. esoteric buddhism and the tantras in east asia 9
history to create anachronistically a “tantric tradition” with Hindu and
Buddhist variants.^12
Exactly how one defines the distinctiveness of the tantras is subject to
debate. Some scholars have adopted a monothetic definition that settles
on one distinctive characteristic as essential.^13 Thus, some scholars have
proposed visualization of oneself in the body of the divinity (Sanskrit
ahaṃkāra) as the sine qua non of the movement. For instance, Michel
Strickmann argued that “le rituel du bouddhisme tantrique est l’union
avec une icône.”^14 Tsuda Shin’ichi used the same criteria to posit a
“critical” disjunction between Mahāyāna Buddhism and tantrism, with
the former culminating in the Mahāvairocanābhisaṃbodhi sūtra (MVS)
and the latter commencing with the Sarvatathāgatatattvasaṃgraha
(STTS).^15 Certainly texts with a bhāvana or visualization section
inserted into otherwise unremarkable vidhis (ritual manuals) appear
in Chinese records during the seventh century. But this definition is
not without problems, chief among them, that visualization is com-
mon in the Mahāyāna (for a discussion see Payne 2006, 1–31). Ronald
Davidson, moving in a more polythetic^16 and historical direction, has
argued that tantra emerged out of Indian sāmanta feudalism and is
structured around the metaphor of the rājādhirāja, and the systematic
use of abhiṣeka, mandala, homa, and injunctions to secrecy.^17 As noted
above, he sees the earliest evidence for a distinctive, and self-conscious
movement appearing in Atikūta’s seventh century ̣ Tuoluoni ji jing
( Dhāraṇīsaṃgraha sūtra, T. 901, 654 C.E.).
We should also note that many of the texts widely reckoned to be
tantras, like the Mañjuśrīnāmasaṃgīti, do not even have the term “tan-
tra” in their titles, and the transliteration of the term tantra is extremely
(^12) Strickmann 1996, 24: “les agama du śivaïsme médiéval et les tantra du bouddhisme
médiéval représentent simplement différentes versions, différentes rédactions d’une
seule et même chose.” Despite Urban there is some logic in Strickmann’s statement. 13
For a discussion of definitions that emerged out of the Tantric Studies Seminar
of the American Academy of Religion see White 2000, 3–38. 14
Strickmann 1998, 203.
(^15) Tsuda 1978, 167–231. Both Davidson and Giebel argue that Vajraśekhara is
probably incorrect and should be rendered Vajra-usṇīsa. See Giebel 1995, 109 and
Davidson, “Sources and Inspirations,” in this volume. The editors have left individual
author’s usage intact.
(^16) Definitions that employ a variety of overlapping characteristics are often termed
“polythetic” definitions. These reflect the influence of Wittgenstein’s notion of “family
resemblances.” 17
Davidson 2002, 114, 117.