a recent Pafica Dana: "Just as one paddy
grain given here sprouts and produces a great harvest, so will the good
effects of this dana produce good fortune for the householder." This
analogy has been recorded throughout the modern Buddhist world
(Moerman 1966, 159; Gombrich 1971), as it has been since antiquity
(e. g. Takakusu 1896,45-6).
In conformity with another popular text domesticated into Newar
practice, individuals sponsor veneration of Svayambhii by a special
blijan during In keeping with the text's narrative describing
the reuniting of a married couple in their next existence through per-
forming a special ritual before a caitya, mourning families sponsor
similar offerings. The regular performance of caitya ven-
eration is now made by a special vihara guthi at Svayambhii: several
young boys circumambulate the Svayambhii hilltop complex playing
buffalo horns each morning during Their service is usually
contracted for by families at the start of the month in a short ceremony
dedicating the merit to the deceased. The full moon day of GuIp.1a is
usually chosen for the family to accompany the musical procession.
(See Lewis 1993c for a translation of this text and a discussion of its
Nepalese domestication.)
32. The Mahavastu recounts this same story, but the one who makes the vow
to Buddhahood is the Brahman Megha (Basak 1963, I, lxviii)
33. Locke reports that Pmlca Dana in Patan, held on sukla two weeks
earlier, is thought to commemorate a visit of Dlpffi!1kara Buddha to the
Kathmandu Valley (Locke 1980, 234). -
340 nABS 16.2
Hindu-Buddhist Relations During GU1J1la Dharma
The peak season for Newar Mahayana devotionalism coincides with
two distinctively Hindu festivals, and the Buddhist community's inter-
action with these events must be factored into the full understanding of
GUI!llil Dharma's Nepalese Buddhist domestication.
Only Hindus observe the distinctly Newar festival called Saparu or
Gai Jatra (Cow Festival) which is held in all major Newar towns on
the first day after the full moon in SravaJ)a. According to the local his-
tory, the Newar king Pratap MalIa (ruled 1641-1674) established this
one-day observance to assuage his queen's grief -after the death of her
son. To demonstrate the ubiquity of death and grief in the WOrld, all
households in which a death occurred were summoned to pass in a
procession by the palace. This royal history is also the reason, say .
informants, for the appearance of male "jokers" in the processions: to
lighten the burden of mourning they are free to satirize anything, and
dress as women, clowns, or performers.
The festival's religious roots lie in the belief that dead individuals
must cross many rivers to reach the realm of death ruled by Yama.
According to pan-Indic traditjons, cows are of invaluable assistance in
this journey. Both Hindu and Buddhist Newars act on their belief that
making a gift of a cow (godana) to a brahmaI)a insures this service to
the departed.
With this purpose in mind, Hindu families in mourning
dress up one or more sons in cow costumes and complete a procession
throughout the town.
Some may also lead a real cow who has been
groomed and garlanded. The women of the house extend the
meritorious service to cows by taking a position along this route near
their mmes to make food offerings to all the other "cow groups"
which pass. The spectacle draws crowds all along the route and espe-
cially at the royal palace.
34. Buddhist Newars perform a life-cycle rite for elders ("Bura Hikwo") that
enhances their karma and establishes them in an exalted status at seventy-
seven years old. This includes making a cow gift to a BrahmID:la as part of an
otherwise purely Vajrayana ritual ceremony (Lewis 1984,299-307).
35. The Buddhist community specifies another festival for its mourning
families to make a similiar town circumambulation: lndra Iatra, \yhich falls
soon after GUIpla. Instead of making offerings to cows, however, the
Buddhist fa9J.ilies make offerings to stiipas and caityas, making their way
around town to greet friends while making pUlJ.ya dedicated to the recently
Newar Buddhists do not join in celebrating this festival. 36 Saparu in
fact is very often the day chosen for BahI Dya[.1 Svayegu by the Asan
Uray. (In 1981, the Asan bajan passed the royal palace area seemingly
. oblivious to the Gill Jatra crowds there.) According to local tradition,
not only should Buddhists abstain from participating: they should also
not even witness the processions for it is "Mara vala" ("Mara has
come"). By identifying Hindu practice with the Buddha's defeated
foe, this polemic reflects a classical ideological assertion by which
Buddhist tradition subordinates Hindu deities and observances. Most
adult Buddhist lay folk know of this contention.
Hindu-Buddhist competition is also evident during the yearly festival
dedicated to s Eighth Day). Although in
modern times Newar Hindus of Kathmandu have fallen away from a
stronger devotionalism that marked Nepal's pre-modern era, this
two-day spring festival is still the occasion of the yearly palanquin
festival to "the Dark Lord." The local custom for Hindu groups is to
display devotional pictures of and other Hindu deities outside of
homes and at prominent public places. At the Annapur1)a temple rest
house in northeast Kathmandu (Bhotahity), for example, local Hindu
shopkeepers hang over 100 pictures, almost all of which are from the
orthodox Hindu pantheon: Siva, Ga1).esa, and, most com-
monly, scenes from the life of
But in another courtyard, the same Hindu pantheon is present, but
about half of the paintings are Buddhist: as part of their GurpJa devo-
tions, prominent Buddhist families set up competing displays alongside
their Hindu neighbors. In yet other courtyards, the content is almost
completely Buddhist in subject matter: in 1981 we found a series of
paintings in one courtyard depicting: Mahasattva Raja Kumara Jataka,
Tara, MafijusrI, the S.n.lgabherl AvadZina, AvalokiteSvara, and tantric
deities. Yet another display was totally Buddhist, presenting the
SilJ'thalasartabilhu Avadana and the life of the Buddha in framed
lithographs. There were also other framed deities from the Mahayana
36. The Sat!lg ha piijl1ris in J ana BaM do perform a special pflja petitioning
Avalokitesvara to intercede for any members reborn in the hells.
342 JIABS 16.2
We have surveyed a host of Mahayana observances held each summer
by Newar Buddhists in Nepal: music-enlivened devotional processions
and pujas given to images of buddhas, bodhisattvas, stiipas, local
deities; daily pilgrimage to Svayambii; presentation of dana offerings
. to many viharas; intensive sutra or dharal)I recitations; special occa-
sions to present dana to the sarpgha, including events involving
extraordinary munificence; ceremonies (puja) generating pUlJya for the
recently dead; public and private recitations of the dharma from
Buddhist literature; Mahayana vratas orchestrating many (of the
above) activities; special avadana-related rituals; cults to Mahayana
sutras; displays of treasured, empowered Dlparpkara images shared
with the community. In Newar Gurpla Dharma, one sees a broad
sample of Indic Mahayana traditions in practice.
Conforming to the desiderata of the five cardinal precepts and the
classical Mahayana seven-fold worship,37 the .ritual traditions of
Gurpla Dharma have clearly been crafted and accumulated to serve the
devout's seeking both practical blessings and final salvation (cf.
Dargyay 1986, 179-80). Indeed, the Newar evidence suggests the im-
portance of ritual traditions expressing and shaping Buddhist history,
while orchestrating exchanges insuring the local sarpgha's mainte-
nance. Buddhism has always been engaged in this domestication pro-
cess. As John Strong has recently noted, "Buddhism, as it is popularly
practiced, consists primarily of deeds done and stories told, that is of
ritualsthat regulate life both inside and olltside the monastery, and of
legends, myths, and tales that are recalled by, for, and about the faith-
ful" (1992, xi). With the addition of dharaI)I (or paritta) recitation as
the "practical religion" and contextualizing these attributes within a
nexus of community exchange, this description aptly describes the
nature of surviving Mahayana tradition in the Newar context.
The traditions of Gurpla Dharma allow further characterizations of
Mahayana Buddhism in practice: it has been primarily through ritual
that individuals express Buddhist identity and seek their spiritual aspi-
37. Often cited in the Mahayana literature is the seven-fold puja: 1. honor the
Buddha; 2. serve the Buddha; 3. confession of misdeeds; 4. delight in good
actions of beings; 5. invitation of budd has to preach the dharma; 6. arouse the
thought of enlightenment; 7. dedication of merit to all beings (Lamotte 1988,
rations (Beyer 1973, xii). Through paja, vrata, chanting, bajan and
bhajan, devotees conform to ideals set forth in canonical texts; though
varying in style of cultural performance, all these rituals seek the COffi-
. passion of budd has and bodhisattvas, planting karmic seeds fostering
bodhi, and these blessings alter destiny both in this world and beyond.
The meaning that Newar Buddhists place on their rituals finds
clear expression in a popular story collection, the
Bakh(Jl}1-. This text recounts the salvation work of A valokitesvara pri-
marily through refuge in bodhisattva rituals:
If those who are born in Nepal observe the gUf!!la dharma, if they show
devotion to Svayambhii, if they play five traditional instruments at the
jtltras, if they revere the paficatathtigatas, they will get the four fruits:
dharma, artha, kama, and be freed from all pap. When they die
they will be free from the fear of Yamaraja and go for rebirth in
Sukbavati bhuvana. (Vajracarya 1972, 6)
In practice, the Mahayana-like the HInayana-has held up the
Dharma as the path that "shows the lay folk the way toheaven." It
was not sublime philosophical exegesis nor meditative rapture but
ritual acts directed to heavenly rebirth that inspired the practice of
most Buddhists throughout history.
Hindu-Buddhist Competition and Boundaries
Newar Dharma tradition in Nepal gives insight regarding the
later history of Indian Buddhism: once the tradition came to uphold
and depend upon elaborate ritual events to unite the community, then
proper form, procedure, and pollution-purity regulation also became
Buddhist concerns. (This is quite evident, for example, in I-Tsing's
account.) Buddhist ritualists had to conform to the logic of brahmani-
cal aesthetics and paja procedures, sanctioning one major avenue of
assimilation. The developmental history of Newar Buddhism likewise
illustrates this trajectory of Mahayana evolution: as an immense
agenda of buddha and bodhisattva image veneration and temple life
developed, Buddhist paja, yatra, and sarp.skara evolved to adapt to the
Brahmanical ritual context. To adopt the highest Indic standards of
cleanliness and image purity, Buddhists thereby adopted Brahmanical
ritual views, including caste reckonings of individual purity. In so
doing, the tradition also imported an ongoing and increasing problem:
the laity's confusion between Buddhist and Hindu cults given identical
344 JIABS 16.2
ritual veneration.
To contend with this, the contestation traditions of
Newar Buddhism $"e'likewise instructive: to endure amidst a Hindu
majority required the elaboration of sharp contradistinctive traditions
such as those drawn up around the Cow and festivals. (Many
others endure in the Newar context [Lewis 1984, 468-481; Gellner
1992, 73-104].)
Lay Buddhism, Domestication, and Redistribution
Although there are few historical' sources available for reconstructing
the exact nature of the Newar saIJIgha's evolution to its present house-
holder and caste-defined Mahayana-Vajrayana form, (with periods of
hypothesized to explain the departure from the classical
norms), what is clear is the reworking of earlier celibate monastic
customs, as .David Gellner has so brilliantly demonstrated, (1992).
GUIJIla month seems to have been a part of this development in several
respects. The, rainy season period has become a time when the com-
munity forgoes marriage ceremonies and focuses on Buddhist practice
(textual study, meditation, construction, piija) , in conformity with
aspects of the classical ideal.
Certain GUIJIla Dharma practices highlight other transformations
evident in the domestication of Newar Mahayana Buddhism. The
emphasis on textual ownership and recitation by the modern saIJIgha
underlines their place as heirs and holders of the spiritual powers culti-
vated by the Mahayana. Conducting bodhisattva rituals and transmit-
ting meditation traditions (tantric and non-tantric) became the right of
certain Newars lineages exclusively. Esoteric Mahayana-Vaj ray ana
initiations in Nepal are now, as in traditional Tibet, open only to those
of high birth status who can give'the necessary dana, making the tra-
dition-in part-a commodity. Thus, a high-caste saIJIgha now holds
the monopoly right to mediate the Mahayana pantheon to the laity
through the rituals of veneration, protection, and initiation.
38. This is true in Nepal today in the devolution of traditional Mahayana
culture (Lewis 1984,555-589), a situation also discussed by Mus (1964).
39. On the possible role of disease in Newar Buddhist history, we now know
of the effects of a severe pestilence in the town of Kathmandu about 1724
recounted in the writings of Situ Panchen. See Lewis and Jamspal 1988, 199.
The prevalence of disease in pre-modern societies should cause scholars of
Buddhism to underline the importance 'of apotropaic ritualism in securing the
tradition's success (Lewis 1994a).
Guip.1a Parka Dana traditions have evolved to ensure the essential
contributions to this sarpgha. We have painted out how this day of
dana is congruent with the modem Newar monastic initiation by allud-
. ing to the classical monasticism but ultimately affirming a Mahayana
superiority.' Newar tradition has domesticated Diparpkara Buddha
imagery and stories to have him as witness and to legitimate' the great-
est yearly redistribution ritual. In Kathmandu until qllite recently,
Pafica Dana was quite a lavish windfall for individual sarpgha mem-
bers.4o Whatever else we might surmise about Buddhism's variegated
history, the Newar Gurpla Dharma traditions point to the central
propensity to ritualize spiritual ideals and to adapt buddha and bod-
hisattva traditions into the cultic and festival practices' of specific
localities. In Nepal, as in every other venue of Buddhism's successful
missionization, rituals evolved to accomplish two crucial and related
tasks: doinesticate the slisana and redistribute the laity's wealth to the
Allen, Michael. 1973. "Buddhism Without Monks: The Vajrayana Religion of
the Newars of the Kathmandu Valley." South Asia 3: 1-14.
______ . N. d. "Processions and Pilgrimage in Newar Religion."
Anderson, Mary. 1971. Festivals of Nepal, London: Allen and Unwin.
Bagchi, P. C. 1941. "The Eight Great Chaityas and their Cult." Indian
Historical Quarterly 17: 223-235.
Bailey, H. W. 1974. "The Pradaksina-Sutra ofChang-Kuin." Buddhist Studies
in Honour of 1. B. Homer. Ed. L. Cousins, et al. Dordrecht Reidel. 15-18.
Bajracarya, Ratna Kaji. 1980. Ye1Jl Deya Bauddha paja Kriyaya Halaf!!j-
valaf!!. Kathmandu: Sankata Printing Press.
Bajracharya. Puma Harsha. 1979. "Than Bahil, An Ancient Centre for San-
skrit StUdy." Indologica Taurinensia Vol. 7: 62-64.
Bajracarya, Tejratna. 1991. Aryya Namasaf!!giti. Kathmandu: Siddhivar Press.
Bangdel, Lain Singh. 1991. The Stolen Images of Nepal. Kathmandu: n. p.
Basak, Radhagovinda. 1963. Mahlivastu Avadana. 3 Vols. Calcutta: Sanskrit
40. This largesse was formerly so great that the Vajracaryas had to hire
farmers to cany their gift loads home.
346 JIABS 16.2
Beal, Samuel. 1970. Si-Yu-Ki: Buddhist Records of the Western World. 2
Vols. New York: Paragon Book Reprint Corp.
Beyer, Stephan. 1973. The Cult of Tara. Berkeley: University of California
Birnbaum, Raoul. 1989-90. "Secret Halls of the Mountain Lords: The Caves
of Wu-Tai Shan." Cahiers d'Extreme-Asie 5: 115-140.
Coedes, G. 1971. The Indianized States of Southeast Asia. Honolulu: Univer-
sity of Hawaii Press.
Cone, Margaret, and Richard F. Gombrich. 1977. The Perfect Generosity of
Prince Vessantara. Oxford: Clarendon Press.
Conze, Edward. 1967. Buddhist Thought in India. Ann Arbor: The University
of Michigan Press.
Cowell, E. B., et al. 1969. Buddhist Mahayana Texts. New York: Dover
Dallapiccola, Anna, ed. 1980. The StOpa: Its Religious, Historical, and
Architectural Significance. Wiesbaden: Franz Steiner Verlag. 112-126.
Dargyay, Eva K. 1986. "Merit-Making and Ritual Aspects in the Religious
Life of Zanskar (West Tibet)." Karma and Rebirth. Ed. Ronald W.
Neufeldt. Albany, New York: SUNY Press. 179-189.
Das, Sukumar Ranjan. 1928. "Hindu Calendar." Indian Historical Quarterly
4: 483-511.
Davidson, Ronald M. 1981. "Litany of Names of ManjusrJ:: Text and Trans-
lation of the Melanges Chinois et Bouddhiques 20:
Dayal, Har. 1932. The Bodhisattva Doctrine in Buddhist Sanskrit Literature.
London: Routledge and Kegan Paul.
DeBary, William Theodore. 1972. The Buddhist Tradition. New York:
Dehejia, Vidya. 1972. Early Buddhist Rock Temples: A Chronology. Ithaca:
Cornell University Press.
______ . 1989. "Stupas and Sculptures in Early Buddhism." Asian
Art 11.3: 7-31.
Dutt, Nalinaksha. 1945a. "Popular Buddhism." Indian Historical Quarterly
21.4: 245-270.
______ . 1945b. "The Place of Laity in Early Buddhism." Indian
Historical Quarterly 21.3: 163-183.
______ .1977. Mahayana Buddhism. Delhi: Motilal Banarsidass.
Dutt, Sukumar. 1962. Buddhist Monks and Monasteries of India. London:
Allen and Unwin.
Edgerton, Franklin. 1953. Buddhist Hybrid Sanskrit Doctionary. New Haven:
Yale University Press.
Eliade, Mircea. 1959. The Sacred and the Profane. New York: Harcourt,
Ellingson, Ter: 1990. "Nasa: dya: Newar God of Music." Selected Reports in
Ethnomusicology 8: 221-272.
Emmerick, R. E. 1968. The Book of Zambasta: A Khotanese Poem on Bud-
dhism. London: Oxford University Press.
Falk, Nancy. 1977. "To Gaze on the Sacred Traces." History of Religions 16:
Foucber, Alfred. 1949. La vie du Bouddha d'apres les textes et les monuments
de l'Intie. Paris: n. p.
Gellner, David N. 1992. Monk, Householder and Tantric Priest: Newar
Buddhism and Its Hierarchy of Ritual. Cambridge University Press.
______ . 1987. "Monk, Householder and Priest: Newar Buddhism
and Its Hierarchy of Ritual." Diss. Oxford University.
______ . 1987a. "The Newar Buddhist Monastery: an Anthroplog-
ical and Historical Typology." The Heritage of the Kathmandu Valley. Ed.
N. Gutschow and A. Michaels. Sankt Augustin: VGH Wissenschaftsverlag.
______ . 1988. "Monastic Initiation in Newar Buddhism." Oxford
University Papers on India. Ed. R. F. Gombrich. 2.1: 42-112.
______ . 1989. "Monkhood and Priesthood in Newar Buddhism."
Purushtirtha 12: 165-191.
Gokbale, Balkrishna Govind. 1980. "Bhakti in Early Buddhism." Journal of
Asian andAfrican Studies 15.1-2: 16-27.
______ . 1966. "Early Buddhist Kingship." Journal of Asian Studies
26.1: 15-22.
Gombrich, Richard F. 1971. Precept and Practice: Traditional Buddhism in
the Rural Highlands of Ceylon. N. p.: Oxford University Press.
Greenwold, Stephen M. 1974. "Buddhist Brahmins." Archives Europeennes
de Sociologie 15: 483-503.
______ . 1978. "The Role of the Priest in Newar Society." Himala-
yanAnthropology. Ed . .Tames Fisher. The Hague: Mouton. 483-503.
Harvey, P. 1984. "The Symbolism of the Early Stupa." Journal of the Inter-
national Asociation of Buddhist Studies 7: 67-93.
Harris, Marvin. 1989. Our Kind. New York: Harper and Row.
Hodgson, Brian H. 1972. Essays on the Languages, Literature, and Religion
of Nepal and Tibet. New Delhi: Manjusri. (Original edition, 1874).
348 JIABS 16.2
Holmberg, David H. 1984. "Ritual Paradoxes in Nepal: Comparative
Perspectives on Tamang Religion." Journal of Asian Studies 43: 697-722.
______ . 1989. Order in Paradox: Myth, R i t ~ a l , and Exchange
Among Nepal's Tmnang. Ithaca: Cornell University Press.
Kern, H. 1884. Saddhartna-Pundarika or Lotus of the True Law. Oxford:
Clarendon Press.
Keyes, Charles F. 1975. "Buddhist Pilgrimage Centers and the Twelve-Year
Cycle: Northern Thai Moral Orders in Space and Time." History of
Religions 15: 71-89.
Kloppenborg, Ria. 1977. "Theravacla Buddhism in Nepal." Kailash 5.4: 301-
Kosambi, D. D. 1965. Ancient India: A History of Its Culture and Civilization.
New York: Meridian.
Lamotte, Etienne. 1988. History of Indian Buddhism: From the Origins to the
Saka Era. Trans. Sara Webb-Boin. Louvain: Institut Orientaliste.
______ .1984. "The Buddha, His Teachings, and His Sangha." The
World of Buddhism. Ed. Heinz Bechert and Richard Gombrich. New York:
Facts on File. 41-58.
Lancaster, Lewis R. 1974. "An Early Mahayana Sennon about the Body of
the Buddha and the Making of Images." Artibus Asiae 36.4: 287-291.
Legge, James. 1965. A Record of Buddhist Kingdoms. New York: Dover.
Lessing, Ferdinand D. 1976. "Structure and Meaning of the Rite Called Bath
of the Buddha According to Tibetan and Chinese Sources." Collected
Essays on Lamaism and Chinese Symbolism. Taipei: Shihlin. 45-57.
Lester, Robert. 1987. Buddhism. San Francisco: Harper and Row.
Levi, Sylvain. 1905-8. Le Nepal. 3 Vols. Paris: Leroux.
Levy, Robert 1. 1990. Mesocosm: Hinduism and the Organization of a Tradi-
tional Newar City in Nepal. Berkeley: University of California Press.
Lewis, Todd T. 1984. The Tuladhars of Kathmandu: A Study of Buddhist Tra-
dition in a Newar Merchant Community. Ann Arbor: University Microfilms
______ . 1989a. "MahayanaVratas in Newar Buddhism." The
Journal of the International Association of Buddhist Studies 12.1: 109-138.
______ . 1989b. "Childhood and Newar Tradition: Chittadhar
Hriclaya's Jhi Maca." Asian Folklore Studies 48.2: 195-210.
______ . 1989c. "Newars and Tibetans in the Kathmandu Valley:
Ethnic Boundaries and Religious History." Journal of Asian and African
Studies 38: 31-57.
_-----. 1993a. "A Chronology of Newar-Tibetan Relations in the
Kathmandu Valley." Publicazioni del Centro Piemontess di Studi sui
Medio ed Estremo Oriente. Ed. Sigfried Lienhard. Turin. Forthcoming.
'--___ ---'_. 1993b. "The Nepal Jana Jivan Kriya Paddhati, a Modern
Newar Guide for Vajrayana Life-Cycle Rites." Indo-Iranian Journal 36:
______ . 1993c. "Contributions to the History of Buddhist Ritual-
ism: a Mahayana A vadana on Caitya Veneration from the Kathmandu
Valley." Journal of Asian History 27.2. Forthcoming.
______ . 1993d. "Newar-Tibetan Trade and the Domestication of the
Si1J1.halasarthabahu Avadana." History of Religions 33.2: 135-60.
______ . 1994a. "Mahayana Protection Rituals: The
Kathasara." Forthcoming.
______ . 1994b. "Buddhist Merchants in Kathmandu: The Asan Tol
Market and Uray Social Organization." Newar Society. Ed. Gellner and
Quigley. Forthcoming.
Lewis, Todd T., and L. Jamspal, 1988. "Newars and Tibetans in the
Kathmandu Valley: Three New Translations from Tibetan Sources."
Journal of Asian and African Studies 36: 187-211.
Lienhard, Siegfried. 1985. Die Abenteuer des Kaufmanns Simhala. Berlin:
Museums fur Indische Kunst.
______ . 1984. "Nepal: The Survival of Indian Buddhism in a
Himalayan Kingdom." The World of Buddhism. Ed. Heinz Bechertand
Richard Gombrich. New York: Facts on File. 108-114.
______ . 1978. "Religionssynkretismus in Nepal." Buddhism in
Ceylon and Studies on Religious Syncretism in Buddhist Countries. Ed.
Heinz Bechert. Gottingen: Vandenhoeck and Ruprecht. 146-177.
Liu, X. 1988. Ancient India and Ancient China. Bombay: Oxford University
Locke, John K. 1975. "Newar Buddhist Initiation Rites." Contributions to
Nepalese Studies 2: 1-23.
______ . 1980. Karunamaya. Kathmandu: Sahiyogi.
______ , 1985. Buddhist Monasteries of Nepal. Kathmandu:
______ ,. 1987. "Uposadha Vrata of Amoghapasa Lokesvara in
Nepal." l'Ethnographie 83.100-101: 159-189.
Macdonald, A. W., and Anne Vergati StahL 1979. Newar Art: Nepalese Art
During the Malta Period. Warminster: Aris and Phillips.
350 nABS 16.2
Manandhar, Thakur Lal. 1986. Newari-English Dictionary. Delhi: Agam Kala
Matsunanti, Kodo.1976. Introducing Buddhism. Tokyo: Charles Tuttle.
Mitra, Rajendralala. 1971. The Sanskrit Buddhist Literature of Nepal.
Calcutta: Sanskrit Pustak Bhandar reprint.
Moerman, Michael. 1966 .. "Ban. Ping's Temple: The Center of a Loosely
Structured Society." Theravada Buddhism: Anthropological Studies. Ed.
Manning Nash. New Haven: Yale Southeast Asia Studies. 138-174.
Monier-Williams, Sir Monier. 1956 A Sanskrit-English Dictionary. 2nd ed.
London: Oxford University Press ..
Mumford, Stan R. 1989. Himalayan Dialogue. Madison: University ofWis-
consin Press.
Murphy, Robert F. 1971. The Dialectics of Social Life: Alarms and
Excursions in Anthropological Theory. New York: Basic Books.
Mus, Paul. 1964. "Thousand-Armed Kannon: A Mystery or a Problem?"
Jndogaku Bukkyogaku Ken":yil 12.1: 470-438.
Novak, Charles M. 1986. Catalogue of Selected Buddhist Manuscripts in the
Asha Saphu Kuthi. Kathmandu: lore Ganesh Press.
Ortner, Sherry B. 1989. High Religion: A Cultural and Political History Of
Sherpa Buddhism. Princeton: Princeton University Press.
Owens, Bruce. 1989. The Politics of Divinity in the Kathmandu Valley: The
Festival of Bunga DyalRato Matsyendranath. Diss. Columbia University.
Pal, Pratapaditya. 1974. The Art of Licchavi Nepal. Bombay: Marg Pub-
Prebish, Charles S. 1975. Buddhist Monastic Discipline: The Sanskrit Pmti-
moksa Sutras of the Mahasamghikils and Mulasarvastivadins. University
Park: The Pennsylvania State University Press.
Pye, Michael. 1978. SkillfUl Means: A Concept in Mahayana Buddhism.
London: Duckworth.
Reynolds, Frank. 1972. "The Two Wheels of Dhamma: A Study of Early
Buddhism." The Two Wheels of Dhamma: Essays on the Theravada
Tradition in India and Ceylon. Ed. Bardwell Smith. Studies in Religion 13.
American Academy of Religion.
Rhys-Davids,Caroline F. 1901. "Notes on the Early Economic Conditions in
Northern India." Journal of the Royal Asiatic Society 53: 859-888.
Rhys-Davids, Thomas W. 1969. Buddhist Suttas. New York: Dover.
______ . 1963. The Questions of King Milinda. New York: Dover
reprint. '
Riccardi, Theodore. 1979. "Buddhism in Ancient and Early Medieval Nepal."
Studies in the History of Buddhism. Ed. A. K. Narain. New Delhi: Agam.
Rosser, Colin. 1 9 6 6 ~ "The Newar Caste System." Caste and Kin in Nepal,
India, and Ceylon. Ed. Christoph von Furer-HaimeIidorf. Bombay: Asia
Publishing House. 68-139.
Sakya, Ananda. 1971. GU1Jlla Bajanya Boli. Kathmandu: Jorganesh.
Sakya, Hemaraj. 1977. Sri SvayambhU Mahlicaitya. Kathmandu: Vikas
______ . 1979. Samyak Mahadana Guthi. Kathmandu: Samkata
Sassoon, David. 1989. "Considering the Perspectives of the Victim: The
Antiquities of Nepal." The Ethics of Collecting Cultural Property: Whose .
Culture? Whose Property? Ed. Phyllis Messenger. Albequerque: University
of New Mexico Press. 61-72.
Schopen, Gregory. 1975. "The Phrase 'sa prthiv1pradesas caityabhiito bhavet'
in the Vajracchedika: Notes on the Cult of the Book in Mahayana." Indo-
Iranian Journal 17: 147-181.
______ . 1987. "Burial 'ad sanctos' and the Physical Presence of the
Buddha in Early Indian Buddhism: A Study in the Archaeology of
Religions." Religion 17: 193-225.
______ . 1989. "The Stupa Cult and the Extant Pali Vinaya."
Journal of the Pali Text Society 13: 83-100.
______ . 1991a. "An Old Inscription from Amaravati and the Cult of
the Local Monastic Dead in Indian Buddhist Monasteries." The Journal of
the International Association of Buddhist Studies 14.2 : 281-329.
______ . 1991b. "The Ritual Obligations and Donor Roles of Monks
in the Pali Vinaya." Journal of the Pali Text Society 16: 1-21.
______ . 1991c. "Monks and the Relic Cult in the Mahaparinibbana-
sutta: An Old Misunderstanding in Regard to Monastic Buddhism." From
Benares to Beijing: Essays on Buddhism and Chinese Religion in Honor of
Jan Yun-hua. Ed. G. Schopen and K. Shinohara. Oakville: 187-201.
______ . 1992a. "On Avoiding Ghosts and Social Censure: Monastic
Funerals in the Mulasarvastivada-Vinaya." Journal of Indian Philosophy
20: 1-39.
______ . 1992b. "Stupa and Tirtha: Tibetan Mortuary Practices and
an Unrecognized Form of Burial Ad Sanctos at Buddhist Sites in India."
Forthcoming mss.
Shastri, Hara Prasad, ed. 1894. The Vrihat Svayambhu Puranam. Calcutta:
Asiatic Society of Bengal.
- ~ - - - - - - - - - - - .. -----------
352 JIABS 16.2
Sbrestha, Durgalal. 1983. Bhagavan DasabaZaya Tutah. Kathmandu: Smriti
Press. . .
Shukla, N. S. 1975. "The Qualities of an Acarya on the Basis of the .Acarya-
kriyasamuccaya of JagaddarpaI)a." Buddhist Studies in'India. Ed. R. C.
Pandeva. New Delhi: Manohar. 126-129.
Singer, Milton. 1972. When a Great Tradition Modernizes. New York:
Skilling, Peter. 1992. "The Literature of the Sravakayana." Journal of
the Pali Text Society 16: 109-182.
Slusser, Mary. 1982. Nepal Mandala. Princeton: Princeton University Press.
______ . 1987. "The Cultural Aspects of Newar Painting." Nepalica:
Heritage of the Kathmandu Valley. Ed. Bernhard Kolver and Siegfried
Lienhard. Sankt Augustin: VHG Wissenschaftsverlag. 15-27.
Snodgrass, A. 1985. The Symbolism of the Stupa. New York: Cornell Uni-
versity Press.
Spiro, Melfred. 1986. Buddhism and Society. 2nd ed. Berkeley: University of
. California Press.
Stablein, William. 1973. "A Medical-Cultural System among the Tibetan and
Newar Buddhists: Ceremonial Medicine." Kailash 1.3: 193-203.
____ 1978. "A Descriptive Analysis of the Content of Nepalese
Buddhist Pujas as a Medical-Cultural System, with References to Tibetan
Parallels." Himalayan Anthropology. Ed. James Fisher. The Hague:
Mouton. 403-411.
Strickmann, Michel. 1990. "The Consecration Sutra: A Buddhist Book of
Spells." Chinese Buddhist Apocrypha. Ed. Robert Buswell, Jr. Honolulu:
University of Hawaii Press. 75-118.
StrI;mg, John S. 1979. Transforming Gift: an Analysis of Devotional
Acts of Offering in Buddhist Avadana Literature." History of Religions
18.3: 221-237.
______ . 1983. The Legend of King Asoka. Princeton University
______ . 1992 The Legend and Cult of Upagupta: Sanskrit Bud-
dhism in North India and Southeast Asia. Princeton University Press.
Suzuki, D. T. 1965. The Training of a Zen Buddhist Monk. NY: University
Swearer, Donald K. 1987. "Buddhist Religious Year." The Encyclopedia of
Religion. Vol. 2. New York: Macmillian. 547-554.
Takaoka Hidenobu, ed. 1981. A Microfilm Catalogue of the Buddhist Manu-
scripts in Nepal. Nagoya: Buddhist Library.
Takakusu, J. 1896. A Record of the Buddhist Religion. Oxford: Clarendon
Tambiah, Stanley J. 1970. Buddhism and the Spirit Cults of Northeast
Thailand. Cambridge University Press.
____ -'-_. 1984. Buddhist Saints of the Forest and the Cult of
Amulets. Cambridge: Cambridge University Press.
Teiser, Stephen F. 1988. The Ghost Festival in Medieval China. Princeton:
Princeton University Press.
Tewari, Ramesh Chandra. 1983. "Socio-Cultural Aspects of Theravada
Buddhism in Nepal." Journal of the International Association of Buddhist
Studies 6.2: 67-93.
Thunnan, Robert A. F. 1976. The Holy Teaching of Vimalakirti: A Mahayana
Scripture. University Park: University of Pennsylvania Press.
Tingey, Carol. 1990. The Heartbeat of Nepal: The Paiicai Baja. Kathmandu:
Royal Nepal Academy.
Tissa, Rajapatirana. 1974. "SuvarI)avarI)avadana: Translated and edited to-
gether with its Tibetan translation and the Diss.
Australian National University.
Toffin, Gerard. 1984. Societe et Religion chez les Newar du Nepal. Paris:
Tucci, Giuseppe. 1988. Stupa: Art, Architectonics and Symbolism (Indo-
Tibetica I) (English translation in Sata-Pitaka Series 347, Lokesh Chandra
ed.) New Delhi: Aditya Prakashan.
Tuladhar, Labh Ratna. 1986. Jhigu Tutah. Kathmandu: Pasa Muna.
Vajracarya, Kamalananda. 1972. JanabahadyalJ Bllkha1[L. Kathmandu: Agam
Warren, Henry Clarke. 1922. Buddhism in Translations. Cambridge: Harvard
University Press.
Wayman, Alex. 1971. "Buddhism." Historia Religidnum 2: 372-464.
______ . 1959. "Studies in Yama and Mara." Indo-Iranian Journal
3.1: 44-75; 3.2: 112-131.
______ . 1984. "The Sarvarahasyatantra." Acta lndologica 6: 521-
______ . 1985. Chanting the Names of Maiijusrl, The Maiijusri-
nllml1samgiti Sanskrit and Tibetan Texts. London: Shambala.
Wegner, Gert-Matthias. 1988. The Nayki1[Lbaja of the Newar Butchers.
Stuttgart: Franz Steiner Verlag.
______ . 1986. The Dhimaybaja of Bhaktapur. Weisbaden: Kom-
missionsverlag Franz Steiner.
354 JIABS 16.2
______ . 1987. "Navadapha of Bhaktapur: Repertoire and Perfor-
mance of the Ten Drums." Heritage of the Kathmandu Valley. Ed. Neils
Gutschow and Axel Michaels. Sankt Augustin: VGH WissenschaftsVerlag.
471-488. '
Welch, Holmes. 1967. The Practice of Chinese Buddhism: 1900-1950.
Cambridge: Harvard University Press.
Wijayaratna, Mohan. 1990. Buddhist Monastic Life. Cambridge University
Wright, Daniel. 1990. History of Nepal. New Delhi: Asian Educational
Zysk, Keneth G. 1991. Asceticism and Healing in Ancient India: Medicine in
the Buddhist Monastery. New York: Oxford University Press, 1991.
A Re-examination of a Kaniska Period Tetradrachm
. Coin Type with an Image of Metrago/Maitreya on
the Reverse (GobI 793.1) and a Brief Notice on the
Importance of the Inscription Relative to Bactro- .
Gandharan Buddhist Iconography of the Period
In an earlier issue of this journal, Joseph Cribb presented an article,
s Buddha Coins-The Official Iconography of Sakyamuni
and Maitreya," in which he published for the first time a number of
coins representing Maitreya Buddha.
However, some of these coins
raised more questions than they answered. Cribb himself corrected the
reading for some of the coin inscriptions from Metrauo Boudo to
Metrago Boudo in his subsequent paper, "The Origin of the Buddha
Image-The Numismatic Evidence."2 In addition, some iconographic
details were still left unresolved and the relationship of the coins to
particular image types was not thoroughly explored.
I am most indebted to Joseph Cribb for allowing me access to the extensive
Kuru:a numismatic holdings in the British Musel.lm, to WilliamF. Spengler
for his suggestions, and to B. N. Mukherjee for sharing his many rich and
creative ideas and his vast enthusiasm. The Encyclopedia of Buddhist Icono-
graphy project, from which this article is a "spinoff," has been and is being
supported by numerous granting agencies and offices of The Ohio State
University. Those granting agencies whose support bears directly on this
article are: The National Endowment for the Humanities, the Kress
Foundation, and the Smithsonian Institution. Those Ohio State University
offices bearing the most direct relationship to this article are the College of
the Arts, the Department of History of Art, the Office of the Vice President
for Research and The Office of In ternational Studies.
1. Joseph Cribb, "Kanika's Buddha Coins-The Official Iconography of
Sakyamuni & Maitreya," JIABS 3.2 (1980): 79-88.
2. Joseph Cribb, "The Origin of the Buddha Image-The Numismatic Evi-
dence," South Asian Archaeology 1981: Proceedings of the Sixth Inter-
national Conference of the Association of South Asian Archaeologists in
Western Europe (Cambridge: Cambridge University Press, 1984) 231-244.
3. One important question is the problem of why boudo, i. e., buddha, is used
as the term for what is clearly a bodhisattva image. While there is no defini-
tive answer for the Bactro-Gandharan usage, it may be suggested that the
Chinese may have learned their usage of the termfo, buddha, for virtually all
of the deities in the Buddhist soteriological system from such a Gandharan
356 JIABS 16.2
One coin in particular, identifiable for sake of convenience as GobI
793.1,4 in the British Museum collection, has been the source of con-
siderable additional study (Figure 1).5 Cribb originally read the in-
sCription as Met[rJauo Bou[doJ and inter-
preted the image as one of Maitreya who was actually known as
Metrago in Greek lexicography (Figure Z). This
variation in orthography was well within the limits of known variants
and posed no particular problem to those of us who are concerned with
the inSCriptions on these coins. Indeed, the coins themselves are usu-
ally inpoor to terrible condition and Cribb had rendered the field of
studies a service by deciphering several very difficult coin in-
sCriptions. However, although his identification of MetragolMaitreya
would prove to be correct, there were several errors in his reading.
It is no criticism of Cribb that this problematic coin soon gave rise
to another reading. B. N. Mukherjee, apparently working only from
published 1: 1 photographs, saw the reading in an entirely different
light and read it as i. e., Amitabha (Figure 3).6 If this
reading was accurate, it would have been an important addition to the
slowly growing body of AmitabhalSukhavatl information for the
Bactro-Gandharan region) Mukherjee is well known for his many
contributions to studies and, in all candor, because I believe
that a clear understanding of the Amitabha cult in Gandhara is key to a
better understanding of initial transmission and early development of
what was to become East Asian "Pure-Land" Buddhism, I frankly
hoped that he was correct. An inscribed tetradrachm of
practice. More study will be necessary before the solution to this intriguing
linguistic question is known.
4. The number is that used by Robert Gobi in his System und Chronologie
der ManzprtLgung des KU'Slinreiche (Wien: Verlag der bsterreichischen
Akademie der Wissenschaften, 1984) pi 79.
5. For those readers who wish to follow the arguments regarding the reading
of the letters on the coin carefully, I have added an appendix of Greek
letters as they appear on coins. Please see Appendix I.
6. B. N. Mukherjee, "Amitabha on Coins," Journal of the Numis-
matic Society of India 49 (1987 [1991]): 44-45.
7. John C. Huntington "A Gandharan Image of Amitabha's SukhavatI,"
Annali dell 'Istituto Orientale di Napoli 40 (n.s. 30) (1980): 651-672; John
Brough, "Amitabha and AvalokiteSvara in an Inscribed Gandharan Sculp-
ture," Indologica Taurinensia: Official Organ of the International Association
of Sanskrit Studies 10 (1982): 65-70.
reign would have been "good" evidence of both the cult and its date in
Bactro-Gandhara. Unfortunately, it was not to be.
. In either case, the reading of the coin was critically important for the
study of Bactro-Gandharan Buddhist iconography, for it could either
identify an image of Amitabha or an important variant type of the
MetragolMaitreya image as found in the Buddhist sculpture of the
region. My current work on the Bactro-Gandharan volume of my
Encyclopedia of Buddhist Iconography has led me to examine as much
of the iconographically relevant inscribed evidence as possible.
Simply stated, my research indicates that there is ample technical evi-
dence that neither Cribb's nor Mukherjee's readings is correct. How-
ever, each reading is problematic for different reasons. My first clue
to the error of Mukherjee's reading was my observation as an icono-
grapher, not as an epigraphist-that the image in the center of the field
is a bodhisattva (Figure 4), not a buddha. Specifically, he wears arm-
bands or armlets on both arms, something that is unknown on images
of buddhas in the Bactro-Gandharan sculptural idiom. On the other
hand, Cribb's interpretation relied on what would have to have been
characters in impossible positions and an erroneous reading of several
of the characters.
In his first article, Cribb read the inscription as l'v1e[r]//auo Bou[do]
by reading counterclockwise from 11 :00 to 9:30, to 7:30 as MET
(HJ-'.7"); then jumping to the top right to 1:00,2:00,2:30, 3:00, and
4:00 as AY[U]OBOY[U] (tA '(o&oY), with the last two letters crammed
together in what, at first impression appears to be a single ligature
(Figure 2). While KU9al!a Greek Oike classical Greek inscriptions on
pottery, for example) is written every which way imaginable: right to
left, left to right, tops of the letter towards the inside, tops of the letters
towards the outside, and so on, a break in the line such as Cribb pro-
posed would be most unusual. Clearly, this reading was questionable.
but, given the frequent irregularities of KU9al!a Greek inscriptions. not
out of the question. Note in particular that Cribb assumes that the rho
for MET[R] and both the delta and the omicron from his reading of
BOU[DO] are missing, giving his interpretation a somewhat demand-
ing reliance on three missing letters across the bottom.
Mukherjee's reading is based on two epigraphical errors and I think
it will be both instructive and possibly cautionary to go through the
process of examining just what went wrong in his interpretation.
Several years ago, at the 1984 KU9al!a Studies conference in
358 JIABS 16.2
Mukherjee announced during the discussions following a
paper that he had identified a K a n i ~ k a tetradrachm with the reverse
inscribed with the name Amitabha.His article was not forthcoming,
however, until 1991.
To fully understand what actually transpired, it
must be understood t.."'1at Robert GobI's monumental corpus of Kusana
numismatics, System und Chronologie der Munzpriigung des
Kusanreiche, had been published in 1984, detailing hundreds of types
of Kusana coins, sometimes with several examples from similar or
even the same dies.lO In GobI's volume, the majority of the coins are
published actual size, in accordance with standard numismatic practice.
However, for technical study, one-to-one reproduction has serious
drawbacks.. When coins are in poor condition, if they have been
poorly photographed, or worse, when they are in both poor condition
and poorly photographed, this can lead to very serious problems.
Indeed, for detailed study of coins it is often necessary to use enlarge-
ments (I prefer about 10 diameters), drawings made from the actual
coin, and photographs made in acute angle raking lighting from several
positions around the coin in order to glean the most information from
the surface of the coin.
Apparently working only from small (1:1) published photographs of
the coin, Mukherjee read the tau at 7:30 as an alpha and what is actu-
ally a gamma at 4:30 as an upsilon (Figure 3). Given the very poor
condition of the coin and the actual appearance of both letters in the
photographs in both Cribb and GobI's publications, I freely admit that,
at first, I did not initially find Mukherjee's reading to be based on un-
reasonable interpretations of the letters in question. On the contrary,
even computer enhancements of the presumed alpha, generated from
the published images in GobI and Cribb's first article, suggested that
Mukherjee was correct. However, by using other resources, it was
clear that Cribb's Original interpretation of the letter at 7:30 as tau was
correct. As for the letter at 4:30, in the reproductions that Mukherjee
apparently used there is visually the appearance of a small tab on the
outside of the bend in the letter gamma, which made it look like an
abbreviated limb of an upsilon, usually rendered U in the translitera-
tion of K u ~ a ~ a Greek although Mukherjee's ''Y'' is not incorrect.
8. Organized and presided over by Dr. R. C. Shanna, then Director of the
Lucknow Musewn. (Regrettably the proceedings have yet to be published.)
9. See note 7.
10. See note 5.
A careful examination of the coin itself,11 and using large photo-
graphic silver halide prints of the coin (Figure 1), a set of eight slides
made of the coin in rotating raking light, and various computer en-
hancements of selected images from this group leaves very little doubt
as to what is and is not present in the inscription (Figure 5). The letter
tau at 7:30 is a tau and there is little chance for it to be anything else.
In comparison to the mu and it is upside down, but, because letter
orientations vary in coins, this is not a problem. At 4:30 the
double ligature is actually two letters, a gamma and an omicron. As
may be seen from Figure 1, the small tab is not there. There simply is
no stroke on the gamma extending into the opening of the omicron as
suggested by the GobI photograph, and, thus, while it is reversed, it is
a gamma followed by an omicron. The beta is, like the gamma, re-
versed, but is an obvious letter as is the omicron following it.
Counterclockwise from those is a clearly rendered upsilon which is
arranged vertically in relation to the figure in the center of the coin,
and after that an obvious delta also arranged vertically. Curiously, the
delta at 1 :00 is very unclear in most photographs, resembling an omi-
cron in some cases and in others simply an engraver's error or possibly
even a damage. However, Figure 1 shows it clearly and that is why I
have selected this particular photograph to illustrate the coin.
According to the foregoing, it is my reading that there is an uninter-
rupted counterclockwise reading, of M-e-t [r] [a] g-o B-o-u-d [0]
(H)-C./[Pl>..]<o &o''(t)[oJ), accepting that there are two missing letters in
the 6:00 position. Thus, we have another Metrago Boudo coin, al-
though not read in quite the way Cribb saw it.
When a preliminary version of this paper was presented at the
studies paneJl2 at the 21st Annual Conference on South Asian
Studies held at The University of Wisconsin at Madison in November
of 1992, a well-known collector of coins, William F. Spengler,
shared from his collection a second specimen apparently from the
same dies as the British Museum example (Figure 6). Significantly, it
was struck in such a manner that the presumed missing alpha and rho
are still present (albeit transposed from the expected spelling). Thus,
the actual reading is "Metargo Boud[o]" (HJ-C./[APJ> 0
(Figure 7). With the gamma and the beta reversed, the tau, upsilon
and delta aligned with different orientation than the other letters which
11. Courtesy of Joseph Cribb, Keeper of Indian coins at the British Museum.
12. Organized by Drs. Martha Carter and Carolyn Schmidt.
360 JIABS 16.2
are radially aligned relative to the circumference of the coin and the
rho and alpha transposed, it is indeed a very difficult inscription.
However, in the discussion about the coins at the meeting, William
Spengler cogently noted that the best die cutters were probably
engaged in producing the dies for the gold staters of the era
and that much less skilled craftsmen were at work on the copper
tetradrachms. As we have just seen, coins of the GobI 793,1 type seem
to have r befm, particularly the work of either a marginally literate or
possibly even illiterate die cutter.
One cannot fault either Cribb or Mukherjee for their readings and
certainly neither had the benefit of a second coin from the same dies
confirming the "missing letters" presented to them. Mukherjee's erro-
neous reading illustrates the problems of dealing with secondary and
even tertiary materials--we all must do it, but one needs to be careful
and, the more problematic the piece, the more careful one must be.
Both Cribb and Mukherjee have contributed greatly to numis-
matics, far more than I ever even hope to, and it is certainly not my
intention to be critical of the work of either of these friends. Quite the
contrary, the inscriptions on these coins are as difficult as any in the
Asian numismatic world, often fragmentary, with, as we have just
seen, misspellings and often in such poor conditions that they are
nearly impossible to read. Even with the computer based analysis
techniques that I am using, there are many problems that are simply
beyond technological solutions. The study of such coins needs the
skills and intuitive sense of someone like either Mukherjee or Cribb to
solve the problems that the coin presents.
Although I have been a relatively serious numismatist for more than
thirty years, my numismatic interests are in other regions of Asia, and,
relative to coins, my intellectual goals are vastly more limited
than either Cribb's or Mukherjee's. I am simply lOOking for contem-
poreaneously inscribed identi'fications of Buddhist images for
inclusion in the Encyclopedia of Buddhist Iconography. In this
context, it must be noted that Cribb's analysis of the iconography of
these coins13 also needs updating in light of the inscriptive evidence
actually present on the coin and in light of what it is now possible to
see in the image on the coins. First of all, his identification of major
Bactro-Gandharan images as the Buddha Sakyamuni must be
13. Cribb, Buddha Coins-The Official Iconography of Sakya-
muni & Maitreya," 81-84.
considered to be very tentative at best and, in many cases, probably
wrong. Images of Buddha Maitreya (Ketumatr Maitreya) exist in
Bactro-Gandharan art and, except for minor details of pedestals, these
. images are .indistinguishable from Sakyamuni images.
Thus, any
image of a standing buddha without a pedestal or an inscription (of
which there is only one known at present) simply cannot be identified
as to whether it is Ketumati Maitreya Buddha or Sakyarnuni Buddha,
or even another Buddha.
Regarding this pOint, it must be understood that images of Maitreya
in both the buddha and bodhisattva forms were very popular in Bactro-
Gandharan Buddhism. Indeed, a survey of the more than 3000 photo-
graphs of Bactro-Gandharan material in The John and Susan
Huntington Photographic Archive at The Ohio State University reveals
that a majority of individual standing bodhisattva images are represen-
tations of the Bodhisattva Maitreya,15 proving beyond any doubt the
popularity of Maitreya among the practitioners of
Buddhism in the Bactro-Gandharan regions. This probably reflects the
popular desire to be reborn in either Paradise or at the time of
Ketumati here in the Saha world some 25,000 years after the death of
Sakyamuni Buddha. While such a Bactro-Gandharan "pure-land" cult
was nbt described by the Chinese pilgrims, nor has it been substan-
tiated by extant historical literature or inscriptions; the evidence of the
cult as known in other contexts, especially China, and the very signifi-
cant number of Maitreya in itself would seem to demonstrate
the presence of the cult in Bactro-Gandhara.
In his first article, Cribb stated that aU the images of
Metrago/Maitreya that he illustrates are depicted displaying
However, this is neither accurate nor is it an accurate
14. John C. Huntington, "The Iconography and Iconology of Maitreya
Images in Gandhara," Journal a/Central Asia 7.1 (July 1984): 149-152.
15. It must be acknowledged ,that if one counts the small life narrative events
and scenes, then the Buddha Sakyamuni does predominate, but in major indi-
vidual images, thy identifiable depictions of Maitreya actually predominate
over the those of Sakyamuni Buddha
16. Cribb, Buddha Coins-The Official Iconography of
Sakyamuni & Maitreya," 82. Actually, Cribb's statement is Qnclear. The first
part of the paragraph says, "The forthcoming study of these coins will show
that only the gesture of reassurance (abhliyamudrli) is depicted on them."
"These coins" appears to refer to MetragolMaitreya images described in the
preceding paragraph. However, the subject of the next sentence is the stand-
362 JIABS 16.2
impression of images of Maitreya on K u ~ a 1 ) a coinage. In the corpus
compiled by Gobl,17 four types of MaitreyaIMetrago coins are listed:
numbers 790, 791, 792, and 793. Of this group, numbers 790, 791 and
792 all have an image of MetragolMaitreya with his right hand raised
and the left hand resting on his left leg. In the 790 and 792 types the
right hand is in abhayamudra and, for example, in the 790.1 coin
traces of MetragolMaitreya's usual attribute of a bottle (kamm).aZu)
can be seen quite clearly in computer enhancements (Figures 8 and 9).
This seems to have been a popular type of image in the Bactro-
Gandharan region and several intact or nearly intact stone examples
survive. One of the best preserved stone examples of the type is
presently housed in the Royal Ontario Museum in Toronto (Figure 10).
However, of special interest to the issue of the varieties of mudra
displayed by Metrago/Maitreya images in Bactro-Gandhara is the coin
illustrated as 791.1 in GobI. The figure of MetragolMaitreya appears
to have been represented with the right hand making vitarkamudra, a
very rare gesture for him in Gandhara. This gesture is clearly visible
in the illustration in GobI. To date, I have not yet found a stone image
of MetragolMaitreya making vitarkamudra and the coin may represent
either a very rare type or a single image that is no longer known.
More importantly, the GobI 793.1 type coin (Figure 1) in the British
Museum example illustrates MaitreyaiMetrago making dharmacakra-
mudra and, in the Spengler example (Figure 6), the thumb of the
proper right hand is clearly visible at the top of the right hand, giving
absolute certainty that the palm of the hand is toward the chest of the
figure and not extending away from the chest as would be necessary in
the abhayamudra gesture. Moreover, even in examples of the coin
type where the thumb cannot be seen, the position of the right hand in
front of the chest and the angle of the left arm, which extends across
the Chest instead of angling down towards the left leg, make it certain
that both hands were in front of the chest-exactly the pOSition that is
demanded by the Bactro-Gandharan version of dharmacakramudra.
Significantly, there is neither a bottle nor a vase associated with
Metrago/Maitreya in the coin image. This is the exact pattern of some
tentatively identified Gandharan stone images of Metrago/Maitreya.
ing buddhas and it is possible that "these coins" refers to the standing bud-
dhas. My observations are based on taking his sentence structure literally.
17. See note 5.
Prior to the correct reading of the inscription and analysis of the image
on this particular coin type, certain stone images were identified as
MetragolMaitreya on the basis of key secondary characteristics, such
as hair am1ngements, 18 but inscriptive evidence has been lacking.
One of the best preserved stone images of the type is in the
Peshawar Museum (Figure 11). It exhibits dharmacakramudra and has
been identified as MetragolMaitreya primarily because of the long
looping hair arrangement of the japimuku.ta, which is often pointed out
to be reminiscent of the hair style on the Apollo Belvedere. This par-
ticular type of hair arrangement has long been identified with
MetragolMaitreya images in Bactro-Gandhara, but the usual attributes
are lacking. With the reading of the GobI 793.1 coin type and the
accurate analysis of the image type on the coin, it is clear that exactly
this type of MetragolMaitreya image is intended. Thus, the coin type
adds another inSCriptional identification to the iconography of Metrago
IMaitreya in the Bactro-Gandhara region.
A note on the drawings accompanying this article:
The drawings of the coins have been prepared by two distinct
means. The first method, for example, used for the GobI 790.1 coin,
was done by tracing a scanned image in a good drawing program on a
Macintosh computer. Using enlargements of the coin with high
resolution computer technology it is possible to work directly over a
photograph of the coin at a magnification of as much as 1600
diameters, which allows for much greater accuracy than the usual eye
copies. The second method of drawing is to make a composite drawing
using the technique described above but from several different sources.
In this case, the drawing does not reflect either a single speciman or
anyone view of the coin, but a composite of several different coins,
or, in the case of the British Museum's GobI no. 793.1, photographs of
the same coin.
18. Huntington, "The Iconography and Iconology of Maitreya Images in
Gandhara," 148-149.
364 JIABS 16.2
Appendix I
Greek letters occurring on KU9flI).a coins
Letters in the following chart have been drawn from inscriptions on the coins
of Kanir;;ka 1. Those to the left occur most commonly; less common variants fallow.
A Alpha AAAA A
B Beta
r Gamma (
A Lambd, A
r K

[ ~ t. ...
f .... -:
Figure 1. Reverse of a copper tetradrachrn of the reign of K a n i ~ k a I.
British Museun Collection. GobI 793.1
366 JIABS 16.2
Drawing of the coin
showing the directions of reading
and position of the missing letters
necessary for Cribb's reading.
hr D D
T [P]
A Y 0 BOY [0]
Cribb, GobI (presumably), etc., reading counterclockwise
starting at 11:00, breaking the line at the double bars at
6:30 and restarting reading clockwise at LOO.
Figure 2. Cribb's reading of the inscription on GobI 793.1.
Drawing of the coin showing
the key problematic letters.
u (Y IJ
AMEDTOBOY (sic. for U), i.e., Amitabha
Mukherjee's reading clockwise starting at the lower left
letter (7:00) based on GobI 793.1. Boxed figure omitted.
The real source of the problem
Appearance of the letters as seen in the
photograph of the coin as published in GobI.
. Alpha in the first
Upsilon in the
final position.
The same letters seen in both British Museum
photograph and JCH photographs.
r Tau
Figure 3. Mukherjee's reading of the inscription on GobI 793.1.
368 JIABS 16.2
Figure 4. Composite drawing of the reverse of GobI 793.1
showing the image of MetragolMaitreya.
MET [P] [A] rOB 0 Y L1 [0]
Figure 5. The reading of the inscription on GobI 793.1
suggested in this article.
Figure 6, Reverse of a copper tetradrachm of the reign of K a n i ~ k a I.
William F. Spengler Collection.
370 JIABS 16.2
Figure 7. Drawing of the British Museum coin (GobI 793.1) with the
letters alpha and rho added in the position in which they occur on the
Spengler coin.
Figure 8. Reverse of a copper tetradrachm of the reign of K a n i ~ k a 1. Berlin
Museum Collection. GobI 793.1. (After GobI.)
372 JIABS 16.2
Figure 9. Drawing of the reverse of a copper
tetradrachm from the reign of K a n i ~ k a I.
(The alpha is lacking in the inscription which,
therefore, reads: MetrDgo Baudo.) GobI 793.1.
Figure 10. The Bodhisattva Metrago/Maitreya originally displaying
abhliyamudra and carrying the kama[lrjalu.
Royal Ontario Museum Collection.
374 JIABS 16.2
Figure 11. The Bodhisattva Metrago/Maitreya displaying a variant of the
Bactro-Gandharan version of dhannacakramudra.
Reinterpreting the Jhiinas
The jhanas, the stages of progressively deepening concentration that
figure so prominently in Buddhist meditation theory, have recently been
the subject of several excellent critical studies.
Two such studies, those of Griffiths (1983) and Stuart-Fox (1989),
have drawn attention to one problem in particular that is demonstrably
crucial in any attempt to understand the jhZina series. Ithas to do with
the composition of the first jhZina. The Pali Abhidhamma and classical
meditation manuals, and with them most present-day accounts of
Theravadin meditation theory, consistently state that the first jhana has
mental onepointedness (cittass' ekaggatZi) as one of its component
"factors." Yet the description which appears repeatedly in the first four
Nikayas (and which, therefore, certainly antedates the Abhidhamma
version) states that mental onepointedness becomes established in the
secondjhZina, not in the first. Stuart-Fox, who discusses this matter in
detail, concludes that the Abhidhamma description of the flIst jhZina is a
secondary development, a result of scholastic editing of the earlier
Nikaya account.
Both Griffiths (briefly and in passing) and Stuart-Fox (at length and
explicitly) draw another closely related conclusion regarding the com-
position of the firstjhZina as described in the Nikayas: vitakka-vicara,
1. See in particular Martin Stuart-Fox, "Jhana and Buddhist Scholasticism,"
Journal of the International Association of Buddhist Studies 12.2 (1989): 79-
110; and Paul Griffiths, "Buddhist Jhana: A fonn-critical study," Religion 13
(1983): 55-68. Also relevant are Winston L. King, Theravada Meditation:
The Buddhist Transformation of Yoga (University Park and London:
Pennsylvania State University Press, 1980) (esp. Chapters 3-6); and Johannes
Bronkhorst, The Two Traditions of Meditation in Ancient India, Alt- und
Neu-Indische Studien 28 (Stuttgart: Franz Steiner, 1986). Such critical stud-
ies contrast with the largely uncritical, though very thorough and useful,
descriptive account by Henepola Gunaratana, The Path of Serenity and In-
sight (Delhi: Motilal Banarsidass, 1985). The present paper was presented in
much abbreviated fonn to the International Congress on Religion, Melbourne,
July 1992.
376 JlABS 16.2
the factor that particularly characterizes the first jhlina, is probably
nothing other than the noimal process of discursive thought, the
but usually unnoticed stream of mental imagery and verbalization. 2
These conclusions conflict with the widespread conception of the fITst
jhlina as a state of deep concentration, a profoundly altered state of con-
sciousness attainable only after long and arduous practice. 3 They can be
shown also to challenge some long-held notions about the jhlina series
as a whole. To investigate the further implications of this revised under-
standing of the firstjhiina is a major objective of the present study.
As to method, this study employs the kind of text-critical approach
adopted by Griffiths and Stuart-Fox, while also taking into account what
is known of the practical-experiential side of jhlina meditation. It care-
fully distinguishes the earliest account of jhlina, found throughout the
Nikayas, from the historically later versions found in some tate suttas,
the Abhidhamma, and Buddhaghosa's Visuddhimagga. Indeed, one of
its specific aims is to clarify the relationship between the earlier and later
The inclusion of meditative experience among the data to be used in
the interpretive process raises some difficult methodological issues.
For present purposes the central problem is that scholars who are non-
meditators, and whQ are therefore in no position to check the accuracy of
accounts of meditative experience, are naturally inclined to have reserva-
tions about interpretive procedures that draw on such accounts. Ade-
quate discussion of this and related methodological issues is clearly
beyond the scope of this paper, so it must suffice here to make just the
following point. In the present case the account of meditative experience
in question is shown to agree substantially with the relevant description
given by Buddhaghosa in his Visuddhimagga-a that should
minimize possible concern on the part of non-meditator scholars.
Whereas previous studies have focused on the first two jhlinas, the
present analysis covers the entire series, comprising the four basic
2. Griffiths 59-60; Stuart-Fox 81-82 and passim.
:to For a typical example of that conception, see Bhikkhu Buddhadasa,
Anapanasati (Mindfulness of Breathing), trans. Bhikkhu Nagasena,
(Bangkok: Sublime Life Mission, 1976). The first 153 pages of
Buddhadasa's book are devoted to the practicalities of attaining the first
4. Some of the methodological issues raised in this paragraph are noted
briefly by Stuart-Fox 94-96. The field of Buddhist Studies will eventually
have to come to terms with such issues if it is ever to do justice to medita-
jhliruis (called, in the Abhiclhamma and Visuddhimagga, rupa-jhlinas,
"materialjhlinas") and the four liruppas (arupa-jhlinas, "non-material
jhlinas").5 For convenience, the separate jhlinas are henceforth referred
. to as "jhlina 1," "jhlina 2," and so on up to "jhlina 8" (neva sannli
The Nikaya account is examined first, followed by
Buddhaghosa's more elaborate version. The two are then considered in
the light of meditative experience. Finally; conclusions are drawnre-
garding the relationship between the two versions, and regarding
identities of the various stages in terms of meditative practices and at-
tainments. These conclusions are seen as indicating a need to revise
some long-established ideas about the jhlinas.
Analysis ofyhe Nikaya Account
The often repeated jhlina formula or "peri cope" may be provisionally,
and rather literally, translated as follows'?
5. All eight are listed at, e. g., M i 40-41; the first four alone (i. e. the rupa-
jhlinas) are listed at, e. g., D i 73-75. (All such source references are to vol-
ume and page numbers in the Pali Text Society's editions of the PaIi texts. D
= Dlgha Nikaya, etc.; Vism = Visuddhimagga; Vibh = Vibhanga.) Griffiths
states (57) that the shorter listing occurs at least 86 times in the first four
Nikayas. Because the ilruppas are often omitted from textual accounts, some
investigators have suggested that they were not part of the Buddha's original
teaching; e. g. Friedrich Heiler, Die Buddhistische Versenkung (Miinchen:
Reinhardt, 1922) 47-51; King 14-15; and Bronkhorst 82-86. That debate is
not pursued here. Instead, the jhilnas, rupa and arupa, are considered to-
gether, as they are in many suttas, as constituting a single series.
6. Cf. Amadeo Sole-Leris, Tranquillity and Insight (Boston: ShambhaIa,
1986) 68-71, where essentially the same nomenclature is adopted.
7. The Pilii reads: 1) vivicceva kilmehi vivicca akusalehi dhammehi
savitakka1[l pitisukha1[l pa/halna1[ljhlina1[l upasampajja
viharati. 2) vitakkavicilrilna'll vupasamil ajjhatta1[l sampasiidana1[l cetaso
ekodibhliva1!l aviCilra'll samiidhija'll pitisukha1[l dutiya1[l jhlina1[l
upasampajja viharati. 3) pitiyil ca virilgil upekhako ca viharati sato ca
sampajilno sukhaii ca kilyena pa/isa1[lvedeti yan ta1[l ariyil ilcikkhanti
upekhako satimil sukhavihlirl ti tatiya'll jhlina1[l upasampajja viharati. 4)
sukhassa ca pahilnil dukkhassa ca pahilnil pubbeva somanas-
sadomanaSSilnal!l atthagamil adukkham asukha1[l ujJekhiisatipilrisuddhi1[l
catuttha'll jhlina1[l upasampajja viharati. 5) sabbaso rupasaiiiiilna1[l
samatikkarna pa/ighasaiiiiilna1[l atthagamil nilnattasaiiiilina1[l amanasiklirli
ananto ilkliso upasampajja viharati. 6) sabbaso
iiklislinaiicliyatana'll samatikkamma ti viiiiiilIJaiicilyatana1[l
upasampajja viharati. 7) sabbaso viiiiiilIJaiicilyatana1[l satnatikkamtna natthi
kiiici ti ilkiiicaiiiiilyatana1[l upasampajja viharati. 8) sabbaso
satnatikkamma nevasaiiiiilnlisaiifiilyatana1[l upasampajja
378 JIABS 16.2
Jhilna 1: Quite separated from sense desires, separated from unwhole-
some mental states, he [the meditator] attains and abides in the fIrst
jhilna, in which are present initial thought (vitakka), sustained thought
(vicara), and separation-born zest (pui) and pleasure (sukha).
Jhana 2: Through the suppression of initial thought and sustained
thought, he attains and abides in the second jMina, in which there is
inner tranquillity and oneness of mind, and in which initial thought
and sustained thought are absent, and concentration-born zest and
pleasure are present.
Jhiina 3: Through the fading away of zest, he abides equanimous,
mindful and discerning; and experiencing pleasure with the body, he
attains and abides in the third jhana, of which the Noble Ones say
"equanimous, mindful, abiding in pleasure."
Jhana 4: Through the relinquishing of pleasure, through the r e l i n ~
quishing of pain, through the previous disappearance of happiness and
sorrow, he attains and abides in the fourth jhana, in which pleasure
and pain are absent, and the purity of equanimity and mindfuless is
Jhana 5: Through the complete transcending of material perceptions,
through the disappearance of impact-perceptions, through non-atten-
tion to variety-perceptions, [aware] that space is endless, he attains and
abides in the realm of endless space (aklisanaiicayatana).
Jhana 6: Through the complete transcending of the realm of endless
space, [aware] that consciousness is endless, he attains and abides in
the realm of endless consciousness (vififiar;aiicayatana).
Jhana 7: Through the complete transcending of the realm of endless
consciousness, [aware] that there is nothing, he attains and abides in
the realm of nothingness (akiiicaiifiayatana).
Jhiina 8: TIlrough the complete transcending of the realm of nothing-
ness, he attains and abides in the realm of neither perception nor non-
perception (n' eva saiifia niisafifiayatana).
This translation is tentative and subject to later revision, particularly in
respect of the major technical terms. Some of the renderings adopted are
based simply on common western usage, for want of more adequate
criteria at this early stage in the investigation. For example, piti is
provisionally given as "zest" because that word is often preferred in
English translations. There are also some syntactic ambiguities in the
PaIi, which will be addressed as the analysis proceeds.
The above standard description of the jha.rws will now be examined
critically within a purely linguistic-textual-doctrinal framework, i. e.
without at this stage making any attempt to link it to meditative practice.
Since it is the Nikaya description that is in question, the later interpreta-
tions and explanations found in the Abhidhamma and the
Visuddhimagga will be referred to only sparingly and with caution.
Attention focuses first on the four rupa-jhifnas (jhilnas 1 to 4).
Each of the first four paragraphs consists essentially in a statement of
(a) the mental factors that are present or absent in each jhifna, and (b) the
factors that are developed or eliminated in making the transition to that
jhi1na from the one preceding it. The mental condition of the monk or
meditator before beginning the jhi1na practice is not described directly.
Indirectly, however, the account does indicate that this pre-jhilna
condition is characterized by the presence of sense desires (kama) and
other unwholesome mental states (akusala dhammas), for it is by
becoming separated or isolated (vivicca) from these that the meditator
attains jhana 1.
It is stated that injMna 1 there exist initial thought (vitakka) and sus-
tained thought (vicara), together with zest (piti) and pleasure (sukha),
both of which are "separation-born" (viveka-ja). 8 The adjective
"separation-born" amounts to a reiteration of the statement that the medi-
tator attains this jhana through becoming separated (vivicca)-i. e. sepa-
rated from sense desires and unwholesome states. Its application to
"zest" and "pleasure" (which immediately follow it in and
not to "initial thought" and "sustained thought" (which immediately pre-
cede it) indicates that it is above all this separation, with resulting zest
and pleasure, that distinguishes jhilna 1 from the pre-jhifna condition. It
indicates that the presence of initial and sustained thought in jhana 1 is
not a consequence of the separation from sense desires and unwhole-
some states; that is, initial and sustained thought are present already in
the pre-jhifna condition and merely perSist through the transition. The
essence of the transition from normal consciousness to jhilna 1 consists,
8. Buddhaghosa suggests that can be seen as qualifying either
p}tisukhal[! or jhtinal[! (Vism 145). I follow former interpretation, as do
NaI)amoli and many others. See Bhikkhu NaI)amoli, trans., The Path of
Pyrification (Visuddhimagga) (Berkeley & London: Shambhala, 1976) 15I.
translation is hereafter denoted Path.)
380 JIABS 16.2
therefore, in (a) the elimination of sense desires and other unwholesome
states, and" (b) the arising of zest and pleasure.
The transition fromjhlina 1 to jhiina 2 is achieved t11!0ugh the sup-
pression or stilling (vupasamli) of initial and sustained thought, and the
estabiishing of inner tranquillity (ajjhatta1J1. sampaSlidana1J1.) and one-
ness of mind (cetaso ekodibhliva1J1.). This is reiterated in the statement
thatjhlina 2 is without initial thought and thought (av"itakka,
aviclira). Zest and pleasure, already established in the preceding jhiina,
are still present but are now described as "concentration-born"
(samlidhi-ja). "Concentration," "inner tranquillity," and "oneness of
mind" are evidently synonyms. 10 The essence of the transition to jhlina
2 is, then, the elimination of initial and sustained thought and the estab-
lishing of concentration.
The transition to jhlina 3 comes about through the fading away of zest
(pUi), as the meditator becomes equanimous or conatively neutral
(upekhako or upekkhako) and also mindful and self-possessed (sato,
sampajiino). Pleasure continues, but is now, for the first time, said to be
experienced with the body (kliyena). As Gunaratana points out, the term
"upekkhli," though having many different applications, always signifies
a midpoint or point of neutrality between extremes. I I In the present case
the reference is clearly to neutrality in the domain of conation, i. e. to a
9. The vague rendering "states" for dhammehi sidesteps the question which
of the many meanings of dhamma is intended here. One important meaning
of dhamma is "mental object" or "mental image," and this could well be tOe
meaning intended in the present context. (See T. W. Rbys Davids and
William Stede, PaZi-English Dictionary (London: Luzac, 1959) 336, dham-
ma.) If it is, then the factors said to be eliminated in the transition from
ordinary consciousness to jhlina 1 are sense desires and unwholesome images.
This would explain what otherwise appears an unnecessary repetition; for
"vivicc' eva klimehi, vivicca akusalehi dhammeht' would then be referring to
two different mental elements. (In Table 1 they would be in two different
columns, "Conation" (kamas) and "Thought" (akusaladhammas), rather than
in the same column as shown.) A further implication would be that the
vitakka-viClira ofjhlina 1, being free of unwholesome thoughts, does after all
differ from the normal flow of thought.
10. Cetaso ekodibhliva is equated at Vibh 258 with cittassa .thiti (steadiness
of mind) and sammlisamlidhi (right concentration); it is defined in the Pali-
English Dictionary (160) as "concentration, fixing one's mind on one point."
The term's equivalence with cittass' ekaggatli is self-evident. Sampaslidana
is explained at Vibh 258 as "saddhli (faith, confidence)"; the Pali-English
Dictionary definition is "tranquilizing" (692). Gunaratana (83) notes these
two meanings, "confidence" and "tranquillity," and opts for the former,
though the latter is clearly more appropriate in the context.
11. Gunaratana 88-90. Cf. Vism 160-161; Path 166-167.
state.of affective detachment. The meditator becomes upekhako through
the disappearance of piti, a conative factor (placed under sankhlira-
khandha in the Abhidhamma classifIcation).12 Thus, the essence of the
. transition from jhlina 2 to jhlina 3 is the replacement of piti (zest?) by
the conatively neutral sati-sampajafifia (mindfulness and self-
possession). That the pleasure (sukha) is now explicitly physical
appears to represent another significant development.
In the transition to jhlina 4, pleasure (sukha) is relinquished or al-
lowed to disappear. The description states that pain (dukkha) disappears
also, though it was not mentioned as present in earlier jhlinas. Since
jhlinas 1,2, and 3 are all described as pleasurable, this disappearance of
pain makes sense only if understood as having been "entailed in the
establishing ofjhlina 1. Such a meaning is the more likely because the
next two factors mentioned, happiness (somanassa) and sorrow
(domanassa), are explicitly stated to have disappeared previously or
earlier (pubbeva).
As Gunaratana pOints out, analysis of the description is complicated
by the existence of two different Nikaya usages of the terms sukha and
First usage: sukha:
Second usage: sukha:
physical and mental pleasUre
physical and mental pain
physical pleasure
physical pain
mental pleasure (happiness)
mental pain (sorrow)
In the description of jhlina 4 all four terms occur, whence it is clear that
the second usage is being followed. Thus the sukha that is relinquished
in attainingjhlina 4 is physical or bodily pleasure, which is in keeping
with the fact that the sukha present in jhlina 3 is experienced "with the
body." The description is not explicit regarding the type of sukha pre-
sent injhlinas 1 and 2.
In the final string of adjectives describing jhlina 4, the pair asukham
adukkha1Jl (without pleasure, without pain) is followed by upekkhli-sati-
12. Gunaratana 60,91.
13. Gunaratana 62-63.
382 JIABS 16.2
parisuddhiTft (having purity of equanimity and mindfulness). 14 Since
upekkha and sati were already present in the preceding jhana, the
addition of the word parisuddhiTft ("purity") evidently signifies that
upekkhii and sati are now no longer associated with sukha; that is,
parisuddhi signifies absence of sukha, just as (in jhiina 3) upekkha sig-
nifies absence of pUi .
The account of the four rupa-jhlinas exhibits a stylistic feature typical
of the Pili canon in general: frequent reiteration through the use of syn-
0nyms and (in negations) antonyms .. For example, the statement that
jhiina 2 is attained through suppression of initial thougnt and sustained
thought (vitakka-vicaranaTft vupasama) is reiterated in the further state-
ments that thatjhlina is without initial and sustained thought (avitakkaTft
avicaraTft), that it is characterized by inner tranquility (ajjhatta7Jt
sampasadanaTft) and oneness of mind (cetaso ekodibhavaTft), and that
the associated zest and pleasure are born of concentration (samadhi-
ja7Jt). Accordingly the above analysis has, in large part, consisted in
identifying such sets of synonyms and antonyms, a procedure that
greatly simplifies the description.
It will be helpful at this point to depict the results' of the analysis dia-
grammatically. This is done in Table 1. Each transition between jhlinas
is represented by a downward-pointing arrow, and the factors respon-
sible for the transition are indicated by the boxed terms attached to the
Table 1 draws attention to some further characteristics of the jhana
description. One evident characteristic is inconsistency in mentioning
the continued existence of a factor in jhanas subsequent to the one in
which that factor first becomes established. For example, equanimity
(upekkhii), which becomes established in jhlina 3, is stated to be present
also injhlina 4. On the other hand, the quality "without initial and sus-
tained thought" (avitakkaTft, avicaraTft)-otherwise "having tranquillity"
(sampasadanaTft), and "having oneness of mind" (cetaso ekodibhava7Jt)
-which is attributed to jhlina 2, is not similarly applied to jhlinas 3 and .
14. This seems more likely to be the meaning of the compound than "having
mindfulness purified by equanimity," because upekkhli (equanimity) was
already present in jhlina 3. However, cf. Path 174; Vism 167-168; Vibh
4, though it is clearly to be understood to apply to them, and indeed al-
ways has been by commentators classical and modem. 15
Another characteristic evident in Ta,ble 1 is that the composition of the
. rupa-jhiinas is specified in terms of three implicit categories. This has
been emphasized by providing the three relevant columns with headings:
. "Thought," "Conation," and "Peeling" (i. e. hedonic tone).
When the above points are taken into account, Table 1 reduces to the
much simpler Table 2. In Table 2 we immediately see the jhlina series
as a process of successively eliminating mental factors. The term below
each arrow is functionally the negation of the one above it; e. g.
ekodibhiiva is the negation of vitakka-vicara .16
Table'2 can in its turn be simplified by replacing each negating term
with a dash, on the understanding that a dash Signifies the absence or
elimination of the factor immediately above it. The result is the maxi -
mally economical representation shown in Table 3.
The terms that appear in Table 3 are the first four of the familiar five
"jhiina factors" (jhiinanglinl): vitakka, vicara, pili, sukha, ekaggata. The
practice of summarizing the composition of the jhlinas by listing the rel-
evant jhlina factors appears sporadically in a few late suttas, and be-
comes well established in the Abhidhamma.
The odd development
whereby the factor ekaggata (= ekodibhava) came to be attributed to
jhlinal is among the problems dealt with by Stuart-Fox.
The analysis can now move on to the arupa-jhlinas, the non-material
jhlinas. The first of these (in our terminology, jhlJna 5) is the realm of
endless space (iiklisanancayatana). It is attained "through the complete
transcending of material perceptions (rupa-sanna), through the disap-
pearance of impact-perceptions (pa,tigha-sannii), through non-attention
to variety-perceptions" (nlinatta-safina), and it entails the awareness that
"space is endless" (ananto iikiiso).
. Of the tbree terms ending in -sanna, the first, rupa-sanna, is familiar
as denoting perception of visual forms, the first of six recognized classes
15. Cf. Buddhaclasa 158: " ... it should be understood that anything dis-
carded in a lower stage remains absent in higher stages and is therefore not
mentioned again."
16. In choosing such negative terms for inclusion in Table 2, I have inten-
tionally avoided the visually self-evident ones (e. g. avitakka as the negation
ofvitakka) in order to make the diagram maximally informative. That
ekodibhtiva is the negation of vitakka-viciira is not immediately apparent and
therefore worth stating explici t1 y .
17. In the suttas it appears (with ekaggatii included) at M i 294, M iii 25-
29, S iv 263. See Stuart-Fox 85 ff.
384 nABS 16.2
of sense perception. 18 However, in the present context it clearly has a
wider scope, justifying the usual translation "material perceptions" or
"perceptions of matter." 19 (Buddhaghosa explains it as perceptions of
the nlpa-jhanas and of their objects-presumably the kasi1}a disks, the
breathing, etc.)2o This ambiguity of rapa-sanna corresponds to an am-
biguity in the word rapa:rupa is sometimes "visible fonn" (the object
of visual perception) and sometimes "matter, materiality" (as when
contrasted with ila11U/, or with arapa).21 In the present context, then,
rapa-sanna covers all but the sixth class of sanna, i. e. aU but dhamma-
safina, the type that has mental images (dhammas) as its objects.
The second of the three terms, pa.tigha-safina ("impact-perception"),
is explained in the Vibhaliga as denoting perceptions of visual forms,
sounds, odors, tastes, and tangible objects. 22 This indicates that
pa.tigha-sanna is identical with the preceding item, rapa-sanna. The
third tenn, nanatta-sanna, ("variety-perception") contains in its literal
meaning little indication just what type of perception is being referred to.
However, the pattern established by rapa-saiina and pa,tigha-sanna
makes it likely that nanatta-sanna is a further synonym, i. e. that it too
signifies "sense-perception," an interpretation explicitly affinned by
18. The six are: rapa-sanna, sadda-, gandha-, rasa-, phoyhabba-, dhamma-
sanna. See D ii 309 and S iii 60.
19. See Path 356, and many other translations of the jhllna description.
20. Vism 328; Path 356-357.
21. See Pali-English Dictionary 574-575, rapa, 1 and 2.
22. Vibh 261. See also Vibh 6 and D if 62, where. pa.tigha-samphassa is
contrasted with adhivacana-samphassa "verbal (or conceptual, i. e. mental)
impression." (Definition from Nyanatiloka, Buddhist Dictionary [Colombo:
Frewin and Co., 1972] 142.) The Vibhanga's explanations of rapa-sanna
and nllnatta-Sanna are uninfonnative.
23. In such a succession of parallel tenns we may expect either that all have
the same meaning (appositional relationship) or that all have different mean-
ings (additive relationship). Clearly the fonner applies here. (An example of
the latter occurs at the beginning -of thejhana 4 fonnula.) Buddhaghosa's
support for this interpretation of nanatta-sanna comes in the following
statement. '''Through the disappearance of impact-perceptions, through non-
attention to variety-perceptions': by this is meant the relinquishing of and
non-attention to all sense-sphere consciousness and its concomitants" (Vism
331). Buddhaghosa implausibly also states that such perceptions were already
abandoned in jhllna 1 (Vism 329-330)--evidently in an attempt to reconcile
the Nikaya account of the jhllnas (which he professes to be explicating) with
the Abhidhainma understanding ofjhllna 1.
We therefore have here a thrice uttered statement that the transition
from jhana 4 to jMina 5 entails the cessation of physical sense percep-
tions. It is appropriate that this cessation of physical or material percep-
tion coincides with the transition out of
the physical or material (rl1pa)jhiinas. The first arl1pa-jhlina (jhlina 5)
can, therefore, be readily incorporated into the condensed table of the
jhOnas by adding a further column, headed "Sense Perception" (see
Table 4).
Jhana 5 is further characterized by the awareness or realization that
"Space (liklisa) is endless." In the Nikayas, liklisa is occasionally ap-
pended to the list of four elements or mahlibhl1tas, and in later times it
assumes the status of a flfih element. 24 The four-earth, water, fire, and
air-are together equated with rl1pa, i. e. materiality or physicality,
sometimes more specifically the human body. Aklisa is what remains
when these four are removed. Thus the awareness that" likiisa is end-
less" amounts to the awareness that" rl1pa is non-existent"; and this
again is an appropriate concomitant to the transition from the material or
rl1pa jhlinas to the non-material or arl1pa jhlina s. The contrast between
'Tl1pa as earth, water, fire, and air, and arl1pa . as the realms of endless
space, endless consciousness, etc., is apparent in the well known Udlina
passage: "There exists, monks, a realm in which there is not earth, nor
water, nor fire, nor air, nor realm of endless space, nor realm of endless
consciousness, nor realm of nothingness, nor realm of neither perception
nor non-perception .... "25
The transition to jhlina 6, the realm of infinite consciousness
(viftftli[laiicayatana), is achieved by transcending the realm of endless
space and realizing that consciousness (viftftli[la) is endless. The type of
analysis applied in earlier jhanas is hardly applicable here. By this stage .
in the series the information given has become so meager that nothing
remains to be considered except the significance of the term viftftli[lG.
24. In the Nikayas the set of four elements occurs frequently, e. g. at D i 55,
M i 53; the set of five occurs only rarely, e. g. at M i 413, S iii 227. On the
seemingly late addition of llkllsa, see G. P. Malalasekera, ed., Encyclopaedia
of Buddhism, vol. 1 (Colombo: Government of Ceylon, 1966) 341.
25. Udana 80. atthi bhikkhave tad llyatanaf!! yattha neva pa!haVl na llpo na
tejo na vllyo na llkllsllnancllyatanaf!! na vinnll1Jancllyatanaf!! na
llkincannllyatanaf!L na nevasannll-nlisannllyatanaf!L . ... See the Vibhanga
analysis of jhllna 5, which explains that llkllsa is "untouched by the four pri-
mary elements, asamphuHhaf!L cattlhi mahllbhlltehi" (Vibh 262).
386 JIABS16.2
That is itself a daunting problem; discussion of which will be deferred
until later in the paper.
The situation becomes even more difficult with the two remaining
jhiinas, the realm of nothingness and the realm of neither'perception nor
non-perception, each of which is attained by "transcending" the realm
that precedes it. The possibilities of the text-analytical approach, as it
can be applied to the Nikaya account, have, therefore, been exhausted for
the present. Accordingly, we now turn to other sources that
provide information on the techniques and experiences associated with
attaining the jhlina s in practice.
The Nikaya account of the jhiinas provides little information for the
practiCing meditator. Suttas such as the Aniipilna-sati Sutta do give
some guidance; however, the standard source of practical information is
the post-canonical manuals, particularly Buddhaghosa's Visuddhimagga
(5th century CE), to which we now turn.
Analysis ofBuddhaghosa's Account
The description of jhilna practice that Buddhaghosa presents in his
Visuddhimagga is widely regarded, rightly or wrongly, as authoritative
on Theravadin meditation. It undoubtedly represents an already well
established tradition, for essentially the same description is found in the
less well known Vimuttimagga of Upatissa, dated a few centuries ear-
(I shall nevertheless, for convenience, refer to this deSCription as
"Buddhaghosa's.") Buddhaghosa's account has been largely respon-
sible for the widespread understanding of jhiina 1 as a state of deep con-
centration. In it he indicates that attainment of jhiina 1 entails a long and
difficult progression through a series of sub-stages, of which the more
advanced clearly do involve deep concentration. His portrayal of jhilna
1 as a deeply concentrated state therefore affirms the Abhidhamma ac-
count (which ascribes ekaggatil to jhiina 1), while conflicting with the
earlier Nikaya account.
The task of sorting out the relationship between these two accounts,
and discovering how the differences may have come about, has already
been tackled in a preliminary way by Griffiths and Stuart-Fox. Here it '
will be dealt with more thoroughly, by first conSidering certain problems
that arise out of the series of sub-stages which Buddhaghosa describes
26. Vimuttimagga of Upatissa (TaishO 1648), transl. by N. R. M. Ebara,
Soma Thera, and Kheminda Thera as The Path of Freedom (Colombo: D.
Roland D. Weerasuria, 1961).
as leading up to jhiina 1 (and to each subsequentjhiIna.) 'This series is
not mentioned in the Nikayas, nor even in the canonical Abhidhamma
texts. Its appearance in the post-canonical Vimuttimagga and Visuddhi-
. magga is evidently associated with the revision whereby ekaggatii was
ascribed to j-hlina 1. Consequently, any elucidation of the significance of
Buddhaghosa's sub-stages may be expected to contribute to an im-
proved understanding of the entire jhlina series. To that end a summary
of Buddhaghosa' s account is now provided. 27
In the example given by Buddhaghosa the meditation object is a spe-
cially prepared "earth kasilJa, " a disk of clay about two spans in diame -
ter. The meditating monk begins by gazing with ,concentrated attention
at this disk, which therefore serves as the "preliminary sign"
(parikamma-nimitta). After long and persistent effort, he becomes able
not only to keep his attention firmly fixed on the disk itself, but also to
retain an accurate mental image of it, i. e. to "see" inwardly a clear mental
replica of the disk when he closes his eyes. 'This replica image is the
"acquired sign" (uggaha-nimitta). The monk thereafter gives up gazing
at the Original disk and concentrates on the replica image instead.
Through this exercise the replica image is progressively stabilized and
reinforced until eventually it gives way to a different type of image, the
"counterpart sign" (Pa.tibhiiga-nimitta). This is an abstract derivative of
the preceding image, bearing a general resemblance to it but lacking its
"faults" and its specific identifying features. Whereas the acquired sign
was a near -perfect mental replica of the original clay "disk, the counter-
part sign is likely to appear as a pure disk of light, for example resem-
bling the full moon or a well polished mirror. The meditator now focus-
es on this counterpart sign, seeking to "extend" it progressively. 'This
exercise is carried out in two stages: "access concentration" (upaciira-
sarriiidhi) and "fixed concentration" (appanii-samiidhi). With the per-
fection of appanii-samiidlzi, the meditator attains the first jhlina.
Once he has fully mastered these practices, the meditator may go on to
develop the second jlzana. This entails, according to Buddhaghosa, the
same series of SUb-stages, but preceded by practice of five "masteries"
(vast). These include reflection on the grossness and undesirability of
the jhiina factor to be eliminated next, which in this case is vitakka
(Buddhaghosa here follows the Abhidhamma division of jhiina 1 into
two separate jlzanas: vitakka and viciira are eliminated successively.)
27. Tbe summary is based on Vism 118-155; Path 122-161. Also, cf.
Vimuttimagga (Ebara et al.) 71-92.
388 JIABS 16.2
Much the same procedure applies for each of the remaining jhi1nas in
tum. Thus, for every one of the jhilnas, nlpa and arupa, the meditator
passes through the same series of sUb-stages: concentration on the
chosen physical object (parikamma-nimitta), development of the
acquired sign (uggaha-nimitta), development of the counterpart sign
(pa.tibhaga-nimitta), access concentration (upacara-samadhi), and
finally fixed concentration (appana-samadhi). On each occasion, the
perfection of appana-samlidhi marks attainment of the relevant jhi'ina.
It can be fairly readily confirmed that B uddhaghosa' s account is gen-
erally accurate as a description of the meditative practice. Numerous
practicing meditators, particularly in the Buddhist countries of southeast
Asia, routinely experience many of the stages Buddhaghosa describes.
They are well able-though not always very willing-to discuss the
process as far as they have experienced it. 28 Such meditators and their
teachers do not necessarily use Buddhaghosa's terminology; however,
some of the stages they describe can be readily recognized and correlated
with his account. In particular, a sequence of three meditation objects-
the original physical object, a replica image of it, and an abstract image
derived from the replica image-is well attested. And for competent
meditators the process culminates in attainment of an imageless state
barely distinguishable from total unconsciousness, which masters iden -
tify as "entry into jhilna. "29
Researchers wishing to investigate the matter at first hand can do so
by taking up intensive meditation themselves. Such experimentation will
support the claim that all meditators pass through essentially the same
sequence of stages, provided they pursue the practice intenSively and
persistently enough, in a suitable environment, and with competent guid-
28. A major difficulty in finding out about meditation practice is that medi-
tators are often very reticent about discussing their experiences and attain-
ments. Such reticence is usually enjoined by their meditation masters on var-
ious grounds, e. g. that to talk about one's attainments could generate conceit
and thereby hinder one's further progress. However, for alternative views on
this question see Winston L. King, "A Comparison of Theravada and Zen
Meditational Methods and Goals," History of Religions 9 (1970): 313; and
Rod Bucknell, "Experiments in Insight Meditation," The Australian Journal
of Transpersonal Psychology 3.2 (1983): 115.
29. Regarding these practical details, I am drawing particularly on a series of
verbal communications with the late Chaokhun Rajasiddhimuni, formerly
meditation master at Khana 5, Wat Mahathat, Bangkok. Though the style of
meditation he taught was purported to be vipassana-bhtivana, insight medita-
tion (in the Mahasi Sayadaw tradition), it entailed a large component of
samatha-bhavana, concentration meditation.
That kasilJa disks are rarely if ever used nowadays isunimpor-
tant, because the sequence is largely the same, whether the concentration
object is a clay disk, a chanted mantra, or the sensation of the breath at
. the nostril.. (Details are given in the next section.) Buddhaghosa's ac-
count therefore deserves acceptance as a reliable description of the stages
in jhlina practice as far as the attainment of what he calls "the first
jhlina. "
However, as an interpretation of those stages in terms of Buddhist
doctrine, Buddhaghosa's account presents several problems. One obvi -
ous problem has to do with the above-noted question concerning the
nature of the first jhlina. Development of a stable mental image as the
object of concentration-whether a replica image (uggaha-nimitta) or an
abstract derived image (pO;tibhZlga-nimitta)-implies well established
mental onepointedness. The final stage, appan a-samZldhi (which
Buddhaghosaidentifies with jMna I-subsequently also jhlina 2, etc.)
is portrayed as an even more advanced stage of samadhi. It follows that
Buddhaghosa's account is in conflict with the Nikaya account; because,
as the Stuart-Fox study makes clear, the jhlina 1 of the Nikaya account
is a rather preliminary stage in which mental onepointedness has not yet
been established. The condition attained by the meditator who has mas-
tered appanZl-samadhi cannot be identical with the stage which the
Nikayas call "the firstjhZlna" (pO;thamaT}1jhlinaT}1).
It could be suggested, in Buddhaghosa's defense, that perfect corres-
pondence is not to be expected: in his account of kasi1:za meditation
Buddhaghosa is referring to the firstjhlina of the Abhidhamma, not the
frrstjhZlna of the Nikayas. (The Abhidhamma version states that the
firstjhZlna has mental one pointedness as a factor; the Nikaya version
does not.) But such an argument would carry no weight, because
Buddhaghosa understands the Abhidhamma and Nikaya descriptions of
"the firstjhana" to be referring to one and the same meditative attain-
ment. He maintains that the verbal discrepancies between the two de-
scriptions are of no consequence, but merely reflect differing perceptions
about what was worth mentioning. 31
30. Such claims entail certain problems, on which see Frank J. Hoffman,
Rationality and Mind in Early Buddhism (Delhi: Motilal Banarsiciass, 1987)
31. On the question whether ekaggatti was worth mentioning as a factor in
jhana 1, see the suggestions by Gunaratana, 67 and 84, and the refutation of
them by Stuart-Fox, 88.
390 JIABS 16.2
Another problem with Buddhaghosa's account is that such details as
the uggaha- andpa.tibhliga-nimittas, and upacara- and appana-samii
dhi are nowhere explicitly mentioned in the Nikayas. There is not even
any indication in the Nikayas that attainment of jhana 1 entails a lengthy
sequence of sub-stages such as Buddhaghosa describes. This raises
questions concerning the transmission of the teaching. If this very basic
information is genuine, why was it not recorded in t.he Nikayas? And
how did commentators like Upatissa and Buddhaghosa manage to come
by it?
It is now evident that the interpretation implicit in Buddhaghosa's
account of kasil}a meditation is problematic. As a description,
Buddhaghosa's account of the sequence of meditative stages as far as
appana-samadhi appears to be accurate; it corresponds with meditative
experience. However, as an interpretation, it is demonstrably in conflict
with the Nikaya account.
We therefore confront the question: How does Buddhaghosa's de-
scription, with its detailed series of sub-stages, relate to the much sim-
pler Nikaya account of the jMnas? This question will be approached
initially by considering in greater detail the techniques and experiences
actually involved in the practice of jhZilla meditation.
The Practice of Concentration
KasiJ}a disks are rarely, if ever, used by present day meditators. The ac-
count that follows therefore describes, instead, the practice of minclful-
ness of breathing (aniiptina-sati), which is probably the most widely
used, and certainly the best documented, Buddhist technique for
jhi1na. 32 The description is based on the standard Theravadin style of
practice, but in respect of the resulting experiences and attainments it is
probably valid for all styles.
The meditator, having found a quiet spot in which to practice, and
having adopted the approved sitting posture, begins by developing an
appropriate mental attitude. This may entail reflecting for a few minutes
on the value and purpose of the practice he or she is about to undertake,
on the virtues of Buddha, Dhamma, and Sangha, or on any similarly up-
32. For textual sources, see Bhikkhu Na1}amoli, Mindfulness of
Breathing (Anliplinasati), (Kandy: Buddhist Publication Society, 1973). For
a very detailed discussion of the practice, see Buddhadasa, op. cit.
Buddbadasa's monastery (Suan Mok, near Chaiya in southern Thailand) is
one of the main centers at which linliplina-sati is currently taught and prac-
ticed on a large scale.
lifting topic. Thus prepared, he or she then closes the eyes and begins
concentrating on the breathing.
This involves focusing attention on the fine tactile sensation experi-
enced at the rim of one nostril as the breath passes in and out. That sen-
sation is the concentration object. At each sitting attention must be fo -
cused on it and restrained from wandering. Invariably, however, atten-
tion does wander. After only a few breaths the meditator realizes that
instead of concentrating on the sensation at the nostril rim, he or she is
involved in a train of thought having no apparent connection with the
practice. He or she immediately returns attention to the concentration
object and begins again, but before long the same thing happens.
Repeatedly, despite all efforts to keep the mind fixed on the concentra-
tion object, thoughts arise; and the trains of mental imagery and inner
speech sometimes continue for a minute or more before the meditator
realizes the digression and is able to cut them short. Only after long and
persistent effort-over weeks or months, depending on individual tem-
perament and the intensity of the success come. Finally,
however, the dedicated meditator does succeed in keeping attention fixed
on the concentration object for up to a minute without any thoughts
With further practice the periods of full concentration and freedom
from thought grow longer and more intense. The meditator becomes
able to sit fully concentrated for several minutes together. With thought
totally absent, there is no sense of boredom; the practice, which had
formerly seemed dull and tiresome in the extreVle, has now become irre-
sistibly interesting.
During this phase of the practice the meditator often finds the body
making strange involuntary movements, for example a pronounced trem-
bling, intermittent jerking, or creeping goose-flesh. The meditation mas-
ter reassures the student that reactions of this kind are common. They
are by-products of the high level of mental energy being developed, and
have no importance other than as signs that progress is being made. The
meditator must merely note their presence and resume the concentration
Following this advice, the meditator finds that the strange movements
do soon cease, and facility in concentration improves accordingly. But
now a new effect appears, in the form of various delightful bodily feel-
ings: a feeling of lightness as if the body were floating some distance
above the seat, or a pervading warmth as if the body were glowing. The
392 JIABS 16.2
meditator may find it possible to bring about an intensification of these
effects; however, the master warns against this. The pleasant feelings
are once again unimportant by-products of the practice; the meditator
must merely acknowledge their existence and return to the 'concentration
With further practice the delightful feelings subside in their turn,
leaving nothing in consciousness but the concentration object. Formerly
faint and barely discernible, the sensation at the nostril rim is now expe-
rienced vividly as a zone of intense tactile sensation. There is now
nothing else in consciousness. As far as the meditator is concerned the
rest of the body is non-existent.
Further prolonged concentration eventually results in a strange trans-
formation of the object. The zone of intense tactile sensation is replaced
by a glowing patch of light of similar shape and orientation, experienced
inwardly as a vivid mental image. (The eyes remain closed throughout
these exercises.) For example, if the zone of sensation at the nostril was
experienced as crescent-shaped, the glowing patch of light that takes its
place is likely to be similarly crescent -shaped. This abstract image is of
variable color, indeed the meditator may find that its color and brightness
can to some extent be modified at will. Its size seems indeterminate,
there being no other content of consciousness with which it might be
compared. Having once developed such an abstract image, the meditator
is instructed to adopt it as the new concentration object. At each sitting
he or she must begin by concentrating on the breath as usual; but as
soon as the abstract image appears, that must be made the concentration
object instead. This has the effect of causing the abstract image to arise
more rapidly ~ a c h time, and, once arisen, to become progressively more
vivid and stable.
The meditator continues practicing in this way, until one day, without
warning, the abstract image suddenly disappears. Thus deprived of the
only content of consciousness, the meditator has the sense of con-
fronting an infinite black vacuum. This strange experience may lead to a
loss of composure, with a consequent abrupt return to normal con-
sciousness. However, the master gives reassurance and advises the stu-
dent to cultivate this state of mental emptiness, entering it at every
opportunity. In addition, the master advocates prolonging its duration
by making a resolution to that effect at the beginning of each meditation
session. Following these instructions, the meditator finds that the state
of emptiness stabilizes and, as promised, lasts progressively longer.
IIi this state of emptiness, as at all previous stages of the practice, the
meditator remains conscious of the condition, retaining a detached
awareness of the state of zero mental content. However, there eventually
comes a time when even this residual consciousness abruptly ceases.
The effect is as if the meditator had suddenly gone under total anesthetic,
or fallen into deep dreamless sleep. It cannot be said of this state that the
meditator experiences it; rather, he or she infers it after the event, per-
haps by referring to a clock or some other indicator of the passage of
It is said that particularly competent meditators develop the ability to
sit in this state of unconsciousness for as long as seven days together.
Some masters set up the less ambitious goal of twenty-four hours, and
tell their students that when they have achieved that they will have gone
as far as this style of practice can take them.
The above account, based on mindfulness of breathing, is broadly
applicable for all forms of concentration meditation (samatha-bhiivanll),
though with some variations in detail depending on the type of object .
used. For example, concentration on the sound of a clock ticking natu-
rally differs in the early stages. (Some meditators find an auditory object
easier to concentrate on than a tactile one; others find it more difficult.)
The abstract image develops in much the same way as with mindfulness
of breathing, though it is likely to be different in appearance, e. g. ex-
hibiting a rhythmic movement in time with the ticking. Thereafter the
sequence of events is identical.
A substantial difference from the course of events described above
exists in the case of a visual object or a chanted mantra. With a visual
object, the meditator begins with the eyes open, but closes them once the
object has so imprinted itself on the memory that it can be visualized
clearly "in the mind's eye." With a mantra, the meditator begins by
repeating the phrase softly, and continues doing so until he or she can
"hear" it inwardly after the voice stops. In either case, the mental replica
-the image of the visual object or the internalized sound of the mantra
-becomes the new concentration object, and in time yields an abstract
image as before.
Practice based on a visual object or a mantra therefore differs from
practice based on the types of Object described earlier (e. g. the breath-
ing) in having a distinct extra stage, that in which the original object is
replaced by a mental replica. However, this difference is perhaps more
apparent than real. It may well be that concentration on the breathing
394 HABS 16.2
does actually give rise to a mental replica of the original tactile sensation;
for such a mental replica would naturally be masked by the original sen-
sation, which itself continues. With a visual object, the o ~ i g i n a l sensa-
tion can be terminated at any time by shutting the eyes, which makes the
replica image clearly distinguishable from it; but one cannot simply stop
breathing at will, whence the apparent Skipping of one stage. It is the
funer sequence of stages that is presented by Buddhaghosa in his
account of the kasbJa practice.
Correlating Doctrine and Practice
Despite the overall correspondence between the above description and
Buddhaghosa's account, there are some evident differences. One that
deserves mention here has to do with the phenomenon of goose-flesh,
trembling, and other involuntary bodily movements, which meditators
commonly experience early in the practice. Present day meditation mas-
ters identify these effects as plti ,a component "factor" (anga) of jhiinas
1 and 2. The main basis for this identification is a vivid description
given by Buddhaghosa.
However, that description occurs not in his
account of the sub-stages leading to jhlina, but rather in his deSCription
of jhflna itself.
Before discussing the significance of this discrepancy, let us note the
potential usefulness of piti as a landmark for correlating the practical
sequence of meditative stages with the textual sequence ofjhlinas. All
accounts of the jh1lnas agree in stating that the jhflna factor piti is present
injhfJnas 1 and 2, but ceases with the attainment of jhana 3. If piti is
correctly identified with the goose-flesh and similar reactions, then the
ceasing of those reactions in the course of meditation should correspond
to the transition from jMina 2 to jhiina 3.
In considering such apparent correspondences, one has to be prepared
to put aside long-held notions about the nature of the jhflnas. The old
understanding ofjhana 1 as a deeply concentrated state has already been
rendered dubious, and that means that both scholars and meditators now
have to be ready to re-think the entire jhana series. In such an enterprise
intellectual flexibility is naturally essential.
33. Vism 143-144; Path 149-150. Mahasi Sayadaw, Practical Insight
Meditation (Kandy: Buddhist Publication Society, 1971) 21, gives the fol-
lowing brief description: "There arises also in him rapture [Plti], causing
'goose-flesh,' falling of tears, tremor in the limbs. It produces in him a subtle
thrill and exhilaration. He feels as if on a swing. He even wonders whether he
is just giddy."
Another potentially useful landmark for correlating meditative stages
withjhiinas is provided by the classification of the jhiinas into two cate-
gories: rapa and arapa, material and non-material. Common sense indi-
cates that this classification would appropriately be applied to the medi-
tative stages as follows: Those stages in which attention is directed to a
'physical object-the actual kasb}a disk, the breathing, a chanted mantra,
etc.-are nlpa, material; and those in which it is directed to a mental
image, or in which there is no specifiable object at all, are anlpa, non-
On this basis, the arising of the mental replica of the medita-
tion object would mark the transition fromjhiina 4 (the last nlpa-jhiina)
to jhiina 5 (the first arapa-jhana).
Here a further conflict with Buddhaghosa's account becomes appar-
ent. We have already noted that one of the earlier sub-stages listed in his
account, namely the arising of the uggaha-nimitta, clearly corresponds
to the arising of the replica image in the medi tation practice. Yet now
we have grounds for inferring that the transition fromjhiina 4 to jhiina 5
corresponds to that same meditative event. This is another problem that
will be deferred until later. For the present, the discussion will focus on
possible correspondences between the meditative series and the Nikaya
jMna series, independently of any connection with Buddhaghosa's sub-
Two points of correspondence between the meditative series and the
jhiina series have already been tentatively identified. Application of
similar reasoning elsewhere in the two series yields the following tenta-
tive pattern of correspondence.
34. Present-day writers on jhlina often translate the nlpa in rupa-jhtina as
"fine-material" (e. g. Gunaratana 108, Nyanaponika 70, 71; contrast Sole-
Leris, 57). This addition of "fine," for which there is no textual justification,
has evidently been felt necessary because of the seeming inappropriateness of
"material" (let alone "physical") to describe the very subtle state thatjhiina 1
is widely assumed to be. Similar considerations no doubt lie behind the
"explanation" (e. g., Gunaratana 92-93, following Vism 163) that the body
referred to injhiina 3 (sukhafi ca kiiyena pa.tiswnvedeti) is actually "the men-
tal body," i. e. the mind, When "body" has to be interpreted as meaning
"mind," there is clearly something seriously wrong.
396 JIABS 16.2
Comparison of Meditative Stages and Jhanas
Stage 1: The meditator's efforts at
concentrating on the assigned.
object fail to stop the flow of
thought, but do bring a pleasant
freedom from affective
Stage 2: The flow of thought
ceases, yielding a pleasant
stillness. Trembling, gooseflesh, .
etc. occur.
Stage 3: The trembling, etc. cease,
as the power of attention becomes
more balanced. Pleasant bodily
feelings of warmth etc. are
Stage 4: The pleasant bodily feel-
ings cease. Balanced attention to
the concentration object continues.
Stage 5: Physical sensation
ceases, giving way to a mental
image which is a replica of the
original concentration object.
Stage 6: There develops a derived
image, an abstract counterpart of
the preceding replica image.
Stage 7: This abstract image dis-
appears, giving way to mental
emptiness, and leaving a sense of
being suspended in an endless
black vacuum.
Jhana 1 : Vitakka and vicara are
present, along with plti and sukha,
both of which are born of sep-
aralion from sense desires and
unwholesome states.
Jhilna 2: Vitakka and vicara cease
with the attaining of ekodibhava.
Piti and sukha are now samadhi-
Jhana 3: Pili ceases, as upekkM
and sati-sampajafifia are
established. Sukha is now felt
with the body.
Jhana 4: Sukha ceases, leaving
pure upekkha and sati.
JMna 5: RJ1pa-lpa,tigha-Intinatta-
sanfia ceases. There comes the
awareness that akilsa is endless.
Jhtina 6: Endless akilsa is tran-
scended and there comes the
awareness that vififililJa is endless.
JMna 7: Endless vififiilna is tran-
scended and there comes the
awareness that nothing whatever
Stage 8: Even the sense of experi-
encing mental emptiness ceases,
as total unconsciousness super-
venes; however, the meditator is
aware of this only in retrospect.
Jhilna 8: Nothingness is tran-
scended and the realm of neither
salina nor non-sanna is attained.
The reasoning behind this proposed pattern of correspondences will
now be spelled out by considering, in order of their occurrence, those
PaIi terms whose meanings are of Significance in defining the different
Vitakka-vicara. The meaning of these paired terms is a key issue in
Stuart-Fox's analysis ofjhanas 1 and 2. Outside of the jhilna context,
vitakka and vicara together mean, as Rhys Davids and Stede note, "just
thought, thinking." 35 The evidence adduced by Stuart-Fox indicates that
this is also what they mean in the standard jhllna formula as we find it in
the Nikayas: vitakka-vicara simply denotes the normal flow of thought,
the stream of imagery and verbalizing which, like a television program
that is rarely switched off, provides a persistent though vague and unob-
trusive background to our everyday waking consciousness.
noticed under normal circumstances, the thought-stream becomes only
too obvious to the meditator when he or she tries to bring it to a halt and
keep all attention focused on the concentration object. Indeed, as practi-
tioners of concentration meditation well know, stopping the flow of
thought is one of the most difficult aspects of the practice. Success in
this task represents a major breakthrough; and the resulting state of pro-
longed freedom from thought (cittass' ekaggatll) constitutes a radically
35. Pali-English Dictionary 620, vitakka; and 615, vicara.
36. For details see Bucknell, "Experiments ... " 103-104. The verbalizing
or "inner speech" aspect of the thought-stream is stressed in the textual expla-
nation of vitakka as vacl-sankhara, "speech-activity," or the precursor of ac-
tual physical speech (M i 301). It is also recognized in the equating of jhtIna
2 with "ariyan silence" (5 ii 273). Reinterpretation of "vitakka-vicara" as
some kind of focused attention was one of the ad hoc adjustments that
became necessary once ekaggata had been attributed to jhana 1. For an
example of the inconsistencies to which this reinterpretation continues to give
rise, see Phra Khantipalo, ed., A Treasury of the Buddha's Discourses from.
the Majjhima-nikaya (Middle Collection), vol. 2 (Bangkok: Mahamakut
Rajavidyalaya Press, n. d.) 62 (translation of Dantabhami-sutta). There
vitakka is translated "thoughts" in one sentence ("Do not think thoughts ... "),
and "initial application" in the next sentence (a description of jhtIna). The
editor acknowledges the inconsistency (note 4), but claims it is unavoidable.
398 JIABS 16.2
altered state of consciousness, a most satisfying and encouraging attain-
It is, therefore, to be expected that the thought-stream, and the task of
suppressing it, should figure prominently in the textual account ofjMna
practice. 'This expectation is fulfilled once one allows that vitakka-vicara
in the jhana description has the same meaning it has in other more gen-
eral contexts in the Nikayas. These various considerations support the
identification of vitakka-vicara with the normal flow of thought; the sup-
pression of vitakka-vicara in the transition fromjhllna 1 to jhllna 2 is the
meditative achievement of bringing the flow of thought to a standstill.
Piti. The jMna deSCription indicates two different varieties of Plti:
separation-born and concentration-born (viveka-ja and samlidhi-ja).37
Accordingly, the "Conation" column of Table 1 presents the following
pre-jhllna :
jhllna 1:
jhllna 2:
jhllna 3:
sense desires and unwholesome states
separation-born piti
concentration-born pUi
equanimous mindfulness and self-possession
Concentration-born pUi, the phenomenon of trembling, gooseflesh, etc.,
is easy to identify; and indeed for an experienced meditator, particularly
one who has also done some insight meditation, the progression through
the entire series is fairly readily perceived, as follows. The practice can
begin only if the meditator is able to curb for a time the mind's habit of
reacting emotionally to the contents of consciousness, i. e. to external
sense objects and mental images. Such affective in its
variety but adequately covered by the broad opposing categories "liking"
and "disliking" -represents a pointless squandering of the energy that is
indispensable for attentive focusing, and thus for the establishing of
mental onepointedness. The beginning meditator, struggling to block the
flow of thought and keep attention fixed on the prescribed concentration
object, applies considerable mental effort, sometimes so much as to
cause sweat to stream from the body. This blOCking and fixing, once
achieved, can be maintained with a much lower level of effort; however,
inexperienced meditators usually fail to make the appropriate adjustment.
Having achieved onepointedness, they continue to put out the same high
level of effort, with the result that the excess manifests in the form of un-
37. On these two types, cf. Buddhadasa 157, 159.
controlled physical movements. With practice, meditators learn to di-
minish the intenSity of the attentive focusing, yielding a state of equilib-
rium which, because it entails no wasteful loss of energy, can be main-
tained for long periods. 38
This view of the process indicates that the relevantjhlina terms are to
be understood as follows: "Sense desires and unwholesome states" are
the varied affective reactions that characterize the pre-jMna condition,
i. e. ordinary consciousness. "Separation-bornpUi" is the high-powered
attentive focusing on the concentration object which the meditator brings
to bear by redeploying the energy normally expended in affective reac-
tion. "Concentration-born piti" is the phenomenon whose outward
maifestation is physical trembling, etc., and whose cause is the maintain-
ing of this high level of attentive focusing after it is no longer needed, i.
e. after onepointedness has been established. And "mindfulness and
self-possession" is the condition of balanced attention that is ultimately
achieved by reducing the intensity of the focusing and establishing the
appropriate equilibrium (upekkNl).
Sukha. As noted in the textual analysis, sukha is said to be present in
jhilnas 1,2, and 3, but is stated to be felt with the body only injhlina 3.
This tallies with the meditator's experience of delightful bodily feelings
following the cessation of the physical forms of pili. In addition it sug-
gests' though not unequivocally, that the sukha of jhiinas 1 and 2 is to
be understood as purely mental pleasure (i. e. somanassa). This again is
in keeping with experience: freedom from affective involvement (jhilna
1) is a pleasurable state of mind, and so too is steady mental onepointed-
ness (jhilna 2). It is doubtful, however, if a phenomenological distinc-
tion between "separation-born sukha" (jhlina 1) and "samadhi-born
sukha" (jhilna 2) can really be drawn.
Akiisa. We have already noted the appropriateness of the term iiklisa
("space") in the title of the first arupa-jhiina: space is all that remains
following cessation of the four material elements (earth, water, fire, and
air), 1. e. following the cessation of nJpa. "Realm of endless space" is
therefore appropriate as a term for the meditative state in which all input
38. If one may invoke a simile worthy of Buddhaghosa, it is like cooking a
stew. The cook at first turns the gas up high in order to bring the contents of
the pot to boiling point. If, being inexperienced, he leaves the flame higb
after that point has been reached, the pot boils over. He then learns to tum
down the flame to a level just sufficient to maintain a steady simmer. Tbe
flame in these three situations corresponds to separation-born pUi injhlina 1,
concentration-born piti injhlina 2, and sati injhlina 3.
400 JIABS 16.2
from the five physical sense organs (rupa-sanntilpa,.tigha-saniitilnanatta
-sanna) has ceased. For the meditator in tIllS state there exists only the
replica image (dharruna-sanna). Here it is well to recall that aklisa is not
emptiness or notllingness, a fact emphasized by the contrast with the
"realm of notllingness" (jhllna 7).
Given the very incomplete state of research into the actual
identities of Buddhist psychological categories, any attempt at interpret-
ing the term vifinli1}a in the jhllna context is necessarily speculative. 39
Nevertheless, some useful observations are possible, especially as re-
gards the distinction between vinnlil}a and sanna. Buddhaghosa likens
sanna to a cllild's perception of a coin (awareness of its color, shape,
texture, etc.), and vilifiol}a to an adult's perception of the same coin
(awareness of its purchasing power and usefulness). 40 This explana-
tion, if valid, indicates that vifiiilil}a is a processed, more abstract deriva-
tive of sanna. Such an understanding of the relationship between sanna
and vinnlil}a makes good sense in the case ofjhanas 5 and 6, for those
two stages can now be interpreted as follows. The awareness of the
replica image (jhllna 5) is an example of the sixth class of sanna
(dhamma-sanna) , while the awareness of the derived abstract image
(jhlina 6) is. an example of the sixth class of viniwl}a (mano-vinnal}a).41
The steady persistence of each type of image, as the only content of the
meditator's consciousness, makes good sense of the phrases "[aware]
that liklisa is endless" (jhlina 5) and "[aware] that vinnlil}a is endless"
(jhllna 6).
Akineanna. This word, meaning "notllingness," indicates a meditative
state having zero content. The description of jhllna 7 includes the state-
ment "n'atthi kine! ti, [aware] that there is nothing," which, like the
parallel" ti" clauses for jhllnas 5 and 6, implies that the meditator is con-
39. For an example of such research, see Rune E. A. Johansson, "CHta,
Mano, VififtaI).a-a Psychosemantic Investigation," University of Ceylon
Review 23 (1965): 165-215.
40. Vism 436-437; Path 480.
41. The six classes of vififia1}a are: cakkhu-vififia1}a, sota-, ghana-, jivha-,
kaya-, mano-vififia1}a. See D ii 308, S iii 61; and cf. the corresponding six
classes of safifia at note 18. In many contexts the words vififia1}a and safifia
appear to be used loosely and almost interchangeably to denote a general,
non-specific awareness or consciousness. Examples are the usage of vififia1}a
at M i 293 (cited by Johansson 196), and the seeming interchangeability of
vififia1}a, safifia, and vedana at M i 293 (Johansson 202). Nevertheless, it is
clearly appropriate to focus on the distinction between safifia and vififia1}a in
the case of jMnas 5 and 6, where the two stand contrasted.
sciolls of the condition. This is, therefore, an accurate description of the
meditative state in which, following the disappearance of the abstract
image, consciousness is empty of all content and the meditator is left
only with a sense of an endless void. 42
Neva sanna nasmlna. Buddhaghosa states that "neither sanna nor
non-sanna" implies also "neither vedana nor non-vedana," "neither citta
nor non-Gitta," and "neither phassa nor non-phassa."43 If he is right,
then the expression" neva sarlrla nasmlna," though specifying only
salina, actually covers all mental components.
Now, this expression
("neither smlna nor non-sanna") has the form of the fourth member of
the Indian tetralemma. To the question "Is there sanna?" Indian logic
allows not only for "There is" and "There is not," but also for "There
both is and is not" and "There neither is nor is not." A connection with
the meditative practice can now be made. In the eighth and final stage the
meditator becomes totally unconscious, but can know this only by infer-
ence after the event. Consequently, it can be argued, the presence of
consciousness, or of any specified mental factor, can be neither affmned
nor denied. Any question about whether there is consciousness can be
answered, strictly speaking, only with "There neither is nor is not." But
42. The question whether consciousness with zero content is possible con-
tinues to be debated within the field of mysticism studies. The case against
such a state of consciousness (variously called "contentless experience," "pure
consciousness," "unmediated consciousness," etc.) is particularly identified
with Steven Katz; the case for it has been no less persuasively put by W. T.
Stace and others. See, for example, Steven T. Katz, "Language, Epistemol-
ogy, and Mysticism," in Steven T. Katz, ed., Mysticism and Philosophical
Analysis (New York: Oxford University Press, 1978) 22-74; W. T. Stace,
Mysticism and Philosophy (London: Macmillan, 1960) (esp. 110); Philip C.
Almond, Mystical Experience and Religious Doctrine (Berlin: Mouton, 1982)
(esp. 174-175); and Robert K. C. Forman, "Mysticism, Constructivism, and
Forgetting," in Robert K. C. Fonnan, ed., The Problem of Pure Conscious-
ness: Mysticism and Philosophy (New York: Oxford University Press, 1990)
3-49. Also see Rode11ck S. Bucknell, "Buddhist Jhtlna as Mystical Experi-
ence," in G. K. Zollschan, J. F. Schumaker, and G. F. Walsh, eds., Explor-
ing the Paranormal (Bridport: Prism Press, 1989) 131-149, where (fore-
shadowing the conclusions reached in the present paper) I identify jhana 7 as
contentless experience, and jhana 8 as what one might call con tentless non-
43. Vism 337; Path 367.
44. The four expressions effectively cover all four mental khandhas: vedana,
sanna, saiLkhtlra (citta), and vinniiJ:za (phassa). In any case, we have the fact
(see note 41) that sanna is sometimes used in a very loose sense to refer to
any consciousness. Also cf. Nyanaponika 164: "Sanna stands sometimes for
consciousness in its entirety, e. g., in neva sanna-nasann' ayatana .... "
402 JIABS 16.2
to non-Indian minds this is philosophical hair-splitting; by generally ac-
cepted standards oflogicality and phenomenological accuracy, the final
meditative stage would be quite correctly described as a state oftotal un-
consCiousness. It is therefore noteworthy that there does exist (in the
po.tJhapada-sutta, belonging to the earliest stratum of the Nikayas) a
single variant version of the account of the eight jhlina s in which the
eighth stage is described straight-forwardly in terms of cessation of
saiina (sanna nirujjhantl). 45
The above discussion has shown that the series of eight jhlinas des-
cribed at numerous places in the Nikayas, correlates well with the series
of eight stages experienced by practitioners of concentration meditation.
One can hardly escape the conclusion that the eightjhlinas are the eight
meditative stages.
This conclusion has serious implications for Buddhaghosa's series of
sub-stages. That series is said to precede attainment of each jhlina; but,
as already noted, some of the sub-stages appear to be identical with cer-
45. D i 184-5. According to this sutta, the monk who has attained the realm
of nothingness recognizes that he is at tile peak of sannli, but that to be with-
out sannli would be a still higher attainment. He therefore practices further
until he "touches cessation" (nirodha phusati). This phrase provides a link
with a common variant of tile jhana description, according to which jhlina 8
is followed by a yet higher attainment wherein the meditator "touches cessa-
tion" (e. g. M i 455-456). As described in the texts, this ninth attainment;
"cessation of perception and feeling" (sannli-vedayita-nirodha) or "attainment
of cessation" (nirodha-samlipatti), tallies well with the state of total uncon-
sciousness already identified with jhtina 8. For several good reasons, includ-
ing its frequent anomalous association Witil "destruction of the lisavas" (e. g.
M iii 28), this ninth attainment is under suspicion of being a later addition to
what was already a complete list of the stages in concentration meditation-
see Paul Griffiths, On Being Mindless: Buddhist Meditation and the Mind-
Body Problem (La Salle, Ill.: Open court, 1986) 16-31; also Bronkhorst 77-
78; and King 17. The evidence, particularly the existence of the PO!.thaplida
version, suggests that the description of jhlina 8 and the description of
nirodha-samlipatti, though usually made to follow each other in accounts of
the jhlinas, were in origin two alternative descriptions of one and the same
meditative attainment. (The Chinese counterpart of the Pilii PO!.thaplida-sutta
[TaishO vol. 1, lID b 12-16] does recognize a discrete ninth stage, its descrip-
tion being identical in wording WiUl the above-mentioned descriptions of
jhlina 8 followed by nirodha-samlipatti. This discrepancy between the PaIi
and Chinese versions of the Po.t.thapadais most readily explained on the
premise that the unique Pali version preserves the "original," since the
Chinese version can then be attributed to editing designed to yield confonnity
with the stereotype.)
tainofthe jhllnas. For example, Buddhaghosa's sub-stage characterized
by the pa.tibhaga-nimitta clearly corresponds to the meditative stage in
which an abstract image becomes established; and that meditative stage
has been shown to correspond also to jhllna 6. The first three of
Buddhaghosa's sub-stages can be fairly positively equated with jhllnas
in this way, which points to the pattern of correspondences shown in
Table 5.
Thus, Buddhaghosa's series of sub-stages duplicates the
series ofjhilnas. What Buddhaghosa portrays as steps on the way to tre
frrstjhllna (and to each subsequentjhilna) are in fact steps on the way to
the last jhllna .
It is now evident thatBuddhaghosa's account is not, as generally sup-
posed, merely a more detailed and precise formulation of the account
found throughout the Nikayas. Rather, it is a fundamentally different
version which is in serious conflict with the Nikaya account. By
Buddhaghosa's day the jhana doctrine had been drastically modified.
The flISt and crucial modification, already introduced, it seems, by the
earliest A.bhidhammikas, consisted in equating the final stage of the
meditative sequence (i. e. the state of total unconsciousness) with attain-
ment of the first jhana rather than the last (jhana 8). Once this new
equation had been set up, two further things became necessary: (1) a set
of terms for the meditative stages passed through on the way to this new
"first jhllna"; and (2) a description of a series of further meditative prac -
tices whereby the remaining jhanas could (allegedly) be attained.
Accordingly, the new set of terms, uggaha-nimitta, etc., was created and
brought into association with a practice consisting in systematic refiee-
46. The correspondence shown in Table 5 is less secure for upacara- and
appana-samadhi than it is for the three nimittas. It is based in part on the
sequence of sub-stages as described in the texts, and that sequence is not
entirely clear. The Vimuttimagga (79) states: "And if the (after-)image
[pa.tibhaga-nimitta] appears in his mind, he gains access-meditation
. [upaclira-samadhi]. And if access-meditation appears in his mind, he, by
means of this, accomplishes fixed meditation [appana-samadhi]." This indi-
cates the sequence: pa.tibhaga-nimitta, upacara-samadhi, appana-samadhi.
The Visuddhimagga appears to indicate the same sequence, but with some
overlap:" ... he should besides extend the counterpart sign [pa.tibhliga-
nimitta] ... for it is possible to extend it on reaching access [upacara-
samadhi] and on reaching absorption [appana-samadhi]" (Vism 152).
However, at another point (Vism 126) the Visuddhimagga refers to " ... the
counterpart sign, which arises together with access concentration [upacara-
samadhi] ... ," suggesting that the pa.tibhliga-nimitta arises simultaneously
with upacara-samiidhi rather than before it. The resulting slight uncertainty
is acknowledged by the query marks in Table 5.
404 HABS 16.2
tion on the need to eliminate the next jhana factor, or (in the case of the
arupa-jhiinds) to move on to the next, more subtle object.
These developments must have been fairly directly linked with the de-
velopments discussed by Stuart-Fox, whereby ekaggata was attributed
to jhiina 1 ,and vitakka-vicara was reinterpreted as some kind of atten-
tive focusing. Only on the basis of such a revised description of jhiina 1
would it have been plausible, and therefore possible, to identify that
jhiina with a deeply concentrated meditative state. Indeed, it may well be
that the seemingly minor step of attributing ekaggatii to jhiina 1 was
what initiated the entire process.
That such modification of the jhana doctrine could corne about may
seem to raise doubts about the meditative credentials of those responsi-
ble for it; it suggests that the authors of the Vimuttimagga and Visuddhi-
magga had little practical acquaintance with meditation. However, this
does not necessarily follow, because it is only the interpretation of the
jhiina doctrine that is at fault in Buddhaghosa's account; the description
of the practice (as far as the first attainment of appana-samadhi) is gen-
erally satisfactory. Indeed, the fact that a new set of names for the medi-
tative stages was developed, centuries after the correspondences with the
original set ofjhiinas had been lost sight of, indicates rather that the tra-
dition ofjhiina practice had survived intact down to Buddhaghosa's day,
and that he at least knew about the stages it entailed.
That the original correspondences between jhana practice and jhiina
doctrine were lost sight of in the first place is in keeping with the now
widely acknowledged development of an early split, within the Sangha,
between meditator-monks and scholar-monks.
The Abhidhamma-like
statements about the jhiinas contained in the Smigiti, Dasuttara, and
other late suttas, are consistent with this split having begun to develop
not long after the founder's death.
Already in the early days of the
Sangha meditators and Dhamma-expounders were going their separate
ways; a serious communication gap was developing.
47. On the split between the scholar-monks and the jhayins or meditators, cf.
A iii 355. Also see Sukumar Dutt, The Buddha and Five After-Centuries
(London: Luzac,1957) 99,116-117; Louis de La Vallee Poussin, "MusIl a et
Narada," Melanges chinois et bouddhiques 5 (1937): 210-222; and Rod
Bucknell and Martin Stuart-Fox, "Did the Buddha Impart an Esoteric
Teaching?" Journal of Indian History 61.1-3 (1983) 14-15.
48. See D iii 219, D iii 274, where vitakka and victira are said to be lost
successively; also cf. M i 294, M iii 25-29, S iv 263, where ekaggata is said
to be present in the firstjhilna.
One negative consequence of Buddhaghosa's complex account of
jhiina was that mastery of the higher jhanas was made to seem a super-
human attainment. With the entire series multiplied by itself, as it were,
the total number of stages was greatly increased; and no genuine instruc-
tions were available for the attainment of any jhana beyond the sup-
posed first one. This effect continues to the present day. To most Bud-
dhist meditators, even "the second jhtina" seems hardly a realistic goal,
while "the anlpa-jhanas" appear impossibly remote. The present re-
vised understanding of the jMnas should, therefore, give encouragement
to practicing meditators. The path of concentration practice is not nearly
as long and arduous as Buddhaghosa made it seem.
jhiina 2:
jhana 3:
(initial and sustained
avitakka, aviciira
(absence of thought)
sampasiidana (tranquility)
cetaso ekodibhiiva
(oneness of mind)
(sense desires)
(pleasure, pain)
akusaladhammii somanassa, domanassa
(unwholesome states) (happiness, sorrow)
pid sukha
(mindfulness & self-
upekkhii, sad
(bodily pleasure)
(absence of pleasure)
. Qq

408 JIABS 16.2
Table 2. Sunimary ofjhlillaS 1 to 4.
jhiina 1 vitakka-vicara . piti sukha
jhiina 2 ekodibhilva piti sukha
J. J.
jhlina 3 ekodibhilva upekkhli sukha
J. J.
jhlina 4 ekodibhllva upekkhli parisuddhi
Table 3. Simplified summary of jhlinas 1 to 4.
jhlina 1 vitakka-vicara piti sukha
jhlina 2 pUi sukha
J. J.
jhlina 3 sukha
J. J.
jhlina 4
Table 4. Simplified summary ofjhlinas 1 to 5.
jhlina 1 vitakka-vicara piti sukha rupasaiiiill
jhlina 2 piti sukha rupasaiiiill
J. J.
jhlina 3 sukha r,upasaiiiia
J. J.
jhlina 4 rupasaiifill
J. J.
jhlina 5
Table 5. Equivalences between sub-stages and jhlinas
parikamma-nimitta 1-4. rupa-jhllnas
(preliminary sign)
uggaha-nimitta 5. akasanaficayatana
(acquired sign) (endless space)
pa.ribhllga-nimitta 6. vififililJaficayatana
(counterpart sign) (endless consciousness)
? upacara-samadhi 7. akificafifiliyatana
(access concentration) (notllingness)
? appana-samlidhi 8. neva safifia nasafifiayatana
(fixed concentration) (neither perception nor non-perception)
The International Assocation of Buddhist Studies
David Seyfort Ruegg
Alexander Macdonald
General Secretary
Lewis R. Lancaster
Regional Secretaries: Leslie S. Kawamura (Americas),
Akira Yuyama (Asia), Alexander Macdonald (Europe)
Members of the Board: Heinz Bechert, Collett Cox,
Richard Gard, Yuichi Kajiyama, Per Kvaerne,
Lokesh Chandra, Donald S. Lopez, A. K. Narain,
Jikido Takasaki, Paul Williams, Hsiang-chou Yo
The International Association of Buddhist Studies, founded in 1976, is
devoted to promoting and supporting scholarship in Buddhist Studies
in all its aspects, past and present, around the world. Membership is
open to scholars of all academic disciplines.
Membership dues are: $35 or 5500 for full members, $20 or 4000
for student members, $1000 or 150,000 for life members. Prospec-
tive members from developing countries may contact the General
Secretary concerning subsidized membership rates. Dues are payable
per calendar year by December 31 of the previous year. Payments may
be sent in US dollars to Professor Lewis Lancaster, Treasurer lABS,
Department of East Asian Languages, 104 Durant Hall, University of
California Berkeley CA 94720 USA or in Japanese Yen to the lABS
account supervised by Dr. Akira Yuyama, Regional Secretary for Asia:
No. 041-4502280, Kokusai Bukkyo Gakkai, Toranomon Branch,
Mitsubishi Bank, Tokyo, JAPAN.
The Eleventh Conference of the lABS
will be held in Mexico City
October 24-28, 1994
For information, please contact:
Professor F10ra Botton
Center for Asian and African Studies, EI Colegio de Mexico
Camino al Ajusco No. 20
Codiga Postal OJOOO
Mexico, D. F., MEXICO