Japanese Journal of Religious Studies 1994 21/4
True Words, Silence, and the Adamantine Dance
On Japanese Mikkyõ and
the Formation of the Shingon Discourse
Fabio RAMBELLI
This paper deals with Japanese esoteric Buddhism (Mikkyõ), in particular
the Shingon tradition, as it relates to the emergence of new and peculiar
epistemological concerns. Through a discussion of the kenmitsu system
outlined by Kuroda Toshio, the paper ³rst situates Mikkyõ within the religious
and institutional framework of medieval Japan, underlining its liminal
and heterological nature as both an institutionalized discourse and a
reservoir of oppositional possibilities. The paper then analyzes the formation
of Shingon orthodoxy as an attempt to systematize the Tantric ³eld in
Japan through a re-organization of preexisting religious doctrines and
practices. Special attention is given to the actual articulation of the kenmitsu
episteme and its orders of signi³cance. Finally, the paper outlines
some fundamental epistemological tenets of Mikkyõ discourse. Though it
focuses on Shingon discourse and orthodoxy, this paper confronts basic
epistemic assumptions and discursive practices common to the multifarious
forms of esoteric Buddhism in Japan.
THE PURPOSE OF THIS PAPER is to describe the discourse of Japanese
esoteric Buddhism (particularly the Shingon Oí tradition) as it
developed in conjunction with the emergence of a distinctive form of
philosophical reµection on signs and the formation of a corpus of
* An earlier draft of this essay, entitled “Kenmitsu Episteme and Mikkyõ Heterology: On
the Semiotic Doctrines and Practices in Medieval Japan,” was presented at the 45th Annual
Meeting of the Association for Asian Studies, Los Angeles, March 1993. I would like to
express my gratitude to Mimi Hall Yiengpruksawan, chair of the panel at which it was presented,
and the other participants, especially Allan Grapard and Neil McMullin, for their
comments and criticism. I wish also to thank Bernard Faure and Yamaguchi Masao. I am
greatly indebted to the editors of, and an anonymous reader for, the Japanese Journal of
Religious Studies, for insightful and valuable suggestions in the process of revising this article.
practices relating to the production of meaning.1
My basic hypothesis
is that esoteric Buddhism (Jpn. mikkyõ Oî, secret teachings, hidden
doctrines) can be understood as a discursive formation that presupposes
a particular cosmology, attitude towards reality, and episteme
(“the attitude that a socio-cultural community adopts in relation to its
own signs”; GREIMAS and COURTÉS 1979, p. 129). It can be seen, in
other words, as an ensemble of knowledge and practices concerned
with the interpretation of reality as well as the production, selection,
conservation, and transmission of knowledge. These things, in turn,
are implemented through interpretive strategies, repertoires of
metaphors, and a general structuring of knowledge. Like every discourse,
that of esoteric Buddhism determines (and is determined by)
distinctive institutions, ideologies, rituals, and relations of power.2
The Mikkyõ semiotic paradigm was extremely inµuential in Japan
for centuries and still operates today on a certain cultural level
(although in a marginalized and nonorganic fashion). An understanding
of this paradigm is thus essential for the study not only of
medieval Japanese religiosity and culture but also of the esoteric ceremonies,
magic rituals, and traditional divination still performed in
contemporary Japan.3
The reconstruction of medieval Mikkyõ4
discourse and its underlying
episteme should, ideally, begin with a consideration of the TantricDaoist
syncretism that occurred mainly, but not exclusively, within the
Chinese Zhenyan Oí lineage during the Tang and Song dynasties,
and then trace its development and transformation in Japan. I
con³ne myself, however, to the early and medieval Japanese Shingon
tradition, not only to set reasonable boundaries to this study but also
to answer in part the urgent need for a cultural history of the Shingon
sect. The lack of such a history has been a major hindrance to the
study of Japanese religiosity in its various manifestations and has left
many questions unresolved, particularly those concerned with the
ways in which Shingon knowledge and practices were codi³ed, trans-
1 According to Charles Sanders Peirce’s de³nition, a sign is “something which stands to
somebody for something in some respect or capacity”—in other words, anything that can be
charged with meaning and interpreted.
2 I agree with James BOON, who considers semiotics “less an integral theory than a
clearinghouse of issues in the complexity of communicational processes” (1982, p. 116). I
see semiotics as an open ³eld of problematics, a network of approaches and theories that
can shed light on basic issues of signi³cation and discursive formations.
3 On these subjects, and on the role of Mikkyõ ideas and practices in contemporary
Japanese magic and religious ritual, see KOMATSU 1988.
4 By the expression “medieval Mikkyõ,” I mean the totality of the forms taken by esoteric
Buddhism from the Insei Š© age at the end of the Heian period (late eleventh–twelfth centuries)
to at least the Nanbokuchõ age (early fourteenth century).
374 Japanese Journal of Religious Studies 21/4
mitted, and diffused, and with the modalities of interaction of the various
esoteric lineages in Japan. Because of this the Shingon tradition
in most major studies on premodern Japanese culture has been obliterated,
or, at best, reduced to a mysti³ed Kðkai W} (Kõbõ Daishi
eÀØ‚; 774–835).5
I use the term “Shingon tradition” for want of a better translation
of the term “Shingon-shð” Oí;. In its medieval usage “Shingon-shð”
indicated a loosely connected network of temples and lineages (ryðha
H$) that shared a myth of Kðkai as founder and a common set of initiatory
knowledge and practices. This complex was de³ned in relation
to other similar “sectarian” denominations, particularly those included
in the Eight Schools system (hasshð k;) and its expanded versions.6
In medieval Japan, the term shð ; referred essentially to a textual corpus
associated with a transmission/foundation lineage in the Three
Lands (India, China, Japan). Such corpora/lineages implied orthodoxy
and legitimacy because they were of³cially recognized by the
emperor and because they were traditionally associated with certain
temples and sacred places (see GYÕNEN). Each shð was thus an
inµuential cultural reality as part of the doctrinal, political, ideological,
and geographical system of the Eight Schools, and at the same
time an “abstract” ideological foundation legitimating the various
locale-speci³c lineages.7
Though I will focus on the creation of Shingon discourse and
orthodoxy, I believe that the basic epistemic assumptions, discursive
practices, and rhetorical strategies discussed here reµect traits common
to all the multifarious forms assumed by esoteric Buddhism in
Japan. By viewing Mikkyõ as a discourse I will try to bring into relief
an important, though often ignored, feature of Japanese medieval culture,
and also counter the ideological mysti³cations of traditional sectarian
scholarship with its stress on speci³c lineages and the ³gures
(myths) of their founders. I hope thereby to avoid con³ning Mikkyõ
to the reassuring boundaries of our received knowledge.
5 The founder of the Japanese Shingon sect.
6 The Eight Schools (Kusha Hà, Jõjitsu ¨×, Ritsu A, Sanron XÇ, Hossõ Ào, Kegon
Tä, Tendai ú×, and Shingon) were the Buddhist scholastic traditions of³cially “imported”
from China and acknowledged by the Japanese imperial system. Such traditions as Zen 7,
Jõdo-shð þF;, Jõdo Shinshð þFO;, and Nichiren-shð Õ¥; were added in the Middle
Ages. The system of the Eight Schools (and its extended versions) constituted the framework
within which each sectarian denomination acquired its status and legitimacy.
7 Properly speaking, Shingon has never had a uni³ed center, and a Shingon “sect” does
not exist even today. Temples af³liated with the Shingon sectarian denomination belong to
either the Kogi ò– Shingon-shð or the Shingi G– Shingon-shð, both of which are further
articulated in many sub-branches.
RAMBELLI: True Words, Silence, and the Adamantine Dance 375
Tantric Heterology and Its Japanese Avatar: The Kenmitsu System
Tantrism, from its very beginnings on the Indian subcontinent, has
constituted a complex heterology, an often successful attempt to confer
centrality to a heterogeneous ensemble of elements that were culturally
marginal and were as such excluded from institutionalized discourses.
This heterology in large part accounts for the dif³culty of
identifying a common substratum to Tantrism’s multifarious historical
and cultural manifestations.
Tantrism was in origin the heterology of what Michel de CERTEAU
calls an “untiring murmur” at the background of Buddhist cultures, a
“consumption” and displacement of “high” culture products and discourses
by marginalized individuals and social groups (1990, p. 53).
James BOON writes, “‘Tantrism’ is a nineteenth-century European
coinage based on an ‘exotic’ term. The ‘ism’ part makes shifting ³elds
of oppositions, differentiations, and plural relations sound substantive,
doctrinaire, and uniform” (1990, p. 159). Tantrism can be characterized
as a complex magico-ritual apparatus that systematically
reverses the renouncement ideals proper to religious institutions,
especially Buddhism (DUMONT 1979, pp. 342–43), although it does
not necessarily conceive of itself as an opposition ideology. As will
become clear later, this characteristic is shared, to some extent, by
Japanese avatars of Tantrism. Ritual based on a principle of reversal
seems, then, to be a fundamental trait of Tantrism. In fact, as BOON
suggests, “Tantrism” is merely “a name for a polymorphous reservoir
of ritual possibilities, continuously µirted with by orthodoxies yet also
the basis of countering them”; it de³nes a ³eld of possibilities against
which “more orthodox positions and transformations become shaped
and motivated” (1990, p. 165).8
Japanese Mikkyõ provides an interesting case of “Tantric heterology.”
As Boon notes with respect to Tantrism in general, the very term
“Mikkyõ” presents Japanese esoteric Buddhism as an apparently uniform
cultural entity. Actually, it covers three quite different aspects of
Japanese Buddhism, among which it is important to distinguish.9
The
³rst aspect is the Tantric substratum as a “reservoir of ritual possibili-
8 Interestingly, BOON sees “a Western parallel” of Tantrism in “that range of hermetic
heterodoxies, a murmur of Gnostic, Neoplatonist, crypto-liturgical positions: from freemasons
to Bohemians, from counterculture to poètes maudits” (1990, p. 165).
9 Although the Tantric ³eld in Japan still needs to be surveyed and charted, I think it
constitutes a continuum ranging from clearly “Tantric” positions to formations that could
be de³ned as “tantroid,” such as the marginal Pure Land movements known as Ichinengi
sç– (sometimes related to the radical Tachikawa-ryð CëH) or the Jishð ´L groups
often associated with Kõya-san ¢Ÿ[ and Shingon institutions.
376 Japanese Journal of Religious Studies 21/4
ties,” a disseminated and nonsystematic cultural entity, a matrix of
anti-institutional potentialities; this is an aspect often downplayed or
ignored by traditional scholarship.10
The second aspect is Tantrism as
“µirted with by orthodoxies,” that is, as a systematic and organized tradition
indissolubly related to non-Tantric forms of Buddhism (kengyõ
ßî, exoteric teachings); this is the most common understanding of
Mikkyõ, since scholars usually stress the systematic aspects of Japanese
Tantrism. Mikkyõ in this second sense is organized into lineages and
possesses textual corpora and ritual practices; it is a vast phenomenon
encompassing various sectarian divisions. The third, and most limited,
aspect is Mikkyõ as the Shingon tradition, conceived of as the purest
form of esoteric Buddhism.11
Tantric Buddhism in its second aspect interacted with other Japanese
Buddhist movements, religious traditions, and philosophical systems
to create a new organism, de³ned by KURODA Toshio (1975) as
an “exoteric-esoteric system” (kenmitsu taisei ßO¿£) with its own ideology
(kenmitsushugi ßOü–, exo-esotericism). Kuroda’s concepts—
formulated to describe the complex Buddhist institutional system in
medieval Japan—have opened the way to understanding Japanese
Buddhism as a global cultural system possessing multiple interrelations
with other religious and cultural systems. His concepts have undergone
various adjustments, but on the whole they are useful tools for
portraying what is an ideological, political, and economic organism.
Kuroda and such followers as Satõ Hiroo, Sasaki Kaoru, and Taira
Masayuki are concerned primarily with the social, institutional, and
ideological aspects of the medieval kenmitsu system,12
while I am concerned
here more with its epistemic aspects. In particular, I see
Mikkyõ discourse as an important part of what I call the “kenmitsu episteme,”
by which I mean the basic epistemic features of Kuroda’s “exoesoteric”
system and ideology.
Kuroda distinguishes three phases in the formation of the kenmitsu
system:
1. Mikkyõ (in the ³rst sense discussed above) uni³ed all religious
movements on an original “magic” background;
10 This aspect of Japanese Mikkyõ has been highlighted, although with different degrees
of explicitness, by KOMATSU and NAITÕ 1985, AMINO 1986, MURAYAMA 1987 and 1990,
KOMATSU 1988, and NAKAZAWA 1988.
11 Most studies on Mikkyõ deal only with Shingon, while most studies on Tendai consider
only its non-Tantric aspects. Tantric elements in other traditions have never been studied in
depth.
12 For a critical appreciation of kenmitsu taisei, see SASAKI 1988, pp. 29–52.
RAMBELLI: True Words, Silence, and the Adamantine Dance 377
2. the Eight Schools established their own doctrines, esoteric practices,
and kenmitsu theories13
on this new esotericized basis;
3. the respective schools, thus organized, were recognized by secular
society as legitimate Buddhism and formed a type of religious
establishment with a strong social impact—a situation that
occurred only in Japan.
KURODA stresses the fact that what underlies the entire kenmitsu system
is not a particular sect, but Mikkyõ in general as a common substratum
of ideas and practices concerned with the ultimate meaning
of reality and the supreme goals of Buddhist cultivation (1975, p.
537). The main characteristic of Japanese Mikkyõ is its capacity to permeate
and unify all religious traditions and to organize the magical
beliefs of the people (pp. 432, 436). It differs from Indian Tantrism in
the importance it assigns to rituals and prayers (kitõ te) for worldly
bene³ts and the protection of the state (p. 433), a difference based
on deeper cultural motivations.14
The kenmitsu system was not just a
religious logic and ideology, but was so closely connected to Japanese
political authority that it acquired the status of an of³cial ideology and
gradually esotericized the state apparatuses (p. 434).15
It constituted
the hegemonic system of thought and practice in medieval Japan (pp.
445–46) and was the reigning orthodoxy and orthopraxy. Shinto was
³tted into this framework as a local and concrete manifestation of
Mikkyõ (p. 537).
It should be noted that Kuroda sees the ensemble of Tendai concepts
and practices known as hongaku hõmon û·À– or hongaku shisõ
û·„` as “the model of kenmitsu ideology” and the Tendai school as
“the representative entity of the kenmitsu system” (1975, p. 445).16
Although Kuroda mentions the central role of Kakuban’s ·Î
Shingon thought in shaping the system (KURODA 1975, p. 475), he
fails to analyze this role and thereby neglects the role of Shingon and
13 Theories delineating the relationship between Tantric and non-Tantric Buddhism.
14 It is possibile to discern in this feature a reversal of the traditional Buddhist outlook,
that is, an awareness that mundane and political activities aimed at establishing a Buddhist
kingdom and constructing a Buddha-land can be closely related to salvation.
15 The present study deals with the question of orthodoxy in relation to the formation
of Shingon discourse; thus the approach taken here differs from that of Kuroda.
16 This, Kuroda argues, is due to the fact that the Tendai tradition (especially the
Sanmon [– lineages) occupied a hegemonic position during the Japanese Middle Ages.
Sasaki Kaoru, on the other hand, indicates that, while Tendai institutions were at the center
of the kenmitsu system in western Japan, the religious system established by the Kamakura
bakufu was essentially based on Zen and Mikkyõ, having its roots in the Rinzai rò Zen, Tõji
X±, and Onjõ-ji Óô± lineages, as well as in Onmyõdõ ‹î‰. SASAKI calls this alternative
system the zenmitsu taisei 7O¿£ (1988, pp. 94–148).
378 Japanese Journal of Religious Studies 21/4
other important esoteric lineages. Kuroda’s treatment leaves it unclear
whether he envisioned a single, Tendai-centered kenmitsu system
shared by all other schools or whether he intended only to present
another inµuential paradigm of a manifold reality.
I am inclined to believe the latter. I see the kenmitsu system, in the
general terms it has been described above, not as the whole institutional
and ideological apparatus of Japanese medieval Buddhism but
as something akin to a “generative scheme” of multiple cultural interventions,
an open framework that the various Buddhist schools and
traditions could actualize on their own terms. In fact, all the Eight (or
Ten) Schools offered the same range of “products” and “services”:
simple formulae for salvation and rebirth, easy practices, relations
with local “Shinto” cults, esoteric doctrines and practices, political ideologies,
services for the protection of the state and the ruling lineages
(chingo kokka ¥D³B), and so forth. These were then personalized
through speci³c doctrines and practices. In this respect, the schools
formed a sort of trust controlling the religious market, and Mikkyõ
was their common religious, epistemic, and ideological substratum.
There are other points in Kuroda’s treatment of kenmitsu requiring
further development. For instance, Kuroda does not mention the fact
that the very notion of kenmitsu resulted from an act, both conceptual
and practical, of articulation and restructuring that affected the entire
Japanese religious and philosophical world. Nor does he deal in depth
with the heterological nature of Tantrism or with the complex process
of creating a Mikkyõ discourse—a necessary requisite for establishing
the kenmitsu system and its distinctive internal logic. Mikkyõ’s evolution
is reduced to the thought of Kðkai and later Tendai developments,
and the esotericization of other schools is presented as an
inevitable outcome.
As we will see in more detail later, “Kengyõ” was constructed simultaneously
with “Mikkyõ” as the Shingon exegetes dissimulated, rearticulated,
displaced, and rewrote preexisting doctrines and practices.
No place was recognized in this process for the ritual rivals of Kðkai’s
Mikkyõ: Onmyõdõ ‹î‰ and the preceding or competing forms of
esoteric Buddhism (zõmitsu PO, taimitsu ×O).17
The ideology of
kenmitsu was introduced by Kðkai in his Ben kenmitsu nikyõ ron as a
means of de³ning the polar relation between the Shingon esoteric system
and preexisting teachings, which he considered super³cial and
provisional. In this respect Kðkai reversed traditional hermeneutical
17 The interaction of Mikkyõ and Onmyõdõ doctrines and practices in Japan has been
described in MURAYAMA (1981, especially pp. 197–241; see also 1987, 1990), HAYAMI 1975,
KOMATSU 1988, and KOMATSU and NAITÕ 1985.
RAMBELLI: True Words, Silence, and the Adamantine Dance 379
criteria,18
turning what was “evident” (ken ß, teachings that are clear
and self-evident without problems of interpretation) into something
“super³cial,” and what was “hidden” or “not immediately evident”
(mitsu O, teachings related to a certain intention of the Buddha and
therefore apparently unclear and requiring interpretation) into something
“profound and true.”
Kðkai’s understanding of the term kenmitsu came to be widely
accepted, and after the late Heian period was commonly used to designate
the whole Buddhist system (although Kðkai’s redistribution of
doctrines and practices was rooted in the old idea of the existence of a
secret transmission of the true teachings and rituals of the Buddha—
an East Asian counterpart of the European hermetic mysteries). In
this manner, Kðkai opened the way for a de³nition of the Mikkyõ discursive
³eld as comprising that which the other doctrines do not teach,
that which the other schools ignore and leave unsaid. The silence of
the Buddha marked the boundaries of Shingon intervention.
Mikkyõ played another important role, functioning as a relay in the
circuit between center and margin. This made the kenmitsu system an
important instrument of power. By controlling and integrating negative
forces that threatened the cultural center from “outside” (KOMATSU
and NAITÕ 1985) and by providing central institutions with an ef³cacious
cosmology and a distinctive epistemic ³eld, Mikkyõ paradoxically
became the dominant paradigm of Japanese medieval culture.19
Systematic Mikkyõ, itself a product of a semantic reversal, succeeded
in reformulating on its own terms and from its own perspective—that
of systematic reversal—the main concepts and practices of Japanese
culture.20
Moreover, monks belonging to esoteric lineages were closely
related to the imperial court and the ruling lineages, so that the
Tendai and Shingon schools exerted a true hegemony (a hegemony
that was economic as well).21
18 See Kðkai, Ben-kenmitsu nikyõ-ron (translated in HAKEDA 1972, pp. 156–57). On the
main criteria of Buddhist hermeneutics, see LOPEZ 1988.
19 YAMAGUCHI Masao (1989) has presented an illuminating interpretation of the ambiguous
and “marginal” nature of the Japanese emperor. This could explain, at least in part, the
political importance of Mikkyõ.
20 In the systematic esotericization of Japan and its culture that was carried out during
the Middle Ages, geographic space was conceived of as a ma«^ala, the Japanese language
was identi³ed with the absolute language of the shingon-darani Oí¼øÍ, and literary production
was assimilated to sacred texts dealing with esoteric truth (this process will be the
subject of a later study). An esoteric dimension was attributed also to death (see KAKUBAN,
Ichigo taiyõ himitsu shð) and birth (see DAIRYÐ; I am grateful to James Sanford for having
brought to my attention this fascinating text).
21 Cases such as that of Kakuban, closely connected to the retired emperor Toba š–,
and Monkan, in the entourage of Emperor Go-Daigo 9ÚE, are well known. Earlier, during
380 Japanese Journal of Religious Studies 21/4
It is, I believe, safe to assume that the real kenmitsu matrix of the
Shingon school emerged during the late Heian period with the
appearance of a new literary genre: the treatises on the distinction
between ken and mitsu by such great scholars and religious ³gures as
Saisen èE (1025–1115), Jitsuhan ×– (?–1144), and Kakuban ·Î
(1095–1143). Generally ignored by scholars today, these men were
directly responsible for the creation of medieval Mikkyõ. Contemporary
events—such as the creation of the cult of Kõbõ Daishi or the emergence
of Kõyasan as an object of popular faith connected with the
quest for immortality and rebirth in paradise—were closely related,
on the one hand, to the cultural mood of the time (the idea of mappõ
=À and the search for methods to counter it), and, on the other, to
the need of religious institutions to gain new sources of income and
wider social support. In this respect, it is interesting that the collection
and study of Kðkai’s works, as well as the attempt to adapt Mikkyõ to
new religious needs and trends, began after the creation of new forms
of cult and religious “consumption.”
Still, Mikkyõ heterology never lost its formidable function of opposition,
precisely because of its special contact with the “outside” and
with “otherness,” and because of its direct links with marginal, heterodoxical,
and ambiguous cultural products (sacred mountain cults,
popular religious practices, and social organizations of marginality).22
Among the expressions of Mikkyõ were the hijiri ¸, marginal religious
³gures that gravitated around central political and religious institutions
and possessed the power to subvert them.23
The number of hijiri
and monks of low status using their esoteric training to get close to
political power was large, and included such ³gures as Gyõki ‘_
(668–749), Genbõ é5 (8th c.), and Dõkyõ ‰ù (d. 770) in the Nara
period, Kðkai and Kakuban in the Heian period, and Monkan k?
(1278–1357) and many of the monks around Emperor Godaigo in the
Nanbokuchõ era. A later example was Tenkai ú} (1536–1643), the
architect of the political and religious cosmology of the Tokugawa
government. An example of a “Tantric” attempt to organize social
the Nara period, esoteric monks such as Genbõ and Dõkyõ were closely associated with
those in political power. On a more orthodox and of³cial level, the Shingon hierarchy has
been close to the emperor since 834, when a Shingon chapel, the Shingon-in OíŠ, was
established inside the Kyoto imperial palace precincts. It is also to be recalled that the development
of Mikkyõ, ³rst in the early Heian period (ninth century) and later in the Insei age
(late eleventh–twelfth centuries), was closely related to more general restructurings of the
Japanese political, social, and economic order.
22 On the cultural role of marginality and its relationship with the center, and on the
principle of exclusion in Japanese culture, see YAMAGUCHI 1975.
23 On the hijiri, see GORAI 1975; SATÕ 1987; SASAKI 1988.
RAMBELLI: True Words, Silence, and the Adamantine Dance 381
marginality was the Shingon Ritsu tradition of Eison µ¨ (1201–1290)
and Ninshõ ݧ (1217–1303)(ÕISHI 1987).
Mikkyõ never became a uni³ed opposition force, but was a reservoir
of nonorganized and asystematic oppositional possibilities. Its history
is a series of attempts to keep an almost impossible balance between
center and periphery, between institutionalized discourses and practices
and their heterological counterparts. A conµictual relation
between center and margin existed throughout the whole of premodern
Japanese history, contributing to the µourishing of the esoteric
tradition. Nevertheless, people apparently did not realize the
questionable compromises such a stance entailed, with perhaps the
only signi³cant exception being the Hossõ monk Tokuitsu ”s at the
beginning of the Heian period.
Tokuitsu’s Criticism of Mikkyõ
That Tokuitsu (fl. ca. 820) was aware of the heterological nature of
Kðkai’s new Mikkyõ is evident from his Shingonshð miketsu-mon, a short
treatise in which he listed his doubts and criticisms concerning
Shingon doctrines and practices (T #2458, 77.862–865). A seemingly
harmless work, it in fact reveals the total incompatibility of Mikkyõ
with the doctrines of the Six Nara Schools (TSUDA 1985). As noted by
TAKAHASHI Tomio, Tokuitsu’s criticism was directed less at the Shingon
school than at Mikkyõ as a distinct new tradition (1990, 181–82). His
criticism encompassed Tendai forms of Mikkyõ as well, so that Tendai
monks were among those who responded to him.
The tenor of the debate was unusual. While disputes among
schools in East Asia were usually over the provisional or ultimate
nature of teachings or lineages, Tokuitsu argued from a Mah„y„na
perspective that Mikkyõ, as explained by Kðkai, was utterly untenable.
His criticism was directed particularly against the features of Kðkai’s
thought connected with the formation of an orthodox esoteric discourse
separate from the Nara Buddhist establishment, features such
as the authenticity of the esoteric lineage, the salvi³c value of its practices,
the idea of sokushin jõbutsu “X¨[ (becoming Buddha in this
very body), and the unconditioned nature of the Sanskrit language.
Since Kðkai saw the salvi³c power of his teachings as lying in the
absolute nature of esoteric words,24
Tokuitsu’s observations threatened
his Shingon system at its very basis: if mantras are not expressions
of an unconditioned language, then the truth they convey is
24 See Kðkai, Shõji jissõgi, Bonji shittan jimo narabini shakugi; see also RAMBELLI 1992 and
1994.
382 Japanese Journal of Religious Studies 21/4
conditioned and the rapid attainment of siddhi (supernatural powers)
is consequently impossible. This would amount to the dissolution of
Mikkyõ. Tokuitsu’s doubts are thus clues to the fundamental alterity
of the esoteric system, and to the impossibility of understanding it on
the basis of Mah„y„na principles.25
Because of Tokuitsu’s perhaps unexpected attack, Kðkai realized
that inµuential ³gures in Nara Buddhism saw the teachings of his new
school as µawed, yet nevertheless as potentially threatening. In order
to confer preeminence upon the Shingon doctrines, therefore, Kðkai
had to ³nd new hermeneutical criteria. He also was at least partly
aware of the fundamental heterogeneity of Mikkyõ, and accordingly
stressed its systematic coherence with Mah„y„na texts. Although Kðkai
never explicitly answered Tokuitsu’s criticisms,26
all of his work can be
understood as an indirect reply (for a different interpretation, see
TSUDA 1985).
Only by raising Shingon Mikkyõ above its marginal and asystematic
background could Kðkai and his successors confer on the Shingon
school a dominant role within the Japanese religious establishment.
In order to bring this about it was necessary, ³rst, to create a new discourse
and orthodoxy that partially concealed Tantrism’s heterogeneity
and underlined its continuity with the dominant forms of state
Buddhism; and, second, to devalue most preceding Tantric forms and
write a new classi³cation of Japanese Buddhist schools. A very dif³cult
agenda, undoubtedly. But Kðkai’s efforts, especially in consolidating
the kenmitsu categorization, constituted an impressive attempt to create
a new tradition. The endeavor required time to bear fruit, and several
centuries passed before convincing replies to Tokuitsu’s objections
were formulated: ³rst it was necessary to build up a solid alternative
point of view grounded in a systematic discourse. Of course, the
debate did not concern only theoretical matters and doctrinal prestige;
25 More recently, TSUDA (1978) has expressed doubts that the two fundamental texts of
the Shingon tradition, the Dari jing (Jpn. Dainichi-kyõ) and the Jinggang ding jing (Jpn.
Kongõchõ-kyõ), can be integrated into a single and noncontradictory system. According to
Tsuda, these two texts epitomize two cosmologies and soteriologies (those of Mah„y„na and
those of Tantric Buddhism) that exist in a “critical” relation to each other, i.e., that are completely
different and incompatible. TSUDA, interestingly, refers to Tokuitsu’s criticism (1985,
pp. 89–91). It should be stressed, however, that Mikkyõ, far from being reducible to the Dari
jing and the Jinggang ding jing, comprises a complex intertext of commentaries on and
explanations of both sðtras, plus numerous other texts that lack direct relations to them.
On a still deeper level, one can recognize a diffuse set of non-systematic knowledge and
ritual actions, many of which are not clearly supported by textual authorities.
26 He directly tackled only Tokuitsu’s eleventh doubt, concerning the Iron Stðpa where
N„g„rjuna, the human patriarch of Mikkyõ, was initiated by Vajrasattva into the esoteric
teachings (KÐKAI, Himitsu mandarakyõ fuhõden).
RAMBELLI: True Words, Silence, and the Adamantine Dance 383
what was really at stake was ideological supremacy and power.
Tokuitsu’s criticisms were not pursued by other members of the
contemporary Buddhist establishment, and Tokuitsu was silenced
even by his own Hossõ colleagues and successors. The Nara establishment
soon realized the ideological and ritual importance of the new
Mikkyõ as an instrument of political and economic control, and
adopted it in a sort of surreptitious paradigm shift. Esoteric Buddhism
became in this way an essential feature of premodern Japanese culture.
It is not by accident, therefore, that Tokuitsu has been canceled
from the of³cial history of Japanese Buddhism, and that most of his
works are no longer extant. Forced to play the role of the loser in the
debates on the kenmitsu matrix, he became a kind of scapegoat of the
kenmitsu system.
Purity and Heterogeneity: The Formation of Mikkyõ Discourse
In his criticism of Mikkyõ, Tokuitsu ignored the important fact that
Nara and early Heian Buddhism already contained numerous esoteric
(Tantric) elements, mainly relating to the ritual and meditative apparatus.
Among these elements were those directed toward the political
center (e. g., rites for the protection of the state) and those expressive
of cultural and political marginality (e.g., individual practices to gain
various siddhi) (see KUSHIDA 1964, 1–54; HAYAMI 1975; MURAYAMA 1987,
1990). We see here a different con³guration of the traits that characterize
Indian Tantrism (DUMONT 1979). The ritual apparatus of Nara
Buddhism, with regard to both central state rites and marginal individual
practices, was Tantric in that it reversed Buddhist ideals of
renunciation by stressing material bene³ts and protection of the state
(symbolized by the imperial lineage).27
Later Shingon scholars stress the “miscellaneous,” “unsystematic,”
and “fragmentary” nature of Nara Mikkyõ, which they label zõmitsu
PO, in contrast to the pure, systematic, and mature esoteric teachings—junmitsu
„O—that were supposedly introduced to Japan by
Kðkai. Although the distinction between zõmitsu and junmitsu is often
taken for granted, its basic criteria are neither clear nor objective, and
it is thus quite problematic as a description of actual doctrinal and ritual
differences.28
MISAKI (1988) has demonstrated the existence of
27 The efforts of esoteric monks like Genbõ toward establishing the Kokubun-ji ³_±
system of state-run provincial temples indicate the importance of Mikkyõ in the formation
of Nara State Buddhism (see HAYAMI 1975, pp. 4–5).
28 Even the origin of the terms zõmitsu and junmitsu is obscure, and presumably quite
late; according to MISAKI (1988, pp. 146–47), the ³rst person to use the words was Ekõ ŠM
384 Japanese Journal of Religious Studies 21/4
multiple esoteric trends in Tang China, and of numerous attempts to
construct orthodoxies. These efforts were continued in Japan by
Shingon and Tendai monks. The junmitsu/zõmitsu distinction was the
product of just such an effort, one that rewrote Mikkyõ’s history to
magnify Kðkai’s lineage, downplay Tantric practices and rites prior to
Amoghavajra,29
and belittle subsequent developments in rival lineages.
These efforts, animated by a certain “volonté d’orthodoxie” (a term
used by Bernard Faure), were in large part successful, though the
translation and production of so-called zõmitsu texts did not cease
(MISAKI 1988, pp. 146–47). Tantric multiplicity also continued to µourish
in marginal cults like Tachikawa-ryð CëH, local traditions like
Shugendõ @à‰, and even “orthodox” Mikkyõ as institutionalized lineages
proliferated and sometimes integrated heterodoxical practices.
The Mikkyõ daijiten de³nes junmitsu as a synonym for ryõbu XH
mikkyõ, a form of Mikkyõ that combines the doctrines and practices of
the Womb (taizõ ̉) system and the Diamond or Vajra (kongõ D¤)
system. Junmitsu is believed to be the direct expression of the enlightenment
of Dainichi Nyorai ØÕØû (Mah„vairocana), the personi³cation
of the Dharmak„ya (MD, p. 1108). Zõmitsu is synonymous with
zõbu PH mikkyõ, that is, everything in Mikkyõ that cannot be reduced
to junmitsu. It comprises conditioned doctrines and rituals propounded
by Dainichi Nyorai’s three communicational and transformational
bodies (the tajuyðshin ¬1äX, hengeshin ˆ5X, and tõrushin fHX),
and as such is explicitly inferior to ryõbu. This is a major difference
with respect to Tendai Mikkyõ (taimitsu ×O), according to which the
zõbu is the very space where the nondualism of the Womb and the
Diamond systems is realized.
The term zõbu was ³rst used by Kðkai in the Shingonshð shogaku kyõritsu-ron
mokuroku, his catalogue of esoteric texts (also known as the
Sangakuroku) compiled in 823. This work, perhaps the ³rst systematic
attempt to classify Mikkyõ texts (MISAKI 1988, p. 150), utilizes the
three traditional categories of sðtras (daikyõ Ø™), precepts (ritsu A),
and treatises (ron Ç). The Shingon sðtras are then classi³ed as
Diamond-lineage, Womb-lineage, or miscellaneous (zõbu). Problems
with criteria and modalities appeared even in this early classi³cation,30
(1666–1734). On the mysti³cations in the traditional sectarian treatment of the junmitsuzõmitsu
distinctions, see ORZECH 1989 (especially pp. 88–92), and MISAKI 1988.
29 Bukong #W (705–774), a Tang „c„rya with direct lineal contacts with Kðkai.
30 For instance, a sðtra such as the Suxidi-jieluo-jing MÒGŠø™ is included among the
precepts and regulations; the Dari jing does not conform to the classi³catory criteria, containing
as it does many explicit references to genze riyaku ê›2Ê; texts that are not sðtras
(i.e., that do not contain doctrinal elements and concern genze riyaku) are included in the
Diamond textual lineage (MISAKI 1988, pp. 150–52). In general, the Shingon school appears
RAMBELLI: True Words, Silence, and the Adamantine Dance 385
however, and later attempts were not much more successful.31
The criteria
tended to be arbitrary and overly inµuenced by the desire to support
the claims to orthodoxy of the compiler’s own lineage. It is not
surprising then that the ryõbu/zõbu distinction is related within
Shingon to the more general kenmitsu articulation.
It is nevertheless possible to trace a distinction between Nara
Mikkyõ and later Mikkyõ. In the latter one ³nds an attempt to develop
a systematic discourse, different from and sometimes antithetical to
“normal” Buddhist discourse. Although very few differences can be
detected between junmitsu and zõmitsu with regard to cosmology and
soteriology, Heian Mikkyõ presents a more systematic aspect, and
devotes a large amount of attention to semiotic and discursive problems
(usually connected, again, with its need to establish its own
orthodoxy). It may be that such a discursive self-awareness was also
present in late Nara Mikkyõ, an interesting point requiring further
research. But, though of interest for the history of Japanese culture
and the establishment of the esoteric orthodoxy, this possibility does
not affect the characteristics of the full-µedged Mikkyõ discourse.
Esoteric elements in pre-Heian Japan were assembled into a literary
and ritual genre, a loose corpus called the darani-zõ ¼ø͉, one of
the ³ve sections of the Buddhist Canon in the prajñ„-p„ramit„ tradition
(Dasheng liqu liuboluomituo jing, T. 8.868b; see also Kðkai’s treatment of
the subject in the Ben-kenmitsu nikyõ-ron). The esoteric formulae, variously
called darani, ju 2, and mitsugo OB, are discussed in many
Mah„y„na texts (UJIKE 1984; MISAKI 1988, pp. 18–25). The wide diversity
of approaches and interpretations shows that dharanic expressions
made up a heterogeneous ³eld not organically integrated within
H‡nay„na and Mah„y„na traditions.
According to UJIKE Kakushõ (1984), who describes in detail the
development of dharanic thought in China and Japan, spells designed
to facilitate the understanding and usage of Mah„y„na doctrines
developed into instruments of power, and later became a kind of
microcosm that offered the chance to “become a Buddha in this very
body” (sokushin jõbutsu “X¨[). UJIKE points out that, after the age of
the great Tang „c„ryas, increasing attention to linguistic problems
together with a new vision of salvation caused the transformation of
the darani-zõ into the Shingon vehicle (1984; see also RAMBELLI 1992,
to have followed Amoghavajra’s method of including into the Diamond lineage authoritative
(and useful) texts and rituals of miscellaneous origin.
31 See, for instance, MISAKI’s analysis of the classi³cation proposed by Gõhõ #µ
(1306–1362), the great scholar monk of Tõ-ji X± (1988, pp. 157–58), and Yðkai’s »r
(1345–1416) attempt as described in MD (s.v. Ryõbu zõbu: 2284).
386 Japanese Journal of Religious Studies 21/4
pp. 189–93). And just as mysticism separated from theology in Europe
in a process studied by CERTEAU (1982), so the dharanic ideas and
practices of the darani-zõ detached themselves from the Mah„y„na corpus
to form an independent discourse. This movement “is related to a
sharper consciousness of a speci³c and original language. The word
that referred to an experience developed to designate a language”
(Michel de Certeau, quoted in the introduction to the Italian translation
of CERTEAU 1982 [Bologna: Il Mulino, 1987]).
Tantrism was also concerned with the operations performed on the
terms it invested with meaning. It thus possessed pragmatic and metalinguistic
signi³cance: it speci³ed both how to use and how to interpret
its expressions. It speci³ed, in other words, how to practice language.
These linguistic and semiotic practices, when they became complex
and explicit enough, established a ³eld of their own: junmitsu Mikkyõ.
Mikkyõ proposed a unitary and organic vision of esoteric linguistic
phenomena, thus performing a restructuring of Buddhist discursivity.
Denomination marked the will to unify all the operations until then
dispersed, to organize, select, and regulate them. A new discipline was
born from this attempt to systematize discursive practices (see also
CERTEAU 1982).
In this process, undoubtedly connected to more general cultural
factors, junmitsu emerged as (Shingon) Mikkyõ orthodoxy; thus
“pure” Mikkyõ was the result of a mysti³ed idea—an ideology—of
orthodoxy, purity, and uncontamination. The very concept of a
Shingon “school,” with its overtones of unity and group identity, conceals
the manifold moves made over the centuries to exploit new and
different possibilities of representation. Bernard FAURE has deconstructed
traditional views of lineage and orthodoxy through a critique
of their arborescent model: “Orthodoxy takes its shape not from its
kernel—a lineage—but from its margins, the other trends against
which it reacts by rejecting or encompassing them” (1987, p. 54).
Shingon Mikkyõ, too, developed in rhizome-like fashion as the result
of “an amnesia, an active forgetting of origins” (1991, p. 14), and of
complex interactions with so-called zõmitsu and taimitsu intervention.
This being the case, what is the role of the founder, Kðkai, in this
rhizomatic process? As FAURE explains, “Individuals…are not the
source of tradition, but rather its products, its nodal points, its textual
paradigms or points of reemergence” (1987, p. 54). Contrary to traditional
myths, Kðkai is to be considered the emergence of peculiar
discursive strategies in relation to already extant ideologies, discourses,
and literary genres. His achievement can be seen to lie in his successful
attempt to bring esoteric trends into the proximity of the political,
RAMBELLI: True Words, Silence, and the Adamantine Dance 387
institutional, and cultural center through his construction of a new
Mikkyõ orthodoxy.
A “Space of Interplay”: The Kenmitsu Matrix and Its Surrounding Silence
Let us now turn to the processes whereby orthodox Mikkyõ discourse
was generated. As CERTEAU points out, “The right to exercise language
otherwise is objecti³ed in a set of circumscriptions and procedures”
(1986, p. 83). First, “a spatializing operation which results in the
determination or displacement of the boundaries delimiting cultural
³elds” (pp. 67–68) is necessary. Next, “the spatial divisions which
underlie and organize a culture” will be reworked (p. 68).
As explained above, the ³rst step in the formation of Mikkyõ discourse
(“determination or displacement of the boundaries delimiting
cultural ³elds”) involved the problematic and arti³cial articulation of
the Tantric ³eld into junmitsu and zõmitsu through the constitution of
a new orthodoxy grounded in the myth of a direct transmission of an
original ostension.32
Sources report that Dõji ‰² (?–744), the Nara
monk credited with introducing the Kokðzõ gumonji-hõ ÐW‰¼l³À
to Japan, studied in the Tang capital Changan under the „c„rya Šubhakarasi½ha
(Shanwuwei 3[a, 637–735). In order to counter this and
assert his own claim to orthodoxy, Kðkai had to invent a new, more
powerful, and more appealing lineage, the one that connected him to
Amoghavajra. Thus much of the Shingon textual production is pervaded
by an insistence on the contrast between the old teachings (miscellaneous
and impure and therefore ineffective), and Kðkai’s new
teachings (systematic and pure and therefore extremely effective).
This is not a mere rhetorical topos, but part of the ideological operation
that helped establish Shingon sectarian orthodoxy by declassing earlier
tendencies as zõmitsu and silencing rival lineages like taimitsu.
Although of³cially relegated to the periphery of the Shingon system,
zõmitsu and, to a certain extent, taimitsu were de facto retained as
an essential part of Shingon Mikkyõ. The general ken-mitsu distinction
operated as a “generative scheme,” according to which the fundamental
oppositions common to the whole Mah„y„na tradition could be
32 The ³rst link in the chain of the secret transmission of Mikkyõ doctrines and practices
is Dainichi Nyorai. In order to stress that these teachings were born in the self-presence
of the Dharmak„ya and are themselves unconditioned, a myth of an original ostension was
created in which the esoteric sðtras and ma«^alas appeared in the sky to N„g„rjuna, who
faithfully copied them and handed them down to later disciples. The myth of the manifestation
in the sky, perhaps of Daoist origin, expressed the idea that the esoteric transmission
transcended the arbitrariness of signs, conditioned cultural codes, and ordinary semiotic
strategies. See also RAMBELLI 1991, pp. 20–21.
388 Japanese Journal of Religious Studies 21/4
displaced, relocated, and reinterpreted. Relevant questions included
the “sudden/gradual” soteriological polarity, the Twofold Truth paradigm,
the conditioned/unconditioned nature of the Buddha’s
preaching, and the semantic levels of language (jisõ °o/jigi °–).
Michel de Certeau’s second phase, the more general cultural reorganization,
corresponds to the Tantric restructuring of the whole religious
situation in Japan, an operation—perhaps already completed in
Tang China—that culminated in Kðkai’s articulation of the ten levels
of the kenmitsu system in the Himitsu mandara jðjðshinron.33
Kðkai
“reversed” the classi³cations of the Three Teachings (sangõ Xî) and
traditional Chinese Buddhist panjiao |î hermeneutics, which
ignored esoteric teachings, by placing his new “orthodox” Mikkyõ at
the top—and, at the same time, in the background—of the whole system,
thus strategically situating formerly marginal practices at the center
of the Buddhist establishment.34
Although engaged in articulating
their own system, Shingon commentators stressed the continuity of
their own teachings with those that preceded them: important
authors like Kðkai, Kakuban, and Raiyu þî (1226–1304) untiringly
repeated that the difference between Mikkyõ and Kengyõ lies not in
their ultimate truth, which is identical, but rather in their approach to
it, which is utterly different.
Basically, Kðkai’s doctrinal and ritual system contained few innovative
elements. The Chinese Tiantai ú× and Huayan Tä schools
already recognized the possibility of becoming a Buddha in the present
life, and Tantric elements already existed in most schools. It is
possible to argue that Kðkai’s success was the result of his ability to
provide the emperor and the imperial system with a new ideology and
a new imagery, rooted in a grandiose cosmology and explicitated in
powerful rituals (such Tantric imperial imagery and ritual were very
fashionable at that time in the sinicized world). The truly new
characteristic of junmitsu—the one that ³rmly grounded it—was its
conviction that it was the only true discourse by virtue of its esoteric
ordering of things.
As CERTEAU has explained, the process of articulating and establish-
33 Various Mikkyõ texts (like the Lueshu jinggangding yuqie fenbie shengwei xiuzheng famen)
developed their own hermeneutics, thus confronting the Buddhist establishment. In any
event, Mah„y„na texts already dealt with the ken-mitsu distinction, although in a different
way (see, for instance, the Jie shenmi jing). A major source of Kðkai’s thought on the matter
was the Shi moheyan lun (Jpn. Shaku makaen ron).
34 Such a hermeneutical reversal is most evident in KAKUBAN’s Gorin kuji myõ himitsu
shaku, where all Buddhist schools and all religious traditions are explicitly envisioned as
steps on the path toward the attainment of esoteric goals. In this manner, all salvational
endeavors became parts of a Mikkyõ soteric framework. On panjiao hermeneutics, see LOPEZ
1988.
RAMBELLI: True Words, Silence, and the Adamantine Dance 389
ing a new discourse requires a “space of interplay,”
one that establishes the text’s difference, makes possible its
operations and gives it “credibility” in the eyes of its readers,
by distinguishing it both from the conditions within which it
arose (the context) and from its object (the content).
(1986, p. 68)
Such a “space of interplay,” a kind of meta-discursive level, is to be
found in the kenmitsu generative scheme, where, as explained above,
Buddhism was rearticulated in order to establish the place of Mikkyõ
in the religious discourse. Shingon orthodoxy (junmitsu) lived
between two vast silences, between two kinds of unsaid: it emerged
from an “ideological silence” where its zõmitsu origins were actively
forgotten and its Tantric rivals silenced, and it set its discursive space
on a background of “epistemological silence,” in the sublime realm
that the other traditions considered beyond the reach of language
and thought. Mikkyõ deals with what the other doctrines do not
teach, with what the other schools cannot fathom and are silent
about: the realm of the supreme enlightenment of the Buddha.35
Thus
silence is an important element in the construction of the discourse
of True Words. Mitsu represents a further reversal of perspective: it
deals not with the itinerary of sentient beings toward Buddhahood,
but with discourse from the absolute point of view of the unconditioned
Dharmak„ya.
Kenmitsu Doctrine
Let us now look at the basic doctrinal framework of the kenmitsu
matrix, based on a small corpus of representative texts on the sub-
ject.36
I hope that this short and synchronic account of the core of
Mikkyõ teachings will provide a useful starting point for further
inquiry, despite its neglect of subtle doctrinal distinctions, sectarian
controversies, and important historical developments.
35 Kðkai, Ben-kenmitsu nikyõ-ron, KDZ 1, 482; RAIYU, Shoshð kyõri dõi shaku, DNBZ 29:
5a–b. According to the Dari jing, the essence of the Shingon teachings is to be found where
“the way of language is interrupted and mental activity also vanishes. It is a realm comprehensible
only in the communication between buddhas” (T #848, 18.9b).
36 The texts are, respectively: KÐKAI, Ben-kenmitsu nikyõ-ron; KAKUBAN, Kenmitsu fudõ ju
and Gorin kuji myõ himitsu-shaku; and RAIHÕ, Shingon myõmoku. Each author stresses different
aspects of the kenmitsu paradigm, in accordance with the main trends of debate in his time.
Kðkai is especially concerned with the uniqueness of Mikkyõ in relation to the other
schools, Kakuban underlines the absolute character of the esoteric teachings and shows how
they transcend the idea of mappõ, and Raihõ emphasizes the essentially enlightened nature
of all things.
390 Japanese Journal of Religious Studies 21/4
As explained above, Mikkyõ divides the teachings of the Tath„gata
into two general kinds: super³cial and secret. Super³cial teachings are
the provisional doctrines taught by Š„kyamuni, or, more generally, by
the lower, conditioned manifestations of the Buddha: the Nirm„«ak„ya
and Sa½bhogak„ya. The meaning of these teachings is clear and
easy to comprehend. Secret teachings are “the most profound doctrines
beyond the faculties of sentient beings, dealing with the ultimate
secrets of all Buddhas’ enlightenment” (RAIHÕ, 734c–35a). As an
unconditioned discourse spoken by the Dharmak„ya to itself for the
pure pleasure of the Dharma,37
these teachings are permanent and
immutable and transcend the doctrine of the Decline of the Law
(mappõ =À).38
They are composed of “real words” (shinjitsugo O×B)
free from all communicational, pragmatic, and contextual con-
straints.39
In this way, esoteric teachings elude the logic of up„ya and
are not restrained by their listeners’ expectations and limitations, a
major shortcoming of Mah„y„na from the Mikkyõ point of view.40
Ken and mitsu show also different attitudes towards principle (ri 7)
and phenomena (ji ª).41
This is particularly important for the present
discussion, because these two ontological categories possess a deep
semiotic relevance. According to the Mah„y„na, ri can be seen as the
ideal type of a sign, while ji de³nes its tokens, actual and manifold
occurrences. Ken distinguishes between ri and ji, thereby establishing
two levels: Dharma-essence (hosshõ À§) versus its multifarious
dharmic aspects. Ken thus fails to attain true nondual knowledge.
Mikkyõ, in contrast, states that both ri and ji are absolute and unconditioned:
every single dharma, with all its particularities, is marked by
the “aspect of true reality.” According to the esoteric tradition, the
37 This is the well-known principle of hosshin seppõ ÀXßÀ (the Dharmak„ya’s preaching),
one of the products of Kðkai’s systematizing genius. It is a perfect model of absolute
communication characterized by total circularity. For a semiotic analysis, see RAMBELLI 1994.
38 The concept of mappõ, though not referred to in Kðkai’s texts, became of major
importance in Japanese culture after the eleventh century. Kakuban stressed the negation of
mappõ as one of the characteristics of Mikkyõ, emphasizing its unconditioned nature and
soteric power.
39 This idea probably resulted from the identi³cation of the linguistic thought of the Shi
moheyan lun (605b) with dharanic conceptions and practices.
40 Nara schools were particularly sensitive on this point. The Six Schools taught that the
differences between Š„kyamuni and Mah„vairocana are dissolved in the meta-level of
absolute reality (although Shingon Mikkyõ proposed itself as that very meta-level). They also
recognized that Mikkyõ, as a part of Buddhism, is an offspring of Š„kyamuni’s enlightenment,
the esoteric teachings being the secret doctrines taught by Š„kyamuni upon entering
Mah„vairocana’s sam„dhi. For a direct account of the Nara approach to Mikkyõ, see GYÕNEN.
41 The different conceptions of ri and ji are the main theme of RAIYU’s Shoshð kyõri dõi
shaku, a contrastive analysis of Shingon and the Mah„y„na schools.
RAMBELLI: True Words, Silence, and the Adamantine Dance 391
Dharmak„ya’s modalities of existence (shiju hosshin v)ÀX), its activities
(sanmitsu XO), and its wisdom (gochi 2J) are not different from
the elements of ordinary human cognition (sense organs, objects,
mind apparatus). As a consequence, the esoteric absolute principle
(ri), or tathat„, is in a nondual relation to phenomena (ji), being articulated
in substance (taidai ¿Ø), signs (sõdai oØ), and dynamic manifestations
(yðdai äØ). It does not transcend human intellective faculties,
and the world of enlightenment—the ultimate result of religious
practice (kabun F_)—can be described and explained in the
absolute language of the Dharmak„ya.42
Individual phenomena do not differ from the supreme principle;
an individual entity is no longer a mere token (ji) of a type (ri), but is
itself an absolute, a microcosm. There is ultimately no distinction
between the mind of each ascetic, the global mind of sentient beings,
and the Buddha. Salvation is thus close and easy to attain: the person
who performs Mikkyõ rituals after proper initiation is able to accomplish
the sublime practice of sanmitsu in his or her “body generated by
father and mother and become Buddha instantaneously.” Although
ma«^alas and dh„ra«‡s are not suited to those of low abilities, their
powers and virtues are unfathomable, and even the most super³cial
practice produces bene³ts and blessings. The esoteric cosmos is an
immense salvi³c machine, in which everything is absolute.
As Tokuitsu realized, at least in part, Mikkyõ’s differences with the
rest of Buddhism relate to the nature, structure, and power of signs.
While the Mah„y„na schools describe the Dharmak„ya—the absolute,
the kernel of Buddhist ontology and soteriology—as devoid of signs
and forms, Mikkyõ describes it as the totality of all possible signs. The
Dharmak„ya is thus able to “speak” and explain to all beings its own
enlightenment—an absolute language exists that is able to convey in
some way the ultimate reality (RAMBELLI 1994). The essential identity
of sentient beings (shujõ L´) and Buddhas is the ground for symbolic
practices that lead to the reproduction within the practitioner of the
characteristics and particularities of the absolute.
42 See for instance KAKUBAN’s Kenmitsu fudõ ju, in particular the following verses: “Ken
teachings explain the initial stage [of practice leading to Buddhahood (inbun ƒ_)], mitsu
teachings explain the ³nal stage [of attainment of Buddhahood ¨[ (kabun F_)]”: “Ken
principle (ri) has no relationship with the sense organs [rokkon ÂÍ], mitsu sees them as the
Four [Buddha-]bodies [shishin vX]; ken principle has no relationship with objects [rokkyõ
Âæ], mitsu sees them as the Three Adamantine Mysteries [san(mitsu) kon(gõ) X(O)D(¤)];
ken principle has no relationship to mind apparatus [rokushiki ÂÆ], mitsu knows them to be
the universal wisdom of the Dharmak„ya”; “Ken principle has neither signs [sõ o] nor activities
[yð ä], mitsu Tathat„ [(shin)nyo (O)Ø] is endowed with substance-signs-dynamic manifestations
[sandai XØ].” On the sandai doctrine, see RAMBELLI 1991, pp. 4–5.
392 Japanese Journal of Religious Studies 21/4
Semiosophia, Semiognosis, Semiopietas: Mikkyõ Orders of Signi³cance
It is now necessary to outline the internal structure of the kenmitsu
episteme. An account of the actual articulation of the kenmitsu epistemic
³eld should take into account the following considerations:
1 the diachronic transformation of Buddhist semiotics;
2 the complex epistemic relations within Buddhism as both a
“high” culture and a “popular” phenomenon;43
3 the presence of other inµuential models of semiotics and semiosis
(Confucian, Daoist, and later, “Western”) that coexisted and
interacted in various ways with and within the kenmitsu epistemic
³eld.
On a super³cial level, the most evident feature of Mikkyõ texts
(both Shingon and Tendai) is their phonetic and graphic exoticism,
in which the foreign is considered closer to the Origin. This is
reµected in the large number of Sanskrit terms and in the wide usage
of siddha½ (Jpn. shittan Ò·) characters. It could be said that the core
of Mikkyõ texts is formed by shingon/shittan, and that everything else
exists only to create a context so that they might be correctly prac-
ticed.44
This reµects an idea of language and signs typical of Tantrism.
As we have seen, ancient zõmitsu texts were a heterogeneous part of
the Mah„y„na paradigm: their language was an up„ya to convey
meaning or induce certain actions. In the Mah„y„na philosophy of
language, linguistic expression has value only insofar as it is able to
convey its contents, to which it has an arbitrary connection. As
Etienne LAMOTTE puts it, “The letter indicates the spirit just as a
³ngertip indicates an object, but since the spirit [that is, the meaning]
is alien to syllables…the letter is unable to express it in full” (1988, p.
15). With the formation of a Tantric discourse in East Asia, basic linguistic
conceptions changed. Language was transformed from an
up„ya into an absolute and unconditioned entity, something that
could not be translated without losing its essential character. Kðkai
believed that the Indian phonemes and script were endowed with a
unique nature. He wrote:
Mantras, however, are mysterious, and each word is profound
in meaning. When they are transliterated into Chinese, the
43 These have traditionally been the objects of inquiry of two different disciplines: the
history of ideas, and anthropology. For a critical presentation of some theoretical positions
concerning the meaning of “popular” religion in East Asia, see FAURE 1991, pp. 79–95.
44 On the importance of re-creating the original context of mantras, see LOPEZ 1990, pp.
369–72.
RAMBELLI: True Words, Silence, and the Adamantine Dance 393
original meanings are modi³ed and the long and short vowels
confused.
(KÐKAI, Shõrai mokuroku, translated in HAKEDA 1972, p. 144)
Correct interpretation and use depend upon correct transmission.
Kðkai mentions that Amoghavajra, aware of the limits of translation,
initiated his disciples using Indian words only (KÐKAI, Bonji shittan jimo
narabini shakugi, T 84.361). He thereby lent epistemic relevance to the
esoteric concept of an unaltered transmission based upon an original
ostension (a necessary part of founding an orthodoxy).
Mikkyõ semiotics is what governs the expression of that which transcends
ordinary language (cf., RAMBELLI 1992). It is possible to recognize
within Mikkyõ three different modes of semiotic knowledge and
interpretive practice of reality: semiosophia, semiognosis, and semio-
pietas.45
Semiosophia refers to exoteric forms of the knowledge of signs (sõ
o), according to which language and signs are considered to be arbitrary
and illusory, but nevertheless usable as up„ya to indicate the
truth. I use this term instead of semiotics in order to distinguish it from
both semiotics as common sense and semiotics as metalanguage.46
Various ken types of semiotics can be classed as semiosophia, including
Kusha, Hossõ, Sanron, Tendai, and Kegon. Although there seem to be
basically three epistemological models (Abhidharma, M„dhyamika,
and Yog„c„ra), each school developed its own concept of the sign in
relation to its view of ultimate reality and its hermeneutical strategies.
In the kenmitsu paradigm, mitsu semiotics presupposes ken semiotics;47
semiosophia thus constitutes the super³cial level (senryakushaku
òFö) on which the esoteric interpretive structure (jinpishaku L¸ö)
is built.
Semiognosis denotes esoteric semiotic doctrines and practices as
45 I am indebted here to Allan Grapard’s threefold categorization of the orders of
signi³cance in Japanese representations of sacred space (geosophia, geognosis, and geopiety)
(forthcoming).
46 It is very dif³cult to evaluate the role of common sense in ideas and practices relating
to signs in the esoteric episteme, especially in light of the almost total lack of research on
this subject. Buddhist setsuwa ßÊ collections, for instance, suggest that signs are clues to a
hidden reality and at the same time instruments for action: they not only foretell and
express events but also give rise to them (see RAMBELLI 1990). It is not clear, however,
whether these texts reµected widespread popular ideas on signs and semiosis or were vehicles
for the diffusion of a new, Buddhist-continental semiotic mentality.
47 According to Kakuban, without the super³cial interpretation of signs (jisõ °o), the
deeper truth (jigi °–) cannot be conveyed, but the esoteric truth cannot be taught to
people lacking the status or the capacity to receive it—this is why it is called “secret” (himitsu
¸O).
394 Japanese Journal of Religious Studies 21/4
something akin to a type of soteriological knowledge (i.e.,
leading to salvation) that is gained through speci³c practices
of a predominantly ritual and/or mystical character…. [B]oth
[semio]sophia and [semio]gnosis are connected with systems
of symbolic representation, but their epistemological frameworks
and intentionality differ.
(GRAPARD forthcoming; the original uses geosophia and
geognosis instead of semiosophia and semiognosis)
Semiognosis refers to speci³c knowledge and practices that are
“claimed to have been extracted from [signs themselves], to correspond
in mysterious ways to sacred scriptures and to divine rule, and
to lead either to mystical achievement or to religious salvation”
(GRAPARD forthcoming). The initiatory knowledge concerning structure,
function, and power of the esoteric symbols (especially mantric
expressions) is considered the kernel of enlightenment and the key to
“becoming Buddha in this very body.”48
In consequence, one of the fundamental activities of the Mikkyõ
exegete is “remotivating” language and signs, that is, overcoming the
arbitrariness of language and signs by ³nding a special “natural” relation
between expression, meaning, and referential object. Remotivation
is accomplished by reorganizing each expression’s semantic
structure and thereby making the expression “identical” to its meaning.
In this process an esoteric symbol becomes a kind of replica of its
object, and the practice in which it occurs is deemed identical to its
goal. Mikkyõ salvi³c practices consist mainly in visualization and
manipulation of mantric expressions (shingon-darani) and other complex
symbols of various kinds, whose very structure, organized on
three deeper levels (jinpi L¸, hichð no jinpi ¸_îL¸, hihichð no jinpi
¸/_îL¸), appears to the initiated person as the inscription of the
path both to salvation and to the attainment of siddhi.49
Related to semiognosis is honji suijaku ûGs), an expression of
the realm of meaning of Shinto and Buddhism that is itself a result
and a displacement of the kenmitsu epistemic ³eld. The combinatory
48 Kðkai equates the monji k° (expressive symbols, signs) of the “Dharmak„ya’s preaching”
(hosshin seppõ) with the three mysteries (sanmitsu) pervading the Dharmadhatu; thus
language and signs (sõdai) cannot exist separately from the cosmic substratum (taidai) of
original enlightenment. Kðkai then adds: “Therefore Dainichi Nyorai, by expounding the
meaning of [the relations between] language and reality, arouses sentient beings from their
long slumber.” Mikkyõ semiotics thus has a direct soteric relevance: “Those who realize this
are called Great Enlightened Ones, those who are confused about it are called ‘sentient
beings’” (Shõji jissõ-gi, 401c). See also RAMBELLI 1992, pp. 163–85; and 1994.
49 On mantric expressions as inscriptions of soteriology, see LOPEZ 1990. For an analysis
of Shingon inscription strategies, see RAMBELLI 1991 and 1992 (pp. 249–55; 265–70; 296–316).
RAMBELLI: True Words, Silence, and the Adamantine Dance 395
logic and practices (shðgõ H§) of honji suijaku concern the relationships
between the Shinto and Buddhist deities, myths, and doctrines
that lie at the basis of Japanese medieval religiosity and ideology, and
obey rules grounded on “associative linguistic phenomena such as
metaphor, paronomasia, and anagogy” (GRAPARD 1988, p. 264; see also
1987, 1992). In other words, operations on the substance (both
graphic and phonetic) of language and meaning governed the esoteric
interpretation of reality.50
According to Grapard, such combinatory
practices brought about a reduction from plurality to singularity
(1987), but I think that they also exposed the plural nature of supposedly
singular entities.51
This kind of esoteric operation on signs is
remarkably evident in a corpus of medieval texts known as engimono
â|], which deal with the history of sites of cult.
The esoteric episteme, in its more conscious and systematic manifestations,
was basically a “high” culture phenomenon. Nevertheless, it
is important to trace the dissemination of esoteric doctrines and practices
among the general populace, and to analyze their transformations
and the counter-practices they produced. This dissemination was
extremely important for the establishment, which saw the “esotericization”
of the lives, activities, and environment of the ordinary people
as a powerful device for controlling them. In general, “popular” texts
dealing with Buddhism (performances, sermons, kana literature, and
narratives) were not directly concerned with esoteric doctrine—one
must recall that, because of Mikkyõ’s belief that it expressed the
absolute point of view of the perfectly enlightened Buddha, it was not
easy for Mikkyõ to translate its doctrines into everyday language and
practice. However, the discourse to which such popular texts belong,
and therefore their semiotic presuppositions, discursive strategies, and
rhetorical devices, are de³nitely esoteric.52
In the engimono genre,
Mikkyõ succeeded in transposing its absolute logic of the unconditioned
(jinen hõni À5À¹) into a narrative of karmic events that
occurred at speci³c historical moments in speci³c places (see KURODA
1989). These widely circulated materials were the major vehicle for
the “popularization” of the esoteric conceptions and the power relations
that they implied.
50 See RAMBELLI 1992 and GRAPARD 1987. Another vivid example of these combinatory
practices can be found in Dairyð’s Sangai isshin-ki, where the stages of the human embryo
are associated to the Thirteen Buddhas via various operations on their names.
51 The absolute value of phenomena and particularities—i.e., of differences—is one of
the major themes of most exoteric and esoteric hongaku (original enlightenment) texts from
the middle Kamakura period; for an introductory account of this subject, see RAMBELLI
1993. Concerning the plural nature of Tantric symbols and entities, see BOON 1990, pp. 79–83.
52 Recent discoveries have revealed the existence of an esoteric genre of setsuwa literature,
an example of which is the Aizen õ shõryð-ki (ô÷ÛNz in KUSHIDA 1979, pp. 819–41.
396 Japanese Journal of Religious Studies 21/4
The diffuse beliefs and practices of the uninitiated concerning
such sacred esoteric objects as images, texts, amulets, and talismans
constitute semiopietas, “a primarily religious mood of relation to sacred
[signs]” (GRAPARD forthcoming). Semiopietas is the esoteric “easy
path” (idõ ^‰) to salvation, represented mainly by the himitsu nenbutsu
¸Oç[ and kõmyõ shingon MgOí practices. For most of these practices
no formal initiation was required—all that was needed was a
transmission with simple explanations, usually called kechien kanjõ
ºâ/™; furthermore, practices pertaining to semiopietas were considered
to be ef³cacious even when not correctly performed, provided
the intention was right, as explained for instance by RENTAI in his
Shingon kaiku-shð. Since the salvi³c power of signs is intrinsic to them,
the uninformed usage of Mikkyõ amulets or talismans (usage that
leaves meaning out of consideration) has its theoretical foundation in
semiognosis, and is legitimated by the weight of tradition and the idea
of an unaltered secret transmission (see also RAMBELLI 1991, pp.
20–21; 1992, pp. 240–42).
Ritual and the Adamantine Dance
I have claimed that at the background of the various avatars of
Tantrism, at least in Japan, lie certain ideas on cosmology and soteriology
that possess a semiotic nucleus de³ning phenomena as manifestations
of the Dharmak„ya and that—above all—deal with the power of
symbolic actions to produce salvation. Mikkyõ envisions the cosmos as
a fractal structure, in which each phenomenon is “formally” similar to
all others and to the totality. This recursive cosmology, unique to
Mikkyõ, is related to a recursive soteriology that attributes enormous
importance to ritual practice and visualization (see ORZECH 1989).
One may assume that certain con³gurations of the Mikkyõ episteme
lay at the basis of the combinatory doctrines and practices that developed
in premodern Japan in a way that was mainly locale-speci³c and
lineage-grounded (GRAPARD 1992).
Allan GRAPARD points to the existence of an “episteme of identity”
(1989, p. 182) underlying Japanese mythology and mountain asceticism,
an episteme that sees “the world (nature) and words (culture)
in the speci³c lights of similitude, reµection, identity, and communication”;
GRAPARD (1989, p. 161) explicitly refers to the preclassical
European episteme as reconstructed by Michel Foucault. I suggest
that such an “episteme of identity,” at least in its more systematic
forms, was ³rst codi³ed on the basis of Mikkyõ doctrine, and that it
then assumed cultural hegemony in medieval Japan. The Mikkyõ episRAMBELLI:
True Words, Silence, and the Adamantine Dance 397
teme appears to be characterized by the workings of what Tsuda
Shin’ichi calls the “logic of yoga,” which asserts the substantial nondifferentiation
of all things on the basis of concepts of analogy and
resemblance. This opens the way, in turn, to a kind of “symbolic
omnipotence,” based on the belief that ritual—indirect “symbolic”
practices—produces numberless powers by virtue of the structure of
the signs involved in the ritual process (TSUDA 1978, 1981).53
It should
be clear, however, that such epistemic constructs, far from being simple
ritual or meditative escamotages, were directly related to the creation
of a ritualized world (closely connected to power and dominant ideology)
in which each event and each phenomenon was cosmologically
marked and played a salvi³c function. Moreover, as forms of visualization
based on a complex semantic and ritual network, symbolic practices
grounded on the logic of yoga produced a cognitive transformation;
when seriously performed, esoteric practices disclosed a different
world.
The logic of yoga thus underlies Shingon ritual practice, which is
often despised as a degeneration of “true” Mikkyõ by scholars who
forget that ritual effort aimed at cosmic integration and political legitimization
is a demonstration of the fundamental principles supporting
the esoteric episteme. As we have seen, basic to Shingon Mikkyõ
are its peculiar semiotics and semiosis. Ritual action is not a degeneration
of “pure” Mikkyõ or a relic of earlier “miscellaneous” forms, as
many scholars insist, but is directly related to the postulates of the esoteric
episteme itself.54
The basic epistemic framework of the Shingon tradition, with its
complex interrelations of cosmology, soteriology, semiotics, and ritual,
was shared by virtually all esoteric lineages in Japan. It should be
stressed, however, that the preceding account applies mainly to those
learned monks (gakuryo ¿Q) who attempted to manifest the esoteric
universe through meditation and ritual and who exploited to the
utmost degree the power that they attributed to esoteric (or esotericized)
signs—a semiotic power that reinforced, and was reinforced by,
economic, social, and political power in the framework of a coherent
sociocosmic order. It is possible to argue, on the basis of diaries and
53 It should be noticed, however, that my treatment of these subjects is different in perspective
and approach from that of Tsuda, which lacks explicit semiotic and ideological con-
cerns.
54 Since the present essay is concerned mainly with the formation of the epistemic space
and the conditions of possibility in the Japanese esoteric Buddhist discourse, all-important
questions concerning ritual practice have remained in the background. Epistemic problems
of esoteric rituals, such as the ritual manipulation of symbolic entities, will be the subject of
a future study.
398 Japanese Journal of Religious Studies 21/4
other textual evidence, that the aristocrats and, to a certain extent,
the ordinary people also lived in such an esotericized, ritual universe.55
They shared the same mentality and ensemble of combinatory beliefs
and practices; at the bottom of their way of life was an awareness—
rarely discussed explicitly or critically—that the cosmos is an unceasing
“adamantine dancing performance” (RAIHÕ, T 77.731a), a continuous
transformation of shapes similar to the endless movement of
waves on the surface of the sea, governed by linguistically grounded
combinatory rules.
This awareness is related to a diffuse heterology/heteropraxy that
pervades the entire Indian tradition (and perhaps the entire Buddhist
world as well) and emerges from what Iyanaga Nobumi calls “mythologie
‘buddhico-ésotérico-šivaïte’.”56
The epistemic aspects of this mentality
have been referred to as “Siwaic Semiotics” (BOON 1990, p. 70).
Medieval Japanese ideals, rituals, and practices of orthodoxy and identity
were thus underlain by a combinative episteme of transformation,
in itself an avatar of Indian šivaitic mentality. The epistemic ³eld manifested
itself and was actualized in at least two ways: in a fully conscious
way through semiognosis, and in a simpli³ed and uninformed way
through semiopietas (semiosophia lying outside the “Tantric” mentality).
Both paradigms were aimed at esoterically framing the lives of the
people, and functioned as powerful means of social control. But when
the incessant “adamantine dance” of shifting forms was properly performed
and ritually controlled, the esoteric cosmos took on the shape
of an immense salvi³c “machine,” where all movements were ritualized
and oriented to individual self-realization and universal salvation.
In the above discussion of Mikkyõ heterology, I mentioned ambiguous,
marginal, and antisystematic forms of Japanese esoteric Buddhism.
These can be seen to represent “Tantric” tendencies aimed at
countering the systematic, “mandalic” Mikkyõ—Mikkyõ as an organic
part of the kenmitsu system—that I have outlined. These trends, all
related in some way to the complex and multifarious hijiri phenomenon,
55 Their lives were probably similar in structure and basic attitudes to that of Jinson c¨
(1430–1508), abbot of the Daijõ-in monzeki ØñŠ–Ô of the Kasuga-Kõfuku-ji rÕöS±
complex, as it has been portrayed by Allan GRAPARD (1992, pp. 171–85). GRAPARD explains:
“To Jinson, the mirrorlike relation between the heavenly bureaucracy and the structure of
the [Kasuga-Kõfukuji] multiplex and of society in general was the manifestation of a preestablished
harmony that could never be discussed, even less called into doubt. Such
preestablished harmony, however, grounded though it may have been in myth and supported
by ritual, needed another type of reinforcement…provided by economic power and, more
precisely, land” (p. 174).
56 Personal communication, 6 April 1993. For a masterful description of the workings of
such a mythology, see IYANAGA 1994.
RAMBELLI: True Words, Silence, and the Adamantine Dance 399
attempted to overcome the symbolic nature of the secret practices, or,
at least, to exploit them in a quest for a more “direct” salvation, either
individual or collective. What follows is a partial list of the most
signi³cant of these movements.
The Shingon Ritsu A school of Eison and Ninshõ attempted to perform
bodhisattva practices within an esoteric context; their activities
were aimed at bringing concrete relief to suffering beings and, at the
same time, at realizing “symbolic”—and therefore indirect—universal
salvation. Shingon Ritsu was also very active in controlling and organizing
the newly rising forces of social marginality—a potential threat
to the kenmitsu establishment (see AMINO 1986 and ¼ISHI 1987).
Shugendõ lineages produced new heterodox and syncretic practices
and spread them throughout Japan, thus contributing to the diffusion
and proliferation of Mikkyõ. The Ji ´ movement of Ippen s’
(1239–1289) at a certain point was virtually in control of Kõya-san,
although its position in the Japanese Tantric ³eld is yet to be analyzed.
Tachikawa-ryð, Genshi kimyõdan éŠof;, and related trends in
other schools developed direct practices grounded on the idea of
absolute nondualism. The peculiar esoteric quest for paradise, a
major esoteric trend since the late Heian period, is interesting
because of its attempt to integrate antithetical Shingon and Nenbutsu
practices. Finally, the case of sokushinbutsu “X[—a sublime and disquieting
murmur pervading the whole East Asian Buddhist tradition—deserves
mention because of the extremes to which the ascetics
involved carried the desire to attain direct and universal salvation.
The doctrines and activities of these and other movements are not
fully compatible with the orthodox Mikkyõ discourse that has been
outlined here; as a kind of “dark side” to the secret teachings, they
require further research.
REFERENCES
PRIMARY SOURCES
DNBZ Dai Nihon bukkyõ zensho ØÕû[î6–, 150 vols. Takakusu
Junjirõ ¢ÈˆµÁ et al., eds. Tokyo: Dai Nihon Bukkyõ Zensho
Kankõkai, 1931. Reprint, 100 vols., Suzuki Gakujutsu Zaidan
Š…¿n(:, ed. Tokyo: Kõdansha, 1970–1973.
KDZ Kõbõ Daishi zenshð eÀØ‚6T, 8 vols. Kõyasan Daigaku Mikkyõ
Bunka Kenkyðjo ¢Ÿ[Ø¿Oîk5ÓÁ‹, ed. Tokyo: Yoshikawa
Kõbunkan, 1965–1968.
MD Mikkyõ daijiten OîØÂø. Mikkyõ Daijiten Hensankai OîØÂø
‹el, ed., 1931. Revised edition by Chishakuin Daigaku
Mikkyõ Gakkainai Mikkyõ Daijiten Saihan Iinkai JΊؿ
400 Japanese Journal of Religious Studies 21/4
Oî¿l»OîØÂøç•W‚l. Kyoto: Hõzõkan, 1970 (reduced
size reprint 1983).
T Taishõ shinshð daizõkyõ رG@؉™, 100 vols. Takakusu
Junjirõ ¢ÈˆµÁ et al., eds. Tokyo: Taishõ Issaikyõ Kankõkai
and Daizõ Shuppan, 1924–1932.
Dari jing ØÕ™ (Mah„vairocana Sðtra; full Chin. title Da pilushena chengfo
shenbian jianchi jing ØÈk캨[Pˆ;³™), T #848, 18.1–55.
Dasheng liqu liuboluomituo jing Øñ7+Â#øP-™. T #261, 8.865–917.
Jie shenmi jing mLO™ (Sa½dhinirmocana Sðtra), T #676, 16.688–711
(also #675, #677).
Jinggang ding jing D¤™™ (Sarvatath„gata tattvasa½graha, original Chin.
title Jinggang ding yiqie rulai zhenshi shedashen xianzheng dajiaowang
jing D¤™s×ØûO×ÚØñêãØî÷™), T #865,
18.207–223.
Lueshu jinggang ding yuqie fenbie shengwei xiuzheng famen FoD¤™î8
_ƒ¸R@ãÀ–(also known in Japan as the Funbetsu shõi-kyõ
_ƒ¸R™), T #870, 18.287–291.
Shi moheyan lun ö#ä-Ç. T #1668, 32.591–668.
DAIRYÐ ØP
Sangai isshin-ki XƒsDz. In Nihon shisõ tõsõ shiryõ Õû„`ymt[,
Washio Junkei Ðň’, ed., pp. 503–40. Tokyo: Tõhõ Shoin,
1930.
GYÕNEN !5
Hasshð kõyõ k;„ê. DNBZ 29: 11–25.
KAKUBAN ·Î
Gorin kuji myõ himitsu-shaku 2sG°g¸Oö. T #2514, 79.11–22.
Ichigo taiyõ himitsu-shð skØê¸OT. In Kõgyõ Daishi senjutsu shð
öîØ‚îoT, vol. 1. Miyasaka Yðshõ ·*»§, ed., pp. 157–76.
Tokyo: Sankibõ Busshorin, 1977 (revised edition, 1989).
Kenmitsu fudõ ju ßO#|†. T #2510, 79.1.
KÐKAI W}
Ben-kenmitsu nikyõ-ron –ßOÌîÇ. KDZ 1: 474–505.
Bonji shittan jimo narabini shakugi ¤°Ò·°ªWö–. T #2701,
84.361–364.
Himitsu mandara jðjðshin-ron ¸ORwøYWDÇ. T #2425, 77.303–362.
Himitsu mandarakyõ fuhõden ¸ORwøî$À). KDZ 1: 1–49.
Shingonshð shogaku kyõ-ritsu-ron mokuroku Oí;‹¿™AÇ‡Æ (Sangakuroku
X¿Æ). KDZ 1.
Shõji jissõ-gi ¹°×o–. T #2429, 77.401–404.
Shõrai mokuroku ¾û‡Æ. KDZ 1: 69–102.
RAMBELLI: True Words, Silence, and the Adamantine Dance 401
RAIHÕ þµ
Shingon myõmoku Oíe‡. T #2449, 77.730–736.
RAIYU þî
Shoshð kyõri dõi shaku ™;î7|bö. DNBZ 29: 4–10.
RENTAI ¥¿
Shingon kaiku-shð OíˆøT. In Shingonshð anjinsho Oí;HD–, vol. 1,
Hase Hõshð ˜úµD, ed., pp. 90–137. Kyoto: Rokudai
Shinpõsha, 1913.
TOKUITSU ”s
Shingonshð miketsu-mon Oí;J·k. T #2458, 77.862–865.
SECONDARY SOURCES
AMINO Yoshihiko }Ÿ3Ò
1986 Igyõ no õken b†u÷Ï. Tokyo: Heibonsha.
BOON, James A.
1982 Other Tribes, Other Scribes. Cambridge: Cambridge University
Press.
1990 Af³nities and Extremes. Chicago and London: University of
Chicago Press.
CERTEAU, Michel de
1982 La fable mystique, XVIe–XVIIe siècle. Paris: Gallimard.
1986 Heterologies: Discourse on the Other. Minneapolis: University of
Minnesota Press.
1990 L’invention du quotidien, 1: Arts de faire. Rev. ed. Paris: Gallimard.
DUMONT, Louis
1979 Homo hierarchicus: Le système des castes et ses implications. Rev. ed.
Paris: Gallimard.
FAURE, Bernard
1987 The Daruma-shð, Dõgen, and Sõtõ Zen. Monumenta Nipponica
42/1: 25–55.
1991 The Rhetoric of Immediacy: A Cultural Critique of Chan/Zen
Buddhism. Princeton: Princeton University Press.
GORAI Shigeru 2û b
1975 Kõya hijiri ¢Ÿ¸. Rev. ed. Tokyo: Kadokawa Sensho.
GRAPARD, Allan G.
1987 Linguistic cubism: A singularity of pluralism in the Sannõ cult.
Japanese Journal of Religious Studies 14: 211–34.
1988 Institution, ritual, and ideology: The twenty-two shrine-temple
multiplexes of Heian Japan. History of Religions 27: 246–69.
1989 The textualized mountain—enmountained text: The Lotus Sutra
in Kunisaki. In The Lotus Sutra in Japanese Culture, George J.
402 Japanese Journal of Religious Studies 21/4
Tanabe, Jr. and Willa Jane Tanabe, eds., pp. 159–89. Honolulu:
University of Hawaii Press.
1992 The Protocol of the Gods: A Study of the Kasuga Cult in Japanese
History. Berkeley, Oxford: University of California Press.
forthcoming Geosophia, geognosis, and geopiety: Orders of
signi³cance in Japanese representations of space. In Now-here.
Space and Social Theory, Deirdre Boden and Roger Friedl, eds.
Los Angeles, Berkeley: University of California Press.
GREIMAS, J.A., and J. COURTÉS, eds.
1979 Sémiotique: Dictionnaire raisonné de la théorie du langage. Paris:
Hachette.
HAKEDA, Yoshito S.
1972 Kðkai: Major Works. New York: Columbia University Press.
HAYAMI Tasuku ™v X
1975 Heian kizoku shakai to bukkyõ rH{Ÿçlo[î. Tokyo:
Yoshikawa Kõbunkan.
IYANAGA Nobumi ‚½=Ë
1994 Daikokuten. Hõbõgirin 7, pp. 839–920.
KOMATSU Kazuhiko ·ÇÉÒ
1988 Nihon no noroi Õûu2J. Tokyo: Kõbunsha.
KOMATSU Kazuhiko and NAITÕ Masatoshi »n±p
1985 Oni ga tsukutta kuni—Nihon …RkUjf³3Õû. Tokyo: Kõbunsha.
KURODA Toshio ¸,pÍ
1975 Nihon chðsei no kokka to shðkyõ Õû_›u³Bo;î. Tokyo:
Iwanami Shoten.
1989 Historical consciousness and hon-jaku philosophy in the
medieval period on Mount Hiei. In The Lotus Sutra in Japanese
Culture, George J. Tanabe, Jr. and Willa Jane Tanabe, eds., pp.
143–58. Honolulu: University of Hawaii Press.
KUSHIDA Ryõkõ ^,dt
1964 Shingon mikkyõ seiritsu katei no kenkyð OíOî¨C[ÝuÓÁ.
Tokyo: Sankibõ Busshorin.
1979 Zoku Shingon mikkyõ seiritsu katei no kenkyð ¡OíOî¨C
[ÝuÓÁ. Tokyo: Sankibõ Busshorin.
LAMOTTE, Etienne
1988 The assessment of textual interpretation in Buddhism. In LOPEZ
1988, pp. 11–27.
LOPEZ, Donald S., Jr.
1990 Inscribing the bodhisattva’s speech: On the Heart Sðtra’s
mantra. History of Religions 29: 351–72.
RAMBELLI: True Words, Silence, and the Adamantine Dance 403
LOPEZ, Donald S., Jr., ed.
1988 Buddhist Hermeneutics. Honolulu: University of Hawaii Press.
MISAKI Ryõshð X2d:
1988 Taimitsu no kenkyð ×OuÓÁ. Tokyo: Sõbunsha.
MURAYAMA Shðichi ª[@s
1981 Nihon onmyõdõ-shi sõsetsu Õû‹î‰trß. Tokyo: Hanawa Shobõ.
1987 Shðgõ shisõshi ronshð H§„`tÇT. Tokyo: Hanawa Shobõ.
1990 Henbõ suru kami to hotoketachi—Nihonjin no shðgõ shisõ ˆå`š
Po[fh—Õû^uH§„`. Kyoto: Jinbun Shoin.
NAKAZAWA Shin’ichi _åGs
1988 Akutõteki shikõ 1J턆. Tokyo: Heibonsha.
¼ISHI Masaaki ØÍhØ
1987 Hinin kyðsai to seichõ annon—Risshð Eison no shðkyõ katsudõ
À^ºòo¸†H2—A;µ¨u;îÏ{. In Kokka to tennõ—
Tennõsei ideorog‡ to shite no bukkyõ ³Boúy—úy£©Ï±õµ2
o^mu[î (Bukkyõ to Nihonjin [îoÕû^ 2). Kuroda Toshio
¸,pÍ, ed., pp. 137–82. Tokyo: Shunjðsha.
ORZECH, Charles D.
1989 Seeing Chen-Yen Buddhism: Traditional scholarship and the
Vajray„na in China. History of Religions 29: 87–114.
RAMBELLI, Fabio
1990 Il potere karmico della parola: Elementi per lo studio della concezione
del linguaggio nel Nihon ryõiki. Annali di Ca’ Foscari 29:
271–89.
1991 Re-inscribing ma«^ala: Semiotic operations on a word and its
object. Studies in Central and East Asian Religions 4: 1–24.
1992 Segni di diamante: Aspetti semiotici del buddhismo esoterico
giapponese di Kakuban, 2 vols. Ph.D. diss., University of Venice.
1993 “Piante e alberi diventano buddha.” La natura nel paradigma
tendai. In Atti del XVI convegno di studi dell’Associazione italiana per
gli studi giapponesi, Fosco Maraini et al., eds., pp. 191–209.
Firenze: Associazione italiana per gli studi giapponesi.
1994 The semiotic articulation of hosshin seppõ: An interpretive study
of the concepts of mon and monji in Kðkai’s mikkyõ. In Esoteric
Buddhism in Japan, Ian Astley, ed., pp. 17–36. Copenhagen and
Aarhus: The Seminar for Buddhist Studies.
SASAKI Kaoru Õ/…¢
1988 Chðsei kokka no shðkyõ kõzõ—Taisei bukkyõ to taiseigai bukkyõ no
sõkoku _›³Bu;îr‹—¿£[îo¿£‘[îuo°. Tokyo:
Yoshikawa Kõbunkan.
SATÕ Hiroo Õne&
1987 Nihon chðsei no kokka to bukkyõ Õû_›u³Bo[î. Tokyo:
Yoshikawa Kõbunkan.
404 Japanese Journal of Religious Studies 21/4
TAKAHASHI Tomio ¢ï)Í
1990 Tokuitsu to Saichõ—Mõ hitotsu no seitõ bukkyõ ”so蘗
‘Lsku±j[î. Tokyo: Chðõ Shinsho.
TSUDA Shin’ichi §,Os
1978 A critical Tantrism. The Memoirs of the Toyo Bunko 36: 167–231.
1981 Mikkyõ kenkyð no hõkõ—Mikkyõ shisõ no hihanteki rikai o
mezashite OîÓÁu¾T—Oî„`u-|í7m¤Œ]^m. Tõyõ
gakujutsu kenkyð Xá¿nÓÁ 20/1: 117–36.
1985 The hermeneutics of Kðkai. Acta Asiatica 47: 82–108.
UJIKE Kakushõ ’B·§
1984 Darani no sekai ¼øÍu›ƒ. Osaka: Tõhõ Shuppan.
YAMAGUCHI Masao [SÄC
1975 Bunka to ryõgisei k5oX–§. Tokyo: Iwanami Shoten.
1989 Tennõsei no bunka jinruigaku úy£uk5^{¿. Tokyo: Rippð
Shoten.
RAMBELLI: True Words, Silence, and the Adamantine Dance 405