1 Boisselier, Jean, ‘Garuḍa dans l'art khmèr’, Bulletin de l'Ecole française d'Extrême-Orient, 44 (1950): 57Google Scholar. (My translation, as are all others from French.)

2 Boisselier said the Garuḍa in the Bayon style were undoubtedly Buddhist but their relations with nāga raised questions, and the hypotheses he advanced should not be taken as solutions but as ‘instruments de recherche’; ibid., p. 86.

3 R. Davidson: ‘In erecting the new temple complexes, kings became patrons to the new divinities that commanded the areas under the rulers’ political control. Thus the new temples satisfied many functions. They became testaments to royal legitimacy, with the rulers using the temple walls as a tabula rasa for the epigraphs that communicated royal piety, regal decisions on legal matters, imperial conquests, formal alliances with other houses, and a host of matters rendering them archives of a ruling house.' Davidson, Ronald M., Indian esoteric Buddhism: A Social history of the Tantric movement (New York: Columbia University Press, 2002), p. 73Google Scholar.

4 Claude Jacques has recently questioned George Coedès' inclusion of Jayavarman VII in the ‘Mahīdharapura dynasty’ on the grounds that Jayavarman was raised in the unidentified city of Jayādityapura and probably based his claim to the throne on the august antecedents of his mother Jayarājacūḍāmaṇi, queen of Jayadityapura. (Jacques, C., ‘The Historical development of Khmer culture from the death of Suryavarman II to the 16^th century’, in Bayon, New Perspectives, ed. Clarke, Joyce (Bangkok: River Books, 2007), p. 32Google Scholar). However, despite the undoubted importance of the maternal line in Khmer genealogy, we know that Jayavarman VII's father, paternal grandfather and paternal great grandmother (sister of Jayavarman VI) came from Mahīdharapura (also yet to be located). I therefore find that Coedès' use of the term ‘Mahīdharapura dynasty’ remains a legitimate designator for the preponderantly Buddhist northern Khmers who, when Jayavarman VII and Indravarman II are included, ruled Angkor from 1080 to 1270 (this is not invalidated by the possibility we consider later that there was a hiatus during the obscure reigns of Yasovarman II and Tribhuvanādityavarman, whose genealogies are unknown).

5 The traditional date of Jayavarman's coronation (1181) has recently been modified to 1182 by Sanskrit scholars in a corrected reading of the śaka year in an inscription. Refer to Michael Vickery, ‘Introduction’, in Bayon, New perspectives, ed. Joyce Clarke, p. 13 n8.

6 This is Coedès' translation of the Phimeanakas inscription (K.485 C v.28), written by the king's Sanskritist wife Indradevī (Coedès, George, ‘Inscription du Bàyon K.470’, Inscriptions du Cambodge, vol. III (Hanoi: EFEO, 1942): 177Google Scholar).

7 The Phimeanakas stela says the reign preceding Jayavarman's was marred by multiple parasols being raised by competing monarchs: ‘Under the preceding reign, the earth, although shaded by numerous parasols, suffered from extreme heat; under his [Jayavarman VII's] reign, when there was only one parasol, the earth was, strangely, released from all suffering.’ (K.485 A v.51/2, Coedès, ‘Inscription du Bàyon K.470’, 387). Michael Vickery, in his Introduction to Bayon, New Perspectives, commends Claude Jacques' recent characterisation of this period of internal strife, intensified by a swirling mix of hostile Khmer-Cham alliances:

8 The quandary the discovery posed is well illustrated in Louis Finot's seminal 1925 paper on Lokeśvara in which he records the presence of Bodhisattvas and Buddhas at Angkor Thom, Tà Prohṃ, Neak Pean, Tan Nei, Ta Som, Bantéay Chmar as well as in the Bayon, in what he still takes to be the late ninth-century capital of the Śaiva king Yasovarman I: ‘This conclusion [that Angkor Thom was … a Buddhist city dedicated to Lokeśvara] raises complex historical problems, which we cannot address here.’ (Finot, Louis, ‘Lokeśvara en Indochine’, in Études asiatiques: Publiees a l'occasion du vingt-cinquieme anniversaire de francaise d' Extreme-Orient par ses members et ses collaborateurs (Paris: I EFEO, 1925), p. 247Google Scholar).

9 Coedès' view was: ‘What is sure is that he left the country exhausted by his megalomania and thenceforth powerless to resist the attacks of his young and turbulent neighbour to the west.’ Coedès, George, Pour mieux comprendre Angkor (Hanoi : Imprimerie d'Extrême Orient, 1925), p. 205Google Scholar. Philippe Stern was among the first to use this word: ‘The buildings in the style of the Bayon are those most strongly marked, in their dimensions, number and poor quality execution, by a sort of artistic megalomania and desire to astonish …’ Stern, Philippe, Le Bayon d'Angkor et l'évolution de l'art khmer: Étude et discussion de la chronologie des monuments khmers (Paris: Librairie orientaliste Paul Geuther, 1927), p. 182Google Scholar.

10 See, for example, Hawixbrock, Christine, ‘Jayavarman VII ou le renouveau d'Angkor, entre tradition et modernité’, Bulletin de l'Ecole française d'Extrême-Orient, 85 (1998): 64CrossRefGoogle Scholar; and Jacques, ‘The Historical development of Khmer culture from the death of Suryavarman II to the 16^th century’, p. 40: ‘Apart from the eulogies contained in the inscriptions, it is sufficient to say that Jayavarman VII was a king generally honoured in the same fashion as all Khmer kings, with no special “megalomania”’.

11 Li-Kouang, Lin, ‘Punyodaya (N'ati), un propagateur du Tantrisme en Chine et au Cambodge à l’époque de Hsüan-Tsang’, Journal Asiatique (July–Sept. 1935): 83–100Google Scholar.

12 Yijing, , A Record of the Buddhist religion as practised in India and the Malay archipelago (A.D. 671–695), trans. Takakusu, J. (Oxford: Clarendon, 1896), p. 12Google Scholar.

13 Dowling, N., ‘New light on early Cambodian Buddhism’, Journal of the Siam Society, 88 (2000): 122–55Google Scholar.

14 Coedès, , ‘La stèle de Tep Pranam, Cambodge’, Journal Asiatique (Mar.–Apr. 1908): 207 n1Google Scholar.

15 Prapandvidya, Chirapat, ‘The Sab Bāk inscription: Evidence of an early Vajrayāna Buddhist presence in Thailand’, Journal of the Siam Society (1990): 11Google Scholar.

16 Finot, ‘Lokeśvara en Indochine’, p. 235.

17 Coedès, ‘La stèle de Tep Pranam, Cambodge’ (1908): 207.

18 Pelliot, P., Mémoires sur les coutumes du Cambodge de Tcheou Ta-Kouan: Version nouvelle (Paris: Librairie d'Amérique et d'Orient, 1951), p. 11Google Scholar.

19 Jacques, Bayon, p. 41.

20 Coedès, George, ‘Inscription du Bàyon K.470’, Inscriptions du Cambodge, 2 (Hanoi : EFEO, 1942), p. 187Google Scholar.

21 The 274 Buddhist objects, mostly broken Buddhas on nāga, were found carefully buried with sand by Sophia University archaeologists in Mar. 2001 in the grounds of Jayavarman's Banteay Kdei temple (Marui, Masako, ‘The Discovery of Buddhist statues at Banteay Kdei temple’, Journal of Sophia Asian Studies, 19 (2001)Google Scholar; and Ishizawa, Yoshiaki, ‘Special issue of the inventory of 274 Buddhist statues and the stone pillar discovered from Banteay Kdei Temple’, Renaissance culturelle du Cambodge, 21 (2004), 2 vols.Google Scholar) The buried icons include several ‘earth-touching’ Maravijaya Buddhas and standing Buddhas with belt, centre-fold and hand turned forward on the chest, which belong to the Khmer Hīnayāna seen in the temples of Praḥ Pallilay temple and monument 486. As Indravarman II apparently followed Jayavarman VII's Buddhism loyally, and Jayavarman VIII seems to have constructed nothing, these temples and the buried Hīnayāna icons probably belong to the early fourteenth century, when the first use of Pāli in a Khmer inscription occurs in 1308 CE. (I differ here from Woodward, who assigns the buried icons to ‘not … later than about the middle decades of the thirteenth century’, Woodward, ‘Foreword’, in Bayon, New perspectives, ed. Joyce Clarke, p. 8).

22 Sanderson, Alexis, ‘The Śaiva religion among the Khmers: Part I’, Bulletin de l'Ecole française d'Extrême-Orient, 90, 1 (2003–4): 429CrossRefGoogle Scholar.

23 ‘We could ask ourselves whether it was not a matter of good form to convert to Buddhism during the reign of Jayavarman VII, for one cannot take all these minor works as representing royal commands.’ Hawixbrock, ‘Jayavarman VII ou le renouveau d'Angkor, entre tradition et modernité’, BEFEO, 85, p. 74.

24 This is a surmise, not an observation. The Brahmins seem to have remained the writers of inscriptions.

25 Snellgrove, David L., Khmer civilization and Angkor (Bangkok: Orchid Press, 2001), p. 54Google Scholar.

26 Michael Vickery assumes Jayavarman to have been ‘part of the Champa political scene’ in these years and therefore proposes that eventually ‘the real conquest of Angkor was by Jayavarman VII and his Cham allies, probably in the 1170s, at least before 1181 …’, Michael Vickery, Champa Revised, a working paper presented at the conference on Champa, Asia Research Institute, National University of Singapore, Aug. 2004, p. 31.

27 Here I support Claude Jacques' interpretation of the ramparts built around Angkor Thom: ‘Jayavarman VII clearly was anxious to raise high walls around his capital: his power was fragile because his enemies were not far away and he had understood that he needed a well-protected capital’ (Jacques, Bayon, p. 43).

28 Mabbett first identified the Indochinese (as opposed to Indian) monarchs as social engineers. ‘… in Indochina kings possessed and exercised a degree of real control over social organization, by virtue of their ritual position (which was foreign to India): they were social engineers’. Mabbett, I. W., ‘Varṇas in Angkor and the Indian caste system’, Journal of Asian Studies, 36, 3 (1977): 429Google Scholar.

29 Coedès, , ‘Le stèle du Praḥ Khan d'Angkor’, Bulletin de l'Ecole française d'Extrême-Orient, 41 (1941): 294Google Scholar. Further, some 20,400 divinities in gold, silver, bronze and stone were erected within it and 8,176 villages with 208,532 male and female slaves donated by the king (Coedès, ‘Le stèle du Praḥ Khan d'Angkor’ (1941): 297).

30 Finot, M.L., ‘Lokeśvara en Indochine’, in Etudes Asiatiques publiées a l'occasion du 25eme anniversaire de l'EFEO (Paris: EFEO, 1925), p. 239Google Scholar.

31 The 1191 Preah Khan stela commands the presence at an annual spring festival in the capital of the icons of 122 gods (including 23 ‘Jayabuddhamahānāthas’ distributed to major cities: K.908 D v.159, Coedès, ‘Le stèle du Pràḥ Khán d'Angkor’, 41 (1941): 267. This is presented as a new, empire-wide Buddhist festival with a leading place in the ritual calendar of the new Buddhist state. For a controversial but illuminating account, refer to a discussion of the form taken by the spring festival in Wyatt, David K., ‘Relics, oaths and politics in 13th century Siam’, Journal of Southeast Asian Studies, 32, 1 (2001): 18CrossRefGoogle Scholar. The Tà Prohṃ stela prescribes a similar new festival with parades, music and dancing involving large numbers of icons a month later than that at Preah Khan (Coedès, , ‘La stèle de Ta-Prohṃ’, Bulletin de l'Ecole française d'Extrême-Orient, 6 (1906): 77CrossRefGoogle Scholar).

32 Jayavarman's revival of Sanskrit in Khmer inscriptions ‘as an international elite language serving both countries [Cambodia and Champa]’ is noted by Vickery (Champa revised, p. 32). We know from the material record that some Cham temples received Khmer icon donations when the Khmers took control of central and southern Champa from 1203–20, and it would follow that these icons were accompanied by Khmer-style, Sanskrit ritual and liturgy. It may be added here that the outlook of reigning Southeast Asian Buddhist monarchs was apparently made internationalist on a greater scale at this time by Islam's destruction of the most important Buddhist monasteries in the Ganges valley (1197–1207) — as seen for example in the King of Burma building a Bodhgayā temple in Burma in an effort to afford Buddhism a replacement international centre following the overrun of the Indian pilgrimage site marking the Buddha's enlightenment.

33 ‘As in the fifth century, so in the tenth, Sanskrit learning was a prime instrument of power, and the cream of this learning was evidently held by many to be the analytical penetration of Buddhist thought and the catholicity of its Mahāyāna doctrine’, Maxwell, T.S., ‘Religion at the time of Jayavarman VII’, in Bayon, New perspectives (Bangkok: River Books, 2007), p. 90Google Scholar.

34 This is not David Wyatt's view, who sees ‘Jayavarman VII and the Angkorian elite’ as maintaining the state Śaivism of previous centuries under a veneer of Buddhism. But Wyatt is putting the cart before the horse when he sees Jayavarman's twelfth-century Buddhism as ‘a sort of Counter-Reformation’ against the thirteenth-century sweep of a (by analogy Protestant) Theravāda which was to emerge from Sri Lanka, Burma and Sukhothai and reach Cambodia by the fourteenth century. ‘It [Jayavarman's politico-religious strategy] was misguided and doomed to failure, for it was sorely deficient in understanding what the religious change had meant, and why it would amount to what can be called a “religious revolution” ’. Wyatt, David, ‘Relics, oaths and politics in thirteenth century Siam’, Journal of Southeast Asian Studies, 32, 1 (2001): 48CrossRefGoogle Scholar.

35 ‘… Garuḍa plays a role of increasing importance. Having made a modest start in statuary, he rose to occupy the first place in the Bayon period…’ (Boisselier, ‘Garuda dans l'art khmèr’, p. 56).

36 Vogel, J. Ph., Indian serpent-lore or the Nāgas in Hindu legend and art (London: Arthur Probsthain, 1926), plate 10Google Scholar. Reproduced with the kind permission of Arthur Probsthain, London.

37 The story is recounted, among other places, in the Vinaya Pitakam 1.3 Oldenberg reprint Pāli Text Society, 1938–66.

38 Longhurst, A.H., Memoirs of the archaeological survey of India: The Buddhist antiquities of Nāgārjunakoṇḍa, Madras Presidency (Madras: ASI, 1937)Google Scholar, Plate Lb.

39 Dupont noted the multiple similarities between the sixth–eighth century Sri Lankan Buddhas and the first Khmer nāga Buddhas of the tenth century, and concluded that the Mon Buddhists had been the indispensable intermediary in the iconic transfer: ‘It is therefore not possible that at the end of the tenth century, Khmer art borrowed [the traditions] directly from Ceylon and, as an intermediary was indispensable, only Mon art could have played this role. We therefore have to admit that the Khmer image of the Buddha on the nāga was inspired … by the Mon iconography …’ Dupont, , L'archéologie Mône de Dvāravatī (Paris: EFEO, 1959), p. 263Google Scholar. Woodward concurs, noting that the Buddha on the nāga appeared in Dvāravatī art in the eighth century but ‘[t]he source was surely Sri Lanka where the nāga-protected Buddha had become an iconic type slightly earlier.’ Woodward, H., The Sacred sculpture of Thailand (London: Thames & Hudson, 1997), p. 72Google Scholar.

40 Bruno Dagens has also recently detached the Khmer nāga Buddha from the Mucalinda story: ‘At first sight nothing explains the success of the Buddha on the nāga, unheard of elsewhere in the Buddhist world; we would readily think that it is not due to the anecdote of Muciliṇḍa, but rather to the very Khmer character that the presence of the nāga gives him…’, Dagens, Bruno, Les Khmers (Paris: Les Belles Lettres, 2003), p. 196Google Scholar.

41 Pal, P., ‘An Unusual naga-protected Buddha from Thailand’, in Buddhist Art, form & meaning (Mumbai: Pal Mang Publications, 2007), p. 55 onwardsGoogle Scholar.

42 Frédéric, Louis, Buddhism (Paris: Flammarion, 1989), p. 279Google Scholar.

43 Getty, Alice, The Gods of northern Buddhism (Oxford: Oxford University Press, 1914), p. 155Google Scholar.

44 Coedès, G., Pour mieux comprendre Angkor (Hanoi: Imprimerie d'Extrême Orient, 1943), p. 98Google Scholar.

45 Boisselier, ‘Garuda dans l'art khmèr’, p. 57.

46 ‘Uncrested, [the nāgas] strongly remind us of the nāga Mucilinda on which the Buddha sits … Their similarity to the nāgas of the corniches leads us to suppose that these are the “good” nāgas protected by Garuḍa’, ibid., p. 74.

47 Boisselier, ‘Garuda dans l'art khmèr’, p. 56.

48 I thank Elizabeth Moore for this designation of the late Bayon style Garuḍa icon.

49 Boisselier, ‘Vajrapāṇi dans l'art du Bàyon’, p. 326.

50 Boisselier, ‘Garuda dans l'art khmèr’, p. 56.

51 Lamotte, Étienne, ‘Vajrapāṇi en Inde’, Mélanges de sinologie offerts à M. Paul Demiéville (Paris: PUF, 1966), pp. 133–8Google Scholar.

52 Lalou, M., ‘Four notes on Vajrapāṇi’, Adyar Library Bulletin (Adyar, 1956), p. 289Google Scholar.

53 Ibid.

54 Burton Watson summarises:

Watson, Burton, The Lotus Sutra (New York: Columbia University Press, 1993), p. xviiiGoogle Scholar.

55 For the Buddhist texts sought abroad by tenth-century Khmer Buddhist guru Kīrtipaṇḍita, refer to Sanderson, A., ‘The Śaiva religion among the Khmers: Part I’, BEFEO, 90–1 (Paris: 2003–4), p. 427 n284Google Scholar. In his recent reflections on the Wat Sithor inscription, Maxwell suggests that in retaining the existing framework of Brahmanical titles, rites and practices, ‘Kīrtipanṇḍita was a forerunner of Jayavarman VII, probably one of many’ (Maxwell, ‘Religion at the time of Jayavarman VII’, p. 90).

56 Snellgrove, D.L., ‘Introduction’, in Sarva-tathāgata-tattva-samgraha, facsimile reproduction of a tenth-century Sanskrit manuscript from Nepal, ed. Chandra, Lokesh and Snellgrove, David L. (New Delhi, 1981), pp. 5–67Google Scholar.

57 Ibid., p. 40.

58 Linrothe, R., Ruthless compassion: Wrathful deities in early Indo-Tibetan esoteric Buddhist art (London: Serindia, 1999), p. 178 and following, 243 and followingGoogle Scholar.

59 Lamotte, ‘Vajrapāṇi en Inde’, p. 151.

60 Boisselier, ‘Vajrapāṇi dans l'art du Bàyon', p. 325.

61 Iyanaga cites the Ninnō-nenju-giki (‘probably composed by Liangpi, a disciple of Amoghavajra’) on the duel with Śiva in which Vajrapāṇi is described as ‘Trailoyavijaya-vajra (Gōsanze-kongō) whose anger is terrible, with four faces and eight arms, and who subjugates Maheśvara (Makeishura-daijizaiten) and the armies of Māra’. Iyanaga, Nobumi, ‘Récits de la soumission de Maheśvara par Trailokyavijaya d'après les sources chinoises et japonaises’, in Tantric and Taoist studies in honour of R. A. Stein, ed. Strickmann, Michel, vol. 2 (1983)Google Scholar, Mélanges chinois et bouddhiques, vol. 21 (Brussels: Institut belge des hautes etudes chinoises: Distributeur, Office international de librairie, 1981–85), p. 671 n54.

62 STTS 18.882.369b–373b.

63 Woodward, ‘The Kārandavyūha Sutra and Buddhist art in tenth century Cambodia’ in Buddhist art, form and meaning, ed. Pratapaditya Pal (New Delhi: Marg Publications), p. 77.

64 Woodward, , The Art and architecture of Thailand from prehistoric times through the thirteenth century (Leiden/Boston: Brill, 2003), p. 148Google Scholar.

65 Coedès, , ‘Etudes cambodgiennes XVII – L’épigraphie du temple de Phimai', Bulletin de l'Ecole française d'Extrême-Orient, 24 (1924): 349Google Scholar.

66 Davidson, Indian esoteric Buddhism, p. 143.

67 Boisselier, ‘Garuda dans l'art khmèr’, p. 325.

68 Mallmann, , Introduction à l'Iconographie du Tântrisme bouddhique (Paris: Maisonneuve, 1986), p. 420Google Scholar. The mudrā was hitherto rare in icons from this region. A photograph in Christie's 2001 New York catalogue shows a regal 42cm bronze Vajradhara with six arms, probably four faces and with vajra and ghantā crossed at the chest, sitting on a high throne with elephants and lions. From the parasol and drapes above the throne, it probably belongs with the ‘Bengali-influenced’ bronzes of the pre-Thai peninsula under the Śailendra, which Woodward dates to the 770–780s (Woodward, The Art and architecture of Thailand, p. 93).

69 Boisselier, ‘Garuda dans l'art khmèr’, p. 330. The tantric texts themselves render the naming of Vajrapāṇi difficult. Linrothe observes: ‘In the STTS Vajrapāṇi is also known as Samantabhadra, Vajrasattva, Vajradhara, Vajrahūṃkara and Trailokyavijaya’ (Linrothe, Ruthless compassion, p. 156).

70 Goloubew, Victor, ‘Sur quelques images khmèrs de Vajradhara’, Journal of the Indian Society of Oriental Art, 5 (1937): 97–104Google Scholar.

71 Dalsheimer, N., L'art du Cambodge ancient: Les collections du musée national de Phnom Penh (Paris: EFEO, Magellan & Cie, 2001), p. 176Google Scholar.

72 Eskenazi, J., Images of faith (London, 1995), no. 41Google Scholar; Jose, Regaldado Trota, Images of faith: Religious ivory carvings from the Philippines (Pasadena: Pacific Asia Museum, 1990)Google Scholar.

73 Sotheby's 27 Apr. 2005 (#221 67cm) and 19 Oct. 2005 (#196 84cm). Hiram Woodward kindly brought these to my attention.

74 Finot, ‘Lokeśvara en Indochine’, p. 237.

75 Goloubew, ‘Sur quelques images khmèrs de Vajradhara’, Plate XIIIB.

76 Plan and report of the survey and excavations of ancient monuments in north-eastern Thailand (1959 reprint 1979: fig 38; 1960–1: figs 4, 90) (Bangkok: Fine Arts Department), for which I am grateful to Hiram Woodward, who believes these Vajradhara can be dated to as early as 1186 (Woodward, The art and architecture of Thailand, p. 208).

77 Pelliot, Paul, ‘Le Bhaisajyaguru’, Bulletin de l'Ecole française d'Extrême-Orient, 3 (1903): 33CrossRefGoogle Scholar.

78 Davidson, Indian esoteric Buddhism, p. 154.

79 Skorupski, T., Sarvadurgatipariśodhanatantra (Oxford: Motilal Banarsidass (UK), 1983), p. xxviiGoogle Scholar.

80 Ibid., p. xxviii.

81 Ibid., p. 58.

82 Stern attributes the northern terraces facing the royal plaza to what he calls a ‘troisième période avancée’. Stern, Philippe, Les monuments khmers du style du Bàyon et Jayavarman VII (Paris: Presses Universitaires de France, 1965), p. 161Google Scholar — ‘either contemporary with the last work on the Bàyon or immediately following it’ (Stern, Les monuments khmers du style du Bàyon et Jayavarman VII, p. 165).

83 ‘We know that in the Buddhist tradition of the north, Vajrapāṇi, substitute for Indra, appeared as the rain god and the protector of the Nāga. To fulfil this role, he sometimes took on Garuḍa's own aspect as seen in some Nepalese bronzes. This is the form he took to protect the Nāga from their implacable enemies when they came to listen to the teachings of the historical Buddha. We believe we have identified the episode on a pediment of Tà Prohṃ at Angkor, where, above a frieze of three-headed Nāga with human bodies, their hands joined, another Nāga and a Garuḍa, both in the attitude of prayer, appear on each side of the Buddha.’ Boisselier, , ‘Vajrapāṇi dans l'art du Bàyon’, Proceedings of the 22^nd Congress of International Orientalists (Istanbul, 1951), p. 326Google Scholar.

84 Lamotte, Étienne, ‘Vajrapāṇi en Inde’, Mélanges de sinologie offerts à M. Paul Demiéville (Paris: PUF, 1966), p. 115Google Scholar.

85 Boisselier, ‘Vajrapāṇi dans l'art du Bàyon', p. 327. Boisselier went on to claim that ‘the continuity of the Vajrapāṇi cult’ linked the late tenth century to the epoch of the Bayon and he added Vajrapāṇi as another layer to the already dense mix of deities Coedès saw in the Bayon face towers. ‘These data [de La Vallée Poussin's showing that Lokeśvara Samantamukha may teach the law with the face of Vajrapāṇi] would suffice to explain the unusual characteristics of the faces of the Bayon towers, which would so remain the image of Lokeśvara projected in the person of the king, but a complex Lokeśvara possessing the supreme intelligence of Vajrasattva, from whom he proceeds, as much as the virtues of Vajrapāṇi with whom he had the power to identify …’

86 Getty, Alice, The Gods of northern Buddhism (Oxford: Oxford University Press, 1914), p. 48Google Scholar.

87 Macdonald, Ariane, Le Maṇḍala du Mañjuśrīmulakalpa (Paris: CRNS Adrien-Maisonneuve, 1962), p. 40Google Scholar.

88 Przyluski, Jean, ‘Les Vidyārāja: contribution à l'histoire de la magie dans les sectes Mahāyānistes’, Bulletin de l'Ecole française d'Extrême-Orient, 23 (1923): 309CrossRefGoogle Scholar.

89 Macdonald, Le Maṇḍala du Mañjuśrīmulakalpa, p. 40.

90 ‘And last the powerful mantra oṃ, eagle, great eagle! You whose wings are spread like a lotus! Killer of all the serpents … The Bodhisattva who teaches this, svāhā is known under the name Garuḍa. He is the best at converting the beings who are difficult to convert. He destroys the venom of the serpents and if we add the Mahāmudrā, he will conquer the hordes of hostile demons and serve as the antidote for all poisons. He has been taught by me (Mañjuśrī) as the means for converting those susceptible to this method. I came to act with the appearance of Garuḍa, the stunning king of the birds. All the rites developed in the Tantra of Garuḍa were taught by me for the benefit of creatures. Having come to earth as the Bodhisattva Garutma, with the appearance of a bird to convert creatures, I worked to combat the venom of the serpents.’ Ibid., p. 79.

91 For comparative chronological studies, using petrological sampling of magnetic susceptibility to indicate which sandstone was extracted at the same time from the quarries, and so which parts of the temples were built earlier than others, refer to Cunin, Olivier and Uchida, Etsuo, Annual report on the technical survey of Angkor monument (2002), p. 216Google Scholar. These researchers for example found that the sandstone blocks in the ‘halls with dancers’ in Ta Prohṃ and Praḥ Khan produced identical measurements of 1.22×10⁻³ SI Unit.

92 This was noted by Mus, Paul, ‘Le Sourire d'Angkor’, Artibus Asiae, 24 (1961): 380Google Scholar.

93 See also, Christine Hawixbrock's study on the large number of additions of small family chapels inside Jayavarman VII's temple complexes. Hawixbrock, , ‘Jayavarman VII ou le renouveau d'Angkor, entre tradition et modernité’, Bulletin de l'Ecole française d'Extrême-Orient, 85 (1998): 64CrossRefGoogle Scholar.

94 Groslier, B.P., Hommes et pierres (Paris: Arthaud, 1956), p. 153Google Scholar.

95 Boisselier, ‘Garuda dans l'art khmèr’, p. 79.

96 Rossabi, Morris, Khubilai Khan: His life and times (Berkeley: University of California Press, 1988), p. 42Google Scholar; and Pott, P.H., Yoga and Yantra, trans. Needham, R. (The Hague: Nijhoff, 1966), pp. 69–70CrossRefGoogle Scholar.

97 Groslier dated the piece to c. 1200; refer to Groslier, Bernard Philippe, Indochina: Art in the melting pot of races, trans. Lawrence, George (London: Methuen, 1962), p. 186Google Scholar.

98 Groslier, Bernard Philippe, ‘Fouilles du Palais Royal d'Angkor Thom’, in Proceedings of the 23^rd International Congress of Orientalists, Cambridge 1954 (London: Royal Asiatic Society, 1954), p. 229Google Scholar. Of course, this single piece, now in the Phnom Penh museum, does not alone prove the nature of the late royal cult of Jayavarman's reign, but the exactitude of its provenance supports the idea of a special link of the king with this deity. Groslier's reliability as an excavator must remain in question until the paperwork on the dig and the subsequent laboratory analyses are studied in the EFEO archives. Apart from a brief report to the 1954 Congress of Orientalists, Groslier, despite promising a book for several years, released only skimpy details of the pollen analysis done by the National Natural History Museum, Paris, in a paper published by the Siam Society (Groslier, , ‘Our knowledge of the Khmer civilization: A Re-appraisal’, Journal of the Siam Society, XLVIII, 1 (1960): 1–26Google Scholar).

99 Lobo, Wibke, ‘L'image de Hevajra et le bouddhisme tantrique’, Angkor et dix siècles d'art khmer, ed. Jessop, and Zephir, (Paris: Réunion de Musées Nationaux, 1997), p. 73Google Scholar. London art market dealer, Alexander Götz, told me he had handled 25 Khmer Hevajras from private collections over the past 25 years and thought there may 100 such icons in public and private collections.

100 K.908 v.CXXVII (Coedès ‘La stèle du Práḥ Khàn d'Angkor’ (1941), Bulletin de l'Ecole française d'Extrême-Orient, 41 (1979): 279.

101 Maspero, Henri, ‘Rapport sommaire sur une mission archéologique au Tchö-kiang’, Bulletin de l'Ecole française d'Extrême-Orient, 14 (1914): 69CrossRefGoogle Scholar.

102 ‘The production of tablets was open to all, and, as we shall see, many were produced by kings, members of the court, and senior monks. The sealings were products of a ritual ideology of mass production – the augmentation of merit by multiplication of images. The mass-production technology led to the earliest known printing of texts (the ye dharmā stanza and dhāraṇī) in India and Southeast Asia. Oddly, it is not usually recognised as such – perhaps because the impression was done on clay, perhaps because text and figure were often produced together from a single mould, perhaps because the impressed texts were not as such meant to be read.’ Skilling, P., ‘Buddhist sealings in Thailand and Southeast Asia: Iconography, function, and ritual context’, in Interpreting Southeast Asia's past: Monument, image and text, EurASEAA, 10, 2, ed. Bacus, Elisabeth, Glover, Ian and Sharrock, Peter (Singapore: NUS Press, 2008)Google Scholar.

103 Woodward, H., ‘Tantric Buddhism at Angkor Thom’, Ars Orientalis, 12 (1981): 57–67Google Scholar.

104 ‘Many additions are made to the monuments that were most important in the first phase. Typical additions were outer galleries, halls with dancers etc.’ Stern, Philippe, Les monuments Khmers du style du Bàyon et Jayavarman VII (Paris: PUF, 1965), p. 147Google Scholar. Stern's evaluation is supported by the recent technical studies of Olivier Cunin and Etsuo Uchida, Annual report on the technical survey of Angkor monument 2002, ed. Takeshi Nakagawa, Japanese team for safeguarding Angkor (JSA/UNESCO/Japanese Trust Fund for the Preservation of the World Cultural Heritage), Tokyo, p. 216. 

105 Claude Jacques now suggests the ‘five great Jinas’ found in the ‘salle aux danseuses’ of Praḥ Khan of Angkor and in Praḥ Khan of Kompong Svāy, as well as the five Buddhas of the northern lintel of Phimai's sanctuary, may have been called the ‘five Śrīghana Buddhas’ by the Khmers because the Sab Bāk inscription (K.1158) of 1066 CE, mentioned above, opens with an invocation of śrīpañcasugatā yādau śrīghanāṃ vibhāvikāḥ… ‘Those who, from the beginning, were the five Śrī Sugata, the creators of Śrīghana…’ and then invokes the sixth Buddha Vajrasattva (Jacques, C., ‘The Sect of Śrīghana in ancient Khmer land’ in Buddhist legacies in mainland Southeast Asia (Paris: EFEO, 2006), p. 73Google Scholar). Jacques lists nine Khmer inscriptions which use this rare epithet for the Buddha. Peter Skilling, who reads the epithet Śrīghana as a Sanskrit tatpuruṣa compound meaning ‘mass of glory’, has traced other occurrences of the word in Nālandā, Nepal, Amarāvatī, Bodh Gaya etc. Refer to Skilling, ‘Random jottings on Śrīghana: An Epithet of the Buddha’, Annual Report of the International Research Institute for Advanced Buddhology ARIRIAB at Soka University, vol. 7 (Mar. 2004), pp. 147–58.

106 Tantric Buddhist consecrations ordered by an emperor of Tang China, following earlier texts than the hevajra-tantra, are recorded as lasting two weeks and involved thousands of people. Orlando records that in 768 CE, tantric Buddhist patriarch Amoghavajra celebrated a ceremony that lasted 14 days. ‘The eunuch attendants, the ministers, and all the commanders of the imperial army were ordered by the emperor to receive abhiṣeka at the ceremony. Altogether more than 5,000 monks and laymen attended.’ Orlando, R., A Study of Chinese documents concerning the life of the tantric Buddhist patriarch Amoghavajra (A.D. 705–774) (Princeton University, 1981), p. 147Google Scholar. Chou records Amoghavajra conducting an abhiṣeka in which he ‘converted in succession hundreds, thousands, and myriads of people’ (Chou, Y., ‘Tantrism in China’, Harvard Journal of Asiatic Studies, 8 (1945): 280Google Scholar).

107 Chau Ju-Kua: His work on the Chinese and Arab trade in the twelfth and thirteenth centuries, entitled Chu-fan-chi, trans. F. Hirth and W. Rockhill (St Petersburg: Imperial Academy of Sciences, 1911), p. 53. The inserted notes are Hirth's.

108 Macdonald, Le Maṇḍala du Mañjuśrīmulakalpa, p. 69.

109 Snellgrove, D., The Hevajra-Tantra, a critical study (London: Oxford University Press, 1959), p. 134Google Scholar.

110 Sharrock, P., ‘The Yoginīs of the Bayon’, in Interpreting Southeast Asia's past: Monument, image and text, EurASEAA, 10, 2, ed. Bacus, Glover, Sharrock (Singapore: NUS Press, 2008), p. 260Google Scholar. 

111 ‘In consequence all classes of society adopted the cult. This was the most prosperous era of esoteric Buddhism in China’ (Tajima, R., Étude sur le Mahavairocana-sūtra (Dainichikyō) (Paris: Librarie d'Amérique et d'Orient, Paris, 1936), p. 23Google Scholar).

112 Zürcher, Erik, ‘Beyond the Jade Gate: Buddhism in China, Vietnam and Korea’, in ed. Bechert, and Gombrich, . The World of Buddhism: Buddhist monks and nuns in society and culture (Thames and Hudson, London, 1984), p. 206Google Scholar.

113 Linrothe, Ruthless compassion, p. 222.

114 Ibid., p. 274. I have earlier drawn a comparison with the political application of the Vajrapāṇi and the STTS at the Tang court, rather than with the more limited Heruka cult under the Ming. Linrothe's perspicacity in this passage seems to me to go beyond the reservations that brought the translator of the hevajra-tantra to a halt when he reflected on the meaning of the Khmer Hevajra bronzes: ‘Judging by the number of images of Hevajra found around Angkor and on various sites on the Khorat Plateau in Thailand … it would seem that a cult of this important Tantric divinity was practised from the eleventh century onwards. Since no relevant literature is available, not even a stray reference on a carved inscription, nothing of certainty can be said regarding this cult.’ David Snellgrove, Khmer civilization and Angkor, p. 57.

115 Linrothe, Ruthless compassion, p. 178.

116 Ibid., p. 214.

117 Hevajratanra, I.iii.17, Snellgrove, D., The Hevajra-Tantra, a critical study (London: Oxford University Press, 1959), p. 59Google Scholar.

118 Iyanaga, , Nobumi, , ‘Recits de la soumission de Maheśvara par Trailokyavijaya – d'après les sources chinoises et japonaises’, in ed. Strickland, Michel. Tantric and Taoist studies in honour of R.A. Stein, Mélanges chinois et bouddhiques vol. XXI (Brussels : Institut belge des hautes études chinoises, 1983), pp. 630–745Google Scholar.

119 Davidson, R., ‘Reflections on the Maheśvara subjugation myth: Indic materials, Sa-skya-pa apologetics and the birth of Heruka’, Journal of the International Association of Buddhist Studies, 14, 2 (1991): 215Google Scholar.

120 From Linrothe's following list, the ‘phase three’ wrathful deities known from clear icons or epigraphic mentions to have been adopted under Mahīdhara rule are Hevajra, Heruka and Saṃvara: ‘Other members of the set [of krodha-vignāntaka] seem to be formal metamorphoses of Trailokyavijaya, but for the most part these are names and deities unique to Phase Three: Heruka, Hevajra, Saṃvara, Yamāri, Guhyasamāja and Kālacakra’ (Linrothe, Ruthless compassion, p. 244).

121 ‘For Phase Two Esoteric Buddhism there is no sixth Tathāgata. Mahāvairocana, the central Tathāgata of the STTS, unified the five Tathāgata. Phase Three, by contrast, considers a sixth Buddha to be the supreme unifier, usually called either Vajrasattva or Vajradhara (note: The sixth Buddha is called Vajrasattva in the Hevajratantra and in the Saṃvara cycle).’ Ibid., p. 237.